1 2022-01-05 02:07:59 +0530 Nirbheek Chauhan <nirbheek@centricular.com>
5 meson: Add explicit check: kwarg to all run_command() calls
6 This is required since Meson 0.61.0, and causes a warning to be
8 https://github.com/mesonbuild/meson/commit/2c079d855ed87488bdcc6c5c06f59abdb9b85b6c
9 https://github.com/mesonbuild/meson/issues/9300
10 This exposed a bunch of places where we had broken run_command()
11 calls, unnecessary run_command() calls, and places where check: true
13 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1507>
15 2022-01-04 16:19:44 +0800 He Junyan <junyan.he@intel.com>
17 * gst-libs/gst/vaapi/gstvaapiprofile.c:
18 vaapi: av1dec: Use named profiles to replace the numeric ones.
19 Use named AV1 profiles (i.e., main, high) to replace the old "0"
21 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1491>
23 2021-12-01 15:36:44 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
25 * gst/vaapi/gstvaapidecodebin.c:
26 vaapidecodebin: Misc enhancements.
27 - Use GST_WARNING rather than g_critical
28 - Replace gst_ghost_pad_new_from_template() with
29 gst_ghost_pad_new() to avoid using the template.
30 - Declare extern variable
31 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1405>
33 2021-12-01 12:26:11 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
35 * gst/vaapi/gstvaapidecode.c:
36 vaapidecode: Autogenerate caps template.
37 vaapidecode is used in vaapidecodebin and it exposes all the
38 theoretically supported caps, but that slows down autoplug. With this
39 autplug is negotiated faster, giving more option to decodebin to select
41 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1405>
43 2021-11-16 17:18:28 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
45 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
46 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
47 vaapi: libs: Use unsized arrays as parameters.
48 Instead of defining a sized array for function signature, use it
49 unsized (a pointer alias, basically). In this way clang warning is
51 warning: ‘fill_profiles’ accessing 64 bytes in a region of size 12 [-Wstringop-overflow=]
52 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1357>
54 2021-11-16 17:21:52 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
57 vaapi: meson: Fail if static build.
58 And initialize the plugin array.
59 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1357>
61 2021-11-15 16:38:34 +0800 Ung, Teng En <teng.en.ung@intel.com>
63 * gst/vaapi/gstvaapipostproc.c:
64 vaapipostproc: fix mising metadata when using system
65 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1325>
67 2021-11-03 18:44:03 +0000 Tim-Philipp Müller <tim@centricular.com>
72 === release 1.19.3 ===
74 2021-11-03 15:43:36 +0000 Tim-Philipp Müller <tim@centricular.com>
79 * gstreamer-vaapi.doap:
83 2021-11-03 15:43:32 +0000 Tim-Philipp Müller <tim@centricular.com>
86 Update ChangeLogs for 1.19.3
88 2021-10-18 15:47:00 +0100 Tim-Philipp Müller <tim@centricular.com>
90 * tests/check/meson.build:
91 meson: update for meson.build_root() and .build_source() deprecation
92 -> use meson.project_build_root() or .global_build_root() instead.
93 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1183>
95 2021-10-18 00:40:14 +0100 Tim-Philipp Müller <tim@centricular.com>
98 * gst-libs/gst/vaapi/meson.build:
100 * tests/check/meson.build:
101 meson: update for dep.get_pkgconfig_variable() deprecation
102 ... in favour of dep.get_variable('foo', ..) which in some
103 cases allows for further cleanups in future since we can
104 extract variables from pkg-config dependencies as well as
105 internal dependencies using this mechanism.
106 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1183>
108 2021-10-14 18:38:26 +0100 Tim-Philipp Müller <tim@centricular.com>
111 meson: bump meson requirement to >= 0.59
112 For monorepo build and ugly/bad, for advanced feature
113 option API like get_option('xyz').required(..) which
114 we use in combination with the 'gpl' option.
115 For rest of modules for consistency (people will likely
116 use newer features based on the top-level requirement).
117 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1084>
119 2021-10-15 00:33:39 +0800 He Junyan <junyan.he@intel.com>
122 Docs: Fix a typo for Intel's vaapi driver name.
123 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/1151>
125 2021-09-28 10:11:15 +1000 Brad Hards <bradh@frogmouth.net>
128 doc: update IRC links to OFTC
129 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/945>
131 2021-09-26 01:07:02 +0100 Tim-Philipp Müller <tim@centricular.com>
135 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/925>
137 === release 1.19.2 ===
139 2021-09-23 01:35:54 +0100 Tim-Philipp Müller <tim@centricular.com>
144 * gstreamer-vaapi.doap:
148 2021-08-26 15:06:53 +0800 Zhang Yuankun <yuankunx.zhang@intel.com>
150 * gst/vaapi/gstvaapidecode.c:
151 vaapi: decoder: modify the condition to judge whether dma buffer is supported
152 It seems "GST_VAAPI_PLUGIN_BASE_SRC_PAD_CAN_DMABUF (decode)" will
153 return false even if this platform support the mem_type dma buffer.
154 And media-driver will return GST_VAAPI_BUFFER_MEMORY_TYPE_DMA_BUF2
155 on Gen12(such as TGL).
156 Without this patch, The command such as:
157 gst-launch-1.0 videotestsrc num-buffers=100 ! video/x-raw, format=I420 ! \
158 x264enc ! h264parse ! vaapih264dec ! video/x-raw\(memory:DMABuf\) ! fakesink
159 will return not-negotiated.
160 Signed-off-by: Zhang Yuankun <yuankunx.zhang@intel.com>
161 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/437>
163 2021-05-14 12:04:04 +0800 He Junyan <junyan.he@intel.com>
165 * gst-libs/gst/vaapi/gstvaapidisplay.c:
166 * gst/vaapi/gstvaapivideocontext.c:
167 Display: Add a property to export the VA display handle.
168 Just like what we do in VA plugins. The display can be seen as a
169 generic gst object and we can add a property to get the internal
171 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/435>
173 2021-05-14 11:49:01 +0800 He Junyan <junyan.he@intel.com>
175 * gst/vaapi/gstvaapivideomemory.c:
176 * gst/vaapi/gstvaapivideomemory.h:
177 plugins: video memory: Add a GST_MAP_VAAPI flag to peek the surface.
178 Just like what we do in VA plugins, the GST_MAP_VAAPI can directly
179 peek the surface of the VA buffers. The old flag 0 just peek the
180 surface proxy, which may not be convenient for the users who do not
181 want to include our headers.
182 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/435>
184 2021-07-09 11:01:14 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
186 * gst/vaapi/gstvaapi.c:
187 * gst/vaapi/gstvaapidecode.c:
188 Revert "vaapi: Demote vaapidecodebin to rank NONE."
189 This reverts commit 7a25c5d4ec95aefeca6515ac023b23c5dd330194.
190 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/436>
192 2021-06-23 16:23:00 +0800 He Junyan <junyan.he@intel.com>
194 * gst/vaapi/gstvaapiencode.c:
195 plugins: encode: fix a deadlock because of _drain()
196 We call gst_vaapiencode_drain() in gst_vaapiencode_change_state(),
197 whose context does not hold the stream lock of the encoder. The
198 current gst_vaapiencode_drain inside unlock/lock pair adds a extra
199 lock count to the stream lock of encoder and causes hang later.
200 We just remove the gst_vaapiencode_drain() and expand its logic
201 correctly according to the lock/unlock context.
202 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/433>
204 2021-06-29 15:09:13 +0800 He Junyan <junyan.he@intel.com>
206 * gst-libs/gst/vaapi/gstvaapidecoder_av1.c:
207 libs: decoder: av1: Clean the film_grain_info field.
208 We need to clean all film_grain_info fields when the film grain
209 feature is not enabled. It may have random data because the picture
210 parameter buffer is not cleaned.
211 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/434>
213 2021-06-03 22:12:04 +0800 He Junyan <junyan.he@intel.com>
215 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2.c:
216 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2.h:
217 libs: encoder: mpeg2: Add highP level for 1080@50p/60p.
218 The MPEG2 spec has amendment 3 to introduce a new level highP, which
219 is used for 1080@50p/60p streams. We need to add this level to avoid
220 encoding failure because of the level check.
222 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/432>
224 2021-05-11 16:59:07 +0800 He Junyan <junyan.he@intel.com>
226 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
227 Decoder: H264: Add the support for frame packing arrangement SEI message.
228 Frame packing arrangement SEI message is an alternative simple stereo 3D
229 manner for AVC. We need to recognize that SEI message and report the correct
231 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/420>
233 2021-05-22 18:54:49 +0100 Tim-Philipp Müller <tim@centricular.com>
235 * gst-libs/gst/vaapi/gstvaapifilter.c:
237 Use g_memdup2() where available and add fallback for older GLib versions
238 Alloc size is based on existing allocations and struct sizes.
239 g_memdup() is deprecated since GLib 2.68 and we want to avoid
240 deprecation warnings with recent versions of GLib.
241 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/431>
243 2021-06-01 15:29:16 +0100 Tim-Philipp Müller <tim@centricular.com>
248 === release 1.19.1 ===
250 2021-06-01 00:16:31 +0100 Tim-Philipp Müller <tim@centricular.com>
255 * gstreamer-vaapi.doap:
259 2021-04-23 19:01:54 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
261 * gst/vaapi/gstvaapi.c:
262 * gst/vaapi/gstvaapidecode.c:
263 vaapi: Demote vaapidecodebin to rank NONE.
264 One of the main reasons of vaapidecodebin was because it mitigated the
265 possible surface exhaustion. But that problem is currently
266 solved. Nowadays, vaapidecodebin brings more problems than it
267 solves. Thus this patch demotes vaapidecodebin to NONE rank while
268 bumping PRIMARY + 1 the most common decoders.
269 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/426>
271 2021-05-20 10:28:05 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
273 * gst/vaapi/gstvaapi.c:
274 * gst/vaapi/gstvaapioverlay.c:
275 plugins: Demote rank of vaapipostproc and vaapioverlay.
276 Since almost all video filters have rank NONE, these both elements
278 This is useful for autovideoconvert and other bins, and users might
279 force to use these by setting the environment variable
280 GST_PLUGIN_FEATURE_RANK.
281 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/429>
283 2021-04-21 10:55:45 +0200 François Laignel <fengalin@free.fr>
285 * tests/check/elements/vaapioverlay.c:
286 Use gst_element_request_pad_simple...
287 Instead of the deprecated gst_element_get_request_pad.
288 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/425>
290 2021-04-23 11:02:05 +0800 Zhang yuankun <yuankunx.zhang@intel.com>
292 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
293 libs: encoder: VP9: fix > 4k encode fail issue
294 The VP9 spec defines the MAX_TILE_WIDTH_B64(64), which is the maximum
295 width of a tile in units of superblocks. So the max width of one tile
296 should not be larger than 64x64=4096. When the width exceeds 4k, we
297 need to split it into multiple tiles in columns. The current vp9 encoder
298 does not handle this correctly.
300 gst-launch-1.0 videotestsrc ! video/x-raw,width=7680,height=4320 ! \
301 vaapivp9enc ! fakesink
303 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/424>
305 2021-04-02 15:43:45 +0800 Haihao Xiang <haihao.xiang@intel.com>
307 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
308 libs: display: drm: don't fallback to default device if explicitly specified device can't load/init
309 Otherwise user will be misled that the specified device is using
310 This fixes https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/issues/305
311 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/422>
313 2021-03-22 13:41:13 +0800 He Junyan <junyan.he@intel.com>
315 * gst-libs/gst/vaapi/gstvaapiencoder.c:
316 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
317 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
318 decoder: H265: Enable cu_qp_delta_enabled_flag when ROI
319 If ROI is enabled, the CUs within the ROI region may have different
320 QP from the other part of the picture. This needs us to enable the
321 cu_qp_delta_enabled_flag even in the CQP mode.
322 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/419>
324 2021-03-19 17:42:36 +1100 Matthew Waters <matthew@centricular.com>
326 * gst-libs/gst/vaapi/gstvaapicontext.c:
327 * gst-libs/gst/vaapi/gstvaapicontext.h:
328 * gst-libs/gst/vaapi/gstvaapiencoder.c:
329 * gst-libs/gst/vaapi/gstvaapifilter.c:
330 * gst-libs/gst/vaapi/gstvaapiminiobject.c:
331 * gst-libs/gst/vaapi/gstvaapiminiobject.h:
332 * gst-libs/gst/vaapi/gstvaapitexture_glx.c:
333 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
334 * gst-libs/gst/vaapi/gstvaapiutils_egl.h:
335 * gst-libs/gst/vaapi/gstvaapivalue.c:
336 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
337 * gst/vaapi/gstvaapipluginbase.c:
338 * gst/vaapi/gstvaapisink.c:
339 * gst/vaapi/gstvaapisink.h:
340 * gst/vaapi/gstvaapivideocontext.c:
341 * gst/vaapi/gstvaapivideomemory.c:
342 * tests/internal/simple-decoder.c:
343 gst: don't use volatile to mean atomic
344 volatile is not sufficient to provide atomic guarantees and real atomics
345 should be used instead. GCC 11 has started warning about using volatile
346 with atomic operations.
347 https://gitlab.gnome.org/GNOME/glib/-/merge_requests/1719
348 Discovered in https://gitlab.freedesktop.org/gstreamer/gst-plugins-good/-/issues/868
349 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/418>
351 2021-02-24 17:41:02 +0100 Paul Goulpié <paul.goulpie@ubicast.eu>
353 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
354 gstvaapiencoder_h264: add ENCODER_EXPOSURE on aud propertie
355 forgot during the following mainline commit: https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/commit/bc2f8fd19e924aa0e193708307326acd037691ce#
356 Signed-off-by: Paul Goulpié <paul.goulpie@ubicast.eu>
358 2021-02-04 15:05:55 +0800 He Junyan <junyan.he@intel.com>
360 * gst/vaapi/gstvaapipostproc.c:
361 plugins: postproc: Fix a problem of propose_allocation when passthrough.
362 We should query the downstream element to answer a precise allocation
363 query when the passthrough mode is enabled.
364 The current way still decides the allocation by the postproc itself. The
366 gst-launch-1.0 -v filesrc location=xxx.264 ! h264parse ! vaapih264dec ! \
367 vaapipostproc ! fakevideosink silent=false sync=true
368 will lose some info such as the GST_VIDEO_META_API_TYPE.
369 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/413>
371 2021-01-27 12:05:44 +0800 Haihao Xiang <haihao.xiang@intel.com>
373 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
374 * gst-libs/gst/vaapi/gstvaapidisplay_drm.h:
375 * gst/vaapi/gstvaapivideocontext.c:
376 libs: display: drm: support gst.vaapi.app.Display context for drm backend
377 Attributes for drm backend:
378 - va-display : ponter of VADisplay
379 - drm-device-fd : the DRM device file descriptor
380 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/409>
382 2021-01-13 14:43:20 +0800 Haihao Xiang <haihao.xiang@intel.com>
385 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
386 libs: display: drm: allow user specify a drm device via an env variable
387 Currently the default drm device is always used on a system with
388 multiple drm devices. This patch allows user to specify the required
389 drm device via GST_VAAPI_DRM_DEVICE env variable
391 GST_VAAPI_DRM_DEVICE=/dev/dri/renderD129 gst-launch-1.0 videotestsrc !
392 vaapih264enc ! fakesink
393 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/409>
395 2021-01-25 14:45:47 +0800 Haihao Xiang <haihao.xiang@intel.com>
397 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
398 libs: display: drm: fix set_device_path_from_fd
399 drmGetBusid() (GET_UNIQUE ioctl) won't return a valid bus id when
400 drmSetInterfaceVersion() (SET_VERSION ioctl) hasn't been called(see[1]),
401 so we can't get the right device path. Running test-display will get the
403 ** (test-display:18630): ERROR **: 10:26:00.434: could not create Gst/VA
405 Calling drmSetInterfaceVersion() before drmGetBusid() can't fix this
406 issue because a special permission is required for SET_VERSION ioctl.
407 This patch retrieves the device path from file descriptor via
409 [1] https://github.com/torvalds/linux/blob/master/drivers/gpu/drm/drm_ioctl.c#L48-L104
410 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/412>
412 2021-01-20 10:42:09 +0100 Rafał Dzięgiel <rafostar.github@gmail.com>
414 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
415 glx: Iterate over FBConfig and select 8 bit color size
416 Texture upload mechanism used by gstreamer-vaapi relies on 8 bpc.
417 In latest mesa versions the first fbconfig might not be 8 bit, so iterate
418 over it to find the correct config with supported values.
419 This also adds 8 bit alpha size to the framebuffer configuration which is
420 required to get it working properly.
421 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/410>
423 2021-01-11 09:57:03 +0800 Ung, Teng En <teng.en.ung@intel.com>
425 * gst-libs/gst/vaapi/gstvaapiutils.c:
426 vaapipostproc: fix code style.
427 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/407>
429 2020-12-21 05:42:00 +0000 Ung, Teng En <teng.en.ung@intel.com>
431 * gst-libs/gst/vaapi/gstvaapidisplay.c:
432 * gst-libs/gst/vaapi/gstvaapidisplay.h:
433 * gst-libs/gst/vaapi/gstvaapifilter.c:
434 vaapipostproc: Remove YUV to/from RGB color primary quirk since iHD driver has fixed in https://github.com/intel/media-driver/commit/a39fe9bc051a8c3efa8f35122a1585981ec7f816.
435 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/407>
437 2020-12-21 05:36:29 +0000 Ung, Teng En <teng.en.ung@intel.com>
439 * gst-libs/gst/vaapi/gstvaapiutils.c:
440 vaapipostproc: Added gstreamer BT2020 color standard support.
441 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/407>
443 2021-01-09 16:05:48 +0800 He Junyan <junyan.he@intel.com>
445 * gst-libs/gst/vaapi/gstvaapidecoder_av1.c:
446 decoder: AV1: Fix a static analysis problem of update_state().
447 No need to check the picture pointer after we have already dereferenced it.
449 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/408>
451 2020-12-22 23:43:52 +0800 He Junyan <junyan.he@intel.com>
453 * gst-libs/gst/vaapi/gstvaapidecoder_av1.c:
454 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
455 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
456 libs: decoder: Add decode_with_surface_id for AV1 film_grain.
457 The AV1 film_graim feature needs two surfaces the same time for
458 decoding. One is for recon surface which will be used as reference
459 later, and the other one is for display. The GstVaapiPicture should
460 contain the surface for display, while the vaBeginPicture() need
461 the recon surface as the target.
462 We add a gst_vaapi_picture_decode_with_surface_id API to handle this
464 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/191>
466 2020-08-27 21:46:41 +0800 He Junyan <junyan.he@intel.com>
468 * gst-libs/gst/vaapi/gstvaapidecoder_av1.c:
469 * gst-libs/gst/vaapi/gstvaapidecoder_av1.h:
470 * gst-libs/gst/vaapi/gstvaapiprofile.c:
471 * gst-libs/gst/vaapi/gstvaapiprofile.h:
472 * gst-libs/gst/vaapi/gstvaapiutils.c:
473 * gst-libs/gst/vaapi/meson.build:
474 * gst/vaapi/gstvaapidecode.c:
476 libs: decoder: AV1: Add the av1 decoder support.
477 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/191>
479 2020-08-27 21:39:35 +0800 He Junyan <junyan.he@intel.com>
481 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
482 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
483 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
484 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
485 libs: codecobject: Add number of elements when create codec object.
486 One slice data may need several slice parameter buffers at one time.
487 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/191>
489 2020-12-12 10:30:41 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
491 * gst/vaapi/gstvaapidecodebin.c:
492 * gst/vaapi/gstvaapipluginbase.c:
493 * gst/vaapi/gstvaapivideobufferpool.c:
494 vaapi: use gst_clear_object instead of g_clear_object
495 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/406>
497 2020-12-08 13:34:35 +0800 He Junyan <junyan.he@intel.com>
499 * gst/vaapi/gstvaapiencode.c:
500 plugins: encode: unlock the stream lock before _flush()
501 The current encoder will hang when EOS comes. When we call the
502 gst_vaapi_encoder_encode_and_queue(), we should release the stream
503 lock, just like what we do in gst_vaapiencode_handle_frame().
504 The deadlock happens when: The input thread holding the stream lock
505 is using gst_vaapi_encoder_create_coded_buffer() to acquire a coded
506 buffer, while the output thread which holding the coded buffer resource
507 is acquiring the stream lock in _push_frame() to push the data to
509 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/404>
511 2020-12-09 00:04:33 +0800 He Junyan <junyan.he@intel.com>
513 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
514 libs: encoder: H264: Fix one assert in get_pending_reordered().
515 gst_vaapi_encoder_h264_get_pending_reordered() does not consider the
516 case for HIERARCHICAL_B mode. The pipeline:
517 gst-launch-1.0 videotestsrc num-buffers=48 ! vaapih264enc prediction-type=2 \
518 keyframe-period=32 ! fakesink
520 ERROR:../gst-libs/gst/vaapi/gstvaapiencoder_h264.c:1996:reflist1_init_hierarchical_b:
521 assertion failed: (count != 0)
522 The last few B frames are not fetched in correct order when HIERARCHICAL_B
524 We also fix a latent bug for normal mode. The g_queue_pop_tail() of B frames
525 make the last several frames encoded in reverse order. The NAL of last few
526 frames come in reverse order in the bit stream, though it can still output
528 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/405>
530 2020-06-25 16:25:21 +0800 He Junyan <junyan.he@hotmail.com>
532 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
533 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
534 libs: encoder: H265: Add screen content coding extensions support.
535 In scc mode, the I frame can ref to itself and it needs the L0 reference
536 list enabled. So we should set the I frame to P_SLICE type. We do not need
537 to change the ref_pic_list0/1 passed to VA driver, just need to enable the
538 VAEncPictureParameterBufferHEVC->pps_curr_pic_ref_enabled_flag to notify
539 the driver consider the current frame as reference. For bits conformance,
540 the NumRpsCurrTempList0 should be incremented by one to include the current
541 picture as the reference frame. We manually do it when packing the slice header.
543 gst-launch-1.0 videotestsrc num-buffers=10 ! \
544 capsfilter caps=video/x-raw,format=NV12, framerate=30/1,width=640,height=360 ! \
545 vaapih265enc ! capsfilter caps=video/x-h265,profile="{ (string)screen-extended-main }" ! \
546 filesink location=out.265
547 Can be used to specify that the encoder should use SCC profiles.
548 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/379>
550 2020-07-11 23:37:29 +0800 He Junyan <junyan.he@intel.com>
552 * gst/vaapi/gstvaapiencode_vp9.c:
553 plugin: encode: vp9: Implement the set_config().
554 We store the allowed profiles list to encoder in set_config().
555 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/380>
557 2020-07-11 23:39:40 +0800 He Junyan <junyan.he@intel.com>
559 * gst/vaapi/gstvaapiencode_vp9.c:
560 plugin: encode: vp9: Add the profile into output caps.
561 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/380>
563 2020-07-11 23:27:21 +0800 He Junyan <junyan.he@intel.com>
565 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
566 libs: encoder: vp9: no need to ensure_hw_profile.
567 Once we decide the profile and can get the valid entrypoint for
568 that profile, hw must already support this profile/entrypoint pair.
569 No need to check it again in set_context_info().
570 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/380>
572 2020-07-11 23:22:55 +0800 He Junyan <junyan.he@intel.com>
574 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
575 libs: encoder: vp9: Improve the manner to decide the profile.
576 We should decide the VP9 encoder's profile based on the chroma and
577 depth of the input format, then make sure it is included in the
579 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/380>
581 2020-07-11 23:17:02 +0800 He Junyan <junyan.he@intel.com>
583 * gst-libs/gst/vaapi/gstvaapiutils_vpx.c:
584 * gst-libs/gst/vaapi/gstvaapiutils_vpx.h:
585 libs: util: vpx: add get_chroma_format_idc for VP9
586 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/380>
588 2020-07-11 23:09:59 +0800 He Junyan <junyan.he@intel.com>
590 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
591 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.h:
592 libs: encoder: vp9: Add allowed_profiles.
593 We need the allowed_profiles to store the allowed profiles in down
596 vaapivp9enc ! capsfilter caps=video/x-vp9,profile="{ (string)1, \
598 We need to store GST_VAAPI_PROFILE_VP9_1 and GST_VAAPI_PROFILE_VP9_3
600 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/380>
602 2020-11-30 18:00:30 +0800 He Junyan <junyan.he@intel.com>
604 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
605 libs: decoder: H265: Fix a typo in scc reference setting.
606 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/402>
608 2020-07-17 18:00:30 +0800 He Junyan <junyan.he@intel.com>
610 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
611 * gst-libs/gst/vaapi/gstvaapiprofile.c:
612 * gst-libs/gst/vaapi/gstvaapiprofile.h:
613 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
614 * gst-libs/gst/vaapi/video-format.c:
615 * gst/vaapi/gstvaapidecode.c:
616 libs: decoder: H265: Add MAIN_422_12 profile supporting.
617 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/390>
619 2020-07-31 14:38:42 +0800 He Junyan <junyan.he@intel.com>
621 * gst-libs/gst/vaapi/gstvaapiimage.c:
622 * gst-libs/gst/vaapi/video-format.c:
623 * gst-libs/gst/vaapi/video-format.h:
624 video-format: Add Y212_LE format.
625 It can be used as HEVC YUV_4:2:2 12bits stream's decoder output, and
626 also can be used as the input format for encoding HEVC YUV_4:2:2 12bits
628 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/390>
630 2020-07-30 23:21:06 +0800 He Junyan <junyan.he@intel.com>
632 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
633 * gst-libs/gst/vaapi/gstvaapiprofile.c:
634 * gst-libs/gst/vaapi/gstvaapiprofile.h:
635 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
636 * gst-libs/gst/vaapi/video-format.c:
637 * gst/vaapi/gstvaapidecode.c:
638 libs: decoder: H265: Add MAIN_444_12 profile supporting.
639 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/390>
641 2020-07-30 23:13:10 +0800 He Junyan <junyan.he@intel.com>
643 * gst-libs/gst/vaapi/gstvaapiimage.c:
644 * gst-libs/gst/vaapi/video-format.c:
645 * gst-libs/gst/vaapi/video-format.h:
646 video-format: Add Y412_LE format.
647 It can be used as HEVC YUV_4:4:4 12bits stream's decoder output, and
648 also can be used as the input format for encoding HEVC YUV_4:4:4 12bits
650 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/390>
652 2020-09-17 16:47:43 +0800 He Junyan <junyan.he@intel.com>
654 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
655 libs: decoder: h265: fill missing predictor_palette_size field.
656 The predictor_palette_size of VAPictureParameterBufferHEVCScc is
657 forgotten and need to be filled when streams have palettes.
658 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/396>
660 2020-09-17 15:35:11 +0800 He Junyan <junyan.he@intel.com>
662 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
663 libs: utils: h265: Use get_profile_from_sps to get profile.
664 We now use gst_h265_get_profile_from_sps() to replace the old way
665 of gst_h265_profile_tier_level_get_profile() to get more precise
666 profile. The new function consider the unstandard cases and give
667 a more suitable profile decision.
668 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/395>
670 2020-10-19 13:46:44 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
672 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
673 libs: decoder: vp9: 0xff segment pred probs if no temporal update
674 According to the spec (6.2.11 Segmentation params syntax)
675 segmentation_pred_prob[i] ast to be 0xff if not temporal_update.
676 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/400>
678 2020-10-19 13:42:53 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
680 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
681 libs: decoder: vp9: avoid reference rewriting
682 The removed code set all the reference frames to the current frame it is a key
683 one, but later, all the reference frames were rewritten with the decoded picture
684 buffers or VA_INVALID_SURFACE if they were not available.
685 Basically, all this time the first reference frame assignment has been ignored,
686 and it's not described by the spec, and this patch removes that code.
687 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/400>
689 2020-09-20 09:56:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
691 * gst/vaapi/gstvaapidecode.c:
692 decoder: don't reply src caps query with allowed if pad is fixed
693 If the pad is already fixed the caps query have to be reply with the
694 current fixed caps. Otherwise the query has to be replied with the
695 autogeneratd src caps.
696 This path fix this by falling back to the normal caps query processing
697 if the pad is already fixed. Otherwise it will fetch the allowed src
699 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/397>
701 2020-09-15 00:11:30 +0800 He Junyan <junyan.he@intel.com>
703 * gst/vaapi/gstvaapidecode.c:
704 plugins: decode: fix a DMA caps typo in ensure_allowed_srcpad_caps.
705 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/392>
707 2020-09-01 09:31:33 +0200 Marc Leeman <m.leeman@televic.com>
709 * gst/vaapi/gstvaapisink.c:
710 vaapisink: when updating the caps, reset rotation
711 When an element upstream changes settings (e.g. crop), new caps are sent
712 to vaapisink. When vaapisink was rotating the image, it needs to
713 re-evaluate if the sink needs to rotate the image.
714 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/389>
716 2020-09-08 17:31:02 +0100 Tim-Philipp Müller <tim@centricular.com>
719 ci: include template from gst-ci master branch again
721 2020-09-08 16:59:07 +0100 Tim-Philipp Müller <tim@centricular.com>
726 === release 1.18.0 ===
728 2020-09-08 00:09:51 +0100 Tim-Philipp Müller <tim@centricular.com>
734 * gstreamer-vaapi.doap:
738 2020-09-07 12:15:43 +0300 Sebastian Dröge <sebastian@centricular.com>
740 * gst-libs/gst/vaapi/gstvaapifilter.c:
741 Update for gst_video_transfer_function_*() function renaming
742 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/391>
744 2020-08-22 12:53:12 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
746 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
747 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
748 Revert "libs: decoder: h264, h265: in context at least 16 reference surfaces"
749 This reverts commit b387081a4d77d3da202da72686ab40fb9c83ee1e as discussed in
750 https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/commit/b387081a4d77d3da202da72686ab40fb9c83ee1e
752 === release 1.17.90 ===
754 2020-08-20 16:16:25 +0100 Tim-Philipp Müller <tim@centricular.com>
759 * gstreamer-vaapi.doap:
763 2020-08-17 11:43:53 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
765 * gst-libs/gst/vaapi/gstvaapisurface_egl.c:
766 libs: surface: egl: guard memory type
767 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/384>
769 2020-08-17 19:26:43 +0800 He Junyan <junyan.he@intel.com>
771 * gst/vaapi/gstvaapidecode.c:
772 plugin: decode: Fix two mem leaks because of caps.
773 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/385>
775 2020-08-16 01:57:15 +0800 He Junyan <junyan.he@intel.com>
777 * gst/vaapi/gstvaapivideomemory.c:
778 plugin: allocator: No need to ref allocator when create mem.
779 We do not need to ref the allocator when creating GstVaapiVideoMemory
780 kind memory, and then release it in _free(). The framework already
782 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/383>
784 2020-08-14 10:42:51 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
786 * gst/vaapi/gstvaapiencode_h264.c:
787 vaapiencode: h264: ignore level without breaking negotiation
788 Since commit 9f627ef2 if the user sets level in the encoder src caps
789 the caps negotiation is rejected.
790 But since the same commit the same encoder set the autoconfigured
791 level in caps. Some change in the base class might fixed the operation
792 order so now the caps are set and later negotiated.
793 This patch removes the level check.
795 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/382>
797 2019-07-09 19:17:48 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
799 * gst-libs/gst/vaapi/egl_vtable.h:
800 * gst-libs/gst/vaapi/gstvaapisurface_egl.c:
801 * gst-libs/gst/vaapi/gstvaapisurface_egl.h:
802 * gst-libs/gst/vaapi/gstvaapitexture_egl.c:
803 libs: egl: surface: export EGLImage as DMABuf if GEM not supported
804 This code path is used when frames are rendered as textures through
805 GstVideoGLTextureUploadMeta with EGL, mainly under Wayland.
806 Originally the EGLImage was exported as GEM, which was handled by
807 Intel drivers, but Gallium ones cannot create VA surfaces from
808 GEM buffers, only DMABuf.
809 This patch checks the memory types supported by VA driver to choose
810 the render the EGLImages from GEM or DMABuf, because GEM is still
811 better where supported.
812 DMABuf is well handled either by intel-vaapi-driver and gallium.
814 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/122>
816 2020-05-26 16:18:32 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
818 * gst-libs/gst/vaapi/gstvaapifilter.c:
819 * gst-libs/gst/vaapi/gstvaapifilter.h:
820 libs: filter: gst_vaapi_filter_get_memory_types()
821 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/122>
823 2020-08-12 18:48:59 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
825 * gst/vaapi/gstvaapipluginbase.c:
826 * gst/vaapi/gstvaapipluginbase.h:
827 plugins: remove gst_vaapi_plugin_base_get_allowed_srcpad_raw_caps()
828 Since nobody uses it, just remove it.
829 Thus extract_allowed_surface_formats() is refactored to attend only
830 gst_vaapi_plugin_base_get_allowed_sinkpad_raw_caps().
831 Now a surface is created when the image chorma is different from the
832 previous one. And if the driver has the quirk, it outputs all the
833 supported image formats without trying them.
834 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/381>
836 2020-08-12 17:50:50 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
838 * gst/vaapi/gstvaapidecode.c:
839 vaapidecode: expose raw src caps with same chroma
840 The try-and-error approach for getting the possible image formats from
841 a surface has brought several problems in different drivers, from
842 crashes to drop in performance.
843 Instead of that we change the algorithm to determine the possible
844 image formats based in the surface chroma: only those available image
845 formats with same chroma are exposed as possible raw caps.
846 Do this is important to avoid performance degrading in raw sinks
847 which doesn't handle NV12 but it does YV12 or I420.
848 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/381>
850 2020-07-10 17:05:38 +0800 He Junyan <junyan.he@intel.com>
852 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
853 libs: util: h265: use common parser API to get vaapi profiles.
854 We can reuse H265 parser's API to recognize the correct profile and
855 then just need to convert them to VAAPI profiles.
856 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/356>
858 2020-08-07 16:41:49 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
860 * gst-libs/gst/vaapi/meson.build:
861 * gst/vaapi/meson.build:
863 build: update for gl pkg-config file split
864 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/378>
866 2020-08-06 12:51:27 +0800 Xu Guangxin <guangxin.xu@intel.com>
868 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
869 h264dec: mark remaining frames as unreference before exec_picture_refs_modification
870 8.2.4.2 required this. Some clips will crash if we do not fill the reference list like this.
871 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/376>
873 2020-07-31 18:22:46 +0800 He Junyan <junyan.he@intel.com>
875 * gst-libs/gst/vaapi/gstvaapiencoder.c:
876 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
877 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
878 libs: encoder: H265: Enable Main 12 profile support.
879 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/375>
881 2020-07-31 19:17:39 +0800 He Junyan <junyan.he@intel.com>
883 * gst-libs/gst/vaapi/video-format.c:
884 video format: Fix P012_LE's chrome type typo.
885 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/375>
887 2020-08-04 21:15:01 +0300 Jordan Petridis <jordan@centricular.com>
889 * gst-libs/gst/vaapi/gstvaapiminiobject.c:
890 gstvaapiminiobject.c: fix clang 10 warnings
891 the typesystem checks in g_atomic_pointer_compare_and_exchange
892 seem to trigger some false positives with clang 10
893 similar to gstreamer!584
894 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/377>
896 2020-07-31 11:07:23 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
898 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
899 libs: window: wayland: destroy all wayland buffers during finalize
900 Some buffers and the associated FrameState state may still be pending at
901 that point. If the wayland connection is shared, then messages for the
902 buffer may still arrive. However, the associated event queue is already
903 deleted. So the result is a crash.
904 With a private connection the associated memory is leaked instead.
905 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/342>
907 2020-06-18 20:25:18 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
909 * tests/examples/test-vaapicontext.c:
910 test: vaapicontext: fix draw callback with multiple videos
911 The callback is called for both windows. So make sure that
912 gst_video_overlay_set_render_rectangle() is called for the correct one.
913 Otherwise, the left video will be randomly moved behind the right video.
914 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/342>
916 2020-06-19 09:23:52 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
918 * tests/examples/meson.build:
919 * tests/examples/test-vaapicontext.c:
920 test: vaapicontext: support wayland display
921 On Wayland, The whole gtk window is one Wayland surface. So
922 gtk_widget_get_window() must be called on the top-level widget.
923 For any other widget the following gdk_window_ensure_native() may create a
924 new top-level Wayland surface that is never visible.
925 As a result, the coordinates passed to
926 gst_video_overlay_set_render_rectangle() must be relativ to the top-level
927 window. Otherwise the video is placed incorrectly.
928 Original-Patch-By: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
929 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/342>
931 2017-12-01 20:18:28 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
933 * tests/examples/test-vaapicontext.c:
934 test: vaapicontext: use playbin to test files
935 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/342>
937 2017-11-29 11:11:39 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
939 * tests/examples/test-vaapicontext.c:
940 test: vaapicontext: add PLAY and NULL buttons
941 They only appear when only one sink is instanciated and their purpose
942 is to test the NULL-PLAY use case in context sharing.
943 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/342>
945 2020-06-19 21:26:52 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
947 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
948 libs: wayland: update the opaque region in set_render_rect
949 gst_vaapi_window_wayland_set_render_rect() may be called from an arbitrary
950 thread. That thread may be responsible for making the window visible.
951 At that point another thread will block in gst_vaapi_window_wayland_sync()
952 because the frame callback will not be called until the window is visible.
953 If that happens, then acquiring the display lock in
954 gst_vaapi_window_wayland_set_render_rect() would result in a deadlock.
955 Cache the size of the opaque rectangle separately and create the opaque
956 region right before applying it to the surface.
957 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/342>
959 2020-06-19 09:21:16 +0200 Hyunjun Ko <zzoon@igalia.com>
961 * gst-libs/gst/vaapi/gstvaapiwindow.c:
962 * gst-libs/gst/vaapi/gstvaapiwindow.h:
963 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
964 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
965 * gst/vaapi/gstvaapisink.c:
966 libs: window: implements gst_vaapi_window_set_render_rectangle
967 Implements new vmethod gst_vaapi_window_set_render_rectangle,
968 which is doing set the information of the rendered rectangle set by
970 This is necessary on wayland at least to get exact information of
972 And vaapisink calls this when gst_video_overlay_set_render_rectangle is
974 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/342>
976 2017-11-08 13:23:39 +0900 Hyunjun Ko <zzoon@igalia.com>
978 * gst/vaapi/gstvaapisink.c:
979 vaapisink: implements gst_vaapisink_wayland_create_window_from_handle()
980 Implements gst_vaapisink_wayland_create_window_from_handle() to support
981 using external wl_surface.
982 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/342>
984 2020-06-19 09:11:20 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
986 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
987 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
988 * gst-libs/gst/vaapi/gstvaapiwindow.c:
989 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
990 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.h:
991 libs: wayland: implement video overlay API
992 The Wayland sub-surfaces API is used to embed the video into an application
994 See Appendix A. Wayland Protocol Specification as the following.
996 The aim of sub-surfaces is to offload some of the compositing work
997 within a window from clients to the compositor. A prime example is
998 a video player with decorations and video in separate wl_surface
1000 This should allow the compositor to pass YUV video buffer processing to
1001 dedicated overlay hardware when possible.
1003 Added new method gst_vaapi_window_wayland_new_with_surface()
1004 Original-Patch-By: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1005 Zhao Halley <halley.zhao@intel.com>
1006 changzhix.wei@intel.com
1007 Hyunjun Ko <zzoon@igalia.com>
1008 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/342>
1010 2020-06-19 21:54:52 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
1012 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
1013 doc: libs: wayland: add 'transfer full' to the returnvalye of gst_vaapi_window_wayland_new
1014 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/342>
1016 2020-06-11 08:25:57 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
1018 * gst-libs/gst/vaapi/gstvaapidisplay.c:
1019 libs: display: always call close_display()
1020 All close_display() have their own checks for use_foreign_display and only
1021 destroy locally created objects in that case.
1022 Without this objects other than the actuall foreign display itself are
1024 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/342>
1026 2020-07-30 23:37:10 +0800 He Junyan <junyan.he@intel.com>
1028 * gst-libs/gst/vaapi/video-format.h:
1029 video-format: Add the missing P012_LE into GST_VAAPI_FORMATS_ALL.
1030 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/373>
1032 2020-05-25 17:02:26 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1034 * gst/vaapi/gstvaapidecode.c:
1035 * gst/vaapi/gstvaapipluginutil.c:
1036 * gst/vaapi/gstvaapipluginutil.h:
1037 plugins: add gst_vaapi_caps_set_width_and_height_range()
1038 This utility function is called internally by
1039 gst_vaapi_build_caps_from_formats() and can be used outside.
1040 This function sets frame size and framerates ranges.
1041 Also gst_vaapi_build_caps_from_formats() is simplified.
1042 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/374>
1044 2020-07-31 15:27:38 +0800 He Junyan <junyan.he@intel.com>
1046 * gst-libs/gst/vaapi/gstvaapidecoder.c:
1047 libs: decoder: fix a crash issue when get_surface_formats.
1048 Some context does not report any valid format that we can support.
1049 For example, the HEVC 444 12 bits decoder context, all the formats
1050 it reports is not supported now, which make the formats list a NULL
1051 array. We should check that pointer before we use it.
1052 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/372>
1054 2020-07-03 19:28:28 +0800 He Junyan <junyan.he@intel.com>
1056 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1057 libs: encoder: h265: choose the profile based on allowed list.
1058 We can decide the profile in ensure_profile(), based on allowed list
1059 passed by the encode. We also need to check whether the entrypoint is
1060 available. Once it is decided, no need to check the hw entrypoint
1062 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/349>
1064 2020-07-29 22:05:41 +0800 He Junyan <junyan.he@intel.com>
1066 * gst/vaapi/gstvaapiencode_h265.c:
1067 plugins: encode: h265: set all allowed profiles to encoder.
1068 We should collect all allowed profiles and pass them to the inside
1069 encoder, rather than just calculate the max profile idc.
1070 The allowed profiles should also be supported by the HW.
1071 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/349>
1073 2020-07-29 22:32:55 +0800 He Junyan <junyan.he@intel.com>
1075 * gst-libs/gst/vaapi/gstvaapidisplay.c:
1076 * gst-libs/gst/vaapi/gstvaapidisplay.h:
1077 libs: display: Add a helper function to get profiles by codec.
1078 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/349>
1080 2020-07-03 01:28:28 +0800 He Junyan <junyan.he@intel.com>
1082 * gst/vaapi/gstvaapiencode_h265.c:
1083 plugins: encode: h265: collect all allowed profiles to encoder.
1084 We should collect all allowed profiles and pass them to the inside
1085 encoder, rather than just calculate the max profile idc.
1086 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/349>
1088 2020-07-03 00:53:31 +0800 He Junyan <junyan.he@intel.com>
1090 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1091 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
1092 * gst/vaapi/gstvaapiencode_h265.c:
1093 libs: encoder: h265: modify set_max_profile to set_allowed_profiles.
1094 In h265, bigger profile idc may not be compatible with the small profile
1095 idc. And more important, there are multi profiles with the same profile
1096 idc. Such as main-422-10, main-444 and main-444-10, they all have profile
1098 So recording the max profile idc is not enough, the encoder needs to know
1099 all allowed profiles when deciding the real profile.
1100 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/349>
1102 2020-07-02 23:33:31 +0800 He Junyan <junyan.he@intel.com>
1104 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1105 libs: encoder: h265: No need to check hw_max_profile.
1106 In h265, higher profile idc number does not mean better compression
1107 performance and may be not compatible with the lower profile idc.
1108 So, it is not suitable to find the heighest idc for hw to ensure the
1110 On the other side, when the entrypoint of the selected profile is valid,
1111 it means the hw really support this profile, no need to check it again.
1112 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/349>
1114 2020-04-08 19:41:09 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1116 * gst/vaapi/gstvaapipostproc.c:
1117 vaapipostproc: early return if fixate srcpad caps fails
1118 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/371>
1120 2020-07-29 13:39:44 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1122 * gst/vaapi/gstvaapipluginutil.c:
1123 vaapipluginutil: simplify gst_vaapi_find_preferred_caps_feature()
1124 Generalize the way how the preferred color format is chosen. Also
1125 use new GStreamre API as syntatic sugar.
1126 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/370>
1128 2020-07-29 14:22:18 +0800 He Junyan <junyan.he@intel.com>
1130 * gst-libs/gst/vaapi/gstvaapidecoder.c:
1131 * gst-libs/gst/vaapi/gstvaapiprofile.c:
1132 * gst-libs/gst/vaapi/gstvaapiprofile.h:
1133 * gst/vaapi/gstvaapidecode.c:
1134 * gst/vaapi/gstvaapiencode.c:
1135 libs: profile: Use get_codec_from_caps to get codec type.
1136 There is no need to get a profile from the caps and then convert
1137 that profile into codec type. We can get the codec type by caps's
1139 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/358>
1141 2020-07-12 19:42:40 +0800 He Junyan <junyan.he@intel.com>
1143 * gst-libs/gst/vaapi/gstvaapiprofile.c:
1144 libs: profile: h265: Fix return value of from_codec_data_h265.
1145 profile_from_codec_data_h265() returns wrong GstVaapiProfile for h265.
1146 The codec data of caps contain the profile IDC, but the mapping between
1147 profile IDC and GstVaapiProfile is wrong.
1148 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/358>
1150 2020-07-07 00:46:23 +0800 He Junyan <junyan.he@intel.com>
1152 * gst/vaapi/gstvaapiencode_vp9.c:
1153 plugins: encode: vp9: Implement vp9's allowed_profiles() func.
1154 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/188>
1156 2020-07-08 16:30:17 +0800 He Junyan <junyan.he@intel.com>
1158 * gst/vaapi/gstvaapiencode_h264.c:
1159 * gst/vaapi/gstvaapiencode_h265.c:
1160 * gst/vaapi/gstvaapipluginutil.c:
1161 * gst/vaapi/gstvaapipluginutil.h:
1162 plugin: util: rename h26x_encoder_get_profiles_from_caps().
1163 Change its name to encoder_get_profiles_from_caps(). Other codecs such
1164 as VP9 also needs to use this function.
1165 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/188>
1167 2020-07-06 23:35:12 +0800 He Junyan <junyan.he@intel.com>
1169 * gst-libs/gst/vaapi/gstvaapiutils_vpx.c:
1170 * gst-libs/gst/vaapi/gstvaapiutils_vpx.h:
1171 * gst-libs/gst/vaapi/meson.build:
1172 libs: utils: vpx: Add utils vpx to handle VP8/9 misc things.
1173 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/188>
1175 2020-07-29 10:17:31 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1177 * gst-libs/gst/vaapi/gstvaapicontext.c:
1178 * gst-libs/gst/vaapi/gstvaapidisplay.c:
1179 * gst-libs/gst/vaapi/gstvaapidisplay.h:
1180 libs: display, context: handle broken jpeg decoder for i965 driver
1181 JPEG decoding in i965 driver is pretty much broken, and the driver is
1182 deprecated which mean authors only accept trivial fixes.
1183 Surfaces for JPEG decoder context in i965 only handle IMC3[1] color
1184 format which is not a common format in GStreamer. It can export it to
1185 I420 at mapping raw bytes, but DMABuf exporting is problematic.
1186 This patch artificially adds NV12 to the context format list when it's
1187 JPEG decoder for i965 and force the usage of old VA-API for surface
1188 creation without specifying color format. Also it artificially
1189 disables the DMABuf announcement.
1190 1. https://docs.microsoft.com/en-us/windows/win32/medfound/recommended-8-bit-yuv-formats-for-video-rendering#420-formats-16-bits-per-pixel
1191 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/369>
1193 2020-07-29 12:02:50 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1195 * gst-libs/gst/vaapi/gstvaapicontext.c:
1196 libs: context: change function to internal code style
1197 Instead of a getter the function `get_preferred_format()` to
1198 `ensure_preferred_format()` which aligns to the code style.
1199 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/369>
1201 2020-07-28 20:00:09 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1203 * gst/vaapi/gstvaapidecode.c:
1204 vaapidecode: always merge profile caps in sink caps
1205 This commit fixes a regression of e962069d, where if the profile's
1206 caps doesn't have a caps profile, it's ignored.
1207 This patch add a conditional jump if the caps doesn't have a profile
1210 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/368>
1212 2020-07-28 12:22:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1215 build: request libdrm >= 2.4.98 and fallback
1217 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/367>
1219 2020-05-18 17:32:27 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1221 * gst/vaapi/gstvaapidecode.c:
1222 vaapidecode: dma caps only use reported color format
1223 This fix pipelines without vaapipostproc after vaapi decoder, such as
1224 gst-launch-1.0 filesrc location=~/file.mp4 ! parsebin ! vaapih264dec ! glimagesink
1225 On EGL platforms, so DMABuf is used.
1226 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/366>
1228 2020-02-07 17:10:45 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1230 * gst/vaapi/gstvaapidecode.c:
1231 vaapidecode: use allowed srcpad caps for caps query
1232 Instead of using just the template caps use the current allowed
1233 srcpad caps, which is created considering the current decoder
1235 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/366>
1237 2020-01-22 17:41:28 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1239 * gst-libs/gst/vaapi/gstvaapidecoder.c:
1240 * gst-libs/gst/vaapi/gstvaapidecoder.h:
1241 * gst/vaapi/gstvaapidecode.c:
1242 vaapidecode: build allowed srcpad caps from va context
1243 Instead of generating allowed srcpad caps with generic information,
1244 now it takes the size an formats limits from the decoder's context.
1245 This is possible since srcpad caps are generated after the internal
1247 The patch replaces gst_vaapi_decoder_get_surface_formats() with
1248 gst_vaapi_decoder_get_suface_attributes().
1249 From these attributes, formats are only used for VASurface memory
1250 caps feature. For system memory caps feature, the old
1251 gst_vaapi_plugin_get_allowed_srcpad_caps() is still used, since
1252 i965 jpeg decoder cannot deliver mappable format for gstreamer.
1253 And for the other caps features (dmabuf and texture upload) the
1254 same static list are used.
1255 This patch also adds DMABuf caps feature only if the context
1256 supports that memory type. Nonetheless, we keep the pre-defined
1257 formats since they are the subset of common derive formats formats
1258 supported either by amd/gallium and both intel drivers, since,
1259 when exporting the fd through vaAcquireBufferHandle()/
1260 vaReleaseBufferHandle(), the formats of the derivable image cannot
1261 be retriebable from the driver. Later we'll use the attribute
1262 formats for the DMABuf feature too, when the code be ported to
1263 vaExportSurfaceHandle().
1264 Finally, the allowed srcpad caps are removed if the internal decoder
1265 is destroyed, since context attribues will change.
1266 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/366>
1268 2020-02-07 16:50:52 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1270 * gst/vaapi/gstvaapidecode.c:
1271 vaapidecode: reorder src caps template
1272 Since negotiation depends on caps order, first is VA, then DMABuf,
1273 later GLUploadTexture (deprecated) and finally raw.
1274 Also, for decoders, the possible available color formats for DMABuf
1275 is extended to all the possible VA color formats.
1276 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/366>
1278 2020-07-22 10:01:41 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
1280 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
1281 libs: window: wayland: use dmabuf protocol if available
1282 Currently vaGetSurfaceBufferWl() is used to create wayland buffers.
1283 Unfortunately this is not implemented by the 'media-driver' and Mesa VA-API
1284 drivers. And the implementation provided by 'intel-vaapi-driver' is not
1285 compatible with a Wayland server that uses the iris Mesa driver.
1286 So create the Wayland buffers manually with the zwp_linux_dmabuf_v1 wayland
1287 protocol. Formats and modifiers supported by the Wayland server are taken
1288 into account. If necessary, VPP is enabled to convert the buffer into a
1290 Fall back to vaGetSurfaceBufferWl() if creating buffers via dambuf protocol
1292 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/346>
1294 2020-07-21 10:03:19 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
1296 * gst-libs/gst/vaapi/gstvaapiwindow.c:
1297 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
1298 libs: window: allow choosing the format for the vpp pool
1299 Signed-off-by: Michael Olbrich <m.olbrich@pengutronix.de>
1300 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/346>
1302 2020-06-28 17:42:29 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
1304 * gst-libs/gst/vaapi/video-format.c:
1305 * gst-libs/gst/vaapi/video-format.h:
1306 video-format: add DRM formats to the mapping table
1307 This will be needed for the DMABuf protocol support to map DRM formats to
1308 vaapi and gstreamer formats.
1309 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/346>
1311 2020-07-22 09:36:18 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
1313 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
1314 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
1315 * gst-libs/gst/vaapi/meson.build:
1316 libs: display: wayland: add basic dmabuf protocol support
1317 This is just the basic infrastructure. Hook up the interface and collect
1318 all supported formats.
1319 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/346>
1321 2020-07-06 09:59:40 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
1323 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
1324 libs: window: wayland: wait for configure before committing the first buffer
1325 Committing the first buffer for a surface must not be done before
1326 ack_configure() has been sent for the xdg_surface.
1327 With weston, the commit will fail with "error 3: xdg_surface has never been
1329 Wait in gst_vaapi_window_wayland_show() until configure is done to avoid
1331 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/346>
1333 2020-07-01 14:50:51 +0800 He Junyan <junyan.he@hotmail.com>
1335 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1336 libs: encoder: h265: set no P frame automatically.
1337 The double reference lists may be required by drivers and there should
1338 be no P frames in the of stream. The old way of converting P frames to
1339 B frames is by setting `low-delay-b` property, which is unconvenient
1340 and has bad user experience, since most of the users do not know when
1341 to set this property, and if it is not set correctly, the encoding
1342 pipeline fails or even hangs on some platforms. VA driver now provides
1343 a attribute to query whether both reference lists must be un-NULL for
1344 a profile/entrypoint pair.
1345 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/284>
1347 2020-04-15 16:26:55 +0800 He Junyan <junyan.he@hotmail.com>
1349 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1350 libs: encoder: h265: Deprecate the low-delay-b property.
1351 In HEVC, P and B definitions are different from AVC: P frames have
1352 just one reference list and so 1 MV, while B frames have two reference
1353 lists and so 2 MVs. No matter B or P, ist reference lists can contain
1354 forward/backward reference. So P and B can both have bi-directions
1355 dependency, the difference is just their reference list
1356 number (i.e. MV number). This is different from the AVC.
1357 The *low delay b mode* refers to a special HEVC mode, in which the
1358 stream just contain I and B frames, without P frames, and all B frames
1359 only have forward direction dependencies (i.e. all inter frames have 2
1360 reference lists but no backward reference in both lists). This is
1361 similar to AVC I/P mode, but changing the P to the forward dependent
1363 The `low-delay-b` property is now just used to simply convert all P
1364 frames to B frames when driver does not support P frames (so both
1365 reference lists have the same references frames). This is a little
1366 different from the meaning of low delay b mode (the two ref lists may
1367 have the different reference frames). And the driver now can report
1368 whether it supports P frames correctly, so there is no need to use
1369 this property and deprecate it.
1370 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/284>
1372 2020-07-24 12:54:31 +0200 Marc Leeman <m.leeman@televic.com>
1374 * gst/vaapi/gstvaapipostproc.c:
1375 postproc: reconfigure after changing cropping values
1376 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/365>
1378 2020-07-09 13:49:29 +0800 He Junyan <junyan.he@intel.com>
1380 * gst/vaapi/gstvaapiencode.h:
1381 * gst/vaapi/gstvaapiencode_h264.c:
1382 * gst/vaapi/gstvaapiencode_h265.c:
1383 * gst/vaapi/gstvaapiencode_jpeg.c:
1384 * gst/vaapi/gstvaapiencode_mpeg2.c:
1385 * gst/vaapi/gstvaapiencode_vp8.c:
1386 * gst/vaapi/gstvaapiencode_vp9.c:
1387 plugin: encode: Add static caps for template documentation.
1388 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/354>
1390 2020-07-08 19:03:14 +0800 He Junyan <junyan.he@intel.com>
1392 * gst/vaapi/gstvaapiencode_vp9.c:
1393 plugin: encode: vp9: Use the dynamically built src template caps.
1394 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/354>
1396 2020-07-08 19:02:45 +0800 He Junyan <junyan.he@intel.com>
1398 * gst/vaapi/gstvaapiencode_vp8.c:
1399 plugin: encode: vp8: Use the dynamically built src template caps.
1400 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/354>
1402 2020-07-08 19:02:23 +0800 He Junyan <junyan.he@intel.com>
1404 * gst/vaapi/gstvaapiencode_jpeg.c:
1405 plugin: encode: jpeg: Use the dynamically built src template caps.
1406 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/354>
1408 2020-07-08 19:00:39 +0800 He Junyan <junyan.he@intel.com>
1410 * gst/vaapi/gstvaapiencode_mpeg2.c:
1411 plugin: encode: mpeg2: Use the dynamically built src template caps.
1412 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/354>
1414 2020-07-08 18:59:18 +0800 He Junyan <junyan.he@intel.com>
1416 * gst/vaapi/gstvaapiencode_h265.c:
1417 plugin: encode: h265: Use the dynamically built src template caps.
1418 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/354>
1420 2020-07-08 18:57:26 +0800 He Junyan <junyan.he@intel.com>
1422 * gst/vaapi/gstvaapiencode_h264.c:
1423 plugin: encode: h264: Use the dynamically built src template caps.
1424 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/354>
1426 2020-07-08 18:46:58 +0800 He Junyan <junyan.he@intel.com>
1428 * gst/vaapi/gstvaapiencode.h:
1429 * gst/vaapi/gstvaapiencode_h264.c:
1430 * gst/vaapi/gstvaapiencode_h265.c:
1431 * gst/vaapi/gstvaapiencode_jpeg.c:
1432 * gst/vaapi/gstvaapiencode_mpeg2.c:
1433 * gst/vaapi/gstvaapiencode_vp8.c:
1434 * gst/vaapi/gstvaapiencode_vp9.c:
1435 plugin: encode: Store the coded caps in type's init data.
1436 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/354>
1438 2020-07-08 18:30:00 +0800 He Junyan <junyan.he@intel.com>
1440 * gst/vaapi/gstvaapipluginutil.c:
1441 * gst/vaapi/gstvaapipluginutil.h:
1442 plugin: util: add helper function build_template_coded_caps_by_codec()
1443 Like build_template_raw_caps_by_codec(), this function can detect and
1444 build the caps for specified codec based on the query of the profiles.
1445 The result is coded caps such as video/x-h265, video/x-h264. The result
1446 can be used as the template of encode's src or decode's sink.
1447 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/354>
1449 2020-07-07 17:16:41 +0800 He Junyan <junyan.he@intel.com>
1451 * gst/vaapi/gstvaapiencode.h:
1452 * gst/vaapi/gstvaapipluginutil.c:
1453 * gst/vaapi/gstvaapipluginutil.h:
1454 plugins: utils: rename build_template_caps_by_codec.
1455 Rename the function build_template_caps_by_codec() to the name of
1456 build_template_raw_caps_by_codec(). It can be used to collect all
1457 raw video formats for encode's sink and decode's src.
1458 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/354>
1460 2020-07-21 20:14:57 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1462 * gst/vaapi/gstvaapidecode.c:
1463 * gst/vaapi/gstvaapipluginutil.c:
1464 * gst/vaapi/gstvaapipluginutil.h:
1465 vaapidecode: merge common profiles before setting size range
1466 The synthetic profiles, such as H264 baseline, H265 intra, etc. are
1467 added at the end of processing all available VA profiles. This
1468 generated an non-optimal caps for negotiation, since the synthetic
1469 profiles don't have frame size ranges.
1470 This patch adds those possible synthetic profiles when the associated
1471 profile is processed, with its frame size ranges.
1472 Now allowed sink caps are simpler.
1473 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/363>
1475 2020-07-21 22:05:08 +0800 He Junyan <junyan.he@intel.com>
1477 * gst/vaapi/gstvaapipluginutil.c:
1478 plugin: util: Add the missing DMA buffer input in template caps.
1479 We pass the wrong parameter to gst_vaapi_build_caps_from_formats()
1480 and lose the DMA feature in caps.
1481 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/362>
1483 2020-07-14 18:13:56 +0800 He Junyan <junyan.he@intel.com>
1485 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
1486 * gst-libs/gst/vaapi/gstvaapiprofile.c:
1487 * gst-libs/gst/vaapi/gstvaapiprofile.h:
1488 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
1489 * gst-libs/gst/vaapi/video-format.c:
1490 * gst/vaapi/gstvaapidecode.c:
1491 libs: decoder: H265: Add MAIN_12 profile supporting.
1492 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/359>
1494 2020-07-09 23:07:38 +0800 He Junyan <junyan.he@intel.com>
1496 * gst-libs/gst/vaapi/gstvaapiimage.c:
1497 * gst-libs/gst/vaapi/video-format.c:
1498 video-format: Add P012_LE format.
1499 It can be used as HEVC YUV_4:2:0 12bits stream's decoder output, and
1500 also can be used as the input format for encoding HEVC YUV_4:2:0 12bits
1502 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/359>
1504 2020-07-17 00:45:53 +0800 He Junyan <junyan.he@intel.com>
1506 * gst/vaapi/gstvaapidecode.c:
1507 plugin: decode: correct ensure_allowed_sinkpad_caps's caps.
1508 The decode allowed caps returned by ensure_allowed_sinkpad_caps()
1509 contains all profiles of the whole VAAPI, like:
1510 image/jpeg, width=(int)[ 0, 1638 4 ], height=(int)[ 0, 16384 ];
1511 video/mpeg, mpegversion=(int)2, profile=(string){ simple, main },
1512 width=(int)[ 0, 2048 ], height=(int)[ 0, 2048 ]; video/x-h264,
1513 profile=(string){ main, high, constrained-baseline }, width=(int)[ 0,
1514 4096 ], height=(int)[ 0, 4096 ]; video/x-h264, profile=(string){
1515 constrained-high, progressive-high, baseline }; video/x-h265,
1516 profile=(string){ main, main-intra }, width=(int)[ 0, 8192 ],
1517 height=(int)[ 0, 8192 ]; video/x-vp8, width=(int)[ 0, 4096 ],
1518 height=(int)[ 0, 4096 ]; video/x-wmv, wmvversion=(int)3,
1519 format=(string)WVC1, profile=(string)advanced, width=(int)[ 0, 3840 ],
1520 height=(int)[ 0, 3840 ]; video/x-wmv, wmvversion=(int)3,
1521 profile=(string){ simple, main }, width=(int)[ 0, 3840 ],
1522 height=(int)[ 0, 3840 ]
1523 Which is verbose and may have latent problems. It should only contains
1524 the profiles belong to its codec type. For example, h265 should only
1526 video/x-h265, profile=(string){ main, main-intra },
1527 width=(int)[ 0, 8192 ], height=(int)[ 0, 8192 ]
1528 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/361>
1530 2020-07-13 11:06:18 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1532 * gst/vaapi/gstvaapidecodebin.c:
1533 vaapidecodebin: don't force NV12 since P010_10LE is now possible
1534 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/360>
1536 2020-07-12 20:34:31 +0800 He Junyan <junyan.he@intel.com>
1538 * gst-libs/gst/vaapi/gstvaapiprofile.c:
1539 libs: profile: The VP9 profiles' name should be just "0,1,2,3"
1540 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/357>
1542 2020-07-08 17:33:32 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1544 * gst-libs/gst/vaapi/gstvaapidecoder.c:
1545 * gst-libs/gst/vaapi/gstvaapidecoder.h:
1546 * gst/vaapi/gstvaapidecode.c:
1547 * gst/vaapi/gstvaapidecode.h:
1548 * tests/internal/simple-decoder.c:
1549 vaapidecode: Remove NO_SURFACE error handling
1550 Since surfaces are not bounded to decoding context it makes no sense
1551 to keep the surface semaphore. This patch removes the handling of
1553 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/353>
1555 2020-07-08 17:48:57 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1557 * gst/vaapi/gstvaapidecode.c:
1558 * gst/vaapi/gstvaapidecode.h:
1559 Revert "vaapidecode: drop non-keyframe in reverse playback"
1560 Since the number of surfaces are not bounded to decoder context,
1561 this hack is no longer needed.
1562 This reverts commit 19c0c8a97385ce119440c4aad2d689fc79297435.
1563 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/353>
1565 2019-12-06 14:21:33 +0800 He Junyan <junyan.he@hotmail.com>
1567 * gst-libs/gst/vaapi/gstvaapicontext.c:
1568 libs: decoder: context: remove surfaces binding from context.
1569 The vaCreateContext do not need to specify the surfaces for the
1570 context creation now. So we do not need to bind any surface to the
1571 context anymore. Surfaces should be the resource belong to display
1572 and just be used in encoder/decoder context.
1573 The previous manner has big limitation for decoder. The context's
1574 surface number is decided by dpb size. All the surfaces in dpb will
1575 be attached to a gstbuffer and be pushed to down stream, and the
1576 decoder need to wait down stream free the surface and go on if not
1577 enough surface available. For more and more use cases, this causes
1578 deadlock. For example,
1579 gst-launch-1.0 filesrc location=a.h264 ! h264parse ! vaapih264dec
1580 ! x264enc ! filesink location=./output.h264
1581 will cause deadlock and make the whole pipeline hang.
1582 the x264enc encoder need to cache more than dpb size surfaces.
1583 The best solution is seperating the surfaces number and the dpb size.
1584 dpb and dpb size shoule be virtual concepts maintained by the decoder.
1585 And let the surfaces_pool in context maintain the re-use of all surfaces.
1586 For encoder, the situation is better, all the surfaces are just used
1587 as reference frame and no need to be pushed to down stream. We can
1588 just reserve and set the capacity of the surfaces_pool to meet the
1592 Co-Author: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1593 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/353>
1595 2020-07-08 17:50:51 +0100 Tim-Philipp Müller <tim@centricular.com>
1598 * scripts/extract-release-date-from-doap-file.py:
1599 meson: set release date from .doap file for releases
1600 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/355>
1602 2020-07-08 11:57:52 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1604 * gst/vaapi/gstvaapipluginbase.c:
1605 plugins: use VA allocator by default on raw caps
1606 Instead of using dmabuf allocator in source pad, when raw video caps
1607 are negotiated, it uses VA allocator as before, since it is stable
1608 in more use cases, for example transcoding, and more backend drivers.
1609 Dmabuf allocator is only used when dmabuf caps feature is negotiated.
1610 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/352>
1612 2020-06-06 18:47:35 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1614 * gst/vaapi/gstvaapi.c:
1615 vaapisink: rank it as secondary
1616 iHD doesn't provide a full implemention for rendering surfaces and
1617 i965 has problems in wayland. And I suspect this path is followed
1618 by other driver implementations.
1619 This patch demotes the rank of vaapisink to secondary, so it will
1620 not be autoplugged avoiding bad experience of users.
1621 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/336>
1623 2020-06-19 10:44:50 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1625 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
1626 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
1627 libs: decoder: h264, h265: in context at least 16 reference surfaces
1628 Registering only stream's DBP size number of surfaces for decoding VA
1629 surfaces brings issues for certain streams. This change register all
1630 possible number of reference surfaces in a stream, which is 16.
1633 2020-07-04 21:21:57 +0800 He Junyan <junyan.he@intel.com>
1635 * gst/vaapi/gstvaapiencode_h265.c:
1636 plugins: encode: h265: Add profile,level,tier to output caps.
1637 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/350>
1639 2020-07-04 21:08:20 +0800 He Junyan <junyan.he@intel.com>
1641 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1642 libs: encoder: h265: init tier to GST_VAAPI_TIER_H265_UNKNOWN.
1643 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/350>
1645 2020-07-04 21:05:49 +0800 He Junyan <junyan.he@intel.com>
1647 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1648 libs: encoder: h265: fix a bug to get get_profile_tier_level.
1649 0 is a valid value for h265 tier.
1650 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/350>
1652 2020-07-02 19:19:35 +0800 He Junyan <junyan.he@intel.com>
1654 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1655 libs: encoder: h265: no need to check the high compression tune.
1656 The h265 encoder just support tune mode:
1658 (3): low-power - Low power mode
1659 So, no need to check and set the high compression parameters.
1660 And by the way, the current ensure_tuning_high_compression manner
1661 of choosing the hightest profile idc as the best compression profile
1662 is not correct. Unlike h264, in h265 the higher profile idc number
1663 does not mean it has more compression tools, and so it has better
1664 compression performance. It may even be un-compatible with the lower
1665 profile idc. For example, the SCREEN_CONTENT_CODING profile with idc
1666 9 is not compatible with 3D_MAIN profile with idc 8.
1667 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/348>
1669 2020-07-03 02:04:14 +0100 Tim-Philipp Müller <tim@centricular.com>
1674 === release 1.17.2 ===
1676 2020-07-03 00:36:40 +0100 Tim-Philipp Müller <tim@centricular.com>
1681 * gstreamer-vaapi.doap:
1685 2020-06-23 10:20:46 -0400 Thibault Saunier <tsaunier@igalia.com>
1687 * docs/gst_plugins_cache.json:
1688 * gst/vaapi/gstvaapiencode.c:
1689 docs: Mark parent classes as plugin API
1691 2020-06-23 00:07:57 +0200 Mathieu Duponchelle <mathieu@centricular.com>
1694 meson: mark plugins cache target as always stale
1696 2020-06-19 23:34:11 -0400 Thibault Saunier <tsaunier@igalia.com>
1698 * docs/gst_plugins_cache.json:
1699 doc: Stop documenting properties from parents
1701 2020-06-20 00:28:35 +0100 Tim-Philipp Müller <tim@centricular.com>
1706 === release 1.17.1 ===
1708 2020-06-19 19:27:11 +0100 Tim-Philipp Müller <tim@centricular.com>
1713 * gstreamer-vaapi.doap:
1717 2020-06-19 15:21:56 +0100 Tim-Philipp Müller <tim@centricular.com>
1719 * docs/gst_plugins_cache.json:
1720 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
1721 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
1722 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
1723 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
1724 Update plugin docs and add more plugins
1725 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/344>
1727 2020-06-11 08:32:05 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
1729 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
1730 libs: wayland: display: only handle the first output
1731 Right now, all outputs are handled. The means that the registry object for
1732 all but the last are leaked. As a result the sizes are not used correctly.
1733 With two outputs, at first the mode and physical size of the second output
1734 are used. If the first output changes the mode, then the physical size of
1735 the second output is used in combination with the resolution of the first
1736 output. The resulting pixel aspect ratio is incorrect.
1737 There seems to be no way to determine on which output the window is shown,
1738 so just use the first one to get consistent results.
1739 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/341>
1741 2020-06-11 18:15:17 +0800 He Junyan <junyan.he@hotmail.com>
1743 * gst/vaapi/gstvaapipluginbase.c:
1744 plugins: pluginbase: Do not destroy display when _close()
1745 When the element's state changes to NULL, it can still receive
1746 queries, such as the image formats. The display is needed in such
1747 queries but not well protected for MT safe.
1748 For example, ensure_allowed_raw_caps() may still use the display
1749 while it is disposed by gst_vaapi_plugin_base_close() because of
1751 We can keep the display until the element is destroyed. When the
1752 state changes to NULL, and then changes to PAUSED again, the display
1753 can be correctly set(if type changes), or leave untouched.
1755 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/343>
1757 2020-06-09 21:19:11 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1759 * gst-libs/gst/vaapi/gstvaapicontext.c:
1760 libs: context: use correct printing modifier
1761 GstVaapiID is an alias of gsize, thus its modifier is platform
1763 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/340>
1765 2020-06-06 00:42:46 +0200 Mathieu Duponchelle <mathieu@centricular.com>
1767 * gst-libs/gst/vaapi/gstvaapidisplay.c:
1768 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
1769 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1770 * gst-libs/gst/vaapi/gstvaapifilter.c:
1771 * gst-libs/gst/vaapi/gstvaapivalue.c:
1772 plugins: uddate gst_type_mark_as_plugin_api() calls
1774 2020-03-05 18:12:27 +0800 He Junyan <junyan.he@hotmail.com>
1776 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1777 libs: encoder: h265: Enable tile in VA command.
1778 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/294>
1780 2020-03-05 17:56:51 +0800 He Junyan <junyan.he@hotmail.com>
1782 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1783 libs: encoder: h265: Add ensure_tile to calculate tiles.
1784 We need consider tiles and slices together, separate tiles uniformly
1785 and then assign slices uniformly to each tiles.
1786 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/294>
1788 2020-04-30 14:19:29 +0800 He Junyan <junyan.he@hotmail.com>
1790 * gst-libs/gst/vaapi/gstvaapidisplay.c:
1791 * gst-libs/gst/vaapi/gstvaapidisplay.h:
1792 libs: display: add a quirk for iHD driver tile encoding.
1793 The iHD driver has a requirement that one slice can not span tiles
1794 when tile is enabled, which is not required by hevc spec.
1795 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/294>
1797 2020-03-05 17:40:43 +0800 He Junyan <junyan.he@hotmail.com>
1799 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1800 libs: encoder: h265: Add tile info to bitstream.
1801 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/294>
1803 2020-03-05 17:29:41 +0800 He Junyan <junyan.he@hotmail.com>
1805 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1806 libs: encoder: h265: promote level if tile is enabled.
1807 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/294>
1809 2020-03-05 17:07:28 +0800 He Junyan <junyan.he@hotmail.com>
1811 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1812 * gst-libs/gst/vaapi/gstvaapiutils_h265.h:
1813 libs: encoder: h265: Add num-tile-cols/rows properties.
1814 These properties are used for support of tile encoding. We just
1815 support uniform mode of tile encoding, that is, separating picture
1816 equally by (num-tile-cols X num-tile-rows).
1817 According to HEVC spec A1, the max number of tiles in column is 20
1818 and in rows is 22, so add two constant definitions.
1819 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/294>
1821 2020-03-05 16:21:24 +0800 He Junyan <junyan.he@hotmail.com>
1823 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1824 libs: encoder: h265: extract slice creation from add_slice_headers
1825 extract slice creation details from add_slice_headers, and let the
1826 add_slice_headers just focuses on calculating slice start address
1828 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/294>
1830 2020-03-05 12:44:45 +0800 He Junyan <junyan.he@hotmail.com>
1832 * gst-libs/gst/vaapi/gstvaapiencoder.c:
1833 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
1834 libs: encoder: Add a helper function to check the tile support.
1835 Encoding by tiles separation now is a very common feature for all
1836 relative new codecs, such as HEVC, AV1, and VP9. Just make this
1837 check as a common helper function of the encoder base class.
1838 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/294>
1840 2020-05-13 18:02:07 +0800 He Junyan <junyan.he@hotmail.com>
1842 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
1843 * gst-libs/gst/vaapi/gstvaapiprofile.c:
1844 * gst-libs/gst/vaapi/gstvaapiprofile.h:
1845 * gst-libs/gst/vaapi/gstvaapiutils.c:
1846 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
1847 libs: decoder: H265: Add SCC_MAIN_444_10 profile support.
1848 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/311>
1850 2020-05-13 16:05:59 +0800 He Junyan <junyan.he@hotmail.com>
1852 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
1853 libs: decoder: update reference list for SCC.
1854 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/311>
1856 2020-05-13 15:46:29 +0800 He Junyan <junyan.he@hotmail.com>
1858 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
1859 libs: decoder: H265: Fill picture and slice SCC parameters.
1860 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/311>
1862 2020-05-13 15:00:53 +0800 He Junyan <junyan.he@hotmail.com>
1864 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
1865 libs: util: H265: recognize the SCC profiles.
1866 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/311>
1868 2020-05-13 14:53:46 +0800 He Junyan <junyan.he@hotmail.com>
1870 * gst-libs/gst/vaapi/gstvaapiprofile.c:
1871 * gst-libs/gst/vaapi/gstvaapiprofile.h:
1872 * gst-libs/gst/vaapi/gstvaapiutils.c:
1873 libs: profile: Add screen extended main/main10/main444 define.
1874 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/311>
1876 2020-04-03 14:53:40 +0800 He Junyan <junyan.he@hotmail.com>
1878 * gst-libs/gst/vaapi/gstvaapiencoder.c:
1879 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1880 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
1881 * gst/vaapi/gstvaapiencode_h265.c:
1882 libs: encoder: h265: Add support for MAIN 4:2:2 10 profile.
1883 Using YUY2 as the input of the encoder can generate main 4:2:2 bit
1884 streams and using Y210 as the input of the encoder can generate main
1885 4:2:2 10 bit streams.
1886 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/334>
1888 2020-05-29 16:40:20 +0800 He Junyan <junyan.he@hotmail.com>
1890 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1891 libs: encoder: h265: Use correct index for SubWidthC and SubHeightC.
1892 We need to use the chroma_format_idc as the index for getting the
1893 SubWidthC and SubHeightC values as the spec 6.1(table 6-1) defines.
1894 The wrong SubWidthC or SubHeightC make us calculate a wrong right
1895 or bottom offset for crop size and generate garbage in output.
1896 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/334>
1898 2020-05-29 15:37:24 +0800 He Junyan <junyan.he@hotmail.com>
1900 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
1901 libs: encoder: h265: Fix chrome idc for 444 10 bits
1902 GST_VAAPI_CHROMA_TYPE_YUV444_10BPP should also set chroma_format_idc
1903 to 3 as GST_VAAPI_CHROMA_TYPE_YUV444 does.
1904 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/334>
1906 2020-06-03 18:37:22 -0400 Thibault Saunier <tsaunier@igalia.com>
1909 doc: Require hotdoc >= 0.11.0
1911 2020-06-03 18:49:53 -0400 Thibault Saunier <tsaunier@igalia.com>
1913 * gst/vaapi/gstvaapipostproc.c:
1914 doc: Fix wrong link to GstVideoDirectionMethod
1916 2020-06-03 17:38:10 -0400 Thibault Saunier <tsaunier@igalia.com>
1918 * docs/gst_plugins_cache.json:
1919 * meson_options.txt:
1920 docs: Update plugin cache with the new format
1921 And fix the default URL which should not be inside quotes.
1923 2020-06-03 17:37:48 -0400 Thibault Saunier <tsaunier@igalia.com>
1925 * gst-libs/gst/vaapi/gstvaapidisplay.c:
1926 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
1927 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
1928 * gst-libs/gst/vaapi/gstvaapifilter.c:
1929 * gst-libs/gst/vaapi/gstvaapivalue.c:
1930 * gst/vaapi/gstvaapipostproc.c:
1931 Use gst_type_mark_as_plugin_api() for all non-element plugin types
1933 2020-05-23 22:09:17 +0800 He Junyan <junyan.he@hotmail.com>
1935 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
1936 * gst-libs/gst/vaapi/gstvaapibufferproxy_priv.h:
1937 * gst-libs/gst/vaapi/gstvaapicontext.h:
1938 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
1939 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
1940 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
1941 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
1942 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
1943 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
1944 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
1945 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
1946 * gst-libs/gst/vaapi/gstvaapiimage.h:
1947 * gst-libs/gst/vaapi/gstvaapiobject.c:
1948 * gst-libs/gst/vaapi/gstvaapiobject.h:
1949 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
1950 * gst-libs/gst/vaapi/gstvaapiprofile.h:
1951 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
1952 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
1953 * gst-libs/gst/vaapi/gstvaapisurface.h:
1954 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
1955 * gst-libs/gst/vaapi/gstvaapivideopool.c:
1956 * gst-libs/gst/vaapi/meson.build:
1957 libs: delete all gstvaapiobject related files.
1958 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/331>
1960 2020-05-23 20:48:54 +0800 He Junyan <junyan.he@hotmail.com>
1962 * gst-libs/gst/vaapi/gstvaapisurface.c:
1963 libs: surface: return fail immediately if can not create subpicture
1964 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/331>
1966 2020-05-23 14:00:58 +0800 He Junyan <junyan.he@hotmail.com>
1968 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
1969 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
1970 * gst-libs/gst/vaapi/gstvaapisurface.c:
1971 * tests/internal/image.c:
1972 libs: subpicture: Make subpicture a standard GstMiniObject.
1973 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/331>
1975 2020-05-26 02:19:15 +0800 He Junyan <junyan.he@hotmail.com>
1977 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
1978 libs: decoder: h264: Add ref flags for splited field.
1979 When split one frame into fields, the second field should also
1980 copy the reference flags.
1981 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/333>
1983 2020-05-25 15:46:58 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1985 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
1986 libs: decoder: h264: disallow multiple slice group
1987 As far as we know there are no VAAPI drivers supporting FMO, which
1988 migth be used in baseline streams.
1989 This commit is a continuation of
1990 https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/328
1991 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/332>
1993 2020-05-20 10:50:05 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
1995 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
1996 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
1997 * gst/vaapi/gstvaapidecode.c:
1998 * gst/vaapi/gstvaapidecode_props.c:
1999 * gst/vaapi/gstvaapidecode_props.h:
2000 vaapidecoder: h264: remove baseline as constrained property
2001 From now on always the baseline is going to be treated as constrained without
2002 need of setting a property.
2003 Since the property was added along the development cycle (1.17 / commit
2004 866a9f06) and never released, we assume that it is safe to remove it.
2006 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/328>
2008 2020-05-21 11:37:36 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2011 README: update VP9 decoder and encoder
2012 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/330>
2014 2020-05-21 13:42:47 +0800 Haihao Xiang <haihao.xiang@intel.com>
2016 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
2017 vaapiencoder_h264: set direct_spatial_mv_pred_flag to true by default
2018 This flag is set to true by default in both MediaSDK and FFmpeg-vaapi,
2019 so let's align this plugin with other libraries / softwares.
2020 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/329>
2022 2020-05-18 18:29:05 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2024 * gst/vaapi/gstvaapivideobufferpool.c:
2025 vaapivideobufferpool: fix meta overwrite
2026 commit 7ac2a207 added a regression by erroneously assumed that
2027 GstVaapiVideoMeta is actually a GstMeta, which is not.
2028 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/327>
2030 2020-05-17 09:55:42 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2032 * gst/vaapi/gstvaapivideobufferpool.c:
2033 * gst/vaapi/gstvaapivideometa_texture.c:
2034 * gst/vaapi/gstvaapivideometa_texture.h:
2035 vaapivideopool: Set pooled flag to added metas.
2036 So this could hint filters how to use these metas.
2037 Had to change the return value for texutre upload meta in order
2039 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/326>
2041 2020-05-16 20:49:31 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2043 * gst-libs/gst/vaapi/gstvaapicontext.c:
2044 * gst-libs/gst/vaapi/gstvaapidisplay.c:
2045 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
2046 * gst-libs/gst/vaapi/gstvaapiparser_frame.c:
2047 * gst-libs/gst/vaapi/gstvaapisurface.c:
2048 libs: use array_unref() rather than array_free()
2049 It is more convinience and thread-safe.
2050 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/325>
2052 2020-04-10 22:20:35 +0800 He Junyan <junyan.he@hotmail.com>
2054 * gst/vaapi/gstvaapi.c:
2055 plugin: use register_type to replace get_type for encode init.
2056 xxx_register_type will detect the template sink caps and is needed
2057 to be called at init time.
2058 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/315>
2060 2020-04-10 22:05:50 +0800 He Junyan <junyan.he@hotmail.com>
2062 * gst/vaapi/gstvaapiencode_vp9.c:
2063 * gst/vaapi/gstvaapiencode_vp9.h:
2064 plugins: encode: Modify sink template of vp9 encode.
2065 Use gst_vaapi_detect_codec_caps to get more precise template caps.
2066 Also implement gst_vaapiencode_vp9_register_type, which should be
2067 called at plugin register time.
2068 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/315>
2070 2020-04-10 22:05:18 +0800 He Junyan <junyan.he@hotmail.com>
2072 * gst/vaapi/gstvaapiencode_vp8.c:
2073 * gst/vaapi/gstvaapiencode_vp8.h:
2074 plugins: encode: Modify sink template of vp8 encode.
2075 Use gst_vaapi_detect_codec_caps to get more precise template caps.
2076 Also implement gst_vaapiencode_vp8_register_type, which should be
2077 called at plugin register time.
2078 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/315>
2080 2020-04-10 22:04:34 +0800 He Junyan <junyan.he@hotmail.com>
2082 * gst/vaapi/gstvaapiencode_mpeg2.c:
2083 * gst/vaapi/gstvaapiencode_mpeg2.h:
2084 plugins: encode: Modify sink template of mpeg2 encode.
2085 Use gst_vaapi_detect_codec_caps to get more precise template caps.
2086 Also implement gst_vaapiencode_mpeg2_register_type, which should be
2087 called at plugin register time.
2088 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/315>
2090 2020-04-10 22:03:49 +0800 He Junyan <junyan.he@hotmail.com>
2092 * gst/vaapi/gstvaapiencode_h265.c:
2093 * gst/vaapi/gstvaapiencode_h265.h:
2094 plugins: encode: Modify sink template of h265 encode.
2095 Use gst_vaapi_detect_codec_caps to get more precise template caps.
2096 Also implement gst_vaapiencode_h265_register_type, which should be
2097 called at plugin register time.
2098 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/315>
2100 2020-04-10 22:00:38 +0800 He Junyan <junyan.he@hotmail.com>
2102 * gst/vaapi/gstvaapiencode_h264.c:
2103 * gst/vaapi/gstvaapiencode_h264.h:
2104 plugins: encode: Modify sink template of h264 encode.
2105 Use gst_vaapi_detect_codec_caps to get more precise template caps.
2106 Also implement gst_vaapiencode_h264_register_type, which should be
2107 called at plugin register time.
2108 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/315>
2110 2020-04-10 22:11:34 +0800 He Junyan <junyan.he@hotmail.com>
2112 * gst/vaapi/gstvaapiencode_jpeg.c:
2113 * gst/vaapi/gstvaapiencode_jpeg.h:
2114 plugins: encode: Modify sink template of jpeg encode.
2115 Use gst_vaapi_detect_codec_caps to get more precise template caps.
2116 Also implement gst_vaapiencode_jpeg_register_type, which should be
2117 called at plugin register time.
2118 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/315>
2120 2020-04-23 00:02:02 +0800 He Junyan <junyan.he@hotmail.com>
2122 * gst/vaapi/gstvaapiencode.h:
2123 plugin: encode: add a helper macro to register encode type.
2124 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/315>
2126 2020-04-10 21:44:05 +0800 He Junyan <junyan.he@hotmail.com>
2128 * gst/vaapi/gstvaapipluginutil.c:
2129 * gst/vaapi/gstvaapipluginutil.h:
2130 plugins: util: Add a helper function to detect supported caps.
2131 This helper function iterate all profiles and entrypoints belong
2132 to the specified codec, query the VAConfigAttribRTFormat and list
2133 all possible video formats.
2134 This function is used by each codec to get the template sink caps
2135 (for encode) or src caps(for decode) at register time, when just
2136 all possible formats are listed and no need to be very accurate.
2137 So there is no context created for the performance reason. Most
2138 codecs just use YUV kinds of formats as the input/output, so we do
2139 not include RGB kinds of formats. User can specified more formats
2140 in extra_fmts(For example, jpeg may need BGRA) if needed.
2141 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/315>
2143 2020-04-10 21:27:32 +0800 He Junyan <junyan.he@hotmail.com>
2145 * gst/vaapi/gstvaapiencode.c:
2146 * gst/vaapi/gstvaapipluginutil.c:
2147 * gst/vaapi/gstvaapipluginutil.h:
2148 plugin: encode: extract the allowed caps maker as a helper function.
2149 Extract all logic about making caps for encode's sink as a standalone
2150 helper function. It can be reused.
2151 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/315>
2153 2020-04-10 21:21:43 +0800 He Junyan <junyan.he@hotmail.com>
2155 * gst-libs/gst/vaapi/video-format.c:
2156 * gst-libs/gst/vaapi/video-format.h:
2157 libs: video-format: add a helper function of get_formats_by_chroma.
2158 The function iterates all supported video formats and returns the
2159 formats belong to the specified chroma type.
2160 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/315>
2162 2020-05-16 21:03:32 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2164 * gst-libs/gst/vaapi/gstvaapitexture.c:
2165 * gst-libs/gst/vaapi/gstvaapitexture_priv.h:
2166 libs: texture: remove unused headers include
2167 This is continuation of
2168 https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/317
2169 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/324>
2171 2020-05-16 19:58:25 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2173 * gst-libs/gst/vaapi/gstvaapicodedbuffer.h:
2174 * gst-libs/gst/vaapi/gstvaapiimage.h:
2175 * gst-libs/gst/vaapi/gstvaapisurface.h:
2176 libs: removed duplicated function declarations
2177 Some headers had duplicated inlined function declaration. This was
2178 for gtkdoc, but now GStreamer uses hotdoc and the internal library
2179 documentation is not generated. So let's remove these extra lines.
2180 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/323>
2182 2020-04-18 19:32:24 +0800 He Junyan <junyan.he@hotmail.com>
2184 * gst-libs/gst/vaapi/gstvaapitexture.c:
2185 * gst-libs/gst/vaapi/gstvaapitexture.h:
2186 * gst-libs/gst/vaapi/gstvaapitexture_egl.c:
2187 * gst-libs/gst/vaapi/gstvaapitexture_glx.c:
2188 * gst-libs/gst/vaapi/gstvaapitexture_priv.h:
2189 * gst-libs/gst/vaapi/gstvaapitexturemap.c:
2190 * gst-libs/gst/vaapi/gstvaapiwindow_egl.c:
2191 * gst/vaapi/gstvaapivideometa_texture.c:
2192 * tests/internal/test-textures.c:
2193 libs: texture: Make texture a standard GstMiniObject.
2194 We store GstVaapiTextureGLX and GstVaapiTextureEGL's private data in
2195 the qdata of miniobject and avoid extending the base texture class.
2196 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/317>
2198 2020-04-26 12:33:29 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2202 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/320>
2204 2020-04-21 18:00:26 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2206 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
2207 libs: display: drm: use g_strcmp0 to be null safe
2208 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/321>
2210 2020-04-26 13:30:16 +0800 Haihao Xiang <haihao.xiang@intel.com>
2212 * gst/vaapi/gstvaapipluginutil.c:
2213 vaapipluginutil: Use GST_VAAPI_DISPLAY_TYPE_DRM for Mesa3D GBM
2214 We may build this plugin with window system support but run it without
2215 window system. Without this patch, the following pipeline will trigger a
2216 segfault when running it without window system.
2217 gst-launch-1.0 filesrc location=input.264 ! h264parse ! vaapih264dec ! fakesink
2218 Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/319>
2220 2020-04-21 11:16:37 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2223 build: use join_paths() for driverdir
2225 2020-04-21 09:04:51 +0000 Veerabadhran G <vegopala@amd.com>
2228 README: Update supported hardware
2229 Added the AMD hardware list to the "Hardware Requirements" section.
2231 2020-04-04 13:58:00 +0200 He Junyan <junyan.he@hotmail.com>
2233 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
2234 * gst-libs/gst/vaapi/gstvaapibufferproxy.h:
2235 * gst-libs/gst/vaapi/gstvaapibufferproxy_priv.h:
2236 libs: bufferproxy: remove GstMemory reference
2237 Since bufferproxy and surface are not referenced circularly, there's
2238 no need to keep, in the buffer proxy, a reference to the GstMemory
2239 where it is held. This patch removes that handling.
2241 2020-03-15 23:29:05 +0800 He Junyan <junyan.he@hotmail.com>
2243 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
2244 * gst-libs/gst/vaapi/gstvaapisurface.c:
2245 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
2246 * gst-libs/gst/vaapi/gstvaapisurface_drm.h:
2247 * gst/vaapi/gstvaapivideomemory.c:
2248 libs,plugin: break surface-bufferproxy circular reference
2249 The bufferproxy may reference the surface and the surface may also
2250 reference the bufferproxy, producing a circular reference, which might
2251 lead to serious resource leak problems.
2252 Now make the relationship clearer, the bufferproxy's references is
2253 transfered to surface, while bufferproxy just keeps the surface's
2254 address without increasing its reference count.
2255 The surface can be created through a bufferproxy like in
2256 gst_vaapi_surface_new_with_dma_buf_handle(), and the surface might
2257 get its bufferproxy via gst_vaapi_surface_get_dma_buf_handle(). In
2258 both cases the surface holds a bufferproxy's reference.
2260 2020-04-03 18:43:52 +0200 He Junyan <junyan.he@hotmail.com>
2262 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
2263 * gst-libs/gst/vaapi/gstvaapibufferproxy_priv.h:
2264 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
2265 libs: bufferproxy: rename parent memeber as surface
2267 2020-03-15 22:07:31 +0800 He Junyan <junyan.he@hotmail.com>
2269 * gst/vaapi/gstvaapivideobufferpool.c:
2270 plugin: bufferpool: use hashmap to cache dmabuf mem-surface
2271 The old way of refer memory by bufferproxy is not a good one, since it
2272 make the logic error prone.
2273 Now it is established a map between surface-bufferproxy and its GstMemory,
2274 caching the memory bound by a surface looked for the specified surface.
2276 2020-03-15 21:50:24 +0800 He Junyan <junyan.he@hotmail.com>
2278 * gst/vaapi/gstvaapivideobufferpool.c:
2279 * gst/vaapi/gstvaapivideobufferpool.h:
2280 plugin: bufferpool: Delete ACQUIRE_FLAG_NO_ALLOC flag.
2281 Delete the GST_VAAPI_VIDEO_BUFFER_POOL_ACQUIRE_FLAG_NO_ALLOC flag.
2282 In fact, no one is using that flag, and all vaapi buffers should
2283 have GstVaapiVideoMeta.
2285 2020-02-13 09:43:38 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2287 * gst/vaapi/gstvaapipostproc.c:
2288 * gst/vaapi/gstvaapipostproc.h:
2289 vaapipostproc: enable HDR10 tone mapping
2291 2020-02-13 09:00:18 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2293 * gst-libs/gst/vaapi/gstvaapifilter.c:
2294 * gst-libs/gst/vaapi/gstvaapifilter.h:
2295 libs: filter: HDR10 tone mapping support
2296 Add support for HDR10 tone mapping (since VA-API 1.4.0).
2298 2020-04-02 15:14:15 +0800 He Junyan <junyan.he@hotmail.com>
2300 * gst-libs/gst/vaapi/gstvaapiencoder.c:
2301 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
2302 * gst-libs/gst/vaapi/gstvaapiprofile.c:
2303 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
2304 * gst/vaapi/gstvaapiencode_h265.c:
2305 libs: encoder: h265: Support MAIN 4:4:4 10 profile.
2306 Using Y410 as the input of the encoder can generate main_444_10 bit
2309 2020-04-02 15:19:41 +0800 He Junyan <junyan.he@hotmail.com>
2311 * gst-libs/gst/vaapi/gstvaapiencoder.c:
2312 libs: encoder: fix an inexact trace info in chroma type check.
2314 2020-03-31 12:22:31 +0800 Haihao Xiang <haihao.xiang@intel.com>
2316 * gst-libs/gst/vaapi/gstvaapiencoder.c:
2317 libs: encoder: make sure format array is not NULL when returning TRUE
2318 This fixed segfault when running the pipeline below with iHD driver
2319 (commit efe5e9a) on ICL
2320 gst-launch-1.0 videotestsrc ! vaapivp9enc tune=low-power ! vaapivp9dec ! \
2323 2020-03-18 13:28:00 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2325 * gst-libs/gst/vaapi/gstvaapidecoder.h:
2326 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
2327 * gst-libs/gst/vaapi/gstvaapidecoder_h265.h:
2328 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.h:
2329 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.h:
2330 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.h:
2331 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.h:
2332 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.h:
2333 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.h:
2334 * gst-libs/gst/vaapi/gstvaapidisplay.h:
2335 * gst-libs/gst/vaapi/gstvaapidisplay_drm.h:
2336 * gst-libs/gst/vaapi/gstvaapidisplay_egl.h:
2337 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
2338 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.h:
2339 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
2340 * gst-libs/gst/vaapi/gstvaapiencoder.h:
2341 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
2342 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
2343 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.h:
2344 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
2345 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.h:
2346 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.h:
2347 * gst-libs/gst/vaapi/gstvaapifilter.h:
2348 * gst-libs/gst/vaapi/gstvaapitexturemap.h:
2349 * gst-libs/gst/vaapi/gstvaapiwindow.h:
2350 * gst-libs/gst/vaapi/gstvaapiwindow_drm.h:
2351 * gst-libs/gst/vaapi/gstvaapiwindow_egl.h:
2352 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
2353 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.h:
2354 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
2355 libs: extend g_autoptr support
2357 2020-03-26 22:40:40 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2359 * gst/vaapi/gstvaapivideometa.c:
2360 vaapivideometa: remove compiler warning
2362 2020-03-22 20:59:20 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2365 * meson_options.txt:
2366 build: Add meson's option package-origin.
2367 This options is added to synchronize with other gstreamer packages
2368 build configuration.
2369 Though, to avoid breaking distro configuration it is set, as default,
2370 the issues gitlab's url, instead of the used string
2371 "Unkown package origin".
2372 Also, set_quoted is used for string based cdata.
2374 2020-02-25 13:45:05 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2376 * gst/vaapi/gstvaapipostproc.c:
2377 vaapipostproc: deprecate format, width and size parameters
2378 Since they should only be controlled by caps negotiation.
2380 2020-03-18 16:41:01 +0800 He Junyan <junyan.he@hotmail.com>
2382 * gst-libs/gst/vaapi/gstvaapiprofile.c:
2383 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
2384 * gst/vaapi/gstvaapidecode.c:
2385 libs,plugins: decoder: Add -intra profile support for hevc.
2386 In hevc, we can consider the -intra profile a subset of the none
2387 -intra profile. The -intra profiles just contain I frames and we
2388 definitely can use the none -intra profiles's context to decode
2390 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2392 2020-03-04 12:35:42 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2394 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
2395 libs: encoder: h265: support ICQ/QVBR BRC
2396 Enable support for ICQ and QVBR bitrate control.
2397 The code is essentially the same for h264 ICQ/QVBR support
2398 which was added in commit 9e0c133a2403.
2400 2020-03-19 11:19:18 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
2402 * gst-libs/gst/vaapi/gstvaapiencoder.c:
2403 libs: encoder: set VA HRD param before RC param
2404 This is a workaround for intel-media-driver bug
2405 https://github.com/intel/media-driver/issues/865
2406 The driver will force the RC method to CBR for HEVCe
2407 when it parses the HRD param. Thus, any RC method
2408 param submitted "prior" to the HRD param will be lost.
2409 Therefore, VBR, ICQ and QVBR for HEVCe can't be
2410 effectively enabled if the RC method param "precedes"
2412 To work around this issue, set the HRD param before
2413 the RC method param so the driver will parse the RC
2414 method param "after" the HRD param.
2415 Afaict, other codecs in the driver (and other drivers)
2416 do not appear to be dependent on the order of HRD and
2417 RC param submission.
2419 2019-11-25 14:16:30 +0800 Haihao Xiang <haihao.xiang@intel.com>
2421 * gst-libs/gst/vaapi/gstvaapidisplay.c:
2422 * gst-libs/gst/vaapi/gstvaapidisplay.h:
2423 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
2424 vaapijpegenc: Add a quantization quirk for iHD driver
2425 iHD driver shifts the value by 50 when calculating quantization for JPEG
2426 encoding, so we should add 50 in this plugin for iHD driver too.
2428 2020-03-13 21:49:15 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2430 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
2431 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
2432 * gst-libs/gst/vaapi/gstvaapipixmap.c:
2433 * gst-libs/gst/vaapi/gstvaapipixmap.h:
2434 * gst-libs/gst/vaapi/gstvaapipixmap_priv.h:
2435 * gst-libs/gst/vaapi/gstvaapipixmap_x11.c:
2436 * gst-libs/gst/vaapi/gstvaapipixmap_x11.h:
2437 * gst-libs/gst/vaapi/gstvaapiwindow.c:
2438 * gst-libs/gst/vaapi/gstvaapiwindow.h:
2439 * gst-libs/gst/vaapi/gstvaapiwindow_egl.c:
2440 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
2441 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
2442 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
2443 * gst-libs/gst/vaapi/meson.build:
2445 * tests/internal/output.c:
2446 * tests/internal/output.h:
2447 * tests/internal/simple-decoder.c:
2448 * tests/internal/test-decode.c:
2449 libs: remove GstVaapiPixmap
2450 GstVaapiPixmap is an abstract base class which only implementation
2451 were GstVaapiPixmapX11. This class were used for a special type of
2452 rendering in the tests apps, utterly unrelated in GStreamer.
2453 Since gstreamer-vaapi is no longer a general-user wrapper for VA-API
2454 we should remove this unused API.
2455 This removal drops libxrender dependency.
2457 2020-03-17 18:51:19 +0800 He Junyan <junyan.he@hotmail.com>
2459 * gst-libs/gst/vaapi/gstvaapiutils.c:
2460 libs: utils: Add HEVC Main444 sting in string_of_VAProfile
2461 HEVCMain444_10 is already a supported profile and misses the strings.
2463 2020-03-17 12:47:33 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2465 * gst/vaapi/gstvaapi.c:
2466 plugin: don't error if cannot create display
2467 This might generated errors on automatic tools such as CI. Let's
2468 rather just raise a warning and let continue.
2470 2020-03-09 01:52:57 +0800 He Junyan <junyan.he@hotmail.com>
2472 * tests/check/meson.build:
2473 test: fix a ninja test failure for vaapioverlay.
2474 That test case only works with drm display, so the build such as
2475 meson -Dwith_x11=yes -Dwith_wayland=no -Dwith_drm=no -Dwith_egl=no
2477 gets a failure when run ninja test. Just enable this test when drm
2480 2020-02-07 23:56:13 +0800 He Junyan <junyan.he@hotmail.com>
2482 * gst-libs/gst/vaapi/gstvaapivideopool.c:
2483 libs: videopool: fix a condition race for pool allocate.
2485 2020-03-05 13:22:23 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2487 * gst/vaapi/gstvaapivideobufferpool.c:
2488 vaapivideobufferpool: force video meta if sizes are different
2489 The strides and offsets could be the same, but the allocation
2490 size might be different (e.g. alignment). Thus, ensure we also
2491 set the flag to copy from VA memory to system memory when alloc
2495 2020-03-05 14:18:32 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2497 * gst/vaapi/gstvaapidecode.c:
2498 vaapidecode: don't set base sink caps twice
2499 Base class's sink pad caps are already set when calling set_format().
2500 There's no need to call it again in gst_vaapidecode_negotiate().
2502 2020-03-05 13:26:38 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2504 * gst/vaapi/gstvaapidecode.c:
2505 vaapidecode: unlock stream if caps update fails
2506 If caps update fail a dead lock occurs since the stream mutex is not
2509 2020-03-03 15:24:32 +0800 Xu Guangxin <guangxin.xu@intel.com>
2511 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
2512 libs: decoder: h265: parser state after update dependent slice
2513 If the dependent_slice_segment_flag is true, most slice info derived from last slice.
2514 So we need check the slice type after we call populate_dependent_slice_hdr
2516 2020-02-16 12:21:28 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2518 * gst-libs/gst/vaapi/gstvaapidisplay.c:
2519 * gst-libs/gst/vaapi/gstvaapidisplay.h:
2520 libs: display: force RGBA image format for i965 driver
2521 Since commit 32bf6f1e GLTextureUpload is broken because i965
2522 doesn't report properly RGBA support. It could be possible to use RGBx
2523 but GLTextureUpload only regotiates RGBA.
2524 The simplest fix to this regression is adding synthetically the RGBA
2525 format in the internal format map.
2527 2020-02-14 19:30:54 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2529 * gst-libs/gst/vaapi/gstvaapidisplay.c:
2530 libs: display: iterate all quirks table
2531 Instead of break at the fist foud quirk in the table, iterate all over
2532 so it would be feasible to add several quirks for one driver per
2535 2020-02-25 12:05:28 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2537 * gst-libs/gst/vaapi/gstvaapifilter.c:
2538 libs: filter: handle RGB to/from YUV color primary driver quirk
2539 The intel-media-driver (iHD) can't convert output color
2540 primaries when doing YUV to/from RGB CSC. Thus, we must
2541 keep the output color primaries the same as the input
2542 color primaries for this case.
2545 2020-02-25 12:00:36 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2547 * gst-libs/gst/vaapi/gstvaapidisplay.c:
2548 * gst-libs/gst/vaapi/gstvaapidisplay.h:
2549 libs: display: add YUV to/from RGB color primary quirk
2550 The intel-media-driver (iHD) can't convert output color
2551 primaries when doing YUV to/from RGB CSC.
2553 2020-02-28 11:33:18 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2555 * gst/vaapi/gstvaapivideomemory.c:
2556 Revert "vaapivideomemory: Store surface allocation flags."
2557 This reverts commit dd428cc4a12c2d5c694fcd3303811cf486002c9d because
2558 it rewrites the buffer size whilst surface allocation flags are
2559 stored when allocator_params_init() is called since fab890ce.
2562 2020-01-27 18:19:57 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2564 * gst/vaapi/gstvaapivideomemory.c:
2565 vaapivideomemory: Try surface with allocation flags.
2566 When a vaapi allocator is instantiated, it first try to generate a
2567 surface with the specified configuration.
2568 This patch adds, in this tried buffer, the requested allocation flags.
2570 2020-01-27 18:10:18 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2572 * gst/vaapi/gstvaapivideomemory.c:
2573 vaapivideomemory: Store surface allocation flags.
2574 Store surface allocation flags passed to the vaapi allocator in
2575 GObject's qdata, because it might be used by the vaapivideobufferpool
2576 when recreating the allocator given any resolution change.
2578 2020-01-24 19:32:52 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2580 * gst-libs/gst/vaapi/gstvaapisurface.c:
2581 * gst-libs/gst/vaapi/gstvaapisurface.h:
2582 libs: surface: Add hints to allocation flags.
2583 When creating surfaces it is possible to pass to VA hints of its usage,
2584 so the driver may do some optimizations.
2585 This commit adds the handling of encoding/decoding hints.
2587 2020-01-24 22:08:50 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2589 * gst-libs/gst/vaapi/gstvaapicontext.c:
2590 * gst-libs/gst/vaapi/gstvaapisurface.c:
2591 * gst-libs/gst/vaapi/gstvaapisurface.h:
2592 * gst-libs/gst/vaapi/gstvaapisurface_egl.c:
2593 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
2594 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
2595 * gst-libs/gst/vaapi/gstvaapiwindow.c:
2596 * tests/internal/test-filter.c:
2597 * tests/internal/test-surfaces.c:
2598 libs: surface: surfacepool: Add allocation flags in constructors.
2600 2020-02-22 16:06:13 +0800 He Junyan <junyan.he@hotmail.com>
2603 build: let the build fail if none if X11, wayland or drm.
2604 In fact, gst_vaapi_create_test_display only test x11, wayland and
2605 drm, no glx and egl entries. So if none of them is enabled, no
2606 vaapi element can be detected.
2608 2020-02-21 00:58:47 +0800 He Junyan <junyan.he@hotmail.com>
2610 * tests/internal/test-display.c:
2611 test: avoid unused warning for test-display
2612 meson -Dwith_x11=yes -Dwith_wayland=no -Dwith_drm=no -Dwith_egl=no
2613 -Dwith_glx=no buildir
2614 generate unused warnings.
2616 2020-02-21 00:50:47 +0800 He Junyan <junyan.he@hotmail.com>
2619 build: fix meson build error when without x11.
2620 meson -Dwith_x11=no build_dir
2621 can not success build the project because the glx is still enabled.
2622 We need to disable GLX when X11 is disabled.
2624 2020-02-14 19:53:09 +0800 Haihao Xiang <haihao.xiang@intel.com>
2626 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
2627 libs: h265enc: Set max_transform_hierarchy_depth_{inter, intra} to 2
2628 Intel HW has limitation on max_transform_hierarchy_depth_inter and
2629 max_transform_hierarchy_depth_intra (see [1]). We can provide a quirk for
2630 other HWs if other HWs may support other values
2631 [1] https://01.org/sites/default/files/documentation/intel-gfx-prm-osrc-kbl-vol10-hevc.pdf
2633 2020-02-21 07:37:50 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2635 * gst/vaapi/gstvaapipostproc.c:
2636 * gst/vaapi/gstvaapipostprocutil.c:
2637 vaapipostproc: do not compensate for crop/direction if no VPP
2638 If we do not have functional VPP, then cropping and video
2639 direction is non-functional and we should avoid calling
2640 any of the gst_vaapi_filter* APIs.
2642 2020-02-21 06:54:47 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2644 * gst/vaapi/gstvaapipostproc.c:
2645 vaapipostproc: only set VPP colorimetry when VPP is available
2646 If we don't have functional vpp then we should not call
2647 gst_vaapi_filter_set_colorimetry.
2649 2020-02-16 01:25:37 +0800 He Junyan <junyan.he@hotmail.com>
2651 * gst/vaapi/gstvaapivideobufferpool.c:
2652 * gst/vaapi/gstvaapivideomemory.c:
2653 * gst/vaapi/gstvaapivideomemory.h:
2654 videobufferpool: don't reset surface when created internally
2655 The bug fixing, in commit 89f202ea, just considers the case when
2656 surface's DMABuf is set through gst_buffer_pool_acquire_buffer(),
2657 which is typically a decoder's behavior. But vaapipostproc doesn't
2658 provide any surface when calling gst_buffer_pool_acquire_buffer(),
2659 thus a surface is created when GstMemory is allocated.
2660 If the surface proxy in buffer's meta is reset at
2661 buffer_pool_reset_buffer(), that surface will be destroyed and it
2662 won't be available anymore. But GstBuffers are cached in the buffer
2663 pool and they are reused again, hence only those images are rendered
2667 2020-02-16 17:19:04 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2669 * gst-libs/gst/vaapi/gstvaapiblend.c:
2670 * gst-libs/gst/vaapi/gstvaapifilter.c:
2671 * gst/vaapi/gstvaapi.c:
2672 libs: blend: filter: handle finalize() if display isn't assigned
2673 I've just discovered iHD driver in Skylake doesn't have VideoProc
2674 entry point, hence, in this platform, when vaapioverlay is tried to be
2675 registered, critical warnings are raised because blend doesn't have a
2677 As it is possible to have drivers without EntryPointVideoProc it is
2678 required to handle it gracefully. This patch does that: only tries to
2679 register vaapioverlay if the testing display has VPP and finalize()
2680 vmethods, in filter and blend, bail out if display is NULL.
2682 2020-02-15 11:02:49 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2684 * gst-libs/gst/vaapi/gstvaapifilter.c:
2685 * gst-libs/gst/vaapi/gstvaapiutils.c:
2686 libs: filter: guard all color properties to VA-API 1.2.0
2687 Older VA-API (0.39.0) doesn't have VAProcColorProperties.
2688 Thus, guard all colorimetry -> VA-API support to version
2692 2020-02-17 08:55:36 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2694 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
2695 libs: decoder: h265: set parser info state at decoding codec data
2696 Commit 1168d6d5 showed up a regression: decode_sps() stores the unit's
2697 parser info in sps array. If that parser info comes from decoding
2698 codec data, that parser info will have an undefined state which might
2700 This patch sets the parser info state, at decoding codec data, with
2701 the internal parser state. This is similar with h264 decoder apprach.
2702 Original-patch-by: Xu Guangxin <guangxin.xu@intel.com>
2704 2020-02-07 15:24:18 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2706 * gst/vaapi/gstvaapipostproc.c:
2707 vaapipostproc: demote log message to trace level
2709 2020-02-14 14:45:56 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2711 * gst-libs/gst/vaapi/video-format.c:
2712 libs: video-format: set general vaapi log category
2713 Instead of logging in an unspecified category, set the default vaapi.
2715 2020-02-15 11:44:48 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2717 * gst-libs/gst/vaapi/gstvaapiutils.c:
2718 libs: utils: guard EncSliceLP for VA-API < 0.39.1
2721 2020-02-14 16:17:04 +0800 Haihao Xiang <haihao.xiang@intel.com>
2723 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
2724 libs: h265enc: Set VA_PICTURE_HEVC_INVALID flag for invalid picture
2726 2020-02-11 11:31:31 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2728 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
2729 * gst-libs/gst/vaapi/gstvaapidisplay.c:
2730 * gst-libs/gst/vaapi/gstvaapiencoder.c:
2731 libs: remove crumbs of libva < 0.39
2732 All these guarded code seem like leftovers of commit 920b1ec7a. This
2733 patch completes that missing clean up.
2735 2020-02-11 00:38:40 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2737 * gst-libs/gst/vaapi/gstvaapifilter.c:
2738 * gst-libs/gst/vaapi/gstvaapiutils.c:
2739 libs: VA explicit color standard not supported until 1.2.0
2740 VAProcColorStandardExplicit and associated VAProcColorProperties
2741 (primaries, transfer and matrix) are not supported until
2743 Use VAProcColorStandardNone instead of VAProcColorStandardExplicit
2747 2020-02-10 09:31:15 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2749 * gst-libs/gst/vaapi/gstvaapiutils.c:
2750 libs: utils: WA: use explicit for sRGB colorimetry
2751 Addresses #228 on iHD side. It seems iHD can't handle
2752 VAProcColorStandardSRGB in all situations for vpp. But
2753 it has no problem when we specify the sRGB parameters
2754 via VAProcColorStandardExplicit parameters.
2756 2020-02-07 11:25:31 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2758 * gst-libs/gst/vaapi/gstvaapifilter.c:
2759 libs: filter: set vpp input/output color range
2760 We've always sent VA_SOURCE_RANGE_UNKNOWN to the driver.
2761 And, the [iHD] driver essentially computes the same color
2762 range as gstreamer when we send VA_SOURCE_RANGE_UNKNOWN for
2763 cases were gstreamer computes it automatically. But,
2764 if the user wants to make it explicit, we should try
2767 2020-02-07 11:20:11 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2769 * gst-libs/gst/vaapi/gstvaapiutils.c:
2770 * gst-libs/gst/vaapi/gstvaapiutils.h:
2771 libs: utils: map GstVideoColorRange to VAAPI VPP
2773 2020-02-07 15:28:24 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2775 * gst/vaapi/gstvaapidecode.c:
2776 vaapidecode: don't remove chroma-site nor colorimetry
2777 Since now they can be handled by vaapipostproc.
2779 2020-02-04 14:17:43 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2781 * gst/vaapi/gstvaapipostprocutil.c:
2782 vaapipostproc: use sink resolution to calculate src colorimetry
2783 The default output colorimetry is persuaded by the output
2784 resolution, which is too naive when doing VPP cropping
2785 and/or scaling. For example, scaling 4K(sink)->1080P(src)
2786 resolution (i.e. both YUV) results in bt2020(sink)->bt709(src)
2787 colorimetry selection and some drivers don't support that
2789 Thus, if output (i.e. downstream) does not specify a
2790 colorimetry then we use the input resolution instead of the
2791 output resolution to create the default colorimetry. Also,
2792 note that we still use the output format since it may be a
2793 different color space than the input. As in the example
2794 above, this will result in bt2020(sink)->bt2020(src)
2795 colorimetry selection and all drivers (afaik) should support
2798 2020-02-04 09:57:42 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2800 * gst/vaapi/gstvaapipostprocutil.c:
2801 vaapipostproc: set srcpad colorimetry unconditionally
2802 We always need a srcpad colorimetry for VAAPI VPP
2804 Also, check the return value of _set_colorimetry.
2806 2020-01-30 12:34:07 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2808 * gst/vaapi/gstvaapipostprocutil.c:
2809 vaapipostproc: do not override explicit srcpad colorimetry
2810 If colorimetry has been set by a capsfilter (e.g.
2811 vaapipostproc ! video/x-raw,colorimetry=bt709) then
2812 don't try to override it. Previously, the aforementioned
2813 capsfilter will fail to negotiate if default colorimetry
2814 is not the same as the capsfilter (e.g. 4K resolutions).
2816 2020-01-30 09:37:18 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2818 * gst/vaapi/gstvaapipostproc.c:
2819 vaapipostproc: set vpp filter colorimetry
2820 Set the input and output colorimetry for vpp filter.
2822 2020-01-30 09:34:10 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2824 * gst-libs/gst/vaapi/gstvaapifilter.c:
2825 * gst-libs/gst/vaapi/gstvaapifilter.h:
2826 libs: filter: support vpp input/output color standard
2827 Add API function to allow setting the input and output vpp
2828 color standard from GstVideoColorimetry.
2830 2020-02-04 11:32:54 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2832 * gst-libs/gst/vaapi/gstvaapiutils.c:
2833 * gst-libs/gst/vaapi/gstvaapiutils.h:
2834 libs: utils: map GstVideoColorimetry to VAAPI VPP
2835 Fallback to VAProcColorStandardExplicit if there is no
2838 2020-02-04 10:27:23 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
2840 * gst-libs/gst/vaapi/gstvaapidisplay.c:
2841 * gst-libs/gst/vaapi/gstvaapidisplay.h:
2842 libs: display: add vpp color standard quirk for i965 driver
2843 The i965 does not properly report supported vpp color
2846 2020-02-02 18:04:35 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2848 * gst/vaapi/gstvaapivideobufferpool.c:
2849 vaapivideobufferpool: dmabuf implies allocator
2850 Some code can be optimized since only if the dmabuf allocator is set,
2851 the internal flag of dmabuf is TRUE, thus there's no need to evaluate
2852 the allocator address.
2854 2020-01-29 11:55:39 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2856 * gst/vaapi/gstvaapivideobufferpool.c:
2857 vaapivideobufferpool: reject configuration if allocator isn't vaapi
2858 If the requested allocator in set_config() is not a VAAPI valid one,
2859 reject the configuration, instead of lying and using a private one.
2860 This patch superseeds !254 and !24
2862 2020-01-29 11:55:28 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2864 * gst/vaapi/gstvaapivideobufferpool.c:
2865 vaapivideobufferpool: add explanation for allocator reconfig
2867 2020-01-29 11:54:38 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2869 * gst/vaapi/gstvaapivideobufferpool.c:
2870 vaapivideobufferpool: check for vaapi meta first
2871 If the configured meta doesn't request vaapi meta then it is not a
2872 vaapi buffer pool. Bail out as soon as possible.
2874 2020-01-29 11:52:38 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2876 * gst/vaapi/gstvaapivideobufferpool.c:
2877 vaapivideobufferpool: turn errors into warnings
2878 set_config() vmethod should fail gracefully, thus upstream could
2879 negotiate another pool if possible.
2880 Instead of sending error messages to the bus, let demote the level
2883 2020-01-29 11:51:47 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2885 * gst-libs/gst/vaapi/gstvaapidisplay.c:
2886 libs: display: log out vendor string when available
2887 This is useful while asking for logs to know the used driver.
2889 2020-01-27 11:49:26 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2891 * gst-libs/gst/vaapi/gstvaapisurface.c:
2892 libs: surface: initialize VASurfaceAttribExternalBuffers
2893 Initialize VASurfaceAttribExternalBuffers using compiler's syntax
2894 rather than using memset().
2896 2020-01-27 11:44:49 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2898 * gst-libs/gst/vaapi/gstvaapisurface.c:
2899 libs: surface: merge two loops into one
2900 Merge two loops into one for setting offsets and strides in the
2901 external buffer descriptor.
2903 2020-01-22 10:42:35 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2905 * gst-libs/gst/vaapi/gstvaapisurface.c:
2906 * gst-libs/gst/vaapi/gstvaapisurface.h:
2907 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
2908 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
2909 libs: surface: surfacepool: rename variable for clearity
2910 In order to be readable, the meaningless 'flags' is renamed to
2911 surface_allocation_flags, which is clearer.
2913 2020-01-27 18:40:46 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2915 * gst/vaapi/gstvaapivideobufferpool.c:
2916 vaapivideobufferpool: Reuse internal allocator is possible.
2917 Instead of creating a new allocator when upstream requests a different
2918 allocator, this patch tries to reuse the internal allocator if it was
2919 already initializated.
2920 If the stream changes, then either one will be unref and a new
2921 allocator is created.
2923 2020-01-27 18:05:14 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2925 * gst/vaapi/gstvaapivideobufferpool.c:
2926 vaapivideobufferpool: Log messages in proper category.
2927 The log messages where logged in the GstBufferPool category because
2928 the instance was not properly casted. This fix that situation.
2930 2020-01-16 11:49:21 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2932 * gst-libs/gst/vaapi/gstvaapidisplay.c:
2933 * gst-libs/gst/vaapi/gstvaapidisplay.h:
2934 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
2935 * gst/vaapi/gstvaapipluginbase.c:
2936 libs: display: driver quirks mechanism
2937 This mechanism comes from ffmpeg vaapi implementation, where they have
2939 A specific driver is identified by a substring present in the vendor
2940 string. If that substring is found, a set of bitwise flags are store.
2941 These flags can be accessed through the function
2942 gst_vaapi_display_has_driver_quirks().
2943 The purpose for this first quirks is to disable the put image try for
2944 AMD Gallium driver (see [1]).
2945 1. https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/merge_requests/72
2947 2020-01-15 23:07:29 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2949 * gst/vaapi/gstvaapipluginutil.c:
2950 plugins: renable Mesa Gallium driver
2952 2020-01-24 11:55:22 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2954 * gst/vaapi/gstvaapivideobufferpool.c:
2955 vaapivideobufferpool: validate returned meta
2956 Validate if the meta returned by gst_buffer_get_vaapi_video_meta() in
2957 the acquired buffer is not null.
2958 This situation should be very "pathological", but still it is better
2959 be safe since that meta might be used later to create a new dma
2962 2020-01-22 18:50:36 +0100 Philipp Zabel <p.zabel@pengutronix.de>
2964 * gst/vaapi/gstvaapivideobufferpool.c:
2965 vaapivideobufferpool: always update/release the underlying surface proxy
2966 gst_vaapi_video_buffer_pool_reset_buffer() is called when the sink
2967 releases the last reference on an exported DMA buffer. This should
2968 release the underlying surface proxy. To avoid releasing the wrong
2969 surface due to a stale surface proxy reference in the buffer's
2970 GstVaapiVideoMeta, always update the reference to the correct surface
2971 in gst_vaapi_video_buffer_pool_acquire_buffer().
2973 2020-01-23 16:56:44 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2975 * gst-libs/gst/vaapi/gstvaapicontext.c:
2976 * gst-libs/gst/vaapi/gstvaapicontext.h:
2977 * gst-libs/gst/vaapi/gstvaapisurface.c:
2978 * gst-libs/gst/vaapi/gstvaapisurface.h:
2979 libs: context: select vaCreateSurfaces version according attributes
2980 This commit tries to centralize the selection of vaCreateSurfaces
2981 version, instead of having fallbacks everywhere.
2982 These fallbacks are hacks, added because new drivers use the latest
2983 version of vaCreateSurfaces (with surface attributes) [1], meanwhile
2984 old drivers (or profiles as JPEG decoder in i965) might rather use the
2986 In order to select which method, there's detected hack: each config
2987 context has a list of valid formats, in the case of JPEG decoder the
2988 list only contains "rare" 4:2:2 formats (ICM3, GRAY8) which aren't
2989 handled correctly by the current gstreamer-vaapi code [2].
2990 The hack consist in identify if the format list contains an arbitrary
2991 preferred format (which is suposedly well supported by
2992 gstreamer-vaapi, mostly NV12). If no prefered colour format is found,
2993 the the old version of vaCreateSurfaces is used, and the surfaces wil
2994 be mapped into a image with their own color format.
2995 1. https://bugzilla.gnome.org/show_bug.cgi?id=797143
2996 2. https://bugzilla.gnome.org/show_bug.cgi?id=797222
2998 2020-01-21 19:35:02 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3000 * gst-libs/gst/vaapi/gstvaapicontext.c:
3001 libs: context: add debug category for context
3003 2020-01-22 20:20:30 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3005 * gst-libs/gst/vaapi/gstvaapiblend.c:
3006 libs: blend: guard VA_BLEND_GLOBAL_ALPHA
3008 2020-01-21 14:09:33 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3010 * gst/vaapi/gstvaapiencode_h264.c:
3011 vaapiencode: promote info to warning
3012 Let's notify user about using constrained-baseline instead on
3013 requested basline profile.
3015 2019-06-28 15:41:16 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
3017 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
3018 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
3019 * gst/vaapi/gstvaapidecode.c:
3020 * gst/vaapi/gstvaapidecode_props.c:
3021 * gst/vaapi/gstvaapidecode_props.h:
3022 vaapih264dec: Add a property to assume constrained-baseline
3023 When baseline-as-constrained is set, the decoder will expose support
3024 for baseline decoding and assume that the baseline content is
3025 constrained-baseline. This can be handy to decode streams in hardware
3026 that would otherwise not be possible to decode. A lot of baseline
3027 content is in fact constrained.
3029 2020-01-17 16:19:52 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3031 * gst/vaapi/gstvaapiencode_h264.c:
3032 * gst/vaapi/gstvaapipluginutil.c:
3033 vaapih264enc: accept baseline as constrained baseline compatible
3035 2020-01-17 14:24:37 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3037 * gst/vaapi/gstvaapiencode_h264.c:
3038 vaapih264enc: update level in src caps
3039 And, if downstream requests a specific level, the caps are not
3040 negotiated, because there is no mechanism right now to specify a
3041 custom level in the internal encoder.
3043 2020-01-17 17:12:53 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3045 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
3046 libs: encoder: h264: document gst_vaapi_encoder_h264_supports_avc()
3048 2020-01-17 13:38:29 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3050 * gst/vaapi/gstvaapiencode_h264.c:
3051 vaapih264enc: force byte-stream if avc isn't supported
3052 Removing the validation in gst_vaapiencode_h264_get_caps() since that
3053 ought be handled in gst_vaapiencode_h264_set_config()
3055 2020-01-08 17:37:22 +0100 Philipp Zabel <p.zabel@pengutronix.de>
3057 * gst/vaapi/gstvaapiencode_h264.c:
3058 vaaph264enc: suppress avc if the driver does not support packed headers
3059 Do not negotiate AVC output if the driver does not support it.
3061 2020-01-08 17:16:35 +0100 Philipp Zabel <p.zabel@pengutronix.de>
3063 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
3064 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
3065 libs: encoder: h264: Add gst_vaapi_encoder_h264_supports_avc()
3066 AVC output requires packed header support in the driver.
3068 2020-01-17 13:07:04 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3070 * gst/vaapi/gstvaapiencode_h264.c:
3071 vaapih264enc: intersect the new proposed caps
3072 Instead of just leave to keep the proposed caps, with the best profile
3073 in the allowed caps, is its intersected again.
3075 2020-01-17 12:58:58 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3077 * gst/vaapi/gstvaapiencode_h264.c:
3078 vaapih264enc: propose new profile caps and fixate it
3079 When the available caps doesn't intersect with the allowed caps in the
3080 pipeline, a new caps is proposed rather than just expecting to
3082 Later, the intersected caps (profile_caps) is fixated in order to
3083 extract the configuration.
3085 2020-01-17 13:18:28 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3087 * gst/vaapi/gstvaapiencode_h264.c:
3088 vaapih264enc: common fail for gst_vaapiencode_h264_set_config()
3089 Add a common fail code path for gst_vaapiencode_h264_set_config().
3091 2020-01-17 12:54:21 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3093 * gst-libs/gst/vaapi/gstvaapiencoder.c:
3094 libs: encoder: unref formats array if none
3095 The formats array is always created, in order to keep the logic and
3096 to avoid broken caps, if this formats array doesn't contain any
3097 elements, it has to be unref and the function should return NULL.
3099 2020-01-17 14:22:48 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3101 * gst/vaapi/gstvaapiencode_h264.c:
3102 vaapih264enc: fix log message
3103 Before the log wasn't processed because wrong instance pointer.
3105 2019-06-22 00:44:25 +0800 He Junyan <junyan.he@hotmail.com>
3107 * gst-libs/gst/vaapi/gstvaapiencoder.c:
3108 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
3109 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
3110 * gst/vaapi/gstvaapiencode_h265.c:
3111 plugin: encode: Add H265 main-444 profile.
3112 Expose the main-444 profile to h265enc caps, when the upstream
3113 chooses to use VUYA as input, we choose main 4:4:4 profile to encode
3116 2020-01-15 19:36:00 +0800 He Junyan <junyan.he@hotmail.com>
3118 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
3119 libs: encoder: h265: Consider main-444 profile when encoding.
3120 Add support of main-444 profile for parameter setting and packed header
3123 2020-01-15 23:04:08 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3125 * gst-libs/gst/vaapi/gstvaapiencoder.c:
3126 * gst-libs/gst/vaapi/gstvaapiencoder.h:
3127 * gst-libs/gst/vaapi/gstvaapiprofilecaps.c:
3128 * gst-libs/gst/vaapi/gstvaapiprofilecaps.h:
3129 * gst/vaapi/gstvaapiencode.c:
3130 vaapiencode: DMABuf only if PRIME is available
3131 Add DMABuf capsfeature in encoders' allowed sinkcaps only if PRIME
3132 memory type is available in the VA surface attributes of codec
3135 2020-01-16 09:14:30 +0800 Haihao Xiang <haihao.xiang@intel.com>
3137 * hooks/pre-commit.hook:
3138 Add hooks/pre-commit.hook
3139 meson.build in gstreamer-vaapi requires hooks/pre-commit.hook
3140 Copied and pasted pre-commit.hook from other gstreamer modules to make
3141 sure gstreamer-vaapi follows the same code style
3143 2019-12-26 16:45:51 +0800 He Junyan <junyan.he@hotmail.com>
3145 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
3146 libs: encoder: h265: Set encoder paramters base on entrypoint.
3147 When the tune is NONE, we now can choose entrypoint freely. So the
3148 GST_VAAPI_ENCODER_TUNE macro may not return the correct current
3150 We also delay CTU size calculation after entrypoint has been decided.
3152 2019-12-28 19:18:12 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3154 * gst-libs/gst/vaapi/gstvaapicodedbufferproxy.c:
3155 * gst-libs/gst/vaapi/gstvaapicodedbufferproxy.h:
3156 * gst-libs/gst/vaapi/gstvaapicodedbufferproxy_priv.h:
3157 * gst-libs/gst/vaapi/gstvaapicontext.c:
3158 * gst-libs/gst/vaapi/gstvaapicontext.h:
3159 * gst-libs/gst/vaapi/gstvaapidisplay.c:
3160 * gst-libs/gst/vaapi/gstvaapiencoder.c:
3161 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
3162 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.h:
3163 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
3164 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
3165 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
3166 * gst-libs/gst/vaapi/gstvaapifei_objects.c:
3167 * gst-libs/gst/vaapi/gstvaapifei_objects.h:
3168 * gst-libs/gst/vaapi/gstvaapifei_objects_priv.h:
3169 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
3170 * gst-libs/gst/vaapi/gstvaapifeienc_h264.h:
3171 * gst-libs/gst/vaapi/gstvaapifeipak_h264.c:
3172 * gst-libs/gst/vaapi/gstvaapifeipak_h264.h:
3173 * gst-libs/gst/vaapi/gstvaapifeiutils_h264.c:
3174 * gst-libs/gst/vaapi/gstvaapifeiutils_h264.h:
3175 * gst-libs/gst/vaapi/gstvaapiprofile.c:
3176 * gst-libs/gst/vaapi/gstvaapiprofile.h:
3177 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
3178 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
3179 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
3180 * gst-libs/gst/vaapi/meson.build:
3181 * gst/vaapi/gstvaapi.c:
3182 * gst/vaapi/gstvaapiencode.c:
3183 * gst/vaapi/gstvaapiencode.h:
3184 * gst/vaapi/gstvaapiencode_h264_fei.c:
3185 * gst/vaapi/gstvaapiencode_h264_fei.h:
3186 * gst/vaapi/gstvaapifeivideometa.c:
3187 * gst/vaapi/gstvaapifeivideometa.h:
3188 * gst/vaapi/meson.build:
3190 * tests/internal/meson.build:
3191 * tests/internal/test-fei-enc-in.c:
3192 * tests/internal/test-fei-enc-out.c:
3193 Remove all FEI related
3194 FEI encoders are not actively mantained neither tested, and it is
3195 using infrastructure that is changing and FEI is stopping this
3197 Also it is required to rethink how FEI can be used in GStreamer.
3199 2020-01-14 11:17:49 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3201 * gst/vaapi/gstvaapioverlay.c:
3202 vaapioverlay: ensure sinkpad has current buffer
3203 Use the gst_video_aggregator_pad_has_current_buffer API
3204 to check if the current sinkpad has a queued buffer before
3205 attempting to obtain a input buffer from the base plugin.
3206 If the sinkpad does not have a current buffer, then it is
3207 either not producing them yet (e.g. current time < sinkpad
3208 start time) or it has reached EOS.
3209 Previously, we only handled EOS case.
3211 gst-launch-1.0 videotestsrc num-buffers=100 \
3212 ! vaapipostproc ! vaapioverlay name=overlay \
3213 ! vaapisink videotestsrc timestamp-offset=1000000000 \
3214 num-buffers=100 ! video/x-raw,width=160,height=120 \
3217 2020-01-14 18:57:31 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3219 * gst/vaapi/gstvaapioverlay.c:
3220 vaapioverlay: unroll the recursive call
3221 Recursive functions are elegant but dangerous since they might
3222 overflow the stack. It is better to turn them into a list tranversal
3223 if possible, as this case.
3225 2020-01-14 18:46:49 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3227 * gst-libs/gst/vaapi/gstvaapiblend.c:
3228 * gst/vaapi/gstvaapioverlay.c:
3229 vaapioverlay: add minimal documentation
3231 2020-01-14 18:25:11 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3233 * gst-libs/gst/vaapi/gstvaapiblend.c:
3234 * gst-libs/gst/vaapi/gstvaapiblend.h:
3235 * gst/vaapi/gstvaapioverlay.c:
3236 libs: blend: simplify generator API
3237 Instead of using a parent structure that has to be derived by API
3238 consumers, this change propse a simplification by using the common
3239 pattern of GTK of passing a function pointer and user data which will
3240 be passed as its parameter. That user data contains the state and the
3241 function will be called to update that state.
3243 2020-01-10 10:14:38 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3245 * gst-libs/gst/vaapi/gstvaapiblend.c:
3246 * gst-libs/gst/vaapi/gstvaapiblend.h:
3247 libs: blend: remove begin/render/end API
3248 This API was risky and is superseded by the surface
3249 generator (process) API.
3252 2020-01-10 10:12:36 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3254 * gst/vaapi/gstvaapioverlay.c:
3255 vaapioverlay: use blend surface generator API
3258 2020-01-10 09:54:30 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3260 * gst-libs/gst/vaapi/gstvaapiblend.c:
3261 * gst-libs/gst/vaapi/gstvaapiblend.h:
3262 libs: blend: add surface generator API
3263 This new API allows the user to call a single method (process)
3264 which handles the [display] lock/unlock logic internally for
3266 This API supersedes the risky begin, render, end API.
3267 It eliminates the need for the user to call a lock method
3268 (process_begin) before processing the input buffers
3269 (process_render) and calling an unlock method (process_end)
3273 2019-12-23 14:29:08 +0800 He Junyan <junyan.he@hotmail.com>
3275 * gst-libs/gst/vaapi/gstvaapiencoder.c:
3276 * gst-libs/gst/vaapi/gstvaapiencoder.h:
3277 * gst/vaapi/gstvaapiencode.c:
3278 * gst/vaapi/gstvaapiencode.h:
3279 * gst/vaapi/gstvaapiencode_h264.c:
3280 * gst/vaapi/gstvaapiencode_h265.c:
3281 plugin: encode: List all possible profiles to detect input formats.
3282 The current get_profile just return one possible profile for the encode,
3283 which is not enough. For example, if we want to support HEVC 4:4:4
3284 profile, the input of encode should be VYUA rather than NV12 in HEVC
3285 main profile. So the command line:
3286 gst-launch-1.0 videotestsrc num-buffers=200 ! capsfilter \
3287 caps=video/x-raw,format=VUYA,width=800,height=600 ! vaapih265enc \
3288 tune=low-power init-qp=30 ! fakesink
3289 can not work because vaapih265enc just report NV12 in sink caps, we need
3290 to specify the profile obviously like:
3291 gst-launch-1.0 videotestsrc num-buffers=200 ! capsfilter \
3292 caps=video/x-raw,format=VUYA,width=800,height=600 ! vaapih265enc \
3293 tune=low-power init-qp=30 ! capsfilter caps=video/x-h265, \
3294 profile=main-444 ! fakesink
3295 The encode should have the ability to choose the profile based on input
3296 format automatically. If the input video format is VUYA, the main-444
3297 profile should be auto choosed.
3298 We modify to let get_allowed_profiles of each encode sub class to return
3299 an array of all supported profiles based on downstream's allowed caps, or
3300 return NULL if no valid profiles specified by downstream.
3301 If no allowed profiles found, all profiles which belong to the current
3302 encoder's codec will be the candidates.
3303 The function gst_vaapi_encoder_get_surface_attributes collects the surface's
3304 attributes for that profile list we just get.
3305 So for this case, both NV12 and VUYA should be returned.
3306 TODO: some codec like VP9, need to implement the get_profile() function.
3308 2020-01-08 15:07:36 +0800 He Junyan <junyan.he@hotmail.com>
3310 * gst/vaapi/gstvaapipluginutil.c:
3311 * gst/vaapi/gstvaapipluginutil.h:
3312 plugin: util: add helper function to detect profiles in caps.
3314 2020-01-08 15:04:18 +0800 He Junyan <junyan.he@hotmail.com>
3316 * gst-libs/gst/vaapi/gstvaapiencoder.c:
3317 * gst-libs/gst/vaapi/gstvaapiencoder.h:
3318 libs: encoder: add a helper function to get all supported profiles
3320 2020-01-13 15:34:54 +0900 Seungha Yang <seungha.yang@navercorp.com>
3322 * gst-libs/gst/vaapi/gstvaapidecoder.c:
3323 libs: decoder: Don't unref null object
3324 ** (gst-launch-1.0:9789): CRITICAL **: 15:29:09.330:
3325 gst_vaapi_context_unref: assertion 'context != NULL' failed
3327 2020-01-10 09:26:44 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3329 * gst/vaapi/gstvaapioverlay.c:
3330 plugins: overlay: use proper NULL check on double pointer
3331 Check the address of the variable is not NULL,
3332 not the address of the pointer.
3334 2020-01-08 23:42:21 +0800 He Junyan <junyan.he@hotmail.com>
3336 * gst-libs/gst/vaapi/gstvaapicodedbuffer_priv.h:
3337 libs: codedbuf: delete a useless field.
3338 The context field in GstVaapiCodedBuffer is not inited correctly
3339 and is never used, just delete it.
3341 2019-12-29 17:57:52 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3343 * gst/vaapi/gstvaapipluginutil.c:
3344 plugins: add iHD driver in whitelist
3346 2020-01-02 21:02:40 +0800 He Junyan <junyan.he@hotmail.com>
3348 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
3349 libs: encoder: modify 265 SPS header's profile compatibility flag.
3350 Make the SPS profile compatibility flags more precisely conform to
3353 2020-01-06 19:39:06 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3355 * tests/check/elements/vaapioverlay.c:
3356 test: vaapioverlay: bail test if not available
3357 vaapioverlay is only registered if the VA driver support the blend
3359 This patch only executes the test if vaapioverlay is available,
3360 otherwise the test is bail out without raising an error.
3362 2020-01-06 14:53:28 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3364 * tests/check/elements/vaapioverlay.c:
3365 tests: vaapioverlay: force drm backend
3367 2019-12-22 17:32:19 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3369 * gst-libs/gst/vaapi/gstvaapiblend.c:
3370 libs: blend: update to new mini-object API
3372 2019-11-19 13:48:22 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3374 * tests/check/elements/vaapioverlay.c:
3375 * tests/check/meson.build:
3376 tests: check: add basic vaapioverlay test
3377 Add test_overlay_position test to verify sink_1 input
3378 is overlayed onto sink_0 input at the appropriate
3381 2019-11-14 12:03:57 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3383 * gst/vaapi/gstvaapi.c:
3384 * gst/vaapi/gstvaapioverlay.c:
3385 * gst/vaapi/gstvaapioverlay.h:
3386 * gst/vaapi/meson.build:
3387 plugins: add vaapioverlay plugin
3388 A plugin similar to the base compositor element but
3389 uses VA-API VPP blend functions to accelerate the
3390 overlay/compositing.
3392 gst-launch-1.0 -vf videotestsrc ! vaapipostproc \
3393 ! tee name=testsrc ! queue \
3394 ! vaapioverlay sink_1::xpos=300 sink_1::alpha=0.75 \
3395 name=overlay ! vaapisink testsrc. ! queue ! overlay.
3397 2019-11-14 12:02:19 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3399 * gst-libs/gst/vaapi/gstvaapiblend.c:
3400 * gst-libs/gst/vaapi/gstvaapiblend.h:
3401 * gst-libs/gst/vaapi/meson.build:
3402 libs: add a vaapi blend class
3403 Support for the VA-API VPP blend functions.
3405 2019-11-14 11:54:59 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3407 * gst/vaapi/gstvaapipluginbase.c:
3408 * gst/vaapi/gstvaapipluginbase.h:
3409 plugins: base: add GstVideoAggregator subclass support
3411 2020-01-05 19:32:16 +0800 He Junyan <junyan.he@hotmail.com>
3413 * gst-libs/gst/vaapi/gstvaapiimage.c:
3414 libs: image: init all image fields correctly.
3416 2020-01-06 17:41:53 +0100 Stéphane Cerveau <scerveau@collabora.com>
3418 * gst/vaapi/gstvaapipostproc.c:
3419 doc: fix pipeline typo in vaapipostproc
3421 2020-01-02 21:11:44 +0800 He Junyan <junyan.he@hotmail.com>
3423 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
3424 libs: encoder: modify 265 VPS header fields.
3425 vps_base_layer_internal_flag and vps_base_layer_available_flag
3426 have been clearly defined now.
3428 2020-01-01 19:54:13 +0800 He Junyan <junyan.he@hotmail.com>
3430 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
3431 libs: display: fix a resource leak in X11 pixmap format.
3433 2020-01-02 18:00:21 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3435 * gst-libs/gst/vaapi/gstvaapiprofilecaps.c:
3436 * gst-libs/gst/vaapi/gstvaapiprofilecaps.h:
3437 libs: utils: delete useless gst_vaapi_profile_caps_append_encoder.
3439 2019-12-30 14:09:17 +0800 He Junyan <junyan.he@hotmail.com>
3441 * gst-libs/gst/vaapi/gstvaapiencoder.c:
3442 * gst-libs/gst/vaapi/gstvaapiencoder.h:
3443 * gst/vaapi/gstvaapiencode.c:
3444 libs: encoder: get surfaces resolution the same time with formats.
3445 We can get all the information about the video format at one shot
3446 when we create the test context for getting the supported formats.
3447 The current way to get the width and height ranges are inefficient,
3448 since it calls the function gst_vaapi_profile_caps_append_encoder()
3449 and it creates another temporal context to detect the resolution
3451 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3453 2019-12-28 17:42:55 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3455 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
3456 libs: encoder: vp9: fix code style
3458 2019-12-16 23:19:46 +0800 He Junyan <junyan.he@hotmail.com>
3460 * gst-libs/gst/vaapi/gstvaapiencoder.c:
3461 * gst-libs/gst/vaapi/gstvaapiencoder.h:
3462 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
3463 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
3464 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
3465 libs: encoder: set entrypoint based on tune automatically
3466 Some profile, such as H265_MAIN_444 on new Intel platform, may only
3467 support ENTRYPOINT_SLICE_ENCODE_LP entrypoint. This leads two
3469 1. We need to specify the tune mode like `vaapih265enc tune=low-power`
3470 every time when we need to use this kind of profile. Or we can not
3471 create the encoder context successfully.
3472 2. More seriously, we set the entrypoint to a fixed value in
3473 init_context_info() and so the create_test_context_config() can not
3474 create the test context for these profile and can not get the
3475 supported video formats, either.
3476 We now change the entrypoint setting based on the tune option of the
3477 encoder. If no tune property provided, we just choose the first
3478 available entrypoint.
3480 2019-12-16 23:19:46 +0800 He Junyan <junyan.he@hotmail.com>
3482 * gst-libs/gst/vaapi/gstvaapiencoder.c:
3483 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
3484 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
3485 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
3486 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
3487 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
3488 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
3489 libs: encoder: set context info profile by encoder
3490 Instead of init_context_info() setting the passed profile, it is
3491 assumed that it has to be set by each encoder.
3492 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3494 2019-12-27 18:49:02 +0100 He Junyan <junyan.he@hotmail.com>
3496 * gst-libs/gst/vaapi/gstvaapicontext.c:
3497 * gst-libs/gst/vaapi/gstvaapiprofile.h:
3498 libs: context: add invalid entrypoint symbol
3499 The symbol GST_VAAPI_ENTRYPOINT_INVALID is just a representation of
3500 zero, which was already used as an invalid value tacitly. This patch
3501 only makes it explicit.
3502 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3504 2019-12-29 01:13:29 +0800 He Junyan <junyan.he@hotmail.com>
3506 * gst-libs/gst/vaapi/gstvaapipixmap_x11.c:
3507 libs: pixmap: Fix a pixmap creation crash.
3508 We use GST_VAAPI_OBJECT_NATIVE_DISPLAY with wrong parameter for x11
3509 pixmap creation, which causes crash if we run the internal test case
3511 test-decode --pixmap
3513 2019-12-22 14:35:18 +0800 He Junyan <junyan.he@hotmail.com>
3515 * gst/vaapi/gstvaapiencode.c:
3516 plugin: encode: set allowed_sinkpad_caps to empty.
3517 We now set encode->allowed_sinkpad_caps to NULL if we fail to get
3518 surfaces formats. This causes two problem:
3519 1. gst_video_encoder_proxy_getcaps use NULL as its caps parameter,
3520 which changes its behavior. It will use encode's sinkpad template
3521 rather than empty caps to do the clip job. So even if we fail to set
3522 allowed_sinkpad_caps, gst_video_encoder_proxy_getcaps can still return
3524 2. We should just set the allowed_sinkpad_caps once. The NULL point
3525 make the ensure_allowed_sinkpad_caps function works again and again.
3527 2019-12-22 15:22:57 +0800 He Junyan <junyan.he@hotmail.com>
3529 * gst-libs/gst/vaapi/gstvaapiencoder.c:
3530 libs: encoder: Add NULL pointer check for context when finalize.
3531 Context may be NULL if pipeline fail in early stage, and the
3532 ensure_context will not be called. Need to add a pointer protection
3535 2019-12-20 06:38:42 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3537 * gst/vaapi/gstvaapipluginbase.c:
3538 plugins: base: do not reset can_dmabuf
3539 Don't reset the can_dmabuf field. This restores the
3540 close/reset logic that existed prior to commit
3541 ca2942176b5632e07eebac23336954f9aaf1cb26 in regards to
3543 Plugins only call gst_vaapi_plugin_base_set_srcpad_can_dmabuf
3544 once during startup, but may need to reset the other private
3545 fields multiple times during negotiation. Thus, can_dmabuf
3546 should be exempt from the resets.
3549 2019-12-06 00:21:12 +0800 He Junyan <junyan.he@hotmail.com>
3551 * gst/vaapi/gstvaapiencode.c:
3552 plugin: encode: Refine encode's sink caps.
3553 The old manner to get the encode's sink caps is not correct.
3554 Such as 264 encode, it gets:
3555 video/x-raw(memory:VASurface),
3556 format=(string){ ENCODED, NV12, I420, YV12, YUY2, UYVY, Y210,
3557 P010_10LE, AYUV, Y410, Y444 }, width=(int)[ 32, 4096 ],
3558 height=(int)[ 32, 4096 ], framerate=(fraction)[ 0/1, 2147483647/1 ];
3559 video/x-raw(memory:DMABuf), format=(string){ I420, YV12, RGBA },
3560 width=(int)[ 32, 4096 ], height=(int)[ 32, 4096 ],
3561 framerate=(fraction)[ 0/1, 2147483647/1 ];
3562 video/x-raw, format=(string){ NV12 }, width=(int)[ 32, 4096 ],
3563 height=(int)[ 32, 4096 ], framerate=(fraction)[ 0/1, 2147483647/1 ]
3564 where the formats for memory:VASurface and memory:DMABuf are superfluous.
3565 All the "I420, YV12, YUY2, UYVY, Y210, RGBA" can not be really used as
3566 input format for encoder.
3568 video/x-raw, format=(string){ NV12 }, width=(int)[ 32, 4096 ],
3569 height=(int)[ 32, 4096 ], framerate=(fraction)[ 0/1, 2147483647/1 ];
3570 video/x-raw(memory:VASurface), format=(string){ NV12 },
3571 width=(int)[ 32, 4096 ], height=(int)[ 32, 4096 ],
3572 framerate=(fraction)[ 0/1, 2147483647/1 ]; video/x-raw(memory:DMABuf),
3573 format=(string){ NV12 }, width=(int)[ 32, 4096 ],
3574 height=(int)[ 32, 4096 ], framerate=(fraction)[ 0/1, 2147483647/1 ]
3575 as the correct result.
3577 2019-12-20 08:37:11 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3579 * gst-libs/gst/vaapi/gstvaapidisplay.c:
3580 * gst-libs/gst/vaapi/gstvaapiprofile.h:
3581 libs: display: code clean up
3583 2019-12-12 21:34:21 +0800 He Junyan <junyan.he@hotmail.com>
3585 * gst-libs/gst/vaapi/gstvaapidisplay.c:
3586 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
3587 * gst-libs/gst/vaapi/gstvaapiprofile.h:
3588 libs: display: refine the profile/entrypoint map.
3589 The old way make the one config for each profile/entrypoint pair,
3590 which is not very convenient for description the relationship
3591 between them. One profile may contain more than one entrypoints
3592 to within it, so a set like data structure should be more suitable.
3594 2019-12-19 14:19:10 +0100 He Junyan <junyan.he@hotmail.com>
3596 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
3597 * gst-libs/gst/vaapi/gstvaapibufferproxy_priv.h:
3598 * gst-libs/gst/vaapi/gstvaapicontext.c:
3599 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
3600 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
3601 * gst-libs/gst/vaapi/gstvaapifilter.c:
3602 * gst-libs/gst/vaapi/gstvaapipixmap_x11.c:
3603 * gst-libs/gst/vaapi/gstvaapisurface.c:
3604 * gst-libs/gst/vaapi/gstvaapisurface.h:
3605 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
3606 * gst-libs/gst/vaapi/gstvaapisurface_egl.c:
3607 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
3608 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
3609 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
3610 * gst-libs/gst/vaapi/gstvaapitexture_egl.c:
3611 * gst-libs/gst/vaapi/gstvaapitexture_glx.c:
3612 * gst-libs/gst/vaapi/gstvaapivideopool.c:
3613 * gst-libs/gst/vaapi/gstvaapiwindow.c:
3614 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
3615 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
3616 * gst/vaapi/gstvaapipluginbase.c:
3617 * gst/vaapi/gstvaapivideomemory.c:
3618 * gst/vaapi/gstvaapivideometa.c:
3619 * gst/vaapi/gstvaapivideometa_texture.c:
3620 * tests/internal/image.c:
3621 * tests/internal/test-filter.c:
3622 * tests/internal/test-surfaces.c:
3623 * tests/internal/test-windows.c:
3624 libs: surface: port to GstMiniObject
3625 GstVaapiMiniObject and GstVaapiObject are deprecated.
3626 This is the first step to remove them by porting GstVaapiSurface as
3627 a GstMiniBuffer descendant.
3628 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3630 2019-12-19 18:26:10 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3632 * gst-libs/gst/vaapi/gstvaapisurface.c:
3633 libs: surface: rename create function names to init
3634 There are several internal functions with 'create' name, but they
3635 don't create any new structure, but rather it initializes that
3636 structure. Renaming those function to reflect better their purpose.
3638 2019-12-19 14:17:34 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3640 * gst-libs/gst/vaapi/gstvaapisurface.c:
3641 libs: surface: use macro accessors
3643 2019-12-19 13:46:09 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3645 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
3646 libs: surface: fix internal documentation
3648 2019-12-18 18:00:49 +0100 He Junyan <junyan.he@hotmail.com>
3650 * gst-libs/gst/vaapi/gstvaapicodedbuffer.c:
3651 * gst-libs/gst/vaapi/gstvaapicodedbuffer.h:
3652 * gst-libs/gst/vaapi/gstvaapicodedbuffer_priv.h:
3653 * gst-libs/gst/vaapi/gstvaapicodedbufferproxy.c:
3654 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
3655 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
3656 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
3657 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
3658 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
3659 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
3660 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
3661 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
3662 * gst-libs/gst/vaapi/gstvaapifeipak_h264.c:
3663 * gst-libs/gst/vaapi/gstvaapivideopool.c:
3664 libs: codedbuffer: port to GstMiniObject
3665 GstVaapiMiniObject and GstVaapiObject are deprecated.
3666 This is the first step to remove them by porting GstVaapiCodedBuffer
3667 as a GstMiniBuffer descendant.
3668 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3670 2019-12-18 12:57:01 +0100 He Junyan <junyan.he@hotmail.com>
3672 * gst-libs/gst/vaapi/gstvaapiimage.c:
3673 * gst-libs/gst/vaapi/gstvaapiimage.h:
3674 * gst-libs/gst/vaapi/gstvaapiimage_priv.h:
3675 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
3676 * gst-libs/gst/vaapi/gstvaapisurface.c:
3677 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
3678 * gst-libs/gst/vaapi/gstvaapivideopool.c:
3679 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
3680 * gst/vaapi/gstvaapipluginbase.c:
3681 * gst/vaapi/gstvaapivideomemory.c:
3682 * gst/vaapi/gstvaapivideometa.c:
3683 * tests/internal/image.c:
3684 * tests/internal/simple-encoder.c:
3685 * tests/internal/test-fei-enc-in.c:
3686 * tests/internal/test-filter.c:
3687 * tests/internal/test-windows.c:
3688 libs: image: port to GstMiniObject base class
3689 GstVaapiMiniObject and GstVaapiObject are deprecrated. This is the
3690 first step to remove them, by porting GstVaapiImage as a
3692 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3694 2019-09-24 01:01:22 +0800 He Junyan <junyan.he@hotmail.com>
3696 * gst-libs/gst/vaapi/gstvaapicodedbuffer.c:
3697 * gst-libs/gst/vaapi/gstvaapicodedbufferpool.c:
3698 * gst-libs/gst/vaapi/gstvaapicontext.c:
3699 * gst-libs/gst/vaapi/gstvaapicontext.h:
3700 * gst-libs/gst/vaapi/gstvaapidecoder.c:
3701 * gst-libs/gst/vaapi/gstvaapiencoder.c:
3702 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
3703 * gst-libs/gst/vaapi/gstvaapiprofilecaps.c:
3704 libs: context: port to a plain C structure
3705 The GstVaapiMiniObject is obsolete and we need to replace it. This
3706 patch turns GstVaapiContext into a plain C structure with its own
3707 reference counting mechanism.
3708 Also this patch removes unused overlays attributes.
3709 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3711 2019-12-18 00:40:58 +0800 He Junyan <junyan.he@hotmail.com>
3713 * gst/vaapi/gstvaapiencode_h264.c:
3714 * gst/vaapi/gstvaapiencode_h264_fei.c:
3715 * gst/vaapi/gstvaapiencode_h265.c:
3716 * gst/vaapi/gstvaapiencode_jpeg.c:
3717 * gst/vaapi/gstvaapiencode_mpeg2.c:
3718 * gst/vaapi/gstvaapiencode_vp8.c:
3719 * gst/vaapi/gstvaapiencode_vp9.c:
3720 plugin: encode: change the dmabuf caps to all supported formats.
3721 The encode's dmabuf caps definition is obsolete, it can support
3722 more formats now. Re-define it to include all supported formats
3723 in video format map.
3725 2019-12-17 17:09:37 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3727 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
3728 libs: encoder: h264fei: remove unnecessary check
3729 Issue detected by Coverity
3730 `info_to_pack.h264_slice_header` is always allocated by
3731 gst_vaapi_feipak_h264_encode(), thus checking it to free it afterwards
3732 in doesn't make much sense. But it requires to be free on the error
3734 There may be a null pointer dereference, or else the comparison
3735 against null is unnecessary.
3736 In gst_vaapi_encoder_h264_fei_encode: All paths that lead to this null
3737 pointer comparison already dereference the pointer earlier
3739 2019-12-17 17:05:22 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3741 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
3742 libs: encoder: h264fei: remove unnecessary assert
3743 Issue detected by Coverity
3744 An unsigned value can never be negative, so this test will always
3745 evaluate the same way.
3746 In add_slice_headers: An unsigned value can never be less than 0
3748 2019-12-17 16:57:41 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3750 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
3751 libs: encoder: h264fei: remove unnecessary check
3752 Issue detected by Coverity
3753 There may be a null pointer dereference, or else the comparison
3754 against null is unnecessary.
3755 In gst_vaapi_encoder_h264_fei_encode: All paths that lead to this null
3756 pointer comparison already dereference the pointer earlier
3758 2019-12-17 16:49:47 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3760 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
3761 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
3762 libs: encoder: h264fei: don't free memory on stack
3763 Issue detected by Coverity
3764 `info_to_pak` variable in gst_vaapi_encoder_h264_fei_encode() is
3765 declared in the stack, but it is free in
3766 gst_vaapi_feienc_h264_encode() as if declared on the heap.
3767 This patch initializes the structure and removes the free.
3768 A non-heap pointer is placed on the free list, likely causing a crash
3770 In gst_vaapi_encoder_h264_fei_encode: Free of an address-of
3771 expression, which can never be heap allocated.
3773 2019-12-17 13:22:12 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3775 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
3776 libs: encoder: h264fei: fix surface leak
3777 Issue detected by Coverity
3778 If the FEI mode is not handled the created resources should be
3779 released and return and error code.
3780 The system resource will not be reclaimed and reused, reducing the
3781 future availability of the resource.
3782 In gst_vaapi_encoder_h264_fei_encode: Leak of memory or pointers to
3785 2019-12-17 13:09:58 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3787 * tests/check/elements/vaapipostproc.c:
3788 tests: check return calling of gst_navigation_event_parse.*
3789 This issue was detected by Coverity.
3790 If the function returns an error value, the error value may be mistaken
3792 In cb_mouse_event: Value returned from a function is not checked for
3793 errors before being used
3795 2019-12-16 16:25:02 +0800 He Junyan <junyan.he@hotmail.com>
3797 * gst/vaapi/gstvaapiencode_h264.c:
3798 * gst/vaapi/gstvaapiencode_h264_fei.c:
3799 * gst/vaapi/gstvaapiencode_h265.c:
3800 * gst/vaapi/gstvaapiencode_jpeg.c:
3801 * gst/vaapi/gstvaapiencode_mpeg2.c:
3802 * gst/vaapi/gstvaapiencode_vp8.c:
3803 * gst/vaapi/gstvaapiencode_vp9.c:
3804 plugin: encode: set sink's raw caps to GST_VAAPI_FORMATS_ALL.
3805 Then encode plugin just supports raw formats declared in vaapi video
3806 format map. This modification makes the template caps more precise.
3808 2019-11-14 11:13:51 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3810 * gst/vaapi/gstvaapipluginbase.c:
3811 plugins: base: add GstPad param to internal helper functions
3812 The base plugin public API function implementations determine
3813 which pad should be passed to the internal helper functions.
3814 Currently, only the base plugin static sinkpad and static
3815 srcpad are supported/used. However, this change enables future
3816 API functions to be added that can accept a pad (i.e. request pad)
3817 from an element subclass (e.g. a GstVideoAggregator subclass).
3819 2019-11-12 12:21:52 -0800 U. Artie Eoff <ullysses.a.eoff@intel.com>
3821 * gst/vaapi/gstvaapipluginbase.c:
3822 * gst/vaapi/gstvaapipluginbase.h:
3823 plugins: base: manage pad-specific data in a single struct
3824 Define a struct (GstVaapiPadPrivate) to encapsulate the
3825 pad-specific data (i.e. buffer pool, allocator, info,
3827 Add an interface to retrieve the data struct for a given
3829 Finally, update the base plugin to use the data struct
3830 throughout the implementation.
3831 This will enable us to easily extend the base plugin in the
3832 future to allow for N-to-1 pad subclasses (e.g. overlay/
3835 2019-10-29 15:13:44 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
3837 * gst/vaapi/gstvaapidecode.c:
3838 * gst/vaapi/gstvaapiencode.c:
3839 * gst/vaapi/gstvaapipluginbase.h:
3840 * gst/vaapi/gstvaapipostproc.c:
3841 plugins: use plugin base macros to access pad specific data
3842 Don't access base struct fields directly since the underlying
3843 definition can change. Instead, use the accessor macros.
3845 2019-12-03 00:52:45 +0800 He Junyan <junyan.he@hotmail.com>
3847 * gst/vaapi/gstvaapidecode.c:
3848 libs: decoder: Modify decode src's template raw formats
3849 We do not need to maintain a standalone list of decoder's output
3850 template for raw formats and that is easy to make mistake(for
3851 example, the AYVU is wrong in that list, should be VUYA).
3852 Just use GST_VAAPI_FORMATS_ALL to replace the raw formats list for
3855 2019-12-11 14:11:13 +0800 He Junyan <junyan.he@hotmail.com>
3857 * gst/vaapi/gstvaapipostproc.c:
3858 libs: postproc: Modify src/sink template raw formats
3859 We need to provide more precise template caps for postproc's src
3860 and sink pads. The GST_VIDEO_FORMATS_ALL make all video formats
3861 available which are really superfluous.
3863 2019-12-10 18:40:42 -0300 Thibault Saunier <tsaunier@igalia.com>
3865 * tests/check/elements/vaapipostproc.c:
3866 Do not mix declaration and code
3868 2019-11-03 17:59:01 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3870 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
3871 libs: decoder: h265: skip all pictures prior the first I-frame
3872 Don't try to decode until the first I-frame is received within the
3873 currently active sequence. i965 H265 decoder don't show any artifact
3877 2019-11-27 01:44:05 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
3879 * gst-libs/gst/vaapi/video-format.c:
3880 libs: video-format: remove dead code
3882 2019-10-31 00:59:34 +0800 He Junyan <junyan.he@hotmail.com>
3884 * gst-libs/gst/vaapi/video-format.c:
3885 * gst-libs/gst/vaapi/video-format.h:
3886 * gst/vaapi/gstvaapipluginutil.h:
3887 libs: video-format: add GST_VAAPI_FORMATS_ALL
3888 GST_VAAPI_FORMATS_ALL collects all declared formats in video-format
3889 as a caps template string, and make them available in caps with
3890 memory:VASurface feature.
3893 2019-11-06 22:37:12 +0800 He Junyan <junyan.he@hotmail.com>
3895 * gst-libs/gst/vaapi/video-format.c:
3896 * tests/internal/test-windows.c:
3897 libs: video-format: change GST_VIDEO_FORMAT_AYUV to VUYA.
3898 We only support VUYA format in gst vaapi now, need to correct
3901 2019-10-09 15:13:09 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
3903 * tests/check/elements/vaapipostproc.c:
3904 tests: check: vaapipostproc test_orientation_mouse_events
3905 Test that vaapipostproc properly translates mouse events
3906 when using video-direction (orientation).
3908 2019-10-09 10:11:54 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
3910 * tests/check/elements/vaapipostproc.c:
3911 * tests/check/meson.build:
3912 tests: check: vaapipostproc test_crop_mouse_events
3913 Test that vaapipostproc properly translates mouse events
3916 2019-10-08 12:20:26 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
3919 * meson_options.txt:
3920 * tests/check/elements/vaapipostproc.c:
3921 * tests/check/meson.build:
3922 * tests/meson.build:
3923 tests: check: initial unit test support
3924 Add minimal unit test toolchain files and a simple
3925 vaapipostproc unit test.
3927 2019-10-08 12:19:06 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
3930 * tests/examples/meson.build:
3931 * tests/examples/test-roi.c:
3932 * tests/examples/test-vaapicontext.c:
3933 * tests/examples/test-vaapipostproc.c:
3934 * tests/examples/test-vaapisink.c:
3935 * tests/internal/codec.c:
3936 * tests/internal/codec.h:
3937 * tests/internal/decoder.c:
3938 * tests/internal/decoder.h:
3939 * tests/internal/image.c:
3940 * tests/internal/image.h:
3941 * tests/internal/meson.build:
3942 * tests/internal/output.c:
3943 * tests/internal/output.h:
3944 * tests/internal/simple-decoder.c:
3945 * tests/internal/simple-encoder.c:
3946 * tests/internal/test-decode.c:
3947 * tests/internal/test-decode.h:
3948 * tests/internal/test-display.c:
3949 * tests/internal/test-fei-enc-in.c:
3950 * tests/internal/test-fei-enc-out.c:
3951 * tests/internal/test-filter.c:
3952 * tests/internal/test-h264.c:
3953 * tests/internal/test-h264.h:
3954 * tests/internal/test-jpeg.c:
3955 * tests/internal/test-jpeg.h:
3956 * tests/internal/test-mpeg2.c:
3957 * tests/internal/test-mpeg2.h:
3958 * tests/internal/test-mpeg4.c:
3959 * tests/internal/test-mpeg4.h:
3960 * tests/internal/test-subpicture-data.c:
3961 * tests/internal/test-subpicture-data.h:
3962 * tests/internal/test-subpicture.c:
3963 * tests/internal/test-surfaces.c:
3964 * tests/internal/test-textures.c:
3965 * tests/internal/test-vc1.c:
3966 * tests/internal/test-vc1.h:
3967 * tests/internal/test-windows.c:
3968 * tests/internal/y4mreader.c:
3969 * tests/internal/y4mreader.h:
3970 * tests/meson.build:
3971 tests: move examples and tests to subfolders
3972 This makes way for adding unit (check) tests.
3974 2019-10-14 01:01:27 +0100 Tim-Philipp Müller <tim@centricular.com>
3982 * gst-libs/Makefile.am:
3983 * gst-libs/gst/Makefile.am:
3984 * gst-libs/gst/vaapi/Makefile.am:
3986 * gst/vaapi/Makefile.am:
3988 * tests/Makefile.am:
3989 * tests/elements/Makefile.am:
3990 Remove autotools build
3992 2019-10-10 15:26:36 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
3994 * gst/vaapi/gstvaapipostproc.c:
3995 vaapipostproc: fix off-by-one coord translations
3996 When translating navigation x,y coordinates for
3997 video-direction, it is necessary to subtract 1
3998 when using the video dimensions to compute the
3999 new x,y coordinates. That is, a 100x200 image
4000 should map coordinates in x=[0-99],y=[0-199].
4001 This issue was found with unit tests provided
4004 2019-10-11 17:34:06 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4006 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
4007 libs: window: x11: Avoid usage of deprecated API
4009 2019-10-11 17:13:34 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4012 build: halt meson configuration if no renderer API
4013 We should halt meson configuration if there is no render API
4014 installed (either DRM, Wayland or X11).
4015 That behavior was already in autotools but missed in meson. This patch
4019 2019-10-09 12:12:18 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4021 * gst-libs/gst/vaapi/gstvaapifilter.c:
4022 libs: filter: fix default orientation regression
4023 Fix regression introduced in f232f87f7082
4025 2019-10-07 11:53:23 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4027 * gst-libs/gst/vaapi/gstvaapifilter.c:
4028 libs: filter: use OP_DATA_DEFAULT_VALUE macro
4030 2019-10-07 11:39:19 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4032 * gst-libs/gst/vaapi/gstvaapifilter.c:
4033 libs: filter: use macro for returning op default value
4034 The code is essentially the same for getting all op default
4035 values. Thus, use a macro to help minimize code duplication
4036 and [hopefully] encourage using the same mechanism for all
4039 2019-10-07 09:56:37 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4041 * gst-libs/gst/vaapi/gstvaapifilter.c:
4042 libs: filter: query param spec for default scale method
4045 2019-10-07 09:44:06 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4047 * gst-libs/gst/vaapi/gstvaapifilter.c:
4048 libs: filter: query param spec for default skin-tone values
4051 2019-10-02 12:54:52 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4053 * gst-libs/gst/vaapi/gstvaapifilter.c:
4054 libs: filter: query param spec for default video-direction
4057 2019-10-07 10:23:09 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4062 2019-09-05 16:40:52 +0800 Yan Wang <yan.wang@linux.intel.com>
4064 * gst-libs/gst/vaapi/gstvaapifilter.c:
4065 * gst-libs/gst/vaapi/gstvaapifilter.h:
4066 * gst/vaapi/gstvaapipostproc.c:
4067 * gst/vaapi/gstvaapipostproc.h:
4068 vaapipostproc: Use level value for skin-tone-enhancement filter.
4069 Currently the parameter of skin-tone-enhancement filter is forced
4070 to zero. In fact it could be set different value by the user.
4071 So create a new property named as "skin-tone-enhancement-level"
4072 for accepting the used defined parameter value.
4073 At the same time, skin-tone-enhancement is marked as deprecated.
4074 When skin-tone-enhancement-level is set, skin-tone-enhancement
4077 2019-09-21 13:39:42 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4079 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
4080 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
4081 libs: decoder: h264, h266: fix g_return_val_if_fail() missuse
4082 g_return_val_fail() documentations says:
4083 If expr evaluates to FALSE, the current function should be
4084 considered to have undefined behaviour (a programmer error).
4085 The only correct solution to such an error is to change the
4086 module that is calling the current function, so that it avoids
4087 this incorrect call.
4088 So it was missused in a couple parts of the H264 and H265 internal
4089 decoders. This patch changes that to plain conditionals.
4090 Also, it was included a couple code-style fixes.
4092 2019-09-23 19:52:20 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4094 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4095 * gst-libs/gst/vaapi/gstvaapiencoder.h:
4096 * tests/simple-encoder.c:
4097 * tests/test-fei-enc-in.c:
4098 libs: encoder: remove gst_vaapi_encoder_{ref,unref}()
4099 Since GstVaapiEncoder is a descendant of of GstObject, there is no
4100 need to keep a custom ref()/unref() methods. This patch deletes them.
4102 2019-09-24 01:03:02 +0800 He Junyan <junyan.he@hotmail.com>
4104 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4105 libs: encoder: correct encoder's ref/unref function.
4106 GstVaapiEncoder now is a standard gstobject and need to use
4107 gst_object_ref/unref functions.
4109 2019-09-19 12:09:20 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4111 * gst/vaapi/gstvaapiencode_h264_fei.c:
4112 gst: encode: h264_fei: remove useless comparison
4113 The expression "len >= 0" is always true since "len"
4114 is an unsigned type. And it is clear that the writers
4115 intention was not to write "len > 0" since we handle
4116 len == 0 in the ensuing "if (len < 3)" conditional
4119 2019-09-19 11:17:24 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4121 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
4122 libs: encoder: h264_fei: fix potential overflow before widen
4123 Found by static analysis. encoder->mb_width * encoder->mb_height
4124 is evaluated using 32-bit arithmetic before widen. Thus, cast
4125 at least one of these to guint64 to avoid overflow.
4127 2019-09-19 10:56:13 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4129 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
4130 libs: encoder: h264_fei: remove dead error condition
4131 Found by static analysis. The feipak is always null
4132 when we reach the error target.
4134 2019-09-19 10:49:11 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4136 * gst-libs/gst/vaapi/gstvaapiutils.c:
4137 libs: utils: add missing break in switch
4139 2019-09-11 11:56:35 +0800 He Junyan <junyan.he@hotmail.com>
4141 * gst-libs/gst/vaapi/video-format.c:
4142 libs: video-format: Make all YUV format available
4143 The YUV formats have no ambiguity for drivers, so we can add them all.
4144 Some old driver(i965) does not implement full get/put image functions
4145 but can use derive image funtions for the YUV format. It does not
4146 report that kind of formats correctly in image query, but will derive
4147 that YUV format image from surface. The dynamic mapping of YUV format
4148 will block that manner.
4149 Adding more YUV format mapping has no side effect. So considering the
4150 legacy driver conformance, we add all YUV formats mapping statically
4151 and dynamic mapping RBG formats
4155 2019-09-18 15:30:03 +1000 Matthew Waters <matthew@centricular.com>
4157 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
4158 * gst-libs/gst/vaapi/gstvaapiutils_egl.h:
4159 egl: Fix racyness in display thread creation
4160 Multiple different scenarios could break the display thread creation and
4161 end up blocking waiting for thread o be created. Fix them all by
4162 correctly waiting for a new boolean to become valid.
4164 2019-09-18 15:29:03 +1000 Matthew Waters <matthew@centricular.com>
4166 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
4167 egl: don't advertise a wrapped EGLContext as actually wrapped
4168 It's not actually wrapped as we create a new EGLContext from the passed
4169 in EGLContext. As a result, the created EGLContext was never destroyed.
4171 2019-09-16 23:28:31 +0800 He Junyan <junyan.he@hotmail.com>
4173 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
4174 libs: h264decoder: do not return error for unhandled NAL unit.
4175 Some streams have error data introducing unknown NAL type. There are
4176 also kinds of NAL types we do not want to handle. The old manner will
4177 set a decoder error when encounter this, which cause a latent crash bug.
4178 The decoder may successfully decode the picture and insert it into DPB.
4179 But there are error NAL units after the AU which cause the post unit error
4180 and make that frame dropped. The later output of the picture still want
4181 to ref that frame and crash.
4182 No need to set decoder error when can not recognize or handle the NAL
4183 unit, just skip it and continue.
4186 2019-09-11 14:32:22 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4188 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4189 * gst-libs/gst/vaapi/gstvaapiencoder.h:
4190 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
4191 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
4192 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
4193 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
4194 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
4195 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
4196 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
4197 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
4198 * gst/vaapi/gstvaapiencode.c:
4199 libs: encoders: use GST_PARAM_USER_SHIFT to define internal params
4200 This patch makes use of GST_PARAM_USER_SHIFT to define the internal
4201 param in encoders to decide which parameters to expose. Thus
4202 gstreamer-vaapi will not interfere with any change in GStreamer in the
4204 Also, the internal symbol was change to
4205 GST_VAAPI_PARAM_ENCODER_EXPOSURE to keep the namespacing.
4207 2019-09-09 18:06:51 +0800 He Junyan <junyan.he@hotmail.com>
4209 * gst-libs/gst/vaapi/gstvaapiimage.c:
4210 * gst-libs/gst/vaapi/video-format.c:
4211 libs: Add BGR10A2_LE support for color space conversion.
4214 2019-08-23 14:41:06 +0800 He Junyan <junyan.he@hotmail.com>
4216 * gst-libs/gst/vaapi/gstvaapisurface.c:
4217 libs: surface: add pointer check for surface_new_from_formats.
4219 gst-launch-1.0 filesrc location=some_name.mjpeg ! jpegparse !
4220 vaapijpegdec ! videoconvert ! video/x-raw,format=I420 ! vaapisink
4221 will crash on i965 driver because of no pointer check.
4222 We now generate the video format map between GST format and VA format
4223 dynamically based on the image format returned by vaQueryImageFormats.
4224 i965 driver does to report image format of 444P and Y800 forcc, while
4225 the jpeg decoder context VASurfaceAttribPixelFormat use them. We can
4226 not recognize these format and pass a NULL pointer to
4227 gst_vaapi_surface_new_from_formats.
4228 We need to add a pointer check here and let the fallback logic handle
4229 this case correctly.
4230 Other drivers work well.
4232 2019-09-07 13:23:01 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4234 * gst-libs/gst/vaapi/gstvaapiutils.c:
4235 libs: utils: guard the VAEntrypointFEI symbol
4236 VAEntrypointFEI appeared in libva 2.0.0 (API version 1.0.0)
4238 2019-09-05 14:48:22 +0800 He Junyan <junyan.he@hotmail.com>
4240 * gst-libs/gst/vaapi/gstvaapidisplay.c:
4241 * gst-libs/gst/vaapi/video-format.c:
4242 * gst-libs/gst/vaapi/video-format.h:
4243 libs: video-format: Refine the video format mapping.
4244 Improve the mapping between va format and gst format. The new map
4245 will be generated dynamically, based on the query result of image
4246 format in VA driver. Also consider the ambiguity of RGB color
4249 2019-04-15 16:51:26 +0100 Philippe Normand <philn@igalia.com>
4251 * gst/vaapi/gstvaapipluginutil.c:
4252 pluginutil: Remove Mesa from drivers white list
4253 The Mesa Gallium driver is poorly tested currently, leading to bad user
4254 experience for AMD users. The driver can be added back to the white list at
4255 runtime using the GST_VAAPI_ALL_DRIVERS environment variable.
4257 2019-07-08 14:18:00 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4259 * gst/vaapi/gstvaapipostproc.c:
4260 * gst/vaapi/gstvaapipostproc.h:
4261 * gst/vaapi/gstvaapipostprocutil.c:
4262 vaapipostproc: allow cropping via properties
4263 Add crop-left, crop-right, crop-top and crop-bottom
4264 properties to vaapipostproc.
4266 2019-08-30 17:31:45 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4268 * gst/vaapi/gstvaapipostproc.c:
4269 vaapipostproc: rotate outbuf and crop meta if forwarding
4270 When forwarding crop meta to downstream, the output
4271 buffer and crop meta need to be rotated, too.
4273 for i in 90r 180 90l vert horiz ul-lr ur-ll
4275 gst-launch-1.0 -vf videotestsrc num-buffers=500 \
4276 ! videocrop top=100 bottom=30 left=40 right=20 \
4277 ! vaapipostproc video-direction=$i \
4279 gst-launch-1.0 -vf videotestsrc num-buffers=500 \
4280 ! videocrop top=100 bottom=30 left=40 right=20 \
4281 ! vaapipostproc video-direction=$i \
4282 ! identity drop-allocation=true \
4286 2019-08-30 14:14:30 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4288 * gst/vaapi/gstvaapipostproc.c:
4289 vaapipostproc: fix output buffer WxH for crop meta forwarding
4290 Adding crop meta x,y to w,h only compensates for left,top
4291 cropping. But we also need to compensate for right,bottom
4293 The video meta contains the appropriate w,h (uncropped)
4294 values, so use it instead.
4296 gst-launch-1.0 -vf videotestsrc num-buffers=500 \
4297 ! videocrop top=50 bottom=30 left=40 right=20 \
4298 ! vaapipostproc ! vaapisink & \
4299 gst-launch-1.0 -vf videotestsrc num-buffers=500 \
4300 ! videocrop top=50 bottom=30 left=40 right=20 \
4301 ! vaapipostproc ! identity drop-allocation=1 \
4304 2019-09-04 10:52:51 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4306 * gst/vaapi/gstvaapipostproc.c:
4307 vaapipostproc: handle size and direction together in src events
4308 Mapping a pointer event needs to consider both size and
4309 video-direction operations together, not just one or the other.
4310 This fixes an issue where x,y were not being mapped correctly
4311 for 90r, 90l, ur-ll and ul-lr video-direction. In these directions,
4312 the WxH are swapped and GST_VAAPI_POSTPROC_FLAG_SIZE is set. Thus,
4313 the first condition in the pointer event handling was entered and
4314 x,y scale factor were incorrectly computed due to srcpad WxH
4316 This also fixes all cases where both video-direction and scaling
4317 are enabled at the same time.
4318 Test that all pointer events map appropriately:
4321 GST_DEBUG=vaapipostproc:5 gst-launch-1.0 -vf videotestsrc \
4322 ! vaapipostproc video-direction=${i} width=300 \
4324 GST_DEBUG=vaapipostproc:5 gst-launch-1.0 -vf videotestsrc \
4325 ! vaapipostproc video-direction=${i} width=300 height=200 \
4327 GST_DEBUG=vaapipostproc:5 gst-launch-1.0 -vf videotestsrc \
4328 ! vaapipostproc video-direction=${i} height=200 \
4330 GST_DEBUG=vaapipostproc:5 gst-launch-1.0 -vf videotestsrc \
4331 ! vaapipostproc video-direction=${i} \
4335 2019-08-20 14:22:57 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
4337 * gst/vaapi/gstvaapipostproc.c:
4338 * gst/vaapi/gstvaapipostproc.h:
4339 vaapipostproc: advertise crop meta is handled
4340 Advertise to upstream that vaapipostproc can handle
4342 When used in conjunction with videocrop plugin, the
4343 videocrop plugin will only do in-place transform on the
4344 crop meta when vaapipostproc advertises the ability to
4345 handle it. This allows vaapipostproc to apply the crop
4346 meta on the output buffer using vaapi acceleration.
4347 Without this advertisement, the videocrop plugin will
4348 crop the output buffer directly via software methods,
4349 which is not what we desire.
4350 vaapipostproc will not apply the crop meta if downstream
4351 advertises crop meta handling; vaapipostproc will just
4352 forward the crop meta to downstream. If crop meta is
4353 not advertised by downstream, then vaapipostproc will
4354 apply the crop meta.
4356 1. vaapipostproc will forward crop meta to vaapisink
4357 gst-launch-1.0 videotestsrc \
4358 ! videocrop left=10 \
4361 2. vaapipostproc will do the cropping
4362 gst-launch-1.0 videotestsrc \
4363 ! videocrop left=10 \
4365 ! identity drop-allocation=1 \
4368 2019-08-29 18:44:36 +0800 He Junyan <junyan.he@hotmail.com>
4370 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
4371 libs: encoder: clean two virtual func in encoder class
4372 set_property and get_default_properties functions are no longer
4373 needed for encoder class.
4375 2019-08-29 18:43:30 +0800 He Junyan <junyan.he@hotmail.com>
4377 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
4378 * gst-libs/gst/vaapi/gstvaapifeienc_h264.h:
4379 libs: encoder: delete old set_property and property enum feienc264
4381 2019-08-29 18:39:27 +0800 He Junyan <junyan.he@hotmail.com>
4383 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
4384 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.h:
4385 libs: encoder: delete old set_property and property enum in h264 fei
4387 2019-08-29 18:37:58 +0800 He Junyan <junyan.he@hotmail.com>
4389 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
4390 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.h:
4391 libs: encoder: delete old set_property and property enum in vp9
4393 2019-08-29 18:36:51 +0800 He Junyan <junyan.he@hotmail.com>
4395 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
4396 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.h:
4397 libs: encoder: delete old set_property and property enum in vp8
4399 2019-08-29 18:35:59 +0800 He Junyan <junyan.he@hotmail.com>
4401 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
4402 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
4403 libs: encoder: delete old set_property and property enum in mpeg2
4405 2019-08-29 18:34:57 +0800 He Junyan <junyan.he@hotmail.com>
4407 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
4408 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.h:
4409 libs: encoder: delete old set_property and property enum in jpeg
4411 2019-08-29 18:31:56 +0800 He Junyan <junyan.he@hotmail.com>
4413 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
4414 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
4415 libs: encoder: delete old set_property and property enum in h265
4417 2019-08-29 18:30:07 +0800 He Junyan <junyan.he@hotmail.com>
4419 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
4420 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
4421 libs: encoder: delete old set_property and property enum in h264
4423 2019-08-29 18:17:42 +0800 He Junyan <junyan.he@hotmail.com>
4425 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4426 * gst-libs/gst/vaapi/gstvaapiencoder.h:
4427 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
4428 libs: encoder: delete EncoderPropInfo related functions
4430 2019-08-29 16:13:19 +0800 He Junyan <junyan.he@hotmail.com>
4432 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4433 * gst-libs/gst/vaapi/gstvaapiencoder.h:
4434 libs: encoder: delete encoder_set_property
4435 We no longer need this obsolete set_property function now after
4436 switch to standard gobject's property manner.
4437 Also delete the old encoder's property enum in the header file.
4439 2019-08-29 15:59:43 +0800 He Junyan <junyan.he@hotmail.com>
4441 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4442 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
4443 libs: encoder: delete properties_get_default for base class
4445 2019-08-29 15:52:04 +0800 He Junyan <junyan.he@hotmail.com>
4447 * gst/vaapi/gstvaapiencode.c:
4448 * gst/vaapi/gstvaapiencode.h:
4449 plugin: encode: delete useless init_properties.
4450 Also delete the get_properties function in encode class. We now
4451 use g_object_class_list_properties to get all properties for
4452 internal encoder class.
4454 2019-08-29 15:43:45 +0800 He Junyan <junyan.he@hotmail.com>
4456 * gst/vaapi/gstvaapiencode.c:
4457 * gst/vaapi/gstvaapiencode.h:
4458 plugin: encode: delete set/get_property func in encode class
4459 Use standard gobject's property functions to replace the old way.
4461 2019-08-29 15:31:16 +0800 He Junyan <junyan.he@hotmail.com>
4463 * gst/vaapi/gstvaapiencode.c:
4464 * gst/vaapi/gstvaapiencode.h:
4465 plugin: encode: delete gst_vaapiencode_init_properties
4466 No need to init the properties got by get_default_properties func
4467 now. The properties are inited correctly in internal encoder class.
4469 2019-08-29 15:19:10 +0800 He Junyan <junyan.he@hotmail.com>
4471 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
4472 libs: encoder: delete 3 useless init macro
4474 2019-08-29 15:16:26 +0800 He Junyan <junyan.he@hotmail.com>
4476 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
4477 * gst-libs/gst/vaapi/gstvaapifeienc_h264.h:
4478 libs: encoder: delete get_default_properties of feienc
4480 2019-08-29 15:14:14 +0800 He Junyan <junyan.he@hotmail.com>
4482 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
4483 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.h:
4484 libs: encoder: delete get_default_properties of H264 Fei
4486 2019-08-29 15:07:17 +0800 He Junyan <junyan.he@hotmail.com>
4488 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
4489 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.h:
4490 libs: encoder: delete get_default_properties of VP9
4492 2019-08-29 15:06:25 +0800 He Junyan <junyan.he@hotmail.com>
4494 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
4495 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.h:
4496 libs: encoder: delete get_default_properties of VP8
4498 2019-08-29 15:03:52 +0800 He Junyan <junyan.he@hotmail.com>
4500 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
4501 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
4502 libs: encoder: delete get_default_properties of MPEG2
4504 2019-08-29 15:03:19 +0800 He Junyan <junyan.he@hotmail.com>
4506 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
4507 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.h:
4508 libs: encoder: delete get_default_properties of JPEG
4510 2019-08-30 19:15:38 +0800 He Junyan <junyan.he@hotmail.com>
4512 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
4513 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
4514 libs: encoder: delete get_default_properties of H265
4516 2019-08-29 14:59:12 +0800 He Junyan <junyan.he@hotmail.com>
4518 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
4519 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
4520 libs: encoder: delete get_default_properties of H264
4522 2019-08-29 14:53:59 +0800 He Junyan <junyan.he@hotmail.com>
4524 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4525 libs: encoder: delete the useless constructed func for encoder.
4527 2019-08-26 23:16:33 +0800 He Junyan <junyan.he@hotmail.com>
4529 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
4530 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
4531 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
4532 libs: encoder: implement get_view_ids for h264 encoder.
4534 2019-08-20 23:56:33 +0800 He Junyan <junyan.he@hotmail.com>
4536 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4537 * gst/vaapi/gstvaapiencode.c:
4538 * gst/vaapi/gstvaapiencode_h264.c:
4539 * gst/vaapi/gstvaapiencode_h264_fei.c:
4540 * gst/vaapi/gstvaapiencode_h265.c:
4541 * gst/vaapi/gstvaapiencode_jpeg.c:
4542 * gst/vaapi/gstvaapiencode_mpeg2.c:
4543 * gst/vaapi/gstvaapiencode_vp8.c:
4544 * gst/vaapi/gstvaapiencode_vp9.c:
4545 gst: encode: enable new type of property mechanism.
4547 2019-08-20 22:16:35 +0800 He Junyan <junyan.he@hotmail.com>
4549 * gst/vaapi/gstvaapiencode.c:
4550 * gst/vaapi/gstvaapiencode.h:
4551 gst: encode: add property help functions for encoder properties.
4552 The encoder is a true gstobject now and all the properties are using
4553 gobject's properties mechanism. Add help functions to handle the properties
4554 between encode and encoder class.
4555 The basic idea is mapping the same property between encoder and encode. All
4556 the encoder's properties will have the same name, the same type in encode.
4557 The set/get property function just forward the property setting/getting to
4558 the encoder using the same property name and value. Because the encoder is
4559 created on needed, we need to cache the property setting in encode.
4561 2019-08-30 18:39:32 +0800 He Junyan <junyan.he@hotmail.com>
4563 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4564 * gst-libs/gst/vaapi/gstvaapiencoder.h:
4565 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
4566 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
4567 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
4568 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
4569 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
4570 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
4571 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
4572 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
4573 libs: encoder: add flags to all encoder properties.
4574 G_PARAM_CONSTRUCT make all properties init correctly, we do not
4575 need to init the properties manually.
4576 G_PARAM_FLAG_VAAPI_ENCODER_EXPOSURE is a vaapi encoder specific
4577 flag, means need to expose the property to according encode class.
4579 2019-08-20 17:00:39 +0800 He Junyan <junyan.he@hotmail.com>
4581 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
4582 libs: encoder: Add properties for h264 encoder fei.
4583 Install properties for h264 encoder fei class. Also set the new get/set
4584 property functions for gobject class. Still use the old properties
4585 way now and this new feature will be enabled later.
4587 2019-08-20 15:58:30 +0800 He Junyan <junyan.he@hotmail.com>
4589 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
4590 libs: encoder: Add properties for h264 fei encoder.
4591 Install properties for h264 fei encoder class. Also set the new get/set
4592 property functions for gobject class. Still use the old properties
4593 way now and this new feature will be enabled later.
4595 2019-08-20 15:29:27 +0800 He Junyan <junyan.he@hotmail.com>
4597 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
4598 libs: encoder: Add properties for vp9 encoder.
4599 Install properties for vp9 encoder class. Also set the new get/set
4600 property functions for gobject class. Still use the old properties
4601 way now and this new feature will be enabled later.
4603 2019-08-20 15:01:02 +0800 He Junyan <junyan.he@hotmail.com>
4605 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
4606 libs: encoder: Add properties for vp8 encoder.
4607 Install properties for vp8 encoder class. Also set the new get/set
4608 property functions for gobject class. Still use the old properties
4609 way now and this new feature will be enabled later.
4611 2019-08-20 14:31:58 +0800 He Junyan <junyan.he@hotmail.com>
4613 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
4614 libs: encoder: Add properties for mpeg2 encoder.
4615 Install properties for mpeg2 encoder class. Also set the new get/set
4616 property functions for gobject class. Still use the old properties
4617 way now and this new feature will be enabled later.
4619 2019-08-20 14:53:06 +0800 He Junyan <junyan.he@hotmail.com>
4621 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
4622 libs: encoder: Add properties for jpeg encoder.
4623 Install properties for jpeg encoder class. Also set the new get/set
4624 property functions for gobject class. Still use the old properties
4625 way now and this new feature will be enabled later.
4627 2019-08-20 14:12:36 +0800 He Junyan <junyan.he@hotmail.com>
4629 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
4630 libs: encoder: Add properties for h265 encoder.
4631 Install properties for h265 encoder class. Also set the new get/set
4632 property functions for gobject class. Still use the old properties
4633 way now and this new feature will be enabled later.
4635 2019-08-20 01:33:40 +0800 He Junyan <junyan.he@hotmail.com>
4637 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
4638 libs: encoder: Add properties for h264 encoder.
4639 Install properties for h264 encoder class. Also set the new get/set
4640 property functions for gobject class. Still use the old properties
4641 way now and this new feature will be enabled later.
4643 2019-08-19 15:38:09 +0800 He Junyan <junyan.he@hotmail.com>
4645 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4646 libs: encoder: add properties and prop help functions
4647 Add all common properties to encoder base class. rate-control and
4648 tune are moved to sub class.
4650 2019-08-29 14:38:49 +0800 He Junyan <junyan.he@hotmail.com>
4652 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4653 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
4654 libs: encoder: delete useless gst_vaapi_encoder_new func.
4655 GstVaapiEncoder is a abstract gobject and never be created directly.
4657 2019-07-27 00:55:53 +0800 He Junyan <junyan.he@hotmail.com>
4659 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4660 * gst-libs/gst/vaapi/gstvaapiencoder.h:
4661 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
4662 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
4663 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
4664 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.h:
4665 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
4666 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
4667 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
4668 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.h:
4669 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
4670 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
4671 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
4672 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
4673 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.h:
4674 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
4675 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.h:
4676 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
4677 * gst-libs/gst/vaapi/gstvaapifeienc_h264.h:
4678 lib: encoder: gstobjectfy all vaapi encoders.
4679 Replace all gstvaapiobject in vaapi encoders with standard gstobject.
4680 Let the gstobject common logic to handle all the init and finalize
4681 works. But the property install/set/get still use the old way, need
4682 to be improved later.
4684 2019-08-29 12:11:36 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4686 * gst/vaapi/gstvaapipostproc.c:
4687 vaapipostproc: check for filter before appending caps
4688 While ensuring the allowed sink pad caps, the filter attributes set
4689 the frame size restriction, but it is not ensured, at that moment,
4690 that the filter is already instantiaded.
4691 In order to silence the glib logs, this patch add only calls
4692 gst_vaapi_filter_append_caps() if the filter is instantiated.
4694 2019-08-28 12:49:03 -0400 Thibault Saunier <tsaunier@igalia.com>
4696 * gst/vaapi/gstvaapidecodebin.c:
4697 Classify vaapidecodebin as a hardware decoder
4699 2019-08-27 18:12:45 +0800 He Junyan <junyan.he@hotmail.com>
4701 * gst/vaapi/gstvaapipostproc.c:
4702 libs: postproc: fix a memory leak point.
4703 filter_ops and filter_formats should already have valid value when
4704 the function gst_vaapipostproc_ensure_filter_caps re-enter
4706 2019-08-27 01:30:36 +0800 He Junyan <junyan.he@hotmail.com>
4708 * gst-libs/gst/vaapi/gstvaapiutils_core.c:
4709 libs: util: Fix a memory leak in config_surface_attributes_get
4711 2019-08-22 14:33:54 +0800 Wangfei <fei.w.wang@intel.com>
4713 * gst/vaapi/gstvaapidecode.c:
4714 vaapidecode: support transform ROI meta
4715 This will benefit the use case like:
4716 src ---> encode ---> decode ---> circle ROI ---> sink
4721 2019-08-23 19:10:15 +0200 Mathieu Duponchelle <mathieu@centricular.com>
4723 * gst/vaapi/gstvaapidecodedoc.c:
4724 docstrings: port ulinks to markdown links
4726 2019-08-20 17:05:14 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4728 * gst-libs/gst/vaapi/Makefile.am:
4729 * gst-libs/gst/vaapi/gstvaapicontext.c:
4730 * gst-libs/gst/vaapi/gstvaapicontext_overlay.c:
4731 * gst-libs/gst/vaapi/gstvaapicontext_overlay.h:
4732 * gst-libs/gst/vaapi/meson.build:
4733 libs: remove context's overlay
4734 The context overlay was an optimization to apply a video composition
4735 to all the surfaces bound to a context.
4736 But since commit 18031dc6 this optimization was disabled, so it is
4737 better just get rid of it.
4739 2019-08-20 16:50:46 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4741 * gst-libs/gst/vaapi/gstvaapicontext.c:
4742 * gst-libs/gst/vaapi/gstvaapisurface.c:
4743 * gst-libs/gst/vaapi/gstvaapisurface.h:
4744 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
4745 * gst/vaapi/gstvaapipluginutil.c:
4746 * tests/test-subpicture.c:
4747 libs: remove surface's parent context
4748 In commit 18031dc6 surface's parent context is not assigned because of
4749 circular references. Since then (2013), there's has no issue with
4750 subpictures attached to a context, the current only users of this API.
4751 This patch cleans up all of related code with the unused surface's
4754 2019-08-18 13:53:53 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4756 * gst/vaapi/gstvaapidecode.c:
4757 vaapidecode: guard if no structure is available in caps
4759 2019-08-18 13:53:19 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4761 * gst-libs/gst/vaapi/gstvaapifilter.c:
4762 * gst-libs/gst/vaapi/gstvaapifilter.h:
4763 * gst/vaapi/gstvaapipostproc.c:
4764 vaapipostproc: append frame size restrictions in caps
4766 2019-08-18 13:09:58 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4768 * gst-libs/gst/vaapi/gstvaapiprofilecaps.c:
4769 libs: profilecaps: refactor common code
4771 2019-08-16 19:35:58 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4773 * gst/vaapi/gstvaapiencode.c:
4774 vaapiencode: set frame size restrictions in caps
4777 2019-08-16 19:28:27 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4779 * gst-libs/gst/vaapi/gstvaapiencoder.c:
4780 * gst-libs/gst/vaapi/gstvaapiencoder.h:
4781 * gst/vaapi/gstvaapiencode.c:
4782 vaapiencode: enhance how the profile is defined
4783 This code doesn't define the profile used by the internal encoder, but
4784 it used to "predict" which is going to be used and to get the caps
4786 Before the profile was predicted by checking the donwstream caps, but
4787 sometimes they are not defined, setting an unknown profile. In order
4788 to enhances this situation, the encoder asks to internal encoder if it
4789 has one. If so, it is used.
4790 To ask the internal encoder's profile a new accessor function was
4791 added: gst_vaapi_encoder_get_profile()
4793 2019-08-16 19:26:36 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4795 * gst-libs/gst/vaapi/gstvaapiprofilecaps.c:
4796 * gst-libs/gst/vaapi/gstvaapiprofilecaps.h:
4797 libs: profilecaps: defines gst_vaapi_profile_caps_append_encoder()
4798 Previously it was just a boilerplate. Now it is real implementation.
4800 2019-08-16 19:17:48 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4802 * gst-libs/gst/vaapi/gstvaapiutils.c:
4803 libs: utils: treat va_rt_format as bitwise flag
4804 The return value of vaGetConfigAttributes() of VAConfigAttribRTFormat
4805 is a bitwise flag with *all* the supported chroma types.
4806 Previously it was assumed that the return value was a single value,
4807 thus when returning the GST_VAAPI_CHROMA_TYPE_XXX the code was a
4808 simple case. But it is wrong.
4809 This patch changes the case block with a sequence of ifs testing the
4810 bitwise. For now we assume a "priority" list in the testing sequence.
4812 2019-08-16 18:07:43 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4814 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
4815 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
4816 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
4817 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
4818 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
4819 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
4820 * gst-libs/gst/vaapi/gstvaapiprofile.c:
4821 * gst-libs/gst/vaapi/gstvaapiprofile.h:
4822 * tests/test-display.c:
4823 libs: profile: add gst_vaapi_profile_get_va_name()
4824 gst_vaapi_profile_get_name() returns a proper name for
4825 GstCaps. Nonetheless, there are many profiles which don't have a name
4826 representation for that realm.
4827 gst_vaapi_profile_get_va_name() returns the name of the profile
4828 according to its VAProfile name.
4829 This new funtion is used in the encoder error message.
4831 2019-08-05 19:47:30 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4833 * gst-libs/gst/vaapi/Makefile.am:
4834 * gst-libs/gst/vaapi/gstvaapiprofilecaps.c:
4835 * gst-libs/gst/vaapi/gstvaapiprofilecaps.h:
4836 * gst-libs/gst/vaapi/meson.build:
4837 * gst/vaapi/gstvaapidecode.c:
4838 libs: profilecaps: move caps config into a new file
4839 Implement all the appending of frame size restrictions in caps, for
4840 encoders and decoders, in a new source file.
4842 2019-08-05 19:45:49 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4844 * gst-libs/gst/vaapi/gstvaapidecoder.c:
4845 libs: decoder: ref the caps as property
4847 2019-08-02 16:56:17 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4849 * gst-libs/gst/vaapi/gstvaapicontext.c:
4850 * gst-libs/gst/vaapi/gstvaapicontext.h:
4851 libs: context: add gst_vaapi_context_get_surface_attributes()
4852 This function copies the surface attributes from the context's object
4855 2019-08-02 12:46:55 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4857 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
4858 * gst-libs/gst/vaapi/gstvaapibufferproxy.h:
4859 * gst-libs/gst/vaapi/gstvaapiutils.c:
4860 * gst-libs/gst/vaapi/gstvaapiutils.h:
4861 libs: move memory types conversions to gstvaapiutils
4862 And add more supported memory types by current VA.
4864 2019-08-01 19:48:21 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4866 * gst-libs/gst/vaapi/gstvaapiutils_core.c:
4867 * gst-libs/gst/vaapi/gstvaapiutils_core.h:
4868 libs: utils: remove unused function gst_vaapi_get_surface_formats()
4870 2019-08-01 19:46:17 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4872 * gst-libs/gst/vaapi/gstvaapicontext.c:
4873 * gst-libs/gst/vaapi/gstvaapicontext.h:
4874 * gst-libs/gst/vaapi/gstvaapifilter.c:
4875 libs: context, filter: use new surface attributes API
4877 2019-08-01 19:13:39 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4879 * gst-libs/gst/vaapi/gstvaapiutils_core.c:
4880 * gst-libs/gst/vaapi/gstvaapiutils_core.h:
4881 libs: utils: add gst_vaapi_config_surface_attributes_get()
4882 To extract the surface restrictions per config using a new structure:
4883 GstVaapiConfigSurfaceAttributes
4885 2019-07-31 13:08:43 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4887 * gst-libs/gst/vaapi/gstvaapiutils_core.c:
4888 Split the surface attribute retrieval
4890 2019-07-15 21:51:46 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4892 * gst/vaapi/gstvaapipostproc.c:
4893 vaapipostproc: handle navigation downstream event
4894 When navigation events contains coordiantes those have to be mapped
4895 to the new size and/or orientation.
4897 2019-07-15 21:23:21 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4899 * tests/elements/test-vaapisink.c:
4900 test-vaapisink: also use vaapipostproc to change orientation
4902 2019-07-15 21:27:20 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4904 * gst-libs/gst/vaapi/gstvaapifilter.c:
4905 * gst-libs/gst/vaapi/gstvaapifilter.h:
4906 * gst/vaapi/gstvaapipostproc.c:
4907 * gst/vaapi/gstvaapipostproc.h:
4908 * gst/vaapi/gstvaapipostprocutil.c:
4909 vaapipostproc: handle image-orientation upstream event
4910 Now that vaapipostproc can possible handle video-direction, it
4911 should also handle the image-orientation event from upstream if
4912 video-direction property is set to auto.
4914 2019-07-26 22:09:37 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4916 * gst/vaapi/gstvaapipostproc.c:
4917 vaapipostproc: add missing locks when adding flags
4919 2019-07-26 22:05:29 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4921 * gst/vaapi/gstvaapipostproc.c:
4922 vaapipostproc: update filter before fixate caps
4923 It is requiered to know if postproc is capable to change the video
4924 direction before fixating the source caps.
4925 In order to do it, it'ss required to know if there's a functional VPP,
4926 but that's checked at create() vmethod, which occurs after caps
4928 This patch checks for a functional VPP at fixate caps and, if so,
4929 checks for the enabled filtes and later do the caps fixations.
4931 2019-07-26 19:46:09 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4933 * gst-libs/gst/vaapi/gstvaapifilter.c:
4934 * gst-libs/gst/vaapi/gstvaapivalue.c:
4935 * gst-libs/gst/vaapi/gstvaapivalue.h:
4936 * gst/vaapi/gstvaapipostproc.c:
4937 vaapipostproc: element warning if video direction is unsupported
4938 If the video direction is unsupported by the driver, an element
4939 warning is posted in the bus to notify the application.
4940 gst_vaapi_enum_type_get_nick() was added in the library thus it can
4941 be used elsewhere. It retrives the nick from an enum gtype.
4943 2019-07-26 19:09:54 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4945 * gst-libs/gst/vaapi/gstvaapifilter.c:
4946 libs: filter: check mirror and rotation caps only once
4947 This patch locks the display before querying the pipeline caps and
4948 stores the mirror and rotation capabilities, thus they are not queried
4949 every time the video direction is set.
4951 2019-08-16 19:51:29 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4953 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
4954 libs: encoder: vp9: set VP9_0 profile as default
4955 Commit 0afc8131 introduced a regression and only NV12 format were
4956 admitted, failing in any other valid color format.
4957 This patch sets the profile to GST_VAAPI_PROFILE_VP9_0 by default.
4959 2019-08-16 13:25:06 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4961 * gst-libs/gst/vaapi/gstvaapifilter.c:
4962 libs: filter: fail if first color balance value is invalid
4964 2019-08-06 19:24:08 +0800 Yan Wang <yan.wang@linux.intel.com>
4966 * gst-libs/gst/vaapi/gstvaapifilter.c:
4967 * gst-libs/gst/vaapi/gstvaapiutils.c:
4968 * gst-libs/gst/vaapi/gstvaapiutils.h:
4969 libs: filter: set all color balance values
4970 When set multiple settings of color balance like hue, saturation,
4971 brightness and contrast for vaapipostproc, they should be set as
4972 parameters of color balance filter, at the same color balance
4974 Otherwise, multiple color balance filter calling will cause
4975 previous setting get reset by the last calling with default value.
4977 Signed-off-by: Yan Wang <yan.wang@linux.intel.com>
4979 2019-08-16 11:02:08 +0800 Wangfei <fei.w.wang@intel.com>
4981 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
4982 libs: h265dec: remove limitation of get iq matrix
4983 According hevc spec, scaling_list_data is not related
4984 to chroma_format_idc.
4986 2019-05-30 23:52:51 +0800 He Junyan <junyan.he@hotmail.com>
4988 * gst-libs/gst/vaapi/gstvaapivideopool.c:
4989 libs: videopool: fix undocumented behavior and counting
4990 gst_vaapi_video_pool_reserve_unlocked() hit an undocumented behavoir
4991 because it locks twice the same mutex.
4992 Also, n had different meanings in the current code: as an increase
4993 value and as a new total of allocated surfaces.
4994 This patche removes the undocumented behavoir (usually a deadlock) and
4995 fixes the meaning of n as the new total of allocated surfaces.
4996 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
4998 2019-07-17 11:56:45 +0800 He Junyan <junyan.he@hotmail.com>
5000 * gst-libs/gst/vaapi/gstvaapiutils.c:
5001 libs: utils: Add missing entries for string_of_VAEntrypoint.
5003 2019-07-18 22:01:01 +0800 He Junyan <junyan.he@hotmail.com>
5005 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
5006 libs: encoder: Consider vp9 profiles based on input format.
5007 Only support GST_VAAPI_PROFILE_VP9_0 and GST_VAAPI_PROFILE_VP9_2 now.
5010 2019-08-12 18:41:52 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5012 * gst/vaapi/gstvaapivideomemory.c:
5013 vaapivideomemory: demote error message to info
5014 The main reason to demote the message's level is because it is not an
5015 error, it's a possible output of the trial and there's a code path
5017 Secondly, it's very annoying when using gallium driver for radeon.
5019 2019-07-18 13:32:46 +0800 Wangfei <fei.w.wang@intel.com>
5021 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5022 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
5023 * gst-libs/gst/vaapi/gstvaapitypes.h:
5024 * gst-libs/gst/vaapi/gstvaapiutils.c:
5025 * gst-libs/gst/vaapi/gstvaapivalue.c:
5026 libs: encoder: h264: support ICQ/QVBR bitrate control mode
5027 ICQ is Intelligent Constant Quality. It will use the initial QP
5028 vaule of icq-quality-factor to adjust QP at MB level intelligently
5029 to improve subjective quality.
5030 QVBR is Quality defined VBR. It will use qvbr-quality-factor to
5031 adjust QP for each MB to get enough quality picture without waste
5034 2019-08-05 10:51:24 +0800 Wangfei <fei.w.wang@intel.com>
5036 * gst-libs/gst/vaapi/gstvaapitypes.h:
5037 libs: Let GST_VAAPI_RATECONTROL_MASK return unsigned int
5038 The value return from GST_VAAPI_RATECONTROL_MASK will be used by
5039 GST_VAAPI_POPCOUNT32 as its inpput. GST_VAAPI_POPCOUNT32 can only
5040 deal with unsigned int. Otherwise there may be an error of out of
5041 range of integer if we define few more rate-control mode.
5043 2019-06-07 09:54:12 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5045 * gst/vaapi/gstvaapidecodebin.c:
5046 vaapidecodebin: set queue's max size buffers to 1
5047 Otherwise the queue will swallow all the available decoder's surfaces
5048 reaching a dead-lock.
5049 This setting might impact the bin's peformance, but it's a trade-off.
5051 2019-06-07 09:53:08 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5053 * gst/vaapi/gstvaapidecodebin.c:
5054 vaapidecodebin: set properties default values
5056 2019-05-31 13:12:35 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5058 * gst/vaapi/gstvaapidecode.c:
5059 vaapidecode: don't error if can't push buffers downtream
5060 When the code path goes to push buffers downstream when no surface
5061 available in decoder context, and it fails the code bails out with a
5063 That behavior is wrong, since it shouldn't be fatal. The use case is
5064 when the video stream is disabled.
5065 This patch just ignores the errors in this situation and demotes the
5066 level of a log message.
5068 2019-05-18 13:24:35 +0800 Wangfei <fei.w.wang@intel.com>
5070 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5071 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
5072 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
5073 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.h:
5074 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5075 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
5076 libs: encoder: h264,h265: add new property "max-qp"
5077 Add new property "max-qp" to allow set the maximum quantisation
5080 2019-05-23 10:18:52 -0400 Wangfei <fei.w.wang@intel.com>
5082 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
5083 libs: encoder: vp9: add low power mode encode
5084 By now, this feature only support by media-driver on Ice Lake
5085 platform, more information you can reference:
5086 https://github.com/intel/media-driver
5088 2019-07-15 15:33:07 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
5090 * gst/vaapi/gstvaapipostprocutil.c:
5091 vaapipostproc: update PAR when rotating
5092 When rotating, swap pixel-aspect-ratio during
5096 2019-07-01 15:26:18 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
5098 * gst-libs/gst/vaapi/gstvaapifilter.c:
5099 * gst-libs/gst/vaapi/gstvaapiutils.c:
5100 * gst-libs/gst/vaapi/gstvaapiutils.h:
5101 * gst/vaapi/gstvaapipostprocutil.c:
5102 vaapipostproc: add rotation support
5103 Adds vpp rotation support to vaapipostproc. Uses
5104 property video-direction. Default is identity (no
5108 2019-05-22 10:47:30 -0400 Wangfei <fei.w.wang@intel.com>
5110 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5111 libs: encoder: h265: pass diff_cu_qp_delta_depth flag to driver
5112 Intel media-driver requires enablement of diff_cu_qp_delta_depth when
5113 cu_qp_delta_enabled_flag enabled.
5116 2019-07-01 17:02:33 +0800 Wangfei <fei.w.wang@intel.com>
5118 * gst-libs/gst/vaapi/gstvaapiutils.c:
5119 libs: encoder: Add MB ratecontrol mode to get its string
5121 2019-07-01 16:52:00 +0800 Wangfei <fei.w.wang@intel.com>
5123 * gst-libs/gst/vaapi/gstvaapiutils.c:
5124 libs: encoder: refine guard of bitrate control mode
5125 Remove useless guard of all bitrate control mode's guard except MB
5126 which is define in VA-API version 0.39.1.
5128 2019-06-29 00:08:40 +1000 Jan Schmidt <jan@centricular.com>
5130 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
5131 h264: Update for parse_vui_params parameter removal.
5132 Update calls to the h264 parser lib for removal of the
5133 parse_vui_params parameter.
5135 2019-06-24 16:26:56 -0400 Wang Zhanjun <zhanjunx.wang@intel.com>
5137 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
5138 libs: dec: vp9: do not use display size as decoded size
5139 If display size is smaller than current frame size, then the crop size
5140 will be set as display size, which either crashes the pipeline or the
5141 output MD5 does not match. Rather it should use the actual decoded size.
5142 This patch removes the cropping set. For rendering we can use aspect
5143 ratio to set display size.
5145 Signed-off-by: Wang Zhanjun <zhanjunx.wang@intel.com>
5146 Signed-off-by: Xu Guangxin <guangxin.xu@intel.com>
5148 2019-06-28 16:32:51 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5153 2019-06-25 19:11:12 +0800 He Junyan <junyan.he@hotmail.com>
5155 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
5156 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
5157 * gst-libs/gst/vaapi/gstvaapiutils_h265_priv.h:
5158 libs: dec: h265: Consider chroma_bit_depth to choose chrome type
5159 For some main-10 stream, sometime the luma is 8 bits while chrome is more
5160 than 8 bits, which cause using the wrong NV12 surface as the render target
5164 2019-06-25 10:31:20 +0800 Wangfei <fei.w.wang@intel.com>
5166 * gst/vaapi/gstvaapidecode.c:
5167 vaapidecode: set initial decode format according surface chroma type
5168 For surfaces with different chroma type, it is prefer to initialize
5169 a format which chroma type should be same with surface chroma type
5170 instead of using fixed NV12.
5172 2019-05-30 09:48:51 -0400 Wangfei <fei.w.wang@intel.com>
5174 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
5175 * gst-libs/gst/vaapi/gstvaapiimage.c:
5176 * gst-libs/gst/vaapi/video-format.c:
5177 * gst/vaapi/gstvaapidecode.c:
5178 * gst/vaapi/gstvaapipluginutil.h:
5179 libs: decoder: jpeg: add support 400/411/422/444 chroma type
5180 When create vaapi surface, it is better to use the chroma type get
5181 from jpeg file instead of using fixed 420 format. And the correct
5182 chroma type can be determined by horizontal_factor/vertical_factor
5183 flags that get from jpegparse.
5185 2019-06-22 00:05:24 +0800 He Junyan <junyan.he@hotmail.com>
5187 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
5188 libs: dec: h265: Fix profile_idc mapping.
5189 The old mapping values return by gst_vaapi_utils_h265_get_profile_idc is
5190 wrong, though GST_H265_PROFILE_IDC_MAIN and GST_H265_PROFILE_IDC_MAIN_10
5191 happened to be the correct value.
5192 We only support Annex A profile_idc (1-4).
5194 2019-06-10 20:46:30 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
5196 * gst/vaapi/gstvaapipluginbase.c:
5197 plugins: remove last negotiated video info if caps are same
5198 If the allocation caps and negotiated caps are the same,
5199 then ensure any previously negotiated video info is also
5200 removed. This can occur when multi-resolution video
5201 decoding returns to it's original resolution.
5204 2019-06-10 20:39:28 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
5206 * gst/vaapi/gstvaapivideomemory.c:
5207 vaapivideomemory: allow negotiated info to be removed
5208 Allow NULL negotiated_vinfo to be passed into
5209 gst_allocator_set_vaapi_negotiated_video_info to allow
5210 any previously set info to be removed.
5212 2019-06-06 17:24:30 +0300 Freyr <freyrnjordrson@gmail.com>
5214 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
5215 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
5216 libs: encoder: vp8,vp9: reset frame_counter when input frame's format changes
5217 When input frame's formate changes, vp{8,9} encoders don't reset their frame
5218 counter, hence the newly created frame could become a P-frame, leading to some
5219 major troubles (sigabrt in libdrm in case of vp9). This patch adds some frame
5220 prediction-related reset logic to the `flush' methods of GstVaapiEncoderVP8 and
5221 GstVaapiEncoderVP9 implementations.
5223 2019-05-31 12:30:03 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
5225 * gst-libs/gst/vaapi/gstvaapiencoder.c:
5226 libs: encoder: increase bitrate prop max value
5227 There are many profile levels that can support
5228 more than 102400 kbps. Thus, increase the max
5229 allowed bitrate property value from 102400 kbps
5230 to 2048000 kbps (same as msdk encoder plugins).
5232 2019-06-04 13:27:50 +0800 He Junyan <junyan.he@hotmail.com>
5234 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
5235 libs: mpeg2 encoder: No packed header for SPS and PPS
5236 Dislable passing down packed PPS and PPS to driver if driver does
5240 2019-05-31 23:10:33 +0200 Niels De Graef <niels.degraef@barco.com>
5245 meson: Bump minimal GLib version to 2.44
5246 This means we can use some newer features and get rid of some
5247 boilerplate code using the G_DECLARE_* macros.
5248 As discussed on IRC, 2.44 is old enough by now to start depending on it.
5250 2019-05-31 13:08:39 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5252 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
5253 libs: dec: vp9: clear parser pointer after release
5254 Fix an use-after-release of the parser pointer in VP9 decoder.
5256 2019-05-28 12:09:36 +0300 Freyr666 <sky_rider_93@mail.ru>
5258 * gst/vaapi/gstvaapiencode.c:
5259 vaapiencode: Fixes deadlock in gst_vaapiencode_change_state function
5260 This fixes a deadlock in gst_vaapiencode_change_state, which was due to
5261 srcpad's chain function was locked waiting for available buffers. Since the
5262 coded buffers in codedbuf_queue become available after sinkpad consume the
5263 encoded frames, Paused -> Ready state change leads to deadlock. Coded buffers
5264 are never consumed and marked free, hence gst_vaapiencode_handle_frame waits for
5265 available buffers and holds the stream_lock of the srcpad.
5267 2019-05-29 23:08:22 +0200 Mathieu Duponchelle <mathieu@centricular.com>
5269 * gst-libs/gst/vaapi/gstvaapitypes.h:
5270 * gst/vaapi/gstvaapidecodebin.c:
5271 * gst/vaapi/gstvaapidecodedoc.c:
5272 * gst/vaapi/gstvaapiencode_h264.c:
5273 * gst/vaapi/gstvaapiencode_h264_fei.c:
5274 * gst/vaapi/gstvaapiencode_h265.c:
5275 * gst/vaapi/gstvaapiencode_jpeg.c:
5276 * gst/vaapi/gstvaapiencode_mpeg2.c:
5277 * gst/vaapi/gstvaapiencode_vp8.c:
5278 * gst/vaapi/gstvaapiencode_vp9.c:
5279 * gst/vaapi/gstvaapipostproc.c:
5280 * gst/vaapi/gstvaapisink.c:
5281 doc: remove xml from comments
5283 2019-05-13 16:39:33 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
5285 * gst-libs/gst/vaapi/gstvaapifilter.c:
5286 * gst-libs/gst/vaapi/gstvaapifilter.h:
5287 * gst-libs/gst/vaapi/gstvaapiutils.c:
5288 * gst-libs/gst/vaapi/gstvaapiutils.h:
5289 * gst/vaapi/gstvaapipostproc.c:
5290 * gst/vaapi/gstvaapipostproc.h:
5291 vaapipostproc: add mirror support
5292 Adds vpp mirroring support to vaapipostproc. Use
5293 property video-direction. Valid values are identity,
5294 horiz or vert. Default is identity (no mirror).
5296 v2: Use GstVideoOrientationMethod enum
5297 v3: Don't warn for VA_MIRROR_NONE.
5298 Use GST_TYPE_VIDEO_ORIENTATION_METHOD type.
5299 v4: Query VAAPI caps when setting mirror value
5300 instead of during per-frame processing.
5301 v5: Return TRUE in warning cases when setting mirror value.
5303 2019-05-29 01:35:17 +0200 Mathieu Duponchelle <mathieu@centricular.com>
5305 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
5306 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
5307 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
5308 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
5309 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
5310 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
5311 * gst/vaapi/gstvaapidecodebin.c:
5312 * gst/vaapi/gstvaapisink.c:
5313 doc: fix some incorrect gtk-doc links
5315 2019-05-16 09:22:42 -0400 Thibault Saunier <tsaunier@igalia.com>
5317 * docs/gst_plugins_cache.json:
5318 docs: Update plugin cache
5319 Fixes https://gitlab.freedesktop.org/gstreamer/gst-docs/issues/36
5321 2019-05-16 16:46:43 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5323 * gst-libs/gst/vaapi/gstvaapisurface.h:
5324 libs: surface: fix documentation format
5326 2019-05-16 10:05:17 +0800 Wangfei <fei.w.wang@intel.com>
5328 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5329 libs: enc: h265: reset num_ref_idx_l1_active_minus1 when low delay B.
5330 When enable low delay B, the reference list 1 will be same with
5331 reference list 0, so need reset the num_ref_idx_l1_active_minus1
5332 to num_ref_idx_l0_active_minus1.
5335 2019-05-13 19:05:43 -0400 Thibault Saunier <tsaunier@igalia.com>
5338 meson: Fix call to wrong function
5340 2018-10-22 11:48:29 +0200 Thibault Saunier <tsaunier@igalia.com>
5345 * docs/gst_plugins_cache.json:
5348 * docs/plugins/Makefile.am:
5349 * docs/plugins/gstreamer-vaapi-plugins-docs.xml.in:
5350 * docs/plugins/gstreamer-vaapi-plugins-sections.txt:
5351 * docs/plugins/gstreamer-vaapi-plugins.types:
5352 * docs/plugins/inspect/plugin-vaapi.xml:
5353 * docs/plugins/running.xml:
5355 * docs/version.entities.in:
5356 * gst/vaapi/meson.build:
5358 * meson_options.txt:
5359 docs: Port to hotdoc
5361 2019-05-10 18:29:10 +0800 He Junyan <junyan.he@hotmail.com>
5363 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5364 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5365 libs: encoder: not call ensure_num_slices inside g_assert
5366 g_assert will take no effect when glib's G_DISABLE_ASSERT macro is
5367 defined. The function inside the g_assert will take no effect and
5368 we will fail to set the correct slice number.
5370 2019-04-29 09:52:39 +0800 Wangfei <fei.w.wang@intel.com>
5372 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
5373 libs: h265: dec: Add extension flags setting.
5374 Use VAPictureParameterBufferHEVCExtension&
5375 VASliceParameterBufferHEVCExtension to pass extension setting from
5376 some extension profile clips which may include these information.
5377 The hevc extension setting only supported after libva release 2.2.0
5380 2019-05-01 12:56:55 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
5382 * gst-libs/gst/vaapi/gstvaapiencoder.c:
5383 * gst-libs/gst/vaapi/gstvaapiencoder.h:
5384 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
5385 libs: encoder: add target-percentage property
5386 Allow users to set the target-percentage for
5387 variable rate controls. The default value is
5388 70 (as hard-coded prior).
5389 v2: minimum allowed value changed from 0 to 1
5390 v3: target-percentage unchanged if CBR used
5393 2019-05-09 00:09:21 +0800 He Junyan <junyan.he@hotmail.com>
5395 * gst-libs/gst/vaapi/gstvaapiencoder.h:
5396 libs: encoder: Add a missing comment for DEFAULT_ROI_VALUE property.
5398 2019-05-08 23:39:20 +0800 He Junyan <junyan.he@hotmail.com>
5400 * gst-libs/gst/vaapi/gstvaapiencoder.c:
5401 * gst-libs/gst/vaapi/gstvaapiencoder.h:
5402 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5403 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
5404 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
5405 libs: encoder: Enable trellis quantization method.
5406 The advanced trellis algorithm is supported in VA driver. We add
5407 its support as a property named "trellis" of encoder.
5408 It only works for H264 now, should be more in future.
5410 2019-05-07 11:03:51 +0800 Wangfei <fei.w.wang@intel.com>
5412 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
5413 libs: decoder: vp9: support 422/444 8bit/10bit chroma type.
5414 According to the vp9 sepc, profile 1/3 support 422/440/444 chroma
5415 type, so we need to add subsampling_x&subsampling_y to fix it.
5416 Here is the relationship between chroma type and profile and
5417 subsampling_x&subsampling_y according to vp9 spec:
5418 ------------------------------------------
5419 Profile | Bit depth | Chroma subsampling |
5420 ------------------------------------------
5422 ------------------------------------------
5423 1 | 8 | 422,440,444 |
5424 ------------------------------------------
5426 ------------------------------------------
5427 3 | 10, 12 | 422,440,444 |
5428 ------------------------------------------
5429 -----------------------------------------------
5430 Subsampling_x | Subsampling_y | Chroma format |
5431 -----------------------------------------------
5433 -----------------------------------------------
5435 -----------------------------------------------
5437 -----------------------------------------------
5439 -----------------------------------------------
5441 2019-04-16 18:33:54 +0800 He Junyan <junyan.he@hotmail.com>
5443 * gst-libs/gst/vaapi/gstvaapiimage.c:
5444 * gst-libs/gst/vaapi/video-format.c:
5445 libs: Add packed 24 RGB format support.
5446 Can not find a suitable chrome_type for this GST_VIDEO_FORMAT_RGB
5447 packed 24 format. Just use GST_VAAPI_CHROMA_TYPE_RGB32 as its chrome
5448 type. This kind of surface will just be created by new API with fourcc
5449 and no old style chrome based creation is available.
5452 2019-03-15 14:29:41 +0800 Wangfei <fei.w.wang@intel.com>
5454 * gst/vaapi/gstvaapiencode.c:
5455 * gst/vaapi/gstvaapiencode_h264.c:
5456 * gst/vaapi/gstvaapiencode_h264_fei.c:
5457 * gst/vaapi/gstvaapiencode_h265.c:
5458 * gst/vaapi/gstvaapiencode_jpeg.c:
5459 * gst/vaapi/gstvaapiencode_mpeg2.c:
5460 * gst/vaapi/gstvaapiencode_vp8.c:
5461 * gst/vaapi/gstvaapiencode_vp9.c:
5462 vaapiencode: handle DMABuf caps feature in sink pad
5463 Add DMABuff caps features in all encoders' sink pad.
5465 2019-05-03 10:31:52 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5467 * gst-libs/gst/vaapi/gstvaapiencoder.c:
5468 libs: encoder: continue if roi meta is NULL
5470 If the function actually returns a null value, a null pointer
5471 dereference will occur.
5472 In gst_vaapi_encoder_ensure_param_roi_regions(): Return value of
5473 function which returns null is dereferenced without checking
5475 2019-04-15 19:58:14 +0800 He Junyan <junyan.he@hotmail.com>
5477 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
5478 lib: decoder: vp9: Set chroma_type by VP9 bit_depth
5479 The decoder's surface chroma type should depend on the bit depth
5480 of VP9's parser. For 10bits VP9 stream, we need to use P10LE kind
5481 10 bits surface as the decoder result.
5484 2019-05-02 16:00:57 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5486 * gst/vaapi/gstvaapipostprocutil.c:
5487 vaapipostproc: don't do any color conversion when GL_TEXTURE_UPLOAD
5488 https://bugzilla.gnome.org/show_bug.cgi?id=748184 has resurrected
5489 with commit 3e992d8a
5490 Since gst_vaapi_find_preferred_caps_feature() returns a color format
5491 from caps negotiation, different from the default one (NV12), the
5492 postproc enables the color transformation. But when GL_TEXTURE_UPLOAD
5493 feature is negotiated, no color transformation shall be done.
5494 Nonetheless, with commit 3e992d8a the requested format changes
5495 firstly, because there's no video sink yet, so ANY caps are
5496 negotiated; but later, when there's a video sink and a caps
5497 renegotiation, the GL_TEXTURE_UPLOAD is negotiated though the color
5498 format conversion still ongoing. It is required to reset that
5500 This patch force default color format when GL_TEXTURE_UPLOAD is
5501 selected as preferred, thus avoiding the color conversion.
5504 2019-04-19 15:49:37 -0700 Julien Isorce <jisorce@oblong.com>
5506 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
5507 libs: surface: fix double free when dmabuf export fails
5508 Happens if vaAcquireBufferHandle fails.
5510 2019-04-29 20:10:39 +0800 He Junyan <junyan.he@hotmail.com>
5512 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5513 libs: h264encoder: fix a typo of GstVaapiEncoderH264PredictionType
5515 2019-04-19 10:43:35 +0100 Tim-Philipp Müller <tim@centricular.com>
5519 * docs/plugins/inspect/plugin-vaapi.xml:
5523 === release 1.16.0 ===
5525 2019-04-19 00:38:12 +0100 Tim-Philipp Müller <tim@centricular.com>
5531 * gstreamer-vaapi.doap:
5535 2019-04-19 00:38:12 +0100 Tim-Philipp Müller <tim@centricular.com>
5537 * docs/plugins/inspect/plugin-vaapi.xml:
5540 2019-04-15 19:34:05 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5542 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5543 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
5544 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5545 libs: encoder: h264,h265: guard VA version for max_qp property
5546 This patch fixes a regression from commit 5b1fe9c6.
5547 max_qp, in rate control configuration, appeared in libva release
5548 2.1 (API 1.1), thus it is required to guard the VA API version.
5551 2019-04-08 18:29:35 +0800 He Junyan <junyan.he@hotmail.com>
5553 * gst-libs/gst/vaapi/gstvaapiimage.c:
5554 * gst-libs/gst/vaapi/video-format.c:
5555 libs: Add RGB565 image format support.
5557 2019-04-10 13:59:05 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5560 build: configure: delay USE_GTK conditional until check libva-x11
5561 libva-x11 is used for X11 applications, so it is required to build
5562 any GTK application.
5563 Later, when Wayland test is added, we should change this.
5565 2019-04-10 13:25:10 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5568 build: configure: disable GLX if libva-x11 is not found
5570 2019-04-15 13:55:26 +0200 He Junyan <junyan.he@hotmail.com>
5572 * gst-libs/gst/vaapi/gstvaapiutils.c:
5573 libs: utils: avoid macro evaluation when stringify
5574 string_of_va_chroma_format() gets a wrong string format description.
5575 For example, the YUV420_10BPP get a string of 0x00000100 as output.
5576 It's because VA_RT_FORMAT_xxx values are macro definitions. And
5577 STRINGIFY(VA_RT_FORMAT_xxx) will expand to its real value
5579 To avoid the macro evaluation, it is changed to show only the color
5580 format without VA_RT_FORMAT_ prefix.
5582 2019-04-15 13:54:15 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5584 * gst-libs/gst/vaapi/gstvaapiutils.c:
5585 libs: utils: use glib's macros
5586 Don't reinvent the wheel.
5588 2019-04-11 15:05:02 +0800 Wangfei <fei.w.wang@intel.com>
5590 * gst/vaapi/gstvaapipluginutil.c:
5591 plugins: find the preferred format from right caps.
5592 When the downstream has any caps, then raw video feature will
5593 be used. At this situation, the preferred format should be chose
5594 from caps which contains "vide/x-raw" feature instead of from
5595 the fist allowed caps.
5598 2019-04-10 11:43:33 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
5600 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5601 libs: encoder: h265: fill tier in va seq param buf
5602 Now that tier is calculated in commit 58e74f9440fe (!68),
5603 ensure we fill in the general_tier_flag in the
5604 VAEncSequenceParameterBufferHEVC.
5606 === release 1.15.90 ===
5608 2019-04-11 00:40:03 +0100 Tim-Philipp Müller <tim@centricular.com>
5614 * gstreamer-vaapi.doap:
5618 2019-04-11 00:40:03 +0100 Tim-Philipp Müller <tim@centricular.com>
5620 * docs/plugins/inspect/plugin-vaapi.xml:
5623 2019-04-09 20:42:04 +0800 He Junyan <junyan.he@hotmail.com>
5625 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5626 libs: encoder: h265: Recognize the correct level and tier.
5627 The current manner can not recognize the correct level and always
5628 set the tier to main. Need to add frame rate check to recognize
5629 levels such as 4.1, 6.2, etc. We also add a logic to check main
5630 and high tier based on bitrate.
5633 2019-04-03 14:12:23 +0800 He Junyan <junyan.he@hotmail.com>
5635 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5636 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
5637 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5638 libs: encoder: h264,h265: Set max_qp if min_qp is non-zero.
5639 media-driver currently fails to set a correct value of max_qp when
5640 min_qp is different to zero, in CBR and VBR mode, generating full
5641 quality frames, thus unexpected huge output.
5642 This patch sets max_qp to an arbitrary value to avoid this output
5646 2019-04-09 12:42:56 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5648 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5649 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
5650 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5651 libs: encoder: h264,h265: initial and minimal QP can be zero
5652 Currently the minimal value for either min_qp and init_qp are 1,
5653 but VA documentation specifiy that zero is also valid and means
5654 to ignore the quantiser.
5655 The default value is not changed though to avoid behaivor changes
5658 2019-04-09 09:20:23 +0800 Haihao Xiang <haihao.xiang@intel.com>
5660 * tests/elements/meson.build:
5661 meson: build test-vaapicontext when using X11
5662 x11_dep and libva_x11_dep are optional and meson ignores these
5663 dependencies even if they are added into the dependency list.
5664 This fixes the error below when libva-x11 is not avaiblabe:
5665 cc -Itests/elements/tests@elements@@test-vaapicontext@exe
5666 -Itests/elements -I../../gstreamer-vaapi/tests/elements -I.
5667 -I../../gstreamer-vaapi/ -Igst-libs -I../../gstreamer-vaapi/gst-libs
5668 -I/usr/include/gstreamer-1.0 -I/usr/include/glib-2.0
5669 -I/usr/lib/x86_64-linux-gnu/glib-2.0/include -I/usr/include/orc-0.4
5670 -I/usr/include/at-spi2-atk/2.0 -I/usr/include/at-spi-2.0
5671 -I/usr/include/dbus-1.0 -I/usr/lib/x86_64-linux-gnu/dbus-1.0/include
5672 -I/usr/include/gtk-3.0 -I/usr/include/gio-unix-2.0/
5673 -I/usr/include/libdrm -I/usr/include/harfbuzz -I/usr/include/pango-1.0
5674 -I/usr/include/fribidi -I/usr/include/atk-1.0 -I/usr/include/cairo
5675 -I/usr/include/pixman-1 -I/usr/include/uuid -I/usr/include/freetype2
5676 -I/usr/include/libpng16 -I/usr/include/gdk-pixbuf-2.0
5677 -fdiagnostics-color=always -pipe -D_FILE_OFFSET_BITS=64 -Wall
5678 -Winvalid-pch -O2 -g -fvisibility=hidden -fno-strict-aliasing -pthread
5679 -DHAVE_CONFIG_H -MD -MQ
5680 'tests/elements/tests@elements@@test-vaapicontext@exe/test-vaapicontext.c.o'
5682 'tests/elements/tests@elements@@test-vaapicontext@exe/test-vaapicontext.c.o.d'
5684 'tests/elements/tests@elements@@test-vaapicontext@exe/test-vaapicontext.c.o'
5685 -c ../../gstreamer-vaapi/tests/elements/test-vaapicontext.c
5686 ../../gstreamer-vaapi/tests/elements/test-vaapicontext.c:29:10: fatal
5687 error: va/va_x11.h: No such file or directory
5688 #include <va/va_x11.h>
5690 2019-04-01 12:56:28 +0800 He Junyan <junyan.he@hotmail.com>
5692 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
5693 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
5694 * gst-libs/gst/vaapi/gstvaapifeipak_h264.c:
5695 libs: encoder: h264_fei: Use gst_param_spec_array for view-ids
5696 GValueArray is deprecated. Use GstValueArray instead.
5698 2019-03-30 18:29:31 +0100 Danilo Spinella <danyspin97@protonmail.com>
5700 * gst/vaapi/gstvaapipluginutil.c:
5701 vaapipluginutil: Fix #endif for USE_X11
5703 2019-03-29 18:29:51 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5705 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5706 libs: encoder: h264: simplify the view-ids setting
5708 2019-03-26 14:54:47 +0800 He Junyan <junyan.he@hotmail.com>
5710 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5711 libs: encoder: h264: Use gst_param_spec_array for view-ids
5712 GValueArray is deprecated. Use GstValueArray instead.
5713 gst_param_spec_array can be deserialized from command line using:
5714 vaapih264enc view-ids="<(uint)40,(uint)100>" num-views=2
5715 While the g_param_spec_value_array() can not, and always get
5716 error: "gst_value_deserialize_g_value_array: unimplemented"
5717 Also fixed an out-of-range bug.
5719 2019-03-29 13:33:41 +0800 He Junyan <junyan.he@hotmail.com>
5721 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
5722 libs: Change the parameter setting order when encode picture.
5723 The order in gst_vaapi_enc_picture_encode when encoding one
5724 picture is not very correct. The misc parameters are set before
5725 the picture parameters. Some of the misc parameters such as
5726 ROI may change the current picture parameters. But the later
5727 setting of picture parameter will re-init all picture related
5728 parameters and clear the previous setting. The right order
5729 should be picture parameter first and then misc parameters.
5730 Signed-off-by: He Junyan <junyan.he@hotmail.com>
5732 2019-03-26 14:20:34 +0800 Wangfei <fei.w.wang@intel.com>
5734 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
5735 libs: decoder: jpeg: support dynamic resolution change decode.
5736 Add size_changed flag to watch out resolution. if change, reset
5737 jpeg decoder's context.
5739 2019-03-23 15:34:03 +0800 Wangfei <fei.w.wang@intel.com>
5741 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5742 libs: encoder: h265: add low power mode encode.
5743 By now, this feature only support by media-driver on Ice Lake
5744 platform, more information you can reference:
5745 https://github.com/intel/media-driver
5747 2019-03-15 18:40:21 +0800 He Junyan <junyan.he@hotmail.com>
5749 * gst/vaapi/gstvaapiencode.c:
5750 vaapiencode: gobject's prop_id differ from vaapi encoder
5751 The vaapi internal encoder's property id are negative, thus they are
5752 different from GObject's property ids.
5753 gst_vaapi_encoder_set_property() should map to the internal encoder
5754 property id, assigned in gst_vaapiencode_default_set_property().
5756 2019-03-21 16:56:34 +0000 Tim-Philipp Müller <tim@centricular.com>
5759 meson: disable compiler warnings for unused vars and args if gst debug system is disabled
5761 2019-03-21 13:31:57 +0000 Tim-Philipp Müller <tim@centricular.com>
5764 meson: use new 'python' module instead of deprecated 'python3' one
5766 2019-03-11 18:38:36 -0300 Thibault Saunier <tsaunier@igalia.com>
5769 Update common submodule back to 59cb678164719ff59dcf6c8b93df4617a1075d11
5770 It was wrongly changed in 3d9555a86d45565870c684fe00ec8bbb0fed7205
5772 2019-03-04 09:16:17 +0000 Tim-Philipp Müller <tim@centricular.com>
5777 * docs/plugins/inspect/plugin-vaapi.xml:
5781 2019-03-01 12:33:26 +0800 He Junyan <junyan.he@hotmail.com>
5783 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
5784 libs: Fix a typo in comments.
5785 Fix a typo in function description of
5786 gst_vaapi_surface_pool_new_with_chroma_type.
5787 Signed-off-by: He Junyan <junyan.he@hotmail.com>
5789 2019-02-27 13:02:10 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5791 * gst/vaapi/gstvaapipluginutil.c:
5792 plugin: if any caps in downstream, negotiate raw video
5793 When downstream has any caps, vaapi should not shovel vaapi featured
5794 buffers, but rather plain raw video, assuming always the worst case
5795 scenario (downstream cannot handle featured video memory but raw
5796 system memory buffers).
5797 This patch query the peer caps without any filter, to know if
5798 donwstream just ask for any caps, if so jump to the color space
5799 checking, otherwise do the caps intersection and continue with the
5800 feature selection algorithm.
5803 === release 1.15.2 ===
5805 2019-02-26 12:01:53 +0000 Tim-Philipp Müller <tim@centricular.com>
5811 * gstreamer-vaapi.doap:
5815 2019-02-26 12:01:53 +0000 Tim-Philipp Müller <tim@centricular.com>
5817 * docs/plugins/inspect/plugin-vaapi.xml:
5820 2019-02-05 16:59:40 +0800 He Junyan <junyan.he@hotmail.com>
5822 * gst/vaapi/gstvaapivideomemory.c:
5823 vaapivideomemory: Prefer same format for surface and image
5824 We prefer to use the same format between image and surface for gst
5825 vaapi allocator. The old way may choose different formats between
5826 image and surface. For example, the RGBA image may have a NV12 surface.
5827 So we need to do format conversion when we put/get image to surface.
5828 Some drivers such as iHD can not support such conversion and always
5829 cause a data flow error. There may also have some performance cost
5830 for format conversion when put/get images.
5831 So we prefer to use the same format for image and surface in the
5832 allocator. If the surface can not support that format, we then
5833 fallback to find a best one as the surface format.
5834 Co-authored-by: Víctor Jáquez <vjaquez@igalia.com>
5836 2019-02-15 15:19:51 +0800 He Junyan <junyan.he@hotmail.com>
5838 * gst-libs/gst/vaapi/video-format.c:
5839 libs: Delete the duplicated ARGB video format.
5840 Two ARGB formats with the same format information.
5841 Should be verbose and delete one.
5842 Signed-off-by: He Junyan <junyan.he@hotmail.com>
5844 2019-02-13 10:39:59 -0500 Adam Jackson <ajax@redhat.com>
5847 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
5848 glx: Stop specifying GLX_DEPTH_SIZE
5849 This code is just confused. It's asking for at least as many bits of
5850 (z-axis) depth as the root window has bits of (color) depth. For rgb565
5851 or rgb888 this is harmless, but at 10 bits per channel this demands a
5852 30-bit or deeper Z buffer. While some hardware could in principle do a
5853 32-bit Z buffer, Mesa does not expose such fbconfigs (at least on Intel
5855 We're not actually using the Z buffer, so just stop asking for one.
5857 2019-01-14 11:30:48 +0100 Niels De Graef <niels.degraef@barco.com>
5860 * gst-libs/gst/vaapi/Makefile.am:
5861 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
5862 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
5863 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
5864 * gst-libs/gst/vaapi/meson.build:
5865 * gst/vaapi/gstvaapisink.c:
5867 libs: wayland: add support for XDG-shell protocol
5868 [wl_shell] is officially [deprecated], so provide support for the
5869 XDG-shell protocol should be provided by all desktop-like compositors.
5870 (In case they don't, we can of course fall back to wl_shell).
5871 Note that the XML file is directly provided by the `wayland-protocols`
5872 dependency and generates the protocol marshalling code.
5873 [wl_shell]: https://people.freedesktop.org/~whot/wayland-doxygen/wayland/Client/group__iface__wl__shell.html
5874 [deprecated]: https://github.com/wayland-project/wayland/commit/698dde195837f3d0844b2725ba4ea8ce9ee7518c
5876 2019-02-16 19:09:50 +0100 Niels De Graef <nielsdegraef@gmail.com>
5878 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
5879 libs: window: wayland: Prefix wl_shell_surface field with `wl_`
5880 It will help us to distinguish from other Wayland shell surface
5881 (such as XDG-shell) later on.
5883 2019-01-14 09:58:19 +0100 Niels De Graef <nielsdegraef@gmail.com>
5885 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
5886 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
5887 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
5888 libs: wayland: Prefix wl_shell field with `wl_`
5889 It will help us to distinguish from other Wayland shells (such as
5890 XDG-shell) later on.
5892 2019-02-08 09:21:28 +0300 Denis Nagorny <denis.nagorny@intel.com>
5894 * gst-libs/gst/vaapi/gstvaapidisplay.c:
5895 libs: display: lock ensure_profile()
5896 Thread safety patch for ensure_profile() function
5899 2019-02-08 16:35:39 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5902 meson: bump the minimum wayland version requirement to 1.11.0
5903 This was missed on commit 77bb3424
5905 2019-01-24 21:08:07 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5907 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
5908 * gst/vaapi/gstvaapisink.c:
5909 vaapisink: x11: trap WM_DELETE_WINDOW message
5910 Register the WM_DELETE_WINDOW message from window manager and
5911 trap it to stop the pipeline cleanly.
5912 Fixes: https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/issues/130
5914 2019-01-21 19:22:58 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5916 * gst-libs/gst/vaapi/gstvaapiwindow.c:
5917 libs: window: remove native-id property
5918 native-id property is problematic since the variable that stores it is
5919 gsize, which is platform specific, and in some is bigger than unsigned
5920 long, and there are not way to handle gsize properties.
5921 Also, GST_VAAPI_ID_INVALID is defined in gsize terms, and we would
5922 like to keep using it for this scope.
5923 This patch removes the native-id property and set it manually in
5924 gst_vaapi_window_new_internal().
5926 2019-01-18 10:33:37 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5928 * gst-libs/gst/vaapi/gstvaapiwindow.c:
5929 libs: window: use G_GSIZE_MODIFIER for window id
5930 gsize type is not equal in all platforms, then the 'l' print modifier
5931 shall not be used always.
5932 This issue was found in Debian builds.
5934 2019-01-17 10:27:13 +0800 Wangfei <fei.w.wang@intel.com>
5936 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5937 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5938 encoder: h264/h265: set SPS cbr_flag with correct value.
5939 The flag only set as 1 when the rate-control mode is CBR.
5941 === release 1.15.1 ===
5943 2019-01-17 02:36:52 +0000 Tim-Philipp Müller <tim@centricular.com>
5949 * gstreamer-vaapi.doap:
5953 2019-01-17 02:36:52 +0000 Tim-Philipp Müller <tim@centricular.com>
5955 * docs/plugins/inspect/plugin-vaapi.xml:
5958 2019-01-14 19:35:34 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5960 * gst-libs/gst/vaapi/gstvaapiencoder.c:
5961 libs: encoder: refactor to avoid code duplication
5962 gst_vaapi_encoder_put_frame() and gst_vaapi_encoder_flush() duplicates
5963 the same code segment where the coded buffer is created, the picture
5964 encoded on it and pushed to the async queue.
5965 The function gst_vaapi_encoder_encode_and_queue() refactor this.
5967 2019-01-14 18:21:30 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
5969 * gst-libs/gst/vaapi/gstvaapiencoder.c:
5970 * gst-libs/gst/vaapi/gstvaapiencoder.h:
5971 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5972 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5973 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
5974 libs: encoder: h264/h265: flush pending ordered pictures
5975 In order to flush the pending pictures, a new internal encoder vmethod
5976 is used: get_pending_reordered()
5977 This method follows an iterator pattern which will return the next
5978 picture to encode and push.
5979 The base encoder will call this function in a loop when flush() is called.
5980 For now, only H.264 and H.265 encoders implement this flushing mechanism.
5982 2018-12-06 10:18:53 +0800 Wangfei <fei.w.wang@intel.com>
5984 * gst-libs/gst/vaapi/gstvaapiencoder.c:
5985 * gst-libs/gst/vaapi/gstvaapiencoder.h:
5986 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
5987 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
5988 libs: encoder: h264/h265: fix encode lose frame issue.
5989 Instead of dropping all remain frames in reorder_frame_list during
5990 flush, keep encoding.
5991 https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/issues/97
5993 2019-01-15 14:33:11 +0800 Wangfei <fei.w.wang@intel.com>
5995 * gst/vaapi/gstvaapipostproc.c:
5996 vaapipostproc: before set surface proxy, check if it already been created and exist.
5997 Fix the deinterlace black frame when playing with glimagesink:
5998 gst-launch-1.0 filesrc location=test.264 ! h264parse ! vaapih264dec \
5999 ! vaapipostproc deinterlace-mode=1 deinterlace-method=1 ! glimagesink
6001 2019-01-11 13:48:29 +0800 Wangfei <fei.w.wang@intel.com>
6003 * gst-libs/gst/vaapi/gstvaapiutils.c:
6004 vaapipostproc: clean up USE_VA_VPP macro since it already removed from configure file.
6006 2018-12-26 14:36:23 +0800 Haihao Xiang <haihao.xiang@intel.com>
6008 * gst-libs/gst/vaapi/meson.build:
6009 * gst/vaapi/meson.build:
6011 * tests/meson.build:
6012 meson: build h264 fei encoder if possible
6014 2018-12-26 14:04:08 +0800 Haihao Xiang <haihao.xiang@intel.com>
6017 configure: bump the minimum wayland version requirement to 1.11.0
6019 2018-12-24 12:58:53 +0800 Haihao Xiang <haihao.xiang@intel.com>
6022 * gst-libs/gst/vaapi/Makefile.am:
6023 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
6024 * gst-libs/gst/vaapi/gstvaapibufferproxy_priv.h:
6025 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
6026 * gst-libs/gst/vaapi/gstvaapicompat.h:
6027 * gst-libs/gst/vaapi/gstvaapicontext.c:
6028 * gst-libs/gst/vaapi/gstvaapidisplay.c:
6029 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
6030 * gst-libs/gst/vaapi/gstvaapifilter.c:
6031 * gst-libs/gst/vaapi/gstvaapiprofile.c:
6032 * gst-libs/gst/vaapi/gstvaapisurface.c:
6033 * gst-libs/gst/vaapi/gstvaapiutils.c:
6034 * gst-libs/gst/vaapi/gstvaapiutils_core.c:
6035 * gst-libs/gst/vaapi/meson.build:
6036 * gst/vaapi/Makefile.am:
6037 * gst/vaapi/gstvaapi.c:
6038 * gst/vaapi/gstvaapidecode.c:
6039 * gst/vaapi/gstvaapidecodebin.c:
6040 * gst/vaapi/gstvaapipluginbase.c:
6041 * gst/vaapi/meson.build:
6044 * tests/simple-decoder.c:
6045 vaapi: bump the minimum vaapi version requirement to 0.39.0
6046 And reduce unnecessary API version and structures check as well.
6047 https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/issues/108
6049 2018-12-22 18:07:35 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6051 * gst-libs/gst/vaapi/gstvaapiwindow.c:
6052 * gst-libs/gst/vaapi/gstvaapiwindow.h:
6053 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
6054 * tests/test-decode.c:
6055 * tests/test-filter.c:
6056 * tests/test-subpicture.c:
6057 * tests/test-textures.c:
6058 * tests/test-windows.c:
6059 libs: window: remove custom ref() and unref()
6060 Use gst_object_ref() and gst_object_unref() instead.
6062 2018-12-22 13:25:09 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6064 * gst-libs/gst/vaapi/gstvaapiwindow.c:
6065 * gst-libs/gst/vaapi/gstvaapiwindow_drm.c:
6066 * gst-libs/gst/vaapi/gstvaapiwindow_egl.c:
6067 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
6068 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
6069 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
6070 libs: window: use its own debug category
6072 2018-12-22 18:02:38 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6074 * gst-libs/gst/vaapi/gstvaapiwindow.c:
6075 * gst-libs/gst/vaapi/gstvaapiwindow.h:
6076 * gst-libs/gst/vaapi/gstvaapiwindow_drm.c:
6077 * gst-libs/gst/vaapi/gstvaapiwindow_drm.h:
6078 * gst-libs/gst/vaapi/gstvaapiwindow_egl.c:
6079 * gst-libs/gst/vaapi/gstvaapiwindow_egl.h:
6080 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
6081 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
6082 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
6083 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
6084 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.h:
6085 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
6086 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
6087 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
6088 libs: window: refactor as gobject
6089 This is another step in the gobjectification of the internal library
6090 of gstreamer-vaapi. Now it is the turn of GstVaapiWindow and its
6092 The idea is to minimize the changeset keeping the same design as
6094 GstVaapiWindow is defined as an abstract class with two properties:
6095 the GstVaapiDisplay and the native ID. Thus, many of the
6096 GstVaapiObject macros were copied as GstVaapiWindow macros.
6097 The function gst_vaapi_window_new_internal() is kept as a decorator
6098 of for calling gst_vaapi_window_create() and the possibility of
6100 The descendant classes, such as glx, still use the private
6101 structures, but through the gobject mechanism.
6103 2018-12-03 22:05:29 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6105 * gst-libs/gst/vaapi/gstvaapifilter.c:
6106 libs: filter: use its own debug category
6108 2018-12-24 14:08:42 +0800 He Junyan <junyan.he@hotmail.com>
6110 * gst/vaapi/gstvaapidecode.c:
6111 * gst/vaapi/gstvaapipluginbase.c:
6112 * gst/vaapi/gstvaapipluginbase.h:
6113 * gst/vaapi/gstvaapipostproc.c:
6114 * gst/vaapi/gstvaapisink.c:
6115 plugins: Add more check for allowed raw caps.
6116 The gst_vaapi_plugin_base_get_allowed_raw_caps is used for both sink
6117 pad and src pad, which cause some bugs. For sink pad, we need to verify
6118 vaPutImage() while for the src pad we need to verify vaGetImage().
6119 For vaapidecoderXXX kind of plugins, the case is more complex. We need
6120 to verify whether the decoded result(in some surface, NV12 format most
6121 of the time) can be vaGetImage to some raw image format. Add more check
6122 to fix all these problems.
6123 https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/issues/123
6124 Signed-off-by: He Junyan <junyan.he@hotmail.com>
6126 2018-12-18 10:44:21 +0800 Wangfei <fei.w.wang@intel.com>
6128 * gst/vaapi/gstvaapipostproc.c:
6129 vaapipostproc: fix csc fail when only change width or height.
6131 2018-12-15 09:47:15 +0900 Wonchul Lee <chul0812@gmail.com>
6133 * tests/elements/meson.build:
6134 meson: Add gtk guard
6136 2018-12-15 14:48:03 +0800 Wangfei <fei.w.wang@intel.com>
6138 * gst/vaapi/gstvaapiencode_h264.c:
6139 libs: enc: h264: set max profile idc with correct profile.
6140 Use the highest rank of available profile as the max profile to
6142 https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/issues/124
6144 2018-12-03 13:56:52 +0100 Niels De Graef <nielsdegraef@gmail.com>
6146 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
6147 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
6148 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
6149 Use G_DEFINE_TYPE_WITH_PRIVATE if applicable
6150 This gets rid of the strange `do_init` macro and makes the intent a bit
6153 2018-12-05 17:24:53 -0300 Thibault Saunier <tsaunier@igalia.com>
6156 Automatic update of common submodule
6157 From ed78bee to 59cb678
6159 2018-11-27 09:47:44 -0500 Wangfei <fei.w.wang@intel.com>
6161 * gst-libs/gst/vaapi/gstvaapiimage.c:
6162 * gst-libs/gst/vaapi/gstvaapiprofile.c:
6163 * gst-libs/gst/vaapi/gstvaapiprofile.h:
6164 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
6165 * gst-libs/gst/vaapi/video-format.c:
6166 * gst/vaapi/gstvaapidecode.c:
6167 * gst/vaapi/gstvaapipluginutil.h:
6168 libs: dec: h265: support decode for main-444 10bit streams.
6169 Add 444 10bit yuv format Y410, which can be used to decode
6170 main-444 10bit streams. Currently, this feature is only
6171 supported by media-driver in Icelake.
6173 2018-11-28 05:56:44 +0200 Jordan Petridis <jordan@centricular.com>
6175 * gst/vaapi/gstvaapidecode.c:
6176 * gst/vaapi/gstvaapisink.c:
6177 Run gst-indent through the files
6178 This is required before we enabled an indent test in the CI.
6179 https://gitlab.freedesktop.org/gstreamer/gstreamer-project/issues/33
6181 2018-11-14 13:11:56 +0800 He Junyan <junyan.he@hotmail.com>
6183 * gst/vaapi/gstvaapipluginbase.c:
6184 plugins: modify image check of extract_allowed_surface_formats.
6185 The extract_allowed_surface_formats function just check whether
6186 we can support some kind of surface/image format pair. We just
6187 need to create a surface, create an image with the same video-format
6188 and putImage from image to surface. All these operations success,
6189 that kind of video-format is supported.
6190 The old manner do not work for some kind of video-format. For example,
6191 the RGBA kind of format will create a NV12 surface and RGBA image,
6192 and the putImage will fail because the format is not same. And so
6193 the RGBA format is not supported but actually it is supported.
6195 2018-11-14 11:34:20 +0100 Michael Olbrich <m.olbrich@pengutronix.de>
6197 * gst/vaapi/gstvaapipostproc.c:
6198 vaapipostproc: add some missing locking
6199 gst_vaapi_plugin_base_close() removed the raw caps that are used indirectly
6200 in gst_vaapipostproc_transform_caps(). The usage is already protected by
6202 This is needed when the pipeline is stopped during startup.
6204 2018-11-20 16:07:44 +0800 Xiang, Haihao <haihao.xiang@intel.com>
6206 * gst/vaapi/gstvaapivideomemory.c:
6208 Otherwise it will result in resource leak when failed to create
6211 2018-11-12 13:39:51 +0100 Michael Olbrich <m.olbrich@pengutronix.de>
6213 * gst/vaapi/gstvaapiencode.c:
6214 vaapiencode: don't start src pad task in set_format
6215 Otherwise the task may be restarted during shutdown. Start the task in
6216 gst_vaapiencode_handle_frame() instead.
6218 2018-11-14 13:52:48 +0800 Wangfei <fei.w.wang@intel.com>
6220 * gst-libs/gst/vaapi/gstvaapiprofile.c:
6221 * gst-libs/gst/vaapi/gstvaapiprofile.h:
6222 * gst-libs/gst/vaapi/gstvaapiutils.c:
6223 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
6224 * gst/vaapi/gstvaapidecode.c:
6225 * gst/vaapi/gstvaapipluginutil.h:
6226 libs: dec: h265: support decode for main-444 8bit streams.
6227 Add 444 8bit yuv format AYUV, which can be used to decode
6228 main-444 8bit streams. Currently, this feature is only
6229 supported by media-driver in Icelake.
6230 https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/issues/119
6232 2018-11-12 17:43:54 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6235 Add Gitlab CI configuration
6236 This commit adds a .gitlab-ci.yml file, which uses a feature
6237 to fetch the config from a centralized repository. The intent is
6238 to have all the gstreamer modules use the same configuration.
6239 The configuration is currently hosted at the gst-ci repository
6240 under the gitlab/ci_template.yml path.
6241 Part of https://gitlab.freedesktop.org/gstreamer/gstreamer-project/issues/29
6243 2018-11-09 22:03:43 +0800 He Junyan <junyan.he@hotmail.com>
6245 * gst-libs/gst/vaapi/gstvaapisurface.h:
6246 * gst-libs/gst/vaapi/gstvaapiutils.c:
6247 libs: Sync the GstVaapiChromaType to VA header file.
6248 Add more kinds of chrometype which will be used to describe
6249 new video formats. Sync it with 1.4.0 version header file.
6250 Alse delete useless GST_VAAPI_CHROMA_TYPE_YUV410 chrome type.
6251 Signed-off-by: He Junyan <junyan.he@hotmail.com>
6253 2018-11-09 23:55:05 +0000 Tim-Philipp Müller <tim@centricular.com>
6255 * gst-libs/gst/vaapi/meson.build:
6256 meson: link with -lm
6257 Fixes #117 hopefully.
6259 2018-11-09 23:46:53 +0000 Tim-Philipp Müller <tim@centricular.com>
6262 meson: bump meson required to 0.47 for feature options
6264 2018-11-06 14:38:08 +0800 Junyan He <junyan.he@intel.com>
6266 * gst-libs/gst/vaapi/video-format.c:
6267 libs: Modify the video format of endianness.
6268 We lack some video format because endianness declare.
6269 The video format should not directly relate to endianness. For example,
6270 ARGB on big endian should not be simplely seen as BGRA on little endian
6271 machine. We should provide endianess convert or format convert help
6272 functions if endianness does not match.
6273 https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/issues/112
6274 Signed-off-by: Junyan He <junyan.he@intel.com>
6276 2018-10-17 18:36:52 +0800 Junyan He <junyan.he@intel.com>
6278 * gst/vaapi/gstvaapipluginutil.c:
6279 plugins: Fix build error when GL is enabled while EGL is disabled.
6280 gl_platform_type in gst_vaapi_get_display_type_from_gl_env generate
6281 unused-variable warning and may block build when Werror enabled.
6282 Several functions like gst_vaapi_display_egl_new_with_native_display
6283 have no prototype warning and link error when GL is enabled but EGL
6284 is disabled. Fix all these warning and link error.
6285 https://bugzilla.gnome.org/show_bug.cgi?id=797358
6286 Signed-off-by: Junyan He <junyan.he@intel.com>
6288 2018-11-03 15:06:09 +0800 Wangfei <fei.w.wang@intel.com>
6290 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
6291 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
6292 libs: encoder: h264/h264fei: remove unuseless code.
6293 The variable are set twice, remove previous one.
6294 https://bugzilla.gnome.org/show_bug.cgi?id=797365
6296 2018-11-03 15:28:35 +0800 Wangfei <fei.w.wang@intel.com>
6298 * tests/simple-encoder.c:
6299 * tests/test-fei-enc-in.c:
6300 tests: check return value when using gst_buffer_map.
6301 https://bugzilla.gnome.org/show_bug.cgi?id=797366
6303 2018-11-02 16:50:47 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6306 * meson_options.txt:
6307 * tests/elements/meson.build:
6308 * tests/meson.build:
6309 build: meson: build examples
6311 2018-11-02 16:50:00 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6313 * gst-libs/gst/vaapi/meson.build:
6314 build: meson: declare headers for libgstvaapi
6315 Thus handling its recompilation if needed.
6317 2018-11-05 05:41:13 +0000 Matthew Waters <matthew@centricular.com>
6320 Update common submodule location
6321 Remove the git directory
6323 2018-11-05 13:00:28 +0800 Haihao Xiang <haihao.xiang@intel.com>
6326 * gstreamer-vaapi.doap:
6327 Clone the code from gitlab
6328 This fixes https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/issues/116
6330 2018-10-24 14:18:37 -0400 Wangfei <fei.w.wang@intel.com>
6332 * gst-libs/gst/vaapi/gstvaapiimage.c:
6333 * gst-libs/gst/vaapi/gstvaapisurface.h:
6334 * gst-libs/gst/vaapi/gstvaapiutils.c:
6335 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
6336 * gst-libs/gst/vaapi/video-format.c:
6337 * gst/vaapi/gstvaapidecode.c:
6338 * gst/vaapi/gstvaapipluginutil.h:
6339 libs: dec: h265: support decode for main-10-422 10bit streams.
6340 Add 422 10bit yuv format Y210, which can be used to decode
6341 main-10-422 10bit streams. Currently, this feature is only
6342 supported by media-driver in Icelake.
6343 https://bugzilla.gnome.org/show_bug.cgi?id=797264
6345 2018-10-13 15:00:32 +0800 Wangfei <fei.w.wang@intel.com>
6347 * gst-libs/gst/vaapi/gstvaapicontext.c:
6348 libs: context: roi_rc_qp_delta_support should not be checked when CQP.
6349 VA_ROI_RC_QP_DELTA_SUPPORT return value will be ignored when the
6350 rate control mode is set as CQP. In CQP mode, it shouldn't check
6351 roi_rc_qp_delta_support return value from driver backend.
6352 https://bugzilla.gnome.org/show_bug.cgi?id=797087
6354 2018-10-15 17:55:24 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6356 * gst/vaapi/gstvaapipostproc.c:
6357 vaapipostproc: fix classification string
6358 The classification string is splitted by '/' and then looks for the
6360 This patch removes the ';' by unifying all the components.
6362 2018-10-15 16:05:02 +0100 Philippe Normand <philn@igalia.com>
6364 * gst/vaapi/gstvaapipostproc.c:
6365 vaapipostproc: Add Hardware classifier to metadata
6367 2018-10-12 16:37:34 +0800 Wangfei <fei.w.wang@intel.com>
6369 * gst-libs/gst/vaapi/gstvaapicontext.c:
6370 libs: context: create context first before using it to create surface.
6371 In gst_vaapi_context_reset(), if the context has to be destroyed, make
6372 sure to create it first before allocating its associated surfaces.
6373 This patch fixes a regression introduced in commit 82872f4 because
6374 the formats available in the current context now are ensured before
6375 creating the context's surfaces.
6376 https://bugzilla.gnome.org/show_bug.cgi?id=797277
6378 2018-10-12 15:39:53 +0100 Philippe Normand <philn@igalia.com>
6380 * docs/plugins/inspect/plugin-vaapi.xml:
6381 * gst/vaapi/gstvaapidecode.c:
6382 * gst/vaapi/gstvaapiencode_h264.c:
6383 * gst/vaapi/gstvaapiencode_h264_fei.c:
6384 * gst/vaapi/gstvaapiencode_h265.c:
6385 * gst/vaapi/gstvaapiencode_jpeg.c:
6386 * gst/vaapi/gstvaapiencode_mpeg2.c:
6387 * gst/vaapi/gstvaapiencode_vp8.c:
6388 * gst/vaapi/gstvaapiencode_vp9.c:
6389 gst: Advertise elements interacting with hardware devices
6391 2018-10-01 09:26:05 +0800 Wangfei <fei.w.wang@intel.com>
6393 * gst-libs/gst/vaapi/gstvaapicontext.c:
6394 * gst-libs/gst/vaapi/gstvaapisurface.c:
6395 * gst-libs/gst/vaapi/gstvaapisurface.h:
6396 libs: context: query surface format before context to create surface.
6397 Before using context to create surface, the supported surface format
6398 should be checked first.
6399 https://bugzilla.gnome.org/show_bug.cgi?id=797222
6401 2018-10-09 17:23:55 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6403 * gst-libs/gst/vaapi/gstvaapiimage.c:
6404 * gst-libs/gst/vaapi/gstvaapivalue.c:
6405 libs: replace g_error with GST_ERROR
6406 And handle those errors rather than halting.
6408 2018-10-09 17:23:30 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6410 * gst-libs/gst/vaapi/gstvaapiimage.c:
6411 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
6412 * gst-libs/gst/vaapi/gstvaapisurface.c:
6413 libs: replace g_warning with GST_WARNING
6415 2018-09-26 14:55:32 -0500 Matteo Valdina <matteo.valdina@gmail.com>
6417 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
6418 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
6419 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
6420 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2.c:
6421 libs: Move from g_debug to GST_DEBUG.
6422 https://bugzilla.gnome.org/show_bug.cgi?id=797202
6424 2018-10-04 02:20:10 +0800 Soon, Thean Siew <thean.siew.soon@intel.com>
6426 * gst/vaapi/gstvaapipostproc.c:
6427 vaapipostproc: change the way of handling deinterlace
6428 The current vaapipostproc calls driver's video processing
6429 pipeline for deinterlacing only if it is Advance deinterlacing.
6430 Modify in the way that it always tries with driver's video
6431 processing pipeline for deinterlacing, and falls back to software
6432 method of appending picture structure meta data only if it fails
6433 with driver's method.
6434 https://bugzilla.gnome.org/show_bug.cgi?id=797095
6436 2018-09-24 16:54:29 -0500 Matteo Valdina <matteo.valdina@gmail.com>
6438 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
6439 * gst-libs/gst/vaapi/gstvaapiutils_h264.h:
6440 libs: h264: Update level table to "Recommendation H.264 (04/17)".
6441 Added level 6, 6.1 and 6.2. Reference Table A-1 – Level limits
6442 from T-REC-H.264-201704.
6443 https://bugzilla.gnome.org/show_bug.cgi?id=797202
6445 2018-09-20 09:57:33 +0800 Wangfei <fei.w.wang@intel.com>
6447 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
6448 * gst-libs/gst/vaapi/gstvaapiprofile.c:
6449 * gst-libs/gst/vaapi/gstvaapiprofile.h:
6450 * gst-libs/gst/vaapi/gstvaapisurface.c:
6451 * gst-libs/gst/vaapi/gstvaapiutils.c:
6452 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
6453 * gst-libs/gst/vaapi/gstvaapiutils_h265_priv.h:
6454 * gst-libs/gst/vaapi/video-format.c:
6455 * gst-libs/gst/vaapi/video-format.h:
6456 * gst/vaapi/gstvaapidecode.c:
6457 * gst/vaapi/gstvaapipluginbase.c:
6458 * gst/vaapi/gstvaapipluginutil.h:
6459 libs: dec: h265: add 422 chroma format support.
6460 Add main-422-10 profile which support 422 chroma format stream.
6461 Currently, this feature is only supported by media-driver in Icelake.
6462 https://bugzilla.gnome.org/show_bug.cgi?id=797143
6464 2018-09-26 19:34:06 +0200 U. Artie Eoff <ullysses.a.eoff@intel.com>
6466 * tests/y4mreader.c:
6467 tests: include sysdeps.h in compilation unit
6468 Fixes https://bugzilla.gnome.org/show_bug.cgi?id=797204
6469 Signed-off-by: U. Artie Eoff <ullysses.a.eoff@intel.com>
6471 2018-09-26 18:04:53 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6473 * tests/y4mreader.c:
6474 * tests/y4mreader.h:
6475 tests: fix compilation
6476 https://bugzilla.gnome.org/show_bug.cgi?id=797204
6478 2018-09-25 20:28:02 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6480 * tests/y4mreader.h:
6481 tests: don's use sysdeps.h in header
6483 2018-09-14 19:30:56 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6485 * gst-libs/gst/vaapi/gstvaapiutils.h:
6486 libs: utils: no need of include config.h
6488 2018-09-13 18:12:02 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6492 * tests/test-decode.c:
6493 * tests/test-subpicture.c:
6494 tests: remove already include string.h
6495 Since sysdeps.h includes string.h there's no need to include it again.
6497 2018-09-13 18:11:25 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6499 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
6500 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6501 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
6502 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
6503 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
6504 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
6505 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
6506 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
6507 * gst-libs/gst/vaapi/gstvaapidisplay.c:
6508 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
6509 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
6510 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
6511 * gst-libs/gst/vaapi/gstvaapiimage.c:
6512 * gst-libs/gst/vaapi/gstvaapiprofile.c:
6513 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
6514 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
6515 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
6516 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
6517 libs: remove already include string.h
6518 Since sysdeps.h includes string.h there's no need to include it again.
6520 2018-09-13 18:26:27 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6522 * gst-libs/gst/vaapi/gstvaapiobject.h:
6523 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
6524 * gst/vaapi/gstvaapivideometa_texture.c:
6525 libs: object: separation of internal API and plugins
6526 Removed exposed macros GST_VAAPI_OBJECT_DISPLAY() and
6527 GST_VAAPI_OBJECT_ID() to plugins, keeping them only for internal
6529 The purpose is readability.
6530 https://bugzilla.gnome.org/show_bug.cgi?id=797139
6532 2018-09-13 16:34:54 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6534 * gst-libs/gst/vaapi/gstvaapiparser_frame.h:
6535 libs: parser_frame: change macros for inlined functions
6536 https://bugzilla.gnome.org/show_bug.cgi?id=797139
6538 2018-09-13 16:10:13 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6540 * gst-libs/gst/vaapi/gstvaapicodedbufferproxy.c:
6541 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
6542 * gst-libs/gst/vaapi/gstvaapivideopool.c:
6543 * gst-libs/gst/vaapi/gstvaapivideopool_priv.h:
6544 libs: videopool: remove unneeded code
6545 The removed code comes frome the bad practice of copy&paste. Better
6546 move it as internal function.
6547 https://bugzilla.gnome.org/show_bug.cgi?id=797139
6549 2018-09-13 12:22:42 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6551 * gst-libs/gst/vaapi/Makefile.am:
6552 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
6553 * gst-libs/gst/vaapi/gstvaapibufferproxy_priv.h:
6554 * gst-libs/gst/vaapi/gstvaapiminiobject.c:
6555 * gst-libs/gst/vaapi/gstvaapiminiobject.h:
6556 * gst-libs/gst/vaapi/gstvaapiobject.c:
6557 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
6558 * gst-libs/gst/vaapi/gstvaapipixmap.c:
6559 * gst-libs/gst/vaapi/gstvaapipixmap_priv.h:
6560 * gst-libs/gst/vaapi/gstvaapitexture.c:
6561 * gst-libs/gst/vaapi/gstvaapitexture_priv.h:
6562 * gst-libs/gst/vaapi/gstvaapiwindow.c:
6563 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
6564 * gst-libs/gst/vaapi/meson.build:
6565 libs: remove dependency on IN_LIBGSTVAAPI_CORE
6566 This conditional code was when libgstvaapi was intended to be library
6567 used outside GStreamer. This not the case anymore, thus removing it.
6568 https://bugzilla.gnome.org/show_bug.cgi?id=797139
6570 2018-09-19 10:16:36 +0800 Wangfei <fei.w.wang@intel.com>
6572 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
6573 libs: dec: h265: fix the macros used for IDC profile
6574 profile_idc flag in SPS only indicate the IDC profile, which may
6575 need some other flags together to get the real profile.
6576 https://bugzilla.gnome.org/show_bug.cgi?id=797160
6578 2018-09-12 19:06:22 +0900 Jimmy Ohn <yongjin.ohn@lge.com>
6580 * gst-libs/gst/vaapi/gstvaapidisplay.c:
6581 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
6582 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
6583 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
6584 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
6585 libs: use g_clear_pointer() when possible
6586 https://bugzilla.gnome.org/show_bug.cgi?id=797131
6588 2018-09-03 13:56:52 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6590 * gst-libs/gst/vaapi/gstvaapifilter.h:
6591 libs: filter: add gobject's cleanup function
6593 2018-05-22 14:28:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6595 * gst-libs/gst/vaapi/gstvaapifilter.c:
6596 * gst-libs/gst/vaapi/gstvaapifilter.h:
6597 * gst-libs/gst/vaapi/gstvaapisurface_egl.c:
6598 * tests/test-filter.c:
6599 libs: filter: remove custom ref() and unref()
6600 Replacing them by gst_object_ref() and gst_object_unref()
6601 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6603 2018-05-22 14:26:48 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6605 * gst-libs/gst/vaapi/gstvaapifilter.c:
6606 * gst-libs/gst/vaapi/gstvaapifilter.h:
6607 libs: filter: refactor filter as gobject
6608 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6610 2018-05-21 13:38:00 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6612 * gst-libs/gst/vaapi/gstvaapidecoder.c:
6613 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
6614 libs: decoder: remove destoy() and create() callbacks
6615 They were all replaced by reset()
6616 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6618 2018-05-21 13:26:01 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6620 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
6621 libs: decoder: vp9: implement reset() callback
6622 remove destroy() and create() callback
6623 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6625 2018-05-21 13:25:37 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6627 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
6628 libs: decoder: vp8: implement reset() callback
6629 remove create() and destroy() callbacks
6630 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6632 2018-05-21 13:24:39 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6634 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
6635 libs: decoder: vc1: implement reset() callback
6636 remove destroy() and create() callbacks
6637 use g_clear_pointer for rbdu_buffer
6639 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6641 2018-05-21 13:24:13 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6643 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
6644 libs: decoder: mpeg4: implement reset() callback
6645 remove destroy() and create() callback
6646 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6648 2018-05-21 13:22:45 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6650 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
6651 libs: decoder: mpeg2: implement reset() callback
6652 remove create() and destroy() callbacks
6653 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6655 2018-05-21 13:22:07 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6657 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
6658 libs: decoder: jpeg: implement reset() callback
6659 and remove create() and destroy() callbacks.
6660 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6662 2018-05-21 13:13:31 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6664 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
6665 libs: decoder: h265: implement reset() callback
6666 and remove create() and destroy()
6667 and use g_clear_pointer for dpb structure
6668 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6670 2018-05-21 13:11:41 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6672 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6673 libs: decoder: h264: remove create() and destroy() callbacks
6674 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6676 2018-05-21 11:56:11 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6678 * gst-libs/gst/vaapi/gstvaapidecoder.c:
6679 * gst-libs/gst/vaapi/gstvaapidecoder.h:
6680 * tests/test-decode.c:
6681 * tests/test-subpicture.c:
6682 libs: decoder: remove gst_vaapi_decoder_unref()
6683 Replaced by gst_object_unref() in tests
6684 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6686 2018-05-21 11:51:14 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6688 * gst-libs/gst/vaapi/gstvaapidecoder.c:
6689 * gst-libs/gst/vaapi/gstvaapidecoder.h:
6690 libs: decoder: remove gst_vaapi_decoder_ref()
6691 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6693 2018-05-21 11:50:17 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6695 * gst-libs/gst/vaapi/gstvaapidecoder.c:
6696 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
6697 libs: decoder: remove gst_vaapi_decoder_new()
6698 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6700 2018-05-18 16:09:31 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6702 * gst-libs/gst/vaapi/gstvaapidecoder.c:
6703 * gst-libs/gst/vaapi/gstvaapidecoder.h:
6704 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6705 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
6706 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
6707 * gst-libs/gst/vaapi/gstvaapidecoder_h265.h:
6708 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
6709 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.h:
6710 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
6711 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.h:
6712 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
6713 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.h:
6714 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
6715 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
6716 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.h:
6717 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
6718 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.h:
6719 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
6720 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.h:
6721 libs: decoder: refactor decoders as gobject
6722 https://bugzilla.gnome.org/show_bug.cgi?id=796308
6724 2018-08-31 20:56:13 -0500 Matteo Valdina <matteo.valdina@gmail.com>
6726 * gst/vaapi/gstvaapidecode.c:
6727 vaapidecode: Requests upstream a key unit at parse or decode error.
6728 This is done to resume decoding after a parse error or decode error.
6729 Bugzilla: https://bugzilla.gnome.org/show_bug.cgi?id=797006
6731 2018-08-31 20:48:13 -0500 Matteo Valdina <matteo.valdina@gmail.com>
6733 * gst/vaapi/gstvaapidecode.c:
6734 vaapidecode: sets return value in failure case.
6735 In gst_vaapidecode_handle_frame, when there is a decode error
6736 there is a code path the returns an uninitialized value.
6737 Bugzilla: https://bugzilla.gnome.org/show_bug.cgi?id=797006
6739 2018-08-30 18:56:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6741 * gst-libs/gst/vaapi/gstvaapidisplay.c:
6742 libs: display: lock at extracting available image formates
6743 When running several vaapi elements at the concurrently, at
6744 initialization, there is a race condition when extractin the avaible
6745 formats for images and subpictures.
6746 This patch add a lock when the those arrays are filled.
6747 https://bugzilla.gnome.org/show_bug.cgi?id=797039
6749 2018-08-31 14:47:55 +0530 Nirbheek Chauhan <nirbheek@centricular.com>
6752 meson: Sync libversion and osxversion code from other repos
6753 gstreamer-vaapi does not build any libraries, only plugins, so this is
6754 not used, but sync it just in case someone does add it in the future.
6756 2018-08-29 13:44:44 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6758 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
6759 libs: encoder: h265: trivial documentation fix
6761 2018-08-30 11:08:07 +0800 Wangfei <fei.w.wang@intel.com>
6763 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
6764 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
6765 libs: encoder: h265: add low delay B frame support.
6766 Low delay B frame provide the function of transforming
6767 P frame into low delay B frame which frame type is B, but
6768 only reference predictive frames. This can be used when P
6769 frame unsupported. Especially for P and B both unsupported,
6770 in this case, I and low delay B frame can be encoded in a
6772 https://bugzilla.gnome.org/show_bug.cgi?id=796984
6774 2018-08-27 20:42:15 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
6776 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6777 libs: decoder: h264: Avoid using picture after it has been free
6778 In some cases, the found_picture ended up being evicted and freed, which
6779 would lead to a use after free when accessing picture->base.poc. In this
6780 fix, we take a ref on the picture before calling dpb_evict.
6781 https://bugzilla.gnome.org/show_bug.cgi?id=787124
6783 2018-07-25 17:03:19 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
6785 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6786 h264decoder: Fail decoding slice with missing inter-view reference
6787 Similarly to previous patch, we have no error concealment. As a side
6788 effect, it's better to skip slices with missing references then passing
6789 NULL pointers to the accelerator. Passing NULL pointer would lead to
6790 major visual artifact, a behaviour that is likely undefined.
6791 https://bugzilla.gnome.org/show_bug.cgi?id=787124
6793 2017-09-14 14:25:41 +0900 Hyunjun Ko <zzoon@igalia.com>
6795 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6796 libs: decoder: h264: reset context when the number of view is increased
6797 Usually in case of MVC decoding, dpb size is increasedi if subset sps.
6798 That's why it resets context without this patch.
6799 But for some media it doesn't increase dpb size. Even in this case we
6800 should reset context to deal with MVC decoding.
6801 Otherwise, it leads to assert.
6802 https://bugzilla.gnome.org/show_bug.cgi?id=787124
6804 2018-07-25 13:50:23 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
6806 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6807 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
6808 * gst/vaapi/gstvaapidecode.c:
6809 vaapidecode: Skip unparsable units from adapter
6810 If the unit could not be parsed, just skip this nal and keep parsing
6811 what is left in the adapter. We need to flush the broken unit in the
6812 decoder specific parser because the generic code does not know about
6813 units boundary. This increases error resilliance.
6814 Before this, the broken unit would stay in the adapter and EOS would be
6815 returned. Which stopped the streaming. Just removing the EOS would have
6816 lead to the adapter size growing indefinitely.
6817 https://bugzilla.gnome.org/show_bug.cgi?id=796863
6819 2018-07-24 12:40:00 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
6821 * gst/vaapi/gstvaapidecode.c:
6822 vaapidecoder: Don't error out on decode errors
6823 This is problematic on live pipeline where loosing network can
6824 cause an important amount of errors.
6825 https://bugzilla.gnome.org/show_bug.cgi?id=796832
6827 2018-07-25 15:47:49 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
6829 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
6830 h265decoder: Don't scan empty buffer
6831 Same as what we did for H264 decoder, this is to avoid an assertion
6833 https://bugzilla.gnome.org/show_bug.cgi?id=796832
6835 2018-07-25 20:21:51 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6837 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6838 libs: h264: renable the vaapi category for logging
6839 h264 log messages were logged in default category because a regression
6840 in code. This patch renable the usage of vaapi logging category.
6841 This regression was introduced in commit 7c365bdd.
6843 2018-07-18 13:09:42 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
6845 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6846 h264decoder: Fail decoding slice if modification process failed
6847 This patch chains up failure to executing the modification process. The
6848 end result is that we now fail decoding the slice if this process fails.
6849 This avoid sending a corrupted state to the accelerator. In some special
6850 cases, this could lead to unrecoverable errors.
6851 https://bugzilla.gnome.org/show_bug.cgi?id=796832
6853 2018-07-18 13:07:51 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
6855 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6856 h264decoder: Don't scan empty buffer
6857 gst_adapter_masked_scan_uint32_peek() asserts if size is 0. Don't
6858 try and scan in that case. This fixes assertion that would some times
6859 happen when the stream is corrupted.
6860 https://bugzilla.gnome.org/show_bug.cgi?id=796832
6862 2018-07-04 12:51:10 +0800 Tianhao Liu <tianhao.liu@intel.com>
6864 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
6865 libs: encoder: jpeg: set component id and Tqi
6866 This change is due a problem encoding JPEGs with Intel's
6867 media-driver: green/black image when playback jpeg
6868 This patch sets component identifier and quantization table
6869 destination selector in frame header to support packing headers
6870 by Intel's media-driver that does not accept packed header
6872 https://bugzilla.gnome.org/show_bug.cgi?id=796705
6874 2018-06-25 14:20:32 +0200 Mathieu Duponchelle <mathieu@centricular.com>
6876 * gst/vaapi/gstvaapipluginutil.c:
6877 pluginutil: downgrade unsupported driver logging
6878 On systems with an Nvidia card, this error is output each time
6879 the registry is rebuilt, which happens pretty often when
6880 using gst-build as a development environment.
6881 https://bugzilla.gnome.org/show_bug.cgi?id=796663
6883 2018-06-24 13:07:20 +0200 Tim-Philipp Müller <tim@centricular.com>
6885 * gst/vaapi/gstvaapivideobufferpool.c:
6886 Update for g_type_class_add_private() deprecation in recent GLib
6888 2018-05-30 16:01:36 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
6890 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6891 h264dec: Remove false assumption about parity order
6892 The decoder was trying to detect earlier that a field was lost base
6893 on guessing the parity order. This breaks in streams were the parity
6895 This patch reverts the field order prediction code added by commit
6897 https://bugzilla.gnome.org/show_bug.cgi?id=796169
6899 2018-05-18 17:03:57 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
6901 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
6902 h264dec: Properly set sentinel in ref frame list
6903 This ensure that we always have sentinels set in the reference
6904 pictures arrays. The code wasn't unsafe, this simply improve the
6905 tracing, so instead of printing 32 lines of zeros, va tracer
6906 prints proper empty lists.
6907 https://bugzilla.gnome.org/show_bug.cgi?id=796169
6909 2018-06-13 18:00:18 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6911 * gst-libs/gst/vaapi/gstvaapidecoder.c:
6912 * gst-libs/gst/vaapi/gstvaapidisplay.c:
6913 * gst-libs/gst/vaapi/gstvaapidisplay.h:
6914 * gst-libs/gst/vaapi/gstvaapiencoder.c:
6915 * gst-libs/gst/vaapi/gstvaapifilter.c:
6916 * gst-libs/gst/vaapi/gstvaapiobject.c:
6917 * gst-libs/gst/vaapi/gstvaapivideopool.c:
6918 * gst/vaapi/gstvaapipluginbase.c:
6919 * gst/vaapi/gstvaapivideometa.c:
6920 * tests/test-decode.c:
6921 libs: display: remove gst_vaapi_display_ref()
6922 Replace it with gst_object_ref()
6923 https://bugzilla.gnome.org/show_bug.cgi?id=796470
6925 2018-06-13 17:54:23 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6927 * gst-libs/gst/vaapi/gstvaapidisplay.c:
6928 * gst-libs/gst/vaapi/gstvaapidisplay.h:
6929 * gst/vaapi/gstvaapi.c:
6930 * gst/vaapi/gstvaapidecode.c:
6931 * gst/vaapi/gstvaapiencode.c:
6932 * gst/vaapi/gstvaapipluginbase.c:
6933 * gst/vaapi/gstvaapipluginutil.c:
6935 * tests/simple-encoder.c:
6936 * tests/test-decode.c:
6937 * tests/test-display.c:
6938 * tests/test-fei-enc-in.c:
6939 * tests/test-filter.c:
6940 * tests/test-subpicture.c:
6941 * tests/test-surfaces.c:
6942 * tests/test-textures.c:
6943 * tests/test-windows.c:
6944 libs: display: remove gst_vaapi_display_unref()
6945 Use gst_object_unref() instead.
6946 https://bugzilla.gnome.org/show_bug.cgi?id=796470
6948 2018-06-13 18:10:28 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6950 * gst/vaapi/gstvaapivideobufferpool.c:
6951 vaapibufferpool: declare parameter display as object
6952 We have neglected to update this code since GstVaapiDisplay turned
6953 into a GstObject descendant.
6954 https://bugzilla.gnome.org/show_bug.cgi?id=796470
6956 2018-06-01 12:36:51 +0900 Hyunjun Ko <zzoon@igalia.com>
6958 * gst-libs/gst/vaapi/gstvaapidisplay.c:
6959 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
6960 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
6961 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
6962 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
6963 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
6964 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
6965 libs: display: replace gst_vaapi_display_new() with gst_vaapi_display_config()
6966 Gobjectification for GstVaapiDisplay was almost done by the commit 185da3d1.
6967 But still something breaking GObject code convention remains, which is
6968 calling gst_vaapi_display_new() in each decendants.
6969 This patch replaces it with gst_vaapi_display_config(), defined in private
6971 https://bugzilla.gnome.org/show_bug.cgi?id=796470
6973 2018-06-13 17:05:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6975 * gst-libs/gst/vaapi/gstvaapidisplay.c:
6976 libs: display: redefine gst_vaapi_display_create()
6977 The function name was gst_vaapi_display_create_unlocked(), nonetheless
6978 it wasn't called unlocked. In order to keep the semantics this patch
6979 renames the gst_vaapi_display_create_unlocked() as
6980 gst_vaapi_display_create(), removing the previous function
6981 gst_vaapi_display_create().
6982 https://bugzilla.gnome.org/show_bug.cgi?id=796470
6984 2018-06-12 15:53:04 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6987 Revert "build: meson: libva gst-uninstall friendly"
6988 This reverts commit fc3eef9c432c1628cb92ab56e74924cf1182da30.
6990 2018-06-12 15:13:33 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6992 * gst/vaapi/gstvaapipluginutil.c:
6993 plugins: fix compilation
6994 gstvaapipluginutil.c:171:1: error: old-style function definition [-Werror=old-style-definition]
6996 2018-04-20 18:05:30 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
6999 build: meson: libva gst-uninstall friendly
7000 Make gstreamer-vaapi to use libva uninstalled.
7002 2018-06-10 10:44:35 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7004 * gst/vaapi/gstvaapipluginutil.c:
7005 plugins: refactor gst_vaapi_create_display_from_gl_context()
7006 gst_vaapi_create_display_from_gl_context() was a spaghetti mess.
7007 This path refactors it, in order to make the code readable and
7009 https://bugzilla.gnome.org/show_bug.cgi?id=796564
7011 2018-05-25 12:17:21 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7013 * gst-libs/gst/vaapi/gstvaapidisplay.c:
7014 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
7015 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
7016 libs: display: resurrect parent private member
7017 This is, practically, a revert of commit dcf135e2.
7018 The parent logic is useful for the EGL display, which is a decorator
7019 of the real windowing subsystem (X11 or Wayland). Thus it is avoided
7020 calling vaInitialize() and vaTerminate() twice.
7021 https://bugzilla.gnome.org/show_bug.cgi?id=795391
7023 2018-04-27 18:35:30 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7025 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
7026 libs: display: egl: initialize params structure
7027 Statically initialise the internal params structure.
7028 https://bugzilla.gnome.org/show_bug.cgi?id=795391
7030 2018-04-27 18:34:37 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7032 * gst/vaapi/gstvaapipluginutil.c:
7033 plugins: handle EGL when creating VAAPI display from gl
7034 If GstGL reports a EGL platform force to create a EGL display using
7035 the native EGL display.
7036 https://bugzilla.gnome.org/show_bug.cgi?id=795391
7038 2018-04-24 18:17:24 +0900 Hyunjun Ko <zzoon@igalia.com>
7040 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
7041 * gst/vaapi/gstvaapipluginutil.c:
7042 display: egl: create VaapiDisplayEGL with native EGL display
7043 gst_vaapi_display_egl_new_with_native_display() has been broken since
7045 Currently it's needed to call this API to create a display providing
7046 the EGL display, so it could avoid duplicated calls to the native
7047 display (eg. eglTerminate).
7048 Signed-off-by: Victor Jaquez <vjaquez@igalia.com>
7049 https://bugzilla.gnome.org/show_bug.cgi?id=795391
7051 2018-06-07 09:34:11 +0800 Tianhao Liu <tianhao.liu@intel.com>
7053 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
7054 libs: decoder: release VA buffers after vaEndPicture
7055 This change is due a problem decoding JPEGs with Intel's media-driver:
7056 no image was generated.
7057 This patch relases the VA buffers after vaEndPicture() is called,
7058 and not before (after vaRenderPicture()).
7059 https://bugzilla.gnome.org/show_bug.cgi?id=796505
7061 2018-06-07 19:49:02 +0100 Tim-Philipp Müller <tim@centricular.com>
7063 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
7064 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
7065 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
7066 meson: fix build when xrender or xrandr are not available
7067 HAVE_XRENDER are defined to 1 or 0, not defined or undefined.
7069 2018-05-25 16:47:00 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
7071 * gst/vaapi/gstvaapipostproc.c:
7072 vaapipostproc: don't copy the GstParentBufferMeta if use_vpp
7073 Otherwise a reference to a DMABuf input buffer is kept until the output
7075 https://bugzilla.gnome.org/show_bug.cgi?id=796399
7077 2018-05-22 21:13:08 +0900 Hyunjun Ko <zzoon@igalia.com>
7079 * gst-libs/gst/vaapi/gstvaapidisplay.c:
7080 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
7081 libs: display: remove unnecessary legacy code since gobjectification
7082 https://bugzilla.gnome.org/show_bug.cgi?id=796470
7084 2018-05-22 21:05:54 +0900 Hyunjun Ko <zzoon@igalia.com>
7086 * gst-libs/gst/vaapi/gstvaapidisplay.c:
7087 * gst-libs/gst/vaapi/gstvaapidisplay.h:
7088 libs: display: remove unused code
7089 https://bugzilla.gnome.org/show_bug.cgi?id=796470
7091 2018-06-05 15:16:53 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7093 * gst/vaapi/gstvaapiencode_h264.c:
7094 vaapiencode: h264: log output caps
7096 2018-06-05 22:38:37 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7098 * gst/vaapi/gstvaapiencode_h264.c:
7099 vaapiencode: h264: find profile in available and allowed caps
7100 The commit 67e33d3de225d0e006d7bf606e7abb20d4544eab ("vaapiencode: h264:
7101 find best profile in those available") changed the code to pick a profile
7102 that is actually supported by the hardware. Unfortunately it dropped the
7103 downstream constraints. This can cause negotiation failures under certain
7105 The fix is split in two cases:
7106 1\ the available VA-API caps doesn't intersect with pipeline's allowed
7108 * The best allowed profile (pipeline's caps) is set as the encoding
7109 target profile (it will be adjusted later by the available profiles
7111 2\ the available VA-API caps does intersect with pipeline's allowed
7113 * The intersected caps are fixed, and its profile is set as the
7114 encoding target profile. In this case the is not the best profile,
7115 but the minimal one (if VA-API reports the profiles in order).
7116 Setting the minimal profile of the intersected caps is better for
7118 This patch fixes other tests related with caps negotiation, for
7119 example, it handles baseline profile, even when VA only supports
7120 constrained-baseline.
7121 Original-patch-by: Michael Olbrich <m.olbrich@pengutronix.de>
7122 https://bugzilla.gnome.org/show_bug.cgi?id=794306
7124 2018-06-01 15:27:25 +0900 Hyunjun Ko <zzoon@igalia.com>
7126 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
7127 libs: encoder: h265: increase log2_max_pic_order_cnt range according to spec
7128 The specification says,
7129 "log2_max_pic_order_cnt_lsb_minus4 shall be in the range of 0 to 12, inclusive."
7130 This patch changes the upper limit from 6 to 12.
7131 https://bugzilla.gnome.org/show_bug.cgi?id=796179
7133 2018-05-21 13:27:14 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7135 * gst/vaapi/gstvaapipluginutil.c:
7136 plugins: guard GstGL code
7138 2018-05-18 18:23:18 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7140 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
7141 libs: decoder: h264: use g_clear_pointer()
7143 2018-05-18 17:27:46 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7145 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.h:
7146 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.h:
7147 libs: decoder: mpeg4, vc1: remove unused header
7149 2018-05-18 11:09:58 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7151 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
7152 libs: decoder: remove unused forward declaration
7154 2018-05-07 07:59:25 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
7157 fix configure.ac regression
7158 Fixes regression introduced by 77527d67abe
7159 https://bugzilla.gnome.org/show_bug.cgi?id=795885
7161 2018-05-05 17:57:49 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7163 * gst-libs/gst/Makefile.am:
7164 * gst-libs/gst/base/Makefile.am:
7165 * gst-libs/gst/base/gstbitwriter.c:
7166 * gst-libs/gst/base/gstbitwriter.h:
7167 * gst-libs/gst/base/meson.build:
7168 * gst-libs/gst/meson.build:
7169 * gst-libs/gst/vaapi/Makefile.am:
7170 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
7171 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
7172 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
7173 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
7174 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
7175 * gst-libs/gst/vaapi/gstvaapifeipak_h264.c:
7176 * gst-libs/gst/vaapi/meson.build:
7177 libs: remove gstbitwriter
7178 Since it is deployed in gstreamer-core, there is no need to use
7180 https://bugzilla.gnome.org/show_bug.cgi?id=795848
7182 2018-04-28 16:10:46 +0800 Wang,Fei <fei.w.wang@intel.com>
7184 * gst/vaapi/gstvaapidecode.c:
7185 vaapih264dec: add constrained and progressive profiles
7186 Those profiles have been added in the version 2012-01
7187 and 2011-06 of the AVC spec (A.2.4.1 and A.2.4.2).
7188 Both are supported by VAProfileH264High
7189 https://bugzilla.gnome.org/show_bug.cgi?id=795624
7191 2018-04-26 18:15:47 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7193 * gst/vaapi/gstvaapipluginbase.c:
7194 * gst/vaapi/gstvaapipluginutil.c:
7195 * gst/vaapi/gstvaapivideocontext.c:
7196 * gst/vaapi/gstvaapivideocontext.h:
7197 plugin: remove custom GstGL context handling
7198 Instead of using our own context handling for looking for GstGL
7199 parameters (display, context and other context), this patch changes
7200 the logic to use the utility function offered by GstGL.
7201 https://bugzilla.gnome.org/show_bug.cgi?id=793643
7203 2018-04-26 15:03:23 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7205 * gst/vaapi/gstvaapipluginbase.c:
7206 plugins: GstGL API must use the member variables
7207 This commit basically is a revert of commits 8092537 and fc1c415
7208 https://bugzilla.gnome.org/show_bug.cgi?id=793643
7210 2018-04-25 16:24:32 +0900 Hyunjun Ko <zzoon@igalia.com>
7212 * gst/vaapi/gstvaapipluginbase.c:
7213 plugins: pass members as parameters of gst_gl_ensure_element_data()
7214 The parameters of gst_gl_ensure_element_data() have to be not
7215 local variable since they are going to be used to see if they're
7216 set in gst_element_set_context() inside the API.
7217 This is basically a revert of commit 3d56306c
7218 https://bugzilla.gnome.org/show_bug.cgi?id=793643
7220 2018-04-25 17:50:14 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7223 meson: fix USE_GLES_VERSION_MASK
7224 1. The macro in the code is USE_GLES_VERSION_MASK
7225 2. glesv3 is provided by glesv2 pkg-config, then it's required to
7227 https://bugzilla.gnome.org/show_bug.cgi?id=795391
7229 2018-04-24 18:12:44 +0900 Hyunjun Ko <zzoon@igalia.com>
7231 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
7232 libs: egl: utils: mark context as wrapped when it is
7233 The returning egl context may be null, so we should check the
7235 https://bugzilla.gnome.org/show_bug.cgi?id=795391
7237 2018-04-24 10:02:33 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7239 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
7240 libs: egl: utils: fix usage of GstGL macros
7241 Include gl.h for the required GstGL symbols.
7242 https://bugzilla.gnome.org/show_bug.cgi?id=795391
7244 2018-04-25 11:01:45 +0100 Tim-Philipp Müller <tim@centricular.com>
7247 meson: use -Wl,-Bsymbolic-functions where supported
7248 Just like the autotools build.
7250 2018-04-20 16:01:29 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7253 meson: use get_pkgconfig_variable()
7254 Use get_pkgconfig_variable() method, of dependency class, rather
7255 than using run_command().
7257 2018-04-20 11:50:55 +0100 Tim-Philipp Müller <tim@centricular.com>
7259 * gst-libs/gst/base/meson.build:
7260 * gst-libs/gst/vaapi/meson.build:
7262 meson: fix miscellaneous meson warnings
7263 WARNING: Passed invalid keyword argument "rqeuired".
7264 gst-libs/gst/base/meson.build:11: WARNING: Passed invalid keyword argument "version".
7265 gst-libs/gst/base/meson.build:11: WARNING: Passed invalid keyword argument "soversion".
7266 gst-libs/gst/vaapi/meson.build:223: WARNING: Passed invalid keyword argument "version".
7267 gst-libs/gst/vaapi/meson.build:223: WARNING: Passed invalid keyword argument "soversion".
7269 2018-03-30 13:41:39 +0200 Paul Kocialkowski <paul.kocialkowski@bootlin.com>
7271 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
7272 display: drm: Allow finding DRM paths out of the PCI subsystem
7273 This removes hard-coded checks on the parent subsystem of potential DRM
7274 devices. These checks were set to exlude devices that do not originate
7275 from the PCI bus, which is only a valid approach on x86 devices.
7276 Other devices may have a DRM device originating from the platform
7277 subsystem, so the checks that were previously restricted to PCI are
7278 extended to cover platform devices as well.
7279 https://bugzilla.gnome.org/show_bug.cgi?id=794840
7280 Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
7282 2017-11-08 13:27:06 +0900 Hyunjun Ko <zzoon@igalia.com>
7284 * gst/vaapi/gstvaapivideocontext.c:
7285 videocontext: support wl-display in "gst.vaapi.app.Display"
7286 Through "gst.vaapi.app.Display" context, users can set their own
7287 VADisplay and native display of their backend.
7288 So far we support only X11 display, from now we also support Wayland
7291 - wl-display : pointer of struct wl_display .
7292 https://bugzilla.gnome.org/show_bug.cgi?id=705821
7294 2017-11-08 13:26:38 +0900 Hyunjun Ko <zzoon@igalia.com>
7296 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
7297 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.h:
7298 libs: display: wayland: add gst_vaapi_display_wayland_new_with_va_display()
7299 Implements new API function so that users could create GstVaapiDisplay
7300 with their own VADisplay within a native display as backend.
7301 https://bugzilla.gnome.org/show_bug.cgi?id=705821
7303 2018-04-13 09:28:53 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
7305 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
7306 wayland: don't poll if there are no pending frames
7307 Otherwise the following poll may not return for an arbitrary amount of
7308 time. This can happen if another wayland event queue has flushed and read
7310 https://bugzilla.gnome.org/show_bug.cgi?id=795224
7312 2017-10-16 12:09:08 +0900 Hyunjun Ko <zzoon@igalia.com>
7314 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
7315 libs: encoder: h265: 16 bit rounding of picture width and height
7316 pic_width_in_luma_samples/pic_height_in_luma_samples can be 16-bit rounded
7318 In addition, codedbuf_size must be calculated according to this change.
7319 https://bugzilla.gnome.org/show_bug.cgi?id=753229
7321 2018-04-16 10:53:47 +0100 Tim-Philipp Müller <tim@centricular.com>
7324 Automatic update of common submodule
7325 From 3fa2c9e to ed78bee
7327 2018-03-30 20:39:18 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7329 * gst/vaapi/gstvaapiencode_h264.c:
7330 vaapiencode: h264: find best profile in those available
7331 Instead to look for the best profile in the allowed profiles by
7332 downstream, the encoder should look for the base profile in the
7333 available profile in VA-API.
7334 https://bugzilla.gnome.org/show_bug.cgi?id=794306
7336 2018-03-20 10:49:10 +0000 Tim-Philipp Müller <tim@centricular.com>
7341 * docs/plugins/inspect/plugin-vaapi.xml:
7345 === release 1.14.0 ===
7347 2018-03-19 20:30:28 +0000 Tim-Philipp Müller <tim@centricular.com>
7352 * gstreamer-vaapi.doap:
7356 2018-03-12 16:59:01 +0000 Julien Isorce <jisorce@oblong.com>
7358 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
7359 libs: decoder: h264: ensure num_ref_frames is greater than 0
7360 Even if it is the h264parse fault or bad video file, vaapih264dec
7361 should set a proper value for VAPictureParameterBufferH264.num_ref_frames
7362 as the driver might use it.
7363 Also see "info.ref_frames = dpb_size;" in
7364 gstvaapidecoder_h264.c::ensure_context
7365 https://bugzilla.gnome.org/show_bug.cgi?id=793836
7367 === release 1.13.91 ===
7369 2018-03-13 19:32:05 +0000 Tim-Philipp Müller <tim@centricular.com>
7374 * gstreamer-vaapi.doap:
7378 === release 1.13.90 ===
7380 2018-03-03 22:59:30 +0000 Tim-Philipp Müller <tim@centricular.com>
7385 * gstreamer-vaapi.doap:
7389 2018-03-01 07:33:27 -0600 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7391 * gst/vaapi/gstvaapipostproc.c:
7392 vaapipostproc: change how the metadata is copied
7393 Instead of copying the metada in prepare_output_buffer() vmethod,
7394 it is done in append_output_buffer_metadata() thus deinterlaced
7395 buffers could also have the proper metas.
7396 GstVideoCropMeta now it is copied internally and it is decided via
7397 transform_meta() vmethod.
7398 A new internal method, copy_metadata() was added to handle VPP
7399 transformation where non-GstVideoVaapiMeta metas were lost.
7401 2018-02-27 16:20:15 -0500 Nicolas Dufresne <nicolas.dufresne@collabora.com>
7403 * gst/vaapi/gstvaapipostproc.c:
7404 postproc: Copy meta data from input to output
7405 This will ensure that meta data without memory tags will be copied. This
7406 was noticed when testing ROI.
7407 https://bugzilla.gnome.org/show_bug.cgi?id=768248
7409 2018-02-23 10:48:36 -0600 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7411 * tests/elements/test-roi.c:
7412 tests: element: rewrite ROI test
7413 Rewrote the ROI test to use GstVideoRegionOfInterest meta rather
7414 than injecting GstEvents. These meta are added as a pad probe in
7417 * Use of navigation messages to control de test
7418 * Use signal watch for processing messages
7419 * Change to H265 rather than H264 since current intel-vaapi-driver
7420 only supports ROI on kabylake.
7421 TODO: add a parameter to change the encoder/decoder to test.
7422 https://bugzilla.gnome.org/show_bug.cgi?id=768248
7424 2018-02-22 14:20:42 -0600 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7426 * gst-libs/gst/vaapi/gstvaapicontext.c:
7427 * gst-libs/gst/vaapi/gstvaapiencoder.c:
7428 * gst-libs/gst/vaapi/gstvaapiencoder.h:
7429 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
7430 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
7431 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
7432 libs: encoder: reimplement ROI using meta
7433 Check input buffers for ROI metas and pass them to VA. Also added a
7434 new "default-roi-delta-qp" property in order to tell the encoder what
7435 delta QP should be applied to ROI by default.
7436 Enabled it for H264 and H265 encoders.
7437 https://bugzilla.gnome.org/show_bug.cgi?id=768248
7439 2018-02-22 08:22:35 -0600 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7441 * gst-libs/gst/vaapi/gstvaapiencoder.c:
7442 * gst-libs/gst/vaapi/gstvaapiencoder.h:
7443 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
7444 Revert "libs: encoder: add api gst_vaapi_encoder_add/del_roi"
7445 This reverts commit 7a6f690340dcb3b82c59efa777d4453227851de8.
7446 https://bugzilla.gnome.org/show_bug.cgi?id=768248
7448 2018-02-22 14:29:19 -0600 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7450 * tests/simple-encoder.c:
7451 Revert "tests: simple-encoder: add an option to set ROI"
7452 This reverts commit c21345c4787bb6342adddea1190f53fe62abff04.
7453 https://bugzilla.gnome.org/show_bug.cgi?id=768248
7455 2018-02-21 10:56:47 -0600 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7457 * gst/vaapi/gstvaapiencode.c:
7458 * gst/vaapi/gstvaapiencode_h264.c:
7459 Revert "vaapiencode: handle custom event GstVaapiEncoderRegionOfInterest"
7460 This reverts commit 8f1b88dac0e64a211325cdcb2cda693b80229bd1.
7461 https://bugzilla.gnome.org/show_bug.cgi?id=768248
7463 2018-02-23 09:25:51 -0600 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7465 * gst/vaapi/gstvaapipluginbase.c:
7466 plugins: copy input buffer metas
7467 When importing buffers to a VA-base buffer, it is required to copy
7468 the metas in the original buffer, otherwise information will be
7469 lost, such as GstVideoRegionOfInterestMeta.
7470 https://bugzilla.gnome.org/show_bug.cgi?id=768248
7472 2018-02-27 06:10:09 -0600 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7474 * gst-libs/gst/vaapi/gstvaapisurface.c:
7475 libs: surface: cast to uintptr_t pointer
7476 According to Debian package auto-building, uintptr_t is not an
7477 unsigned long in i386 arch, raising an "incompatible pointer type"
7479 This patch adds a casting for compiler's satisfaction in i386.
7481 2018-02-25 20:46:56 -0600 Matteo Valdina <matteo.valdina@gmail.com>
7483 * gst-libs/gst/vaapi/gstvaapiencoder.c:
7484 libs: encoder: add zero as valid value for periodic keyframe.
7485 Enabled zero as valid value for keyframe-period property.
7486 https://bugzilla.gnome.org/show_bug.cgi?id=793829
7488 2018-02-22 08:24:12 -0600 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7490 * gst-libs/gst/vaapi/gstvaapiencoder.c:
7491 libs: encoder: code-style fix
7493 2018-02-17 18:32:11 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7495 * gst/vaapi/gstvaapidecode.c:
7496 vaapidecode: generate system allocated buffers
7497 Generate system allocated output buffers when downstream doesn't
7498 support GstVideoMeta.
7499 The VA buffer content is copied to the new output buffer, and it
7500 replaces the VA buffer.
7501 https://bugzilla.gnome.org/show_bug.cgi?id=785054
7503 2018-02-15 19:32:37 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7505 * gst/vaapi/gstvaapipostproc.c:
7506 vaapipostproc: handle system allocated buffers when required
7507 When downstream can't handle GstVideoMeta it is required to send
7508 system allocated buffers.
7509 The system allocated buffers are produced in prepare_output_buffer()
7510 vmethod if downstream can't handl GstVideoMeta.
7511 At transform() vmethod if the buffer is a system allocated buffer,
7512 a VA buffer is instanciated and replaces the out buffer. Later
7513 the VA buffer is copied to the system allocate buffer and it
7514 replaces the output buffer.
7515 https://bugzilla.gnome.org/show_bug.cgi?id=785054
7517 2018-02-15 19:32:19 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7519 * gst/vaapi/gstvaapipluginbase.c:
7520 * gst/vaapi/gstvaapipluginbase.h:
7521 plugins: add gst_vaapi_copy_va_buffer()
7522 This helper function aims to copy buffers with VA memory to dumb
7523 buffers, when GstVideoMeta is not available dowstream.
7524 https://bugzilla.gnome.org/show_bug.cgi?id=785054
7526 2018-02-15 19:29:51 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7528 * gst/vaapi/gstvaapipluginbase.c:
7529 * gst/vaapi/gstvaapipluginbase.h:
7530 plugins: add COPY_OUTPUT_FRAME flag
7531 This patch add the member copy_output_frame and set it TRUE when
7532 when downstream didn't request GstVideoMeta API, the caps are raw
7533 and the internal allocator is the VA-API one.
7534 https://bugzilla.gnome.org/show_bug.cgi?id=785054
7536 2018-02-15 19:28:33 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7538 * gst/vaapi/gstvaapipluginbase.c:
7539 * gst/vaapi/gstvaapipluginbase.h:
7540 plugins: store the first downstream allocator if available
7541 The allocator will be required if we need to allocate a buffer
7542 to store the frame with the expected strides.
7543 https://bugzilla.gnome.org/show_bug.cgi?id=785054
7545 2018-02-20 02:25:13 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7547 * gst/vaapi/gstvaapivideobufferpool.c:
7548 * gst/vaapi/gstvaapivideobufferpool.h:
7549 vaapibufferpool: add gst_vaapi_video_buffer_pool_copy_buffer()
7550 This function will inform the element if it shall copy the generated
7551 buffer by the pool to a system allocated buffer before pushing it
7553 https://bugzilla.gnome.org/show_bug.cgi?id=785054
7555 2018-02-15 19:22:08 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7557 * gst/vaapi/gstvaapivideobufferpool.c:
7558 vaapibufferpool: don't change config when forcing video meta
7559 VA-API based buffer might need a video meta because of different
7560 strides. But when donwstream doesn't support video meta we need to
7561 force the usage of video meta.
7562 Before we changed the buffer pool configuration, but actually this
7563 is a hack and we cannot rely on that for downstream.
7564 This patch add a check fo raw video caps and allocator is VA-API,
7565 then the option is enabled without changing the pool configuration.
7566 In this case the element is responsible to copy the frame to a
7567 simple buffer with the expected strides.
7568 https://bugzilla.gnome.org/show_bug.cgi?id=785054
7570 2018-02-20 09:15:05 -0600 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7572 * gst/vaapi/gstvaapipostproc.c:
7573 vaapipostproc: set discont flag at vpp deinterlacing
7574 When deinterlacing with VPP the discont flag was not forwarded to
7575 the new created buffer. This patch sets the discont flag if input
7578 2018-02-20 02:14:37 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7580 * gst/vaapi/gstvaapivideobufferpool.h:
7581 vaapibufferpool: remove wrong gcc annotation
7583 2018-02-15 14:55:42 -0600 Matteo Valdina <matteo.valdina@gmail.com>
7585 * gst-libs/gst/vaapi/gstvaapiencoder.c:
7586 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
7587 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
7588 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
7589 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
7590 * gst-libs/gst/vaapi/gstvaapifeipak_h264.c:
7591 * gst-libs/gst/vaapi/gstvaapiutils_h26x_priv.h:
7592 libs: encoder: h264,h265: extend max periodic keyframe.
7593 Increased max values of periodic key frame for h26x codecs.
7594 This allow more fine tunning of encoder that in certian scenario
7595 want higher periodic key frame.
7596 For example: it doesn't want a key frame each 10 seconds but
7598 https://bugzilla.gnome.org/show_bug.cgi?id=786320
7600 2018-02-15 19:44:35 +0000 Tim-Philipp Müller <tim@centricular.com>
7606 === release 1.13.1 ===
7608 2018-02-15 17:39:16 +0000 Tim-Philipp Müller <tim@centricular.com>
7613 * gstreamer-vaapi.doap:
7617 2018-02-15 18:15:33 +0000 Tim-Philipp Müller <tim@centricular.com>
7619 * gst/vaapi/Makefile.am:
7620 vaapi: dist new header
7622 2018-02-12 17:53:58 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7624 * gst/vaapi/gstvaapi.c:
7625 vaapi: register vaapisink as marginal on wayland
7626 vaapsink, when used with the Intel VA-API driver, tries to display
7627 surfaces with format NV12, which are handled correctly by
7628 Weston. Nonetheless, COGL cannot display YUV surfaces, making fail
7629 pipelines on mutter.
7630 This shall be solved either by COGL or by making the driver to paint
7631 RGB surfaces. In the meanwhile, let's just demote vaapisink as
7632 marginal when the Wayland environment is detected, no matter if it is
7634 https://bugzilla.gnome.org/show_bug.cgi?id=775698
7636 2018-02-12 19:00:36 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7638 * gst/vaapi/gstvaapipluginutil.c:
7639 plugins: update mesa's vendor string in whitelist
7640 Mesa has updated its VA-API Gallium driver vendor string:
7641 https://cgit.freedesktop.org/mesa/mesa/commit/?id=5db29d62ce1fefa3f2ee6e4a4688576fde4bde4a
7642 This patch tries to cover both, the old and the new one.
7643 https://bugzilla.gnome.org/show_bug.cgi?id=793386
7645 2018-02-08 19:22:17 +0000 Tim-Philipp Müller <tim@centricular.com>
7648 meson: make version numbers ints and fix int/string comparison
7649 WARNING: Trying to compare values of different types (str, int).
7650 The result of this is undefined and will become a hard error
7651 in a future Meson release.
7653 2018-02-07 09:13:26 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7655 * gst/vaapi/gstvaapipluginbase.c:
7656 plugins: handle vaapi allocator in allocation query
7657 In propose_allocation() if the numer of allocation params is zero, the
7658 system's allocator is added first, and lastly the native VA-API
7660 In decide_allocation(), the allocations params in query are travered,
7661 looking for a native VA-API allocator. If it is found, it is reused as
7662 src pad allocator. Otherwise, a new allocator is instantiated and
7663 appended in the query.
7664 https://bugzilla.gnome.org/show_bug.cgi?id=789476
7666 2018-02-07 09:06:46 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7668 * gst/vaapi/gstvaapivideomemory.h:
7669 vaapivideomemory: remove unused macro
7670 GST_VAAPI_VIDEO_ALLOCATOR_NAME was added in commit 5b11b8332 but it
7671 was never used, since the native VA-API allocator name has been
7672 GST_VAAPI_VIDEO_MEMORY_NAME.
7673 This patch removes GST_VAAPI_VIDEO_ALLOCATOR_NAME macro.
7674 https://bugzilla.gnome.org/show_bug.cgi?id=789476
7676 2018-02-02 08:54:00 +0000 VaL Doroshchuk <valkov@yandex.ru>
7678 * gst/vaapi/gstvaapisink.c:
7679 vaapisink: don't mask button events for foreign windows
7680 Don't subscribe to button press events when using a foreing window,
7681 because the user created window would trap those events, preveting the
7683 https://bugzilla.gnome.org/show_bug.cgi?id=791615
7685 2018-02-05 08:51:56 +0100 Tim-Philipp Müller <tim@centricular.com>
7688 autotools: use -fno-strict-aliasing where supported
7689 https://bugzilla.gnome.org/show_bug.cgi?id=769183
7691 2018-01-30 20:38:37 +0000 Tim-Philipp Müller <tim@centricular.com>
7694 meson: use -fno-strict-aliasing where supported
7695 https://bugzilla.gnome.org/show_bug.cgi?id=769183
7697 2018-01-30 12:56:49 +0000 Philippe Normand <philn@igalia.com>
7699 * gst/vaapi/gstvaapi.c:
7700 vaapi: add NULL-sentinel to kernel_names
7701 The array needs to be NULL-terminated according to the
7702 gst_plugin_add_dependency() documentation.
7704 2018-01-18 18:53:29 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7706 * gst/vaapi/gstvaapipostprocutil.c:
7707 vaapipostproc: remove spurious code
7708 This assignation is dead code, since gst_video_info_from_caps() set
7710 https://bugzilla.gnome.org/show_bug.cgi?id=790149
7712 2018-01-18 18:51:57 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7714 * gst/vaapi/gstvaapipostprocutil.c:
7715 vaapipostproc: if no p-a-r in out caps define a range
7716 Instead of copying the pixel-aspect-ratio from the sink caps, define
7717 an open range for the src caps pixel-aspect-ratio. Later it will be
7719 https://bugzilla.gnome.org/show_bug.cgi?id=790149
7721 2018-01-18 13:10:59 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7723 * gst/vaapi/gstvaapisink.c:
7724 vaapisink: check for display's color-balance properties
7725 Check for display's color-balance properties, available by the VA-API
7726 driver, before setting them.
7727 Also logs an info message of those unavailable properties.
7728 https://bugzilla.gnome.org/show_bug.cgi?id=792638
7730 2018-01-17 17:30:50 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7732 * gst/vaapi/gstvaapipluginbase.c:
7733 plugins: re-using buffer pool breaks renegotiation
7734 at propose_allocation() we should not reuse the proposed buffer,
7735 because it could break renegotiation.
7736 https://bugzilla.gnome.org/show_bug.cgi?id=792620
7738 2018-01-17 17:26:24 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7740 * gst/vaapi/gstvaapipluginbase.c:
7741 plugins: use g_clear_object() to unref sinkpad_buffer_pool
7742 https://bugzilla.gnome.org/show_bug.cgi?id=792620
7744 2018-01-17 12:42:12 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7746 * gst-libs/gst/vaapi/meson.build:
7747 build: meson: add missing GstGL dependency
7749 2018-01-17 12:41:54 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7751 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
7752 libs: utils: egl: add missing guards for GstGL
7754 2018-01-11 11:48:02 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7756 * gst/vaapi/gstvaapipluginbase.c:
7757 plugins: remove dmabuf-import hack
7758 Remove the hack to check if an upstream element has enabled the
7759 property io-mode enabled as dmabuf-import.
7760 https://bugzilla.gnome.org/show_bug.cgi?id=792034
7762 2017-12-01 15:04:35 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7764 * gst-libs/gst/vaapi/Makefile.am:
7765 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
7766 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
7767 * gst-libs/gst/vaapi/gstvaapiutils_egl.h:
7768 libs: egl: utils: use eglGetPlatformDisplay()
7769 eglGetDisplay() is currently broken in Mesa for Wayland. Also using
7770 eglGetDisplay() is rather fragile, and it is recommended to use
7771 eglGetPlatformDisplay() when possible.
7772 In order to do that, this patch uses the helper in GstGL. If
7773 gstreamer-vaapi is not compiled with GstGL support, eglGetDisplay()
7775 https://bugzilla.gnome.org/show_bug.cgi?id=790493
7777 2017-12-08 14:46:02 +0100 Michael Tretter <m.tretter@pengutronix.de>
7779 * gst/vaapi/gstvaapipostproc.c:
7780 vaapipostproc: lock ensure_filter with postproc_lock
7781 gst_vaapipostproc_ensure_filter might free the allowed_srcpad_caps
7782 and allowed_sinkpad_caps. This can race with copying these caps in
7783 gst_vaapipostproc_transform_caps and lead to segfaults.
7784 The gst_vaapipostproc_transform_caps function already locks
7785 postproc_lock before copying the caps. Make sure that calls to
7786 gst_vaapipostproc_ensure_filter also acquire this lock.
7787 https://bugzilla.gnome.org/show_bug.cgi?id=791404
7789 2018-01-10 17:10:28 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7791 * tests/test-filter.c:
7792 tests: test-filter: fix dereference before null check
7793 Null-checking op_info suggests that it may be null, but it has already
7794 been dereferenced on all paths leading to the check.
7795 There may be a null pointer dereference, or else the comparison
7796 against null is unnecessary.
7798 2018-01-10 17:06:53 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7800 * tests/y4mreader.c:
7801 tests: y4mreader: fix string state checkup
7802 str cannot be null in that moment, but it may be the end of string.
7804 2018-01-10 16:59:56 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7806 * tests/y4mreader.c:
7807 tests: y4mreader: use int for fgetc
7808 Assigning the return value of fgetc to char truncates its value.
7809 It will not be possible to distinguish between EOF and a valid
7812 2018-01-10 16:48:07 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7814 * tests/y4mreader.c:
7815 tests: y4mreader: fix incompatible cast
7816 Passed pointer in parse_int() are unsigned int (32 bits, unsigned) but
7817 they are dereferenced as a wider long (64 bits, signed). This may lead
7818 to memory corruption.
7820 2017-12-19 16:01:10 +0000 Tim-Philipp Müller <tim@centricular.com>
7823 meson: fix fallback for gstreamer-gl-1.0, it's now in -base
7825 2017-12-14 14:53:27 +1100 Matthew Waters <matthew@centricular.com>
7828 Automatic update of common submodule
7829 From e8c7a71 to 3fa2c9e
7831 2017-12-06 16:11:46 -0500 Nicolas Dufresne <nicolas.dufresne@collabora.com>
7833 * gst/vaapi/gstvaapivideocontext.c:
7834 videoconvert: gst_element_post_message() is transfer full on msg
7835 For this reson we need not to unref the message, even if it failed.
7837 2017-12-06 16:11:25 -0500 Nicolas Dufresne <nicolas.dufresne@collabora.com>
7839 * gst/vaapi/gstvaapivideocontext.c:
7840 Revert "vaapivideocontext: possible memleak when no bus attached"
7841 This reverts commit 0438a3e62660e64ed390b6bb83bfb560b91664aa.
7843 2017-12-01 23:03:32 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7845 * tests/elements/test-vaapicontext.c:
7846 test: vaapicontext: process have-context bus message
7848 2017-11-29 18:29:45 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7850 * tests/elements/test-vaapicontext.c:
7851 test: vaapicontext: app context is not persistent
7853 2017-11-29 11:02:03 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7855 * gst/vaapi/gstvaapipluginbase.c:
7856 * gst/vaapi/gstvaapivideocontext.c:
7857 * gst/vaapi/gstvaapivideocontext.h:
7858 vaapivideocontext: only vaapisink process app context
7859 gst.vaapi.app.Display context is made for applications that will
7860 provide the VA display and the native display to used by the
7861 pipeline, when are using vaapisink as overlay. There are no use
7862 case for encoders, decoders, neither for the postprocessor.
7863 In the case of the vaapisink, it shall query for gst.vaapi.Display
7864 upstream first, and then, if there is no reply,
7865 gst.vaapi.app.Display context will be posted in the bus for the
7866 application. If the application replies, a GstVaapiDisplay object
7867 is instantiated given the context info, otherwise a
7868 GstVaapiDisplay is created with the normal algorithm to guess the
7869 graphics platform. Either way, the instantiated GstVaapiDisplay
7870 is propagated among the pipeline and the have-message bus message.
7871 Also only vaapisink will process the gst.vaapi.app.Display, if
7872 and only if, it doesn't have a display already set. This is
7873 caused because if vaapisink is in a bin (playsink, for example)
7874 the need-context is posted twice, leading to an error state.
7875 https://bugzilla.gnome.org/show_bug.cgi?id=790999
7877 2017-12-01 20:21:54 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7879 * gst/vaapi/gstvaapivideocontext.c:
7880 vaapivideocontext: log the name of GstVaapiDisplay
7881 https://bugzilla.gnome.org/show_bug.cgi?id=790999
7883 2017-11-30 14:24:43 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7885 * gst/vaapi/gstvaapivideocontext.c:
7886 vaapivideocontext: possible memleak when no bus attached
7887 https://bugzilla.gnome.org/show_bug.cgi?id=790999
7889 2017-11-27 13:04:24 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7891 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.h:
7892 libs: window: wayland: remove unused header include
7893 Remove wayland-client.h include since there is no exposed symbols from
7896 2017-11-27 12:18:56 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
7898 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
7899 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
7900 libs: encoder: h264,h265: guard rate control's macroblock
7901 macroblock parameter appear on VA-API 1.0.0. It should be guarded.
7903 2017-11-27 20:17:55 +1100 Matthew Waters <matthew@centricular.com>
7906 Automatic update of common submodule
7907 From 3f4aa96 to e8c7a71
7909 2016-07-29 14:58:49 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
7911 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
7912 libs: encoder: h264: Add Hierarchical-B encode
7913 Frames are encoded as different layers. Frame in a particular
7914 layer will use pictures in lower or same layer as references.
7915 Which means decoder can drop the frames in upper layer but still
7916 decode lower layer frames.
7917 B-frames, except the one in top most layer, are reference frames.
7918 All the base layer frames are I or P.
7919 eg: with 3 temporal layers
7923 T1, T2, T3: Temporal Layers
7926 T1: I0->P4 , P4->P8 etc..
7928 T3: I0--> B1 <-- B2, B2 --> B3 <-- P4
7929 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
7930 https://bugzilla.gnome.org/show_bug.cgi?id=788918
7932 2016-07-28 18:33:23 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
7934 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
7935 libs: encoder: h264: Add Hierarchical-P encode
7936 Frames are encoded as different layers. A frame in a particular
7937 layer will use pictures in lower or same layer as references.
7938 Which means decoder can drop the frames in upper layer but still
7939 decode lower layer frames.
7940 eg: with 3 temporal layers
7944 T1, T2, T3: Temporal Layers
7946 P0->P1 , P0->P2, P2->P3, P0->P4......repeat
7947 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
7948 https://bugzilla.gnome.org/show_bug.cgi?id=788918
7950 2016-07-28 16:51:28 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
7952 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
7953 libs: encoder: h264: Fix frame_num generation
7954 The frame_num generation was not correctly implemented.
7955 According to h264 spec, frame_num should get incremented
7956 for each frame if previous frame is a referece frame.
7957 For eg: IPBPB sequece should have the frame numbers 0,1,2,2,3
7958 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
7959 https://bugzilla.gnome.org/show_bug.cgi?id=788918
7961 2016-07-28 15:53:48 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
7963 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
7964 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
7965 libs: encoder: h264: Add new property "prediction-type"
7966 Adds new property "prediction-type" to select different reference
7967 picture selection modes like hierarchical-p, hierarchical-b etc.
7968 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
7969 https://bugzilla.gnome.org/show_bug.cgi?id=788918
7971 2016-07-28 15:12:05 +0300 XuGuangxin <guangxin.xu@intel.com>
7973 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
7974 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
7975 libs: encoder: h264: Add machinery for implementing hierarchical-prediction
7976 Adds some basic building blocks to ease the implementation
7977 of hierarchical prediction modes.
7978 -- add an utility method to find temporal level of each frame
7979 -- define max_ref_frame count based on temporal level count
7980 -- add temporal_level_div[] for finding temporal level each frame
7982 -- find ip_period based on temporal level count
7983 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
7984 https://bugzilla.gnome.org/show_bug.cgi?id=788918
7986 2016-07-28 14:17:53 +0300 XuGuangxin <guangxin.xu@intel.com>
7988 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
7989 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
7990 libs: encoder: h264: Add property "temporal-levels"
7991 Adds new property "temporal-levels" to select the number of
7992 temporal levels to be included in the encoded stream.
7993 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
7994 https://bugzilla.gnome.org/show_bug.cgi?id=788918
7996 2016-07-27 16:41:01 +0300 XuGuangxin <guangxin.xu@intel.com>
7998 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
7999 libs: encoder: objects: Add a reference flag
8000 We can have p-frame as non-ref and also b-frame as ref
8001 which are not supported yet. Reference flag
8002 is the first machinery needed for more advanced
8003 reference picture selection modes.
8004 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8005 https://bugzilla.gnome.org/show_bug.cgi?id=788918
8007 2017-11-02 13:21:34 +0100 Daniel van Vugt <daniel.van.vugt@canonical.com>
8009 * gst-libs/gst/vaapi/gstvaapisurface_egl.c:
8010 libs: surface: egl: add comment
8011 Add a warning comment when using old intel-vaapi-drivers (>1.8.4),
8012 where the creation of surfaces from GEM fd may fail.
8013 https://bugzilla.gnome.org/show_bug.cgi?id=773453
8015 2017-10-10 13:38:21 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8017 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
8018 * gst-libs/gst/vaapi/gstvaapidisplay_egl.h:
8019 * gst-libs/gst/vaapi/gstvaapitexture_egl.c:
8020 libs: display: egl: add gst_vaapi_display_egl_set_current_display()
8021 Adds a new function that changes the internal EGL display to the
8022 current one (eglGetCurrentDisplay()) and sets the current context
8023 too (eglGetCurrentContext()).
8024 This new function is called by gst_vaapi_texture_egl_create() updating
8025 the GstVaapiDisplayEGL with the current EGL display.
8026 https://bugzilla.gnome.org/show_bug.cgi?id=773453
8028 2017-10-09 16:02:11 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8030 * gst-libs/gst/vaapi/gstvaapitexture_egl.c:
8031 libs: texture: egl: update EGL display and context
8032 It is required to use the context of the calling thread when wrapping
8033 a foreign texture. According the documentation of
8034 GstVideoGLTextureUploadMeta:
8035 "The caller of gst_video_gl_texture_upload_meta_upload() must
8036 have OpenGL set up and call this from a thread where it is valid
8037 to upload something to an OpenGL texture."
8038 This patch updates the EGL display and context in GstVaapiDisplay
8039 instance to the one used by te renderer that uploads the texture.
8040 Original-patch-by: Daniel van Vugt <daniel.van.vugt@canonical.com>
8041 https://bugzilla.gnome.org/show_bug.cgi?id=773453
8043 2017-10-10 19:53:04 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8045 * gst/vaapi/gstvaapipluginbase.c:
8046 plugins: centralize assignation of GL objects
8047 Add plugin_set_gst_gl() where the GstGL objects are assigned.
8048 https://bugzilla.gnome.org/show_bug.cgi?id=773453
8050 2017-10-10 19:13:35 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8052 * gst/vaapi/gstvaapipluginbase.c:
8053 plugins: set GL objects if ensured
8054 Only set the GL display and GL other context if they are ensured.
8055 https://bugzilla.gnome.org/show_bug.cgi?id=773453
8057 2017-10-10 17:14:15 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8059 * gst/vaapi/gstvaapipluginbase.c:
8060 plugins: set GL objects if context is handled
8061 Only set the GL display and GL other context if they are extracted
8062 correctly from the gstreamer's context.
8063 https://bugzilla.gnome.org/show_bug.cgi?id=773453
8065 2017-10-10 19:57:45 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8067 * gst/vaapi/gstvaapipluginbase.c:
8068 plugins: fix memory leak when GL context is created
8069 When the GL display and context are created inside an VAAPI element
8070 the created GL context is leaked.
8071 https://bugzilla.gnome.org/show_bug.cgi?id=773453
8073 2017-10-10 14:01:59 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8075 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
8076 libs: display: egl: free leaked memory
8077 The EGL VAAPI display forgot to release the egl display, context and
8078 proxied VAAPI display.
8079 https://bugzilla.gnome.org/show_bug.cgi?id=773453
8081 2017-10-05 19:25:08 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8083 * gst-libs/gst/vaapi/gstvaapitexture_egl.c:
8084 libs: texture: egl: code style
8085 https://bugzilla.gnome.org/show_bug.cgi?id=773453
8087 2017-10-04 13:51:23 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8089 * gst-libs/gst/vaapi/gstvaapisurface_egl.c:
8090 libs: surface: egl: error message if no extension
8091 Instead of silently fail to export the image if there is not available
8092 the EGL_MESA_drm_image, log an error message. Also a code refactoring
8094 https://bugzilla.gnome.org/show_bug.cgi?id=773453
8096 2017-10-31 13:10:50 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8098 * gst/vaapi/gstvaapipluginbase.c:
8099 plugins: direct rendering on memory:VASurface
8100 As buffers negotiated with memory:VASurface caps feature can also be
8101 mapped, they can also be configured to use VA derived images, in other
8102 words "direct rendering".
8103 Also, because of the changes in dmabuf allocator as default allocator,
8104 the code for configuring the direct rendering was not clear.
8105 This patch cleans up the code and enables direct rendering when the
8106 environment variable GST_VAAPI_ENABLE_DIRECT_RENDERING is defined,
8107 even then the memory:VASurface cap feature is negotiated.
8108 https://bugzilla.gnome.org/show_bug.cgi?id=786054
8110 2017-10-04 11:54:31 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8112 * gst/vaapi/gstvaapipluginbase.c:
8113 * gst/vaapi/gstvaapipluginutil.c:
8114 plugins: only dmabuf on srcpad if downstream
8115 Set if source pad can handle dmabuf only if the GstGL context comes
8117 It is possible to know that at two moments:
8118 1\ In the case of GstGLTextureUpload caps feature is negotiated and
8119 downstream pool reports back gst.gl.GstGLContext.
8120 2\ When GstGLContext is found as GstContext from dowstream.
8121 https://bugzilla.gnome.org/show_bug.cgi?id=788503
8123 2017-10-04 11:52:32 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8125 * gst/vaapi/gstvaapivideocontext.c:
8126 vaapivideocontext: add inline documentation
8127 Document function gst_vaapi_find_gl_local_context().
8128 https://bugzilla.gnome.org/show_bug.cgi?id=788503
8130 2017-10-04 11:50:25 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8132 * gst/vaapi/gstvaapipluginutil.c:
8133 * gst/vaapi/gstvaapivideocontext.c:
8134 * gst/vaapi/gstvaapivideocontext.h:
8135 vaapivideocontext: return the direction of gl context
8136 In function gst_vaapi_find_gl_context() add a direction parameter to
8137 return back the direction where the GstGL context was found.
8138 This is going to be useful when checking if downstream can import
8139 dmabuf-based buffers.
8140 https://bugzilla.gnome.org/show_bug.cgi?id=788503
8142 2017-10-04 08:30:51 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8144 * gst/vaapi/gstvaapipluginbase.c:
8145 * gst/vaapi/gstvaapipluginbase.h:
8146 plugins: add gst_vaapi_plugin_base_set_srcpad_can_dmabuf()
8147 This patch refactors the code by adding the function
8148 vaapi_plugin_base_set_srcpad_can_dmabuf(), it determines if the passed
8149 GstGLContext can handle dmabuf-based buffers.
8150 The function is exposed publicly since it is intended to be used later
8151 at GstVaapiDisplay instantiation.
8152 https://bugzilla.gnome.org/show_bug.cgi?id=788503
8154 2017-10-20 12:37:15 +0200 Hyunjun Ko <zzoon@igalia.com>
8156 * gst/vaapi/gstvaapiencode.c:
8157 vaapiencode: allow to set property on runtime
8158 Tis patch, allows some properties that we want to be set on
8159 runtime. (eg. bitrate)
8160 Note that all properties are under control by num_codedbuf_queued.
8161 https://bugzilla.gnome.org/show_bug.cgi?id=786321
8163 2017-09-15 15:38:18 +0900 Hyunjun Ko <zzoon@igalia.com>
8165 * gst-libs/gst/vaapi/gstvaapiencoder.c:
8166 libs: encoder: allow to set bitrate on runtime
8167 In case of streaming, controlling bitrate dynamically for encoder might be
8168 important to manage quality of the streaming.
8169 This patch is to support such a scenario.
8170 https://bugzilla.gnome.org/show_bug.cgi?id=786321
8172 2017-10-10 11:35:24 +0300 Sebastian Dröge <sebastian@centricular.com>
8174 * gst/vaapi/gstvaapi.c:
8175 * gst/vaapi/gstvaapi.h:
8176 * gst/vaapi/gstvaapidecodebin.c:
8177 vaapidecodebin: Register element if no VPP support is available too
8178 VPP support is only needed for advanced deinterlacing, which is not
8179 enabled by default either. Error out if it is selected but VPP is not
8180 supported, and otherwise just work without VPP support.
8181 https://bugzilla.gnome.org/show_bug.cgi?id=788758
8183 2017-10-16 11:57:16 +0200 Thibault Saunier <thibault.saunier@osg.samsung.com>
8185 * gst/vaapi/gstvaapipluginutil.c:
8186 Avoid infinite loop when vaapi_create_display fails
8187 Which might be the case when using, for example, xvfb.
8189 2017-10-02 18:53:21 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8191 * gst-libs/gst/vaapi/gstvaapiutils.c:
8192 libs: utils: log warn if display fail
8193 gstreamer-vaapi initializes the display by trial-and-error, thus
8194 logging an error message if the display initialisation fails the user
8195 may be weary of the error message in the screen, if using VA-API 1.0
8196 This commit set the VA error log handler to GStreamer warning level
8197 while calling vaInitialize() and set it to error after that.
8198 https://bugzilla.gnome.org/show_bug.cgi?id=783169
8200 2017-09-29 20:05:22 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8202 * gst/vaapi/gstvaapipluginutil.c:
8203 plugins: try to create test display in order
8204 When creating the test display for querying capabilites, it try in
8205 certain order: DRM, Wayland and finally X11. GLX nor EGL are tried
8206 since they are either composited with X11 or Wayland.
8207 The reason for this is to reduce the posibility of failure that could
8208 blacklist the plugin.
8209 https://bugzilla.gnome.org/show_bug.cgi?id=782212
8211 2017-09-29 15:07:47 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8213 * gst-libs/gst/vaapi/gstvaapidisplay.c:
8214 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
8215 libs: display: delay getting screen resolution
8216 Instead of extracting the screen resolution at GstVaapiDisplay
8217 creation, this patch delay it until the screen size is requested for
8219 https://bugzilla.gnome.org/show_bug.cgi?id=782212
8221 2017-09-28 18:58:17 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8223 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
8224 libs: display: egl: avoid two vaDisplay instantiates
8225 GstVaapiDisplayEGL is a wrapper of another GstVaapiDisplay, either X11
8226 or Wayland. Nonetheless it created another vaDisplay for it, instead
8227 of using the wrapped one.
8228 This patch enables the reuse of the wrapped vaDisplay avoiding
8230 https://bugzilla.gnome.org/show_bug.cgi?id=782212
8232 2017-09-28 17:45:00 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8234 * gst-libs/gst/vaapi/gstvaapidisplay.c:
8235 * gst-libs/gst/vaapi/gstvaapidisplay.h:
8236 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
8237 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
8238 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
8239 libs: display: remove display_type from display info
8240 Since it's no required to pass the display type in the display info,
8241 the structure member is removed.
8242 https://bugzilla.gnome.org/show_bug.cgi?id=782212
8244 2017-09-28 17:35:01 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8246 * gst-libs/gst/vaapi/gstvaapidisplay.c:
8247 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
8248 libs: display: remove display_type member
8249 It is not used any more since GstVaapiDisplay was ported as a
8250 GstObject-based. This information is part of the class information.
8251 https://bugzilla.gnome.org/show_bug.cgi?id=782212
8253 2017-09-28 16:12:23 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8255 * gst-libs/gst/vaapi/gstvaapidisplay.c:
8256 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
8257 libs: display: remove parent member
8258 Parent was a crumb left from display cache.
8259 https://bugzilla.gnome.org/show_bug.cgi?id=782212
8261 2017-10-03 13:06:33 +0200 Sebastian Dröge <sebastian@centricular.com>
8263 * gst/vaapi/gstvaapi.c:
8264 vaapi: Also register vaapipostproc without VPP support
8265 It can still do simple deinterlacing then.
8267 2017-10-03 10:51:06 +0200 Sebastian Dröge <sebastian@centricular.com>
8269 * gst/vaapi/gstvaapipostproc.c:
8270 vaapipostproc: Allow running without VPP support
8271 We returned FALSE from ::start() if VPP support is not available, but it
8272 is only really needed for complex filters and during transform we check
8273 for that. For simple deinterlacing it is not needed.
8275 2017-09-27 18:35:20 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8277 * gst/vaapi/gstvaapipostproc.c:
8278 vaapipostproc: use scoped variable for return value
8279 Instead of reusing a parameter variable for the return value of
8280 gst_vaapipostproc_transform_caps(), this patch uses the function
8281 scoped pointer. Thus, the code is cleaner.
8282 https://bugzilla.gnome.org/show_bug.cgi?id=785706
8284 2017-09-27 18:32:03 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8286 * gst/vaapi/gstvaapipostproc.c:
8287 vaapipostproc: removed unused parameter
8288 Removed caps parameter from gst_vaapipostproc_transform_caps_impl()
8289 helper function since the it is not used.
8290 https://bugzilla.gnome.org/show_bug.cgi?id=785706
8292 2017-09-27 13:32:34 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8294 * gst/vaapi/gstvaapipostproc.c:
8295 vaapipostproc: use scoped variable for return value
8296 Instead of reusing a parameter variable for the return value of
8297 gst_vaapipostproc_fixate_caps(), this patch uses the function scoped
8298 pointer. Thus, the code is cleaner.
8299 https://bugzilla.gnome.org/show_bug.cgi?id=785706
8301 2017-09-27 11:27:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8303 * gst/vaapi/gstvaapipluginutil.h:
8304 plugins: memory:DMABuf only handles planar formats
8305 When glimagesink negotiates the caps feature memory:DMABuf the
8306 exported dmabufs buffers with NV12 format are not well rendered, thus
8307 setting only planar.
8308 https://bugzilla.gnome.org/show_bug.cgi?id=788229
8310 2017-09-25 17:04:12 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8312 * gst/vaapi/gstvaapiencode.c:
8313 vaapiencode: flush pending frames before set format
8314 Flush pending frames, if any, in the internal encorder, before setting
8315 the new negotiated format.
8316 https://bugzilla.gnome.org/show_bug.cgi?id=786173
8318 2017-09-25 15:50:19 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8320 * gst/vaapi/gstvaapidecode.c:
8321 vaapidecode: drain pending frames before set format
8322 Drain pending frames, if any, in the internal decoder before setting
8323 the new negotiated format.
8324 https://bugzilla.gnome.org/show_bug.cgi?id=786173
8326 2017-09-22 19:35:04 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8328 * tests/test-display.c:
8329 tests: display: use GObject getter
8330 Instead of using the gst_vaapi_display_get_property(), this patch
8331 replaces it with g_object_get_property() to dump the available VA
8333 https://bugzilla.gnome.org/show_bug.cgi?id=788058
8335 2017-09-22 19:25:20 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8337 * gst/vaapi/gstvaapisink.c:
8338 vaapisink: use GObject setter and getter
8339 Instead of using gst_vaapi_display_set_property() or
8340 gst_vaapi_display_get_property(), this patch set replace it usage
8341 with g_object_set() or g_object_get().
8342 Also the internal helper cb_set_value() is removed since it is not
8344 https://bugzilla.gnome.org/show_bug.cgi?id=788058
8346 2017-09-22 18:59:49 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8348 * gst-libs/gst/vaapi/gstvaapidisplay.c:
8349 libs: display: initialize value if they are not yet
8350 This is a difference between the GObject API and the GstVaapi one: the
8351 GValue passed to get a property value, in GObject has to be
8352 initialized with g_value_init(), but in GstVaapi is has not.
8353 In order to overcome this mismatch, this patch call g_value_init()
8354 internally only in the passed one is not already initialized.
8355 https://bugzilla.gnome.org/show_bug.cgi?id=788058
8357 2017-09-22 17:04:17 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8359 * gst-libs/gst/vaapi/gstvaapidisplay.c:
8360 libs: display: optimize properties setters and getters
8361 Shuffled some code to avoid to find the properties descriptor in the
8362 array twice, adding the internal functions _set_property() and
8364 https://bugzilla.gnome.org/show_bug.cgi?id=788058
8366 2017-09-22 16:29:02 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8368 * gst-libs/gst/vaapi/gstvaapidisplay.c:
8369 libs: display: install properties in class
8370 Install the properties in the class as a normal GObject. Implement
8371 set_property() and get_property() vmethods.
8372 https://bugzilla.gnome.org/show_bug.cgi?id=788058
8374 2017-09-22 15:16:34 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8376 * gst-libs/gst/vaapi/gstvaapidisplay.c:
8377 libs: display: remove gst_vaapi_display_properties_init()
8378 Remove gst_vaapi_display_properties_init() since it can be unrolled in
8379 gst_vaapi_display_class_init()
8380 https://bugzilla.gnome.org/show_bug.cgi?id=788058
8382 2017-09-22 15:12:05 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8384 * gst-libs/gst/vaapi/gstvaapidisplay.c:
8385 libs: display: remove libgstvaapi_init_once()
8386 It is not required since it can be unrolled in
8387 gst_vaapi_display_class_init()
8388 https://bugzilla.gnome.org/show_bug.cgi?id=788058
8390 2017-09-22 17:50:15 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8392 * tests/test-display.c:
8393 tests: test-display: remove display cache tests
8394 Since commit ec3e10f6, display cache was removed. This patch removes
8395 this leftovers in the display test.
8397 2017-09-18 14:29:55 +0900 Hyunjun Ko <zzoon@igalia.com>
8399 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
8400 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
8401 libs: decoder: h264/h265: decode codec data only if opened
8402 Fixes regression introduced by commit 2eb2b26a.
8403 There is a use case when the decoder set the src caps and immediatly
8404 tries to process the media codec_data, this happens before decoder is
8405 even opened, thus priv->parser is not instantiated yet.
8406 https://bugzilla.gnome.org/show_bug.cgi?id=787818
8408 2017-09-18 19:11:45 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8410 * gst-libs/gst/vaapi/gstvaapiencoder.c:
8411 * gst-libs/gst/vaapi/gstvaapiencoder.h:
8412 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
8413 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
8414 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
8415 libs: encoder: change mbbrc from uint to enum
8416 Instead of handling the macroblock bitrate control as a integer, this
8417 patch changes it as a enum, which is more self documented in the
8419 https://bugzilla.gnome.org/show_bug.cgi?id=787855
8421 2017-09-18 13:55:49 +1000 Jan Schmidt <jan@centricular.com>
8423 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
8424 Fix a typo in the prop string for compliance-mode
8426 2017-09-15 18:31:49 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8428 * gst-libs/gst/vaapi/gstvaapiencoder.c:
8429 libs: encoder: don't unref properties
8430 This patch fixes a regression introduced in commit 148f867c, since the
8431 props variable is set to object's member variable
8432 encoder->properties. And it is set in the instance initialization,
8433 thus it will not be leaked.
8434 https://bugzilla.gnome.org/show_bug.cgi?id=787733
8436 2017-09-15 15:14:47 +0900 Hyunjun Ko <zzoon@igalia.com>
8438 * gst-libs/gst/vaapi/gstvaapiencoder.c:
8439 * gst/vaapi/gstvaapiencode.c:
8440 vaapiencode/libs: encoder: fix leaks of properties
8441 https://bugzilla.gnome.org/show_bug.cgi?id=786321
8443 2017-08-24 21:51:22 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8445 * gst-libs/gst/vaapi/gstvaapidecoder.c:
8446 libs: decoder: at update_caps() decode codec_data
8447 When updating the caps in decoder, if the caps has codec_data (avC
8448 format), it has to be parsed to update the state of the decoder.
8449 https://bugzilla.gnome.org/show_bug.cgi?id=786173
8451 2017-09-13 15:44:32 +0900 Hyunjun Ko <zzoon@igalia.com>
8453 * gst-libs/gst/vaapi/gstvaapicontext.c:
8454 libs: context: fix wrong counter of the array of attributes
8455 The counter value passed to vaCreateConfig is always +1.
8456 This is a regression caused by commit e42ec3ad.
8457 The present patch fixes wrong counting of the array of attributes.
8458 https://bugzilla.gnome.org/show_bug.cgi?id=787613
8460 2017-09-13 12:23:42 +0900 Hyunjun Ko <zzoon@igalia.com>
8462 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
8463 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
8464 libs: encoder: h265: support I/P/B QP setting seperatedly
8465 Creates 2 properties, qp-ip and qp-ib for setting different QP for P/B
8467 and set slice_qp_delta for each frame according to the value provided.
8468 https://bugzilla.gnome.org/show_bug.cgi?id=785923
8470 2017-09-13 12:22:07 +0900 Hyunjun Ko <zzoon@igalia.com>
8472 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
8473 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
8474 libs: encoder: h264: support I/P/B QP setting seperatedly
8475 Creates 2 properties, qp-ip and qp-ib for setting different QP for P/B
8477 and set slice_qp_delta for each frame according to the value provided.
8478 In addition, remove the limitation of (<= 4) when setting
8480 https://bugzilla.gnome.org/show_bug.cgi?id=785923
8482 2017-09-13 12:15:57 +0900 Hyunjun Ko <zzoon@igalia.com>
8484 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
8485 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
8486 libs: encoder: h264/h265: keep min_qp as is unless it's over init_qp
8487 Creates new variable for QP for I frame and keep it at configuration and
8488 use this for pic_init_qp and slice_qp_delta setting.
8489 Since changing min qp doesn't make sense, keep min qp as is.
8490 https://bugzilla.gnome.org/show_bug.cgi?id=785923
8492 2017-09-13 12:09:45 +0900 Hyunjun Ko <zzoon@igalia.com>
8494 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
8495 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
8496 libs: encoder: h265: Add mbbrc property
8497 This property supports Macroblock level Bitrate Control as the
8498 following (same as h264 encoder):
8502 https://bugzilla.gnome.org/show_bug.cgi?id=785917
8504 2017-09-13 12:02:53 +0900 Hyunjun Ko <zzoon@igalia.com>
8506 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
8507 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
8508 libs: encoder: h264: Add mbbrc property
8509 This property supports Macroblock level Bitrate Control as the
8514 https://bugzilla.gnome.org/show_bug.cgi?id=785917
8516 2017-09-13 11:39:09 +0900 Hyunjun Ko <zzoon@igalia.com>
8518 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
8519 libs: encoder: h265: add multi reference support
8520 This is doing the same as h264 encoder as the following:
8521 Using num_ref_frames provided and the result of the Query
8522 VAConfigAttribEncMaxRefFrames, it determines the size of reference list
8523 and perform encoding with multi reference frames as the following:
8524 1\ The num_ref_frames is being considered as the number of
8525 reference picture list0
8526 2\ Encoder adds 1 reference frame more to the reference picture list1
8527 internally if b-frame encoding.
8528 3\ If num_ref_frames is bigger than the number of refrence frames
8529 supported in the driver, it will be lowered.
8530 Also this patch includes:
8531 - Set num_negative_pics and num_positive_pics according to the number of
8533 - Set delta_poc according to the number of refs.
8534 - Increase max_dec_pic_buffering according to the number of refs
8535 - Change max_num_reorder_pics according to num of bframes
8536 https://bugzilla.gnome.org/show_bug.cgi?id=783804
8538 2017-09-13 11:37:33 +0900 Hyunjun Ko <zzoon@igalia.com>
8540 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
8541 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
8542 libs: encoder: h265: add refs property
8543 Users can provide the number of reference frame by this property,
8544 which is exaclty same as h264.
8545 The value of the property will be considered as the number of
8546 reference picture list0 and will add 1 reference frame more to the
8547 reference picture list1 internally if b-frame encoding.
8548 If the value provided is bigger than the number of refrence frames
8549 supported in the driver, it will be lowered.
8550 The maximum value is aligned to the value of the driver supported now.
8551 https://bugzilla.gnome.org/show_bug.cgi?id=783804
8553 2017-09-13 11:17:26 +0900 Hyunjun Ko <zzoon@igalia.com>
8555 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
8556 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
8557 libs: encoder: h264/5: determine num_ref_idx_active_override_flag according to reference list
8558 Follows the specification as below:
8559 7.4.7.1 in Rec. ITU-T H.265 v4 (12/2016)
8560 num_ref_idx_active_override_flag equal to 1 specifies that the syntax
8561 element num_ref_idx_l0_active_minus1 is present for P and B slices and
8562 that the syntax element num_ref_idx_l1_active_minus1 is present for B
8564 num_ref_idx_active_override_flag equal to 0 specifies that the syntax
8565 elements num_ref_idx_l0_active_minus1 and num_ref_idx_l1_active_minus1
8567 https://bugzilla.gnome.org/show_bug.cgi?id=783804
8569 2017-09-13 11:06:20 +0900 Hyunjun Ko <zzoon@igalia.com>
8571 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
8572 libs: encoder: h265: keep idr_period equal to keyframe period
8573 Remove FIXME code, which makes previous assignation spurious.
8574 This also means to make idr_period equal to keyframe period,
8575 which is same as h264 encoder.
8576 https://bugzilla.gnome.org/show_bug.cgi?id=783804
8578 2017-09-06 14:03:25 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
8580 * gst/vaapi/gstvaapipluginbase.c:
8581 Request minimum buffer even if need_pool is FALSE
8582 When tee is used, it will not request a pool, but still it wants to
8583 know how many buffers are required.
8584 https://bugzilla.gnome.org/show_bug.cgi?id=730758
8586 2017-09-05 10:58:57 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
8588 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
8589 libs: encoder: h264_fei: VA-API 1.0 compat
8590 Use VA_ENC_PACKED_HEADER_H264_SEI compat macro for VA-API 1.0
8592 https://bugzilla.gnome.org/show_bug.cgi?id=787322
8593 Signed-off-by: U. Artie Eoff <ullysses.a.eoff@intel.com>
8595 2017-09-01 13:48:01 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8597 * gst/vaapi/gstvaapisink.c:
8598 vaapisink: Fix rendering in drm display
8599 Make sure vaapisink create a va surface backed buffer pool and all
8600 required attributes get assigned correctly for drm display type.
8601 This is needed to make the below pipeline working:
8602 gst-launch-1.0 filesrc location= raw_video.mov ! videoparse format=uyvy
8603 width=320 height=240 framerate=30/1 ! vaapisink display=drm
8604 https://bugzilla.gnome.org/show_bug.cgi?id=786954
8606 2017-08-09 18:46:09 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8608 * tests/Makefile.am:
8609 * tests/test-fei-enc-in.c:
8610 * tests/test-fei-enc-out.c:
8611 FEI: Add test applications to showcase fei use case
8612 test-fei-enc-out: A simple fei encoding application to output mv, mbcode and distortion
8614 ./test-fei-enc-out -i sample_320x240.nv12 -w 320 -h 240 -o out.264 -v mv.out -d out.dist -m out.mbcode -e 1
8615 test-fei-enc-in: A simple fei encoding application for testing input fei buffers
8617 ./test-fei-enc-in -c h264 -o out.264 -e 4 -q 1 sample_i420.y4m
8618 Fixme: Running test-fei-enc-in in PAK mode with mv and mbcode input buffers
8619 from saved files is still not working
8621 Wang, Yi <yi.a.wang@intel.com>
8622 Leilei <leilei.shang@intel.com>
8623 Zhong, Xiaoxia <xiaoxia.zhong@intel.com>
8624 xiaominc <xiaomin.chen@intel.com>
8625 Li, Jing B <jing.b.li@intel.com>
8626 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8627 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8629 2017-08-09 18:36:13 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8631 * gst/vaapi/Makefile.am:
8632 * gst/vaapi/gstvaapi.c:
8633 * gst/vaapi/gstvaapiencode_h264_fei.c:
8634 * gst/vaapi/gstvaapiencode_h264_fei.h:
8635 FEI: plugin: Add vaapih264feienc element
8636 A new FEI based encoder element for h264 is added: vaapih264feienc
8637 FEI is a an extension to VA-API which is providing low level
8638 advanced control over different stages of encoding.
8639 Extending vaapih264enc with fei support is possible, but it will
8640 make the code too much complicated and will be difficult
8641 to debug. So adding the new encoder element, but keeping
8642 the rank as 0 , vaapih264enc will stay as the primary
8643 encoder for normal use cases.
8644 The vaaih264feienc is mainly useful for customers who want to play
8645 with MotionVectors and Macroblock Predictions. Also user can
8646 do one stage of encoding(eg: only the Motion Vector Calculation)
8647 in software and offload trasformation/entroy-coding etc to
8648 Hardware (which is what PAK module is doing) using FEI element.
8649 vaapih264feienc can work in different modes using fei-mode properoty
8650 eg: gst-launch-1.0 videotestsrc ! vaapih264feienc fei-mode=ENC+PAK ! filesink location=sample.264
8651 Important Note: ENC only mode won't produce any encoded data which is expected.
8652 But ENC alwys requires the output of PAK in order to do the inter-prediction
8653 over reconstructed frames.
8654 Similary PAK mode alway requires MV and MBCode as input, so unless there is an
8655 upstream element providing those buffers, PAK only won't work as expected.
8656 In a nutshell, ENC_PAK and the ENC+PAK modes are the only options we can verify
8657 with vaapih264feienc. But ideally, EN+PAK mode verification is enough to make sure
8658 that ENC and PAK are working as expected since ENC+PAK mode always invoke ENC and PAK
8659 separately in vaapih264feienc.
8661 Wang, Yi <yi.a.wang@intel.com>
8662 Leilei <leilei.shang@intel.com>
8663 Zhong, Xiaoxia <xiaoxia.zhong@intel.com>
8664 xiaominc <xiaomin.chen@intel.com>
8665 Li, Jing B <jing.b.li@intel.com>
8666 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8667 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8668 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8670 2017-08-09 18:32:13 -0700 Yi A Wang <yi.a.wang@intel.com>
8672 * gst/vaapi/gstvaapiencode.c:
8673 * gst/vaapi/gstvaapiencode.h:
8674 FEI: plugin: Add virtual methods to base encode
8675 Two new virtual methods are added to gstvaapiencode.
8676 load_control_data(): load the FEI input buffers set by the upstream elements
8677 save_stats_to_meta(): save the FEI output buffers to Meta for downnstream elements
8678 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8679 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8680 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8682 2017-08-09 18:26:57 -0700 Yi A Wang <yi.a.wang@intel.com>
8684 * gst/vaapi/Makefile.am:
8685 * gst/vaapi/gstvaapifeivideometa.c:
8686 * gst/vaapi/gstvaapifeivideometa.h:
8687 FEI: plugin: Add fei specific video meta
8688 GstVaapiFeiVideoMeta holds the below fei codec objects:
8689 GstVaapiEncFeiMbCode
8691 GstVaapiEncFeiMvPredictor
8692 GstVaapiEncFeiMbControl
8694 GstVaapiEncFeiDistortion
8695 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8696 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8697 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8699 2017-08-09 18:19:06 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8701 * gst-libs/gst/vaapi/Makefile.am:
8702 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.c:
8703 * gst-libs/gst/vaapi/gstvaapiencoder_h264_fei.h:
8704 * gst-libs/gst/vaapi/gstvaapifeienc_h264.c:
8705 * gst-libs/gst/vaapi/gstvaapifeienc_h264.h:
8706 * gst-libs/gst/vaapi/gstvaapifeipak_h264.c:
8707 * gst-libs/gst/vaapi/gstvaapifeipak_h264.h:
8708 FEI: libs: Add FEI encoder
8709 Adding FEI encoder to core lib.
8710 The code is splitted into three session:
8711 1: gstvaapiencoder_h264_fei.{h,c}
8712 This is the replica of gstvaapiencoder_h264.{c,h} but with FEI.
8713 All the modes ENC, PAK and ENC_PAK are running based
8714 the code in these files.
8715 2: gstvaapifeienc_h264.{h,c}
8716 Abstract implementation intended for ENC (only VME) operation.
8717 3: gstvaapifeipak_h264.{h,c}
8718 Abstrct implementation intended for PAK (only the PAK module)
8719 Right now ENC_PAK, ENC and PAK are running based on code
8720 in gstvaapiencoder_h264_fei.{h,c}. The abstract implementations
8721 in gstvaapifeienc_h264.{h,c} and gstvaapifeipak_h264.{h,c} are
8722 needed if user request for ENC+PAK mode operation.
8723 ENC+PAK: Here we need to invoke two sequence of
8724 vaBeginPicture/vaRenderPicutre/vaEndPicture for each frame,
8725 first for the ENC only and the second for PAK only.
8726 Each mode associated with separate context ,but same pool of surfaces are
8727 shared between the modes.
8728 This is more useful once we have custom BRC algorithms.
8730 Wang, Yi <yi.a.wang@intel.com>
8731 Leilei <leilei.shang@intel.com>
8732 Zhong, Xiaoxia <xiaoxia.zhong@intel.com>
8733 xiaominc <xiaomin.chen@intel.com>
8734 Li, Jing B <jing.b.li@intel.com>
8735 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8736 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8738 2017-08-09 17:54:27 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8740 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
8741 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
8742 FEI: libs: Add fei codec objects to GstVaapiEncPicture
8743 All the codec objects(vaapi buffers) supposed to be
8744 submited in vaRenderPicutre are associated with a GstVaapiEncPicture
8745 for each frame, follow the same design for FEI too.
8746 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8747 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8749 2017-08-09 16:05:13 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8751 * gst-libs/gst/vaapi/gstvaapicodedbufferproxy.c:
8752 * gst-libs/gst/vaapi/gstvaapicodedbufferproxy.h:
8753 * gst-libs/gst/vaapi/gstvaapicodedbufferproxy_priv.h:
8754 FEI: libs: Add fei codec objects in codedbufferproxy
8755 MbCode, MV and Distortion buffers (fei codec objects)
8756 can be treated as output of different fei modes based user request.
8757 For eg: MbCode and MV are the output of ENC only. MbCode, MV and Dist
8758 can be dumped as output in ENC_PAK mode for analysis purpose.
8759 So treating them as a part of CodedBufferProxy too.
8760 Here we avoided Qp, MbControl and MvPredictor codec objects since
8761 there is no practical use case of treating them as "output buffers".
8763 Zhong, Xiaoxia <xiaoxia.zhong@intel.com>
8764 xiaominc <xiaomin.chen@intel.com>
8765 Leilei <leilei.shang@intel.com>
8766 Li, Jing B <jing.b.li@intel.com>
8767 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8768 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8770 2017-08-09 15:49:21 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8772 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
8773 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
8774 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
8775 FEI: libs: Add fei codec objects to surface proxy
8776 Add fei codec objects to surface proxy since handling the
8777 fei buffers(codec objects here) external to gstvaapisurfaceproxy
8778 will make the code complicated. Especially considering the behavior
8779 of encoder where the input frame order from upstream and output
8780 frame order to the downstream are not sequential.
8782 Zhong, Xiaoxia <xiaoxia.zhong@intel.com>
8783 xiaominc <xiaomin.chen@intel.com>
8784 Leilei <leilei.shang@intel.com>
8785 Li, Jing B <jing.b.li@intel.com>
8786 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8787 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8789 2017-08-09 15:35:10 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8791 * gst-libs/gst/vaapi/Makefile.am:
8792 * gst-libs/gst/vaapi/gstvaapifei_objects.c:
8793 * gst-libs/gst/vaapi/gstvaapifei_objects.h:
8794 * gst-libs/gst/vaapi/gstvaapifei_objects_priv.h:
8795 FEI: Add codec objects for fei usecase
8796 There are 6 new va buffer types, each defined as a specific codec object.
8797 Borrowed the code from gstvaapicodecobject , but made a clear separation
8798 to avoid any possible mess-up. Because unlike the other gstvaaicodecobjects,
8799 feicodecobjects can be shared between elements and also can be accessed
8800 from different thread.
8801 Unlike the other fei codecs object, VAEncMiscParameterTypeFEIFrameControl
8802 object is not shared between elements.So we utilize the already
8803 existing gst_vaapi_enc_misc_param_new(), but still keeping the code
8804 in gstvaapfei_objects_priv.h in order to have a better
8807 -- Probably we need _locked_map() and _unlocked_map()
8808 -- Context can be associated with PreEnc(not just Enoder)
8809 once we have the proper support inplace, but for now we don't have
8810 PreEnc support, so should be safe enough to use GstVaapiEncoder.
8811 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8812 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8814 2017-08-09 14:22:12 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8816 * gst-libs/gst/vaapi/Makefile.am:
8817 * gst-libs/gst/vaapi/gstvaapifeiutils_h264.c:
8818 * gst-libs/gst/vaapi/gstvaapifeiutils_h264.h:
8819 FEI: libs: add H264 fei specific utility functions
8820 Added enum/flag type definitions for a number of FEI
8821 input and output parameters.
8822 Original author of the patch: Wang, Yi <yi.a.wang@intel.com>
8823 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8824 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8825 Signed-off-by: Wang, Yi <yi.a.wang@intel.com>
8826 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8828 2017-08-09 14:10:16 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8830 * gst-libs/gst/vaapi/gstvaapiencoder.c:
8831 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
8832 FEI: libs: Add virtual method for secondary context creation.
8833 Add a new vitrual method ensure_secondary_context to the
8834 base encoder which is only required for the FEI entrypoint, that too
8835 only when user configures the ENC+PAK mode. ENC+PAK mode is not something
8836 supported directly by libva or driver, but this can be enabled
8837 from the middleware.
8838 Original Author of this idea: Leilei Shang <leilei.shang@intel.com>
8839 Signed-off-by: Leilei Shang <leilei.shang@intel.com>
8840 Signed-off-by: xiaominc <xiaomin.chen@intel.com>
8841 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8842 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8843 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8845 2017-08-09 14:05:03 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8847 * gst-libs/gst/vaapi/gstvaapiencoder.c:
8848 FEI: libs: make sure the default context creation works as expected.
8849 Current code always guess the entrypoint during init phase in case
8850 if there is no entrypoint already configured in GstVaapiContextInfo.
8851 Make sure FEI Entrypoint is not messing up with this logic.
8852 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8853 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8855 2017-08-09 13:45:40 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8857 * gst-libs/gst/vaapi/gstvaapicontext.c:
8858 * gst-libs/gst/vaapi/gstvaapicontext.h:
8859 * gst-libs/gst/vaapi/gstvaapiencoder.c:
8860 FEI: libs: Add FEI functional mode configuration
8861 FEI Entrypoint can work in either one of the 3 different modes:
8862 VA_FEI_FUNCTION_ENC, VA_FEI_FUNCTION_PAK or VA_FEI_FUNCTION_ENC_PAK.
8863 Add infrastructure in gstvaapicontext and gstvaapiencoder for this
8864 functioal mode configuration.
8865 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8866 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8868 2017-08-09 13:02:24 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8870 * gst-libs/gst/vaapi/gstvaapidisplay.c:
8871 * gst-libs/gst/vaapi/gstvaapiprofile.c:
8872 * gst-libs/gst/vaapi/gstvaapiprofile.h:
8873 FEI: libs: Add FEI Entrypoint mapping
8874 Define the new mapping GST_VAAPI_ENTRYPOINT_SLICE_ENCODE_FEI
8875 for VAEntrypointFEI.
8876 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8877 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8879 2017-08-09 12:58:29 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8882 FEI: Add support for FEI conditional build
8883 FEI(Flexible Encoding Infrastructure) is an extension
8884 to VA API. Define USE_H264_FEI_ENCODER based on
8885 fei header file and required structures availability.
8886 https://bugzilla.gnome.org/show_bug.cgi?id=785712
8887 https://bugzilla.gnome.org/show_bug.cgi?id=784667
8889 2017-08-28 17:34:50 -0700 Orestis Floros <orestisf1993@gmail.com>
8891 * gst/vaapi/gstvaapidecode.c:
8892 vaapidecode: force add h264 SVC profiles in caps
8893 When vaapih264dec's base-only profile is set to TRUE, fake SVC profile
8895 https://bugzilla.gnome.org/show_bug.cgi?id=732266
8896 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8898 2017-08-28 17:32:57 -0700 Orestis Floros <orestisf1993@gmail.com>
8900 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
8901 libs: decoder: h264: decode SVC base layer only
8902 Drops non-base NALs when the base-only property is set to TRUE.
8903 This modifies the behavior for MVC streams with base-only too: All the
8904 non-base units are dropped before they are decoded instead of dropping
8905 the non-base frames.
8906 The relevant part from the H264 spec is:
8907 > Decoders that conform to one or more of the profiles specified in
8908 Annex A rather than the profiles specified in Annexes G or H shall
8909 ignore (remove from the bitstream and discard) the contents of all NAL
8910 units with nal_unit_type equal to 14, 15, or 20.
8911 To eliminate side effects from the offending units:
8912 - PPS's with a broken seq_parameter_set_id (referring to dropped subset
8914 - The NAL parsing is skipped and their flags are set to
8915 GST_VAAPI_DECODER_UNIT_FLAG_SKIP.
8916 - Prefix units are not stored in prev_pi. Otherwise, parse_slice() would
8917 use them even if they are flagged to be skipped. Subset SPS's and slice
8918 extension units are not stored there either.
8919 https://bugzilla.gnome.org/show_bug.cgi?id=732266
8920 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8922 2017-08-28 17:28:04 -0700 Orestis Floros <orestisf1993@gmail.com>
8924 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
8925 libs: decoder: h264: check nalu validity in parser info finalize
8926 https://bugzilla.gnome.org/show_bug.cgi?id=732266
8927 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
8929 2017-08-28 19:20:42 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8931 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
8932 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
8933 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
8934 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
8935 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
8936 libs: encoder: remove unused cast macro
8937 Remove internal macro to cast structure that are already declared
8940 2017-08-28 19:09:07 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8942 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
8943 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
8944 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.h:
8945 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
8946 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.h:
8947 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.h:
8948 Revert "libs: encoders: remove unused cast macros"
8949 This reverts commit fd7d38f7d26b11e592638092b4073b5c1764f255.
8951 2017-08-28 18:32:32 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8953 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
8954 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
8955 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.h:
8956 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
8957 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.h:
8958 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.h:
8959 libs: encoders: remove unused cast macros
8960 They are only used inside the code, where another macro is defined.
8961 Thus these exported macros have no use.
8963 2017-08-24 20:26:11 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8965 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
8966 libs: decoder: h264: improve code-style
8967 https://bugzilla.gnome.org/show_bug.cgi?id=786173
8969 2017-08-25 16:22:59 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8971 * gst-libs/gst/vaapi/gstvaapicompat.h:
8972 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
8973 libs: encoder: h264: handle deprecated enum
8974 In VA-API 1.0 the enum VAEncPackedHeaderH264_SEI is deprecated, and
8975 instead VAEncPackedHeaderRawData should be used.
8976 This patch creates a compatibility symbol,
8977 VA_ENC_PACKED_HEADER_H264_SEI, to expose the used enum according the
8979 https://bugzilla.gnome.org/show_bug.cgi?id=784398
8981 2017-08-25 16:07:34 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8983 * gst-libs/gst/vaapi/gstvaapiprofile.c:
8984 * gst-libs/gst/vaapi/gstvaapiutils.c:
8985 libs: guard deprecated symbols
8986 In VA-API 1.0 the H.264 baseline profile is deprecated. This patch
8987 guards the H.264 baseline usage. Consider this commit as a
8988 continuation of commit e0e0a474
8989 https://bugzilla.gnome.org/show_bug.cgi?id=784398
8991 2017-08-17 12:54:47 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
8995 * meson_options.txt:
8996 Revert "build: check for libva-2.0"
8997 This reverts commit 8f2eb70803099d4b533ecc10fc259041d8714210.
8998 https://bugzilla.gnome.org/show_bug.cgi?id=784398
9000 2017-08-17 12:44:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9002 * gst-libs/gst/vaapi/gstvaapicompat.h:
9003 * gst-libs/gst/vaapi/gstvaapicontext.c:
9004 * gst-libs/gst/vaapi/gstvaapiencoder.c:
9005 libs: macro to get a renamed value in VA-API 1.0
9006 In VA-API 1.0 the union bits in VAEncMiscParameterBufferROI has
9007 renamed one member from roi_value_is_qp_delat to
9008 roi_value_is_qp_delta, which is the correct name.
9009 In order to keep back compatibility a macro has added to access this
9011 https://bugzilla.gnome.org/show_bug.cgi?id=784398
9013 2017-08-22 11:37:28 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9015 * gst/vaapi/gstvaapipluginbase.h:
9016 * gst/vaapi/gstvaapipluginutil.c:
9017 plugins: include main gstgl header
9018 Instead including particular gstgl header files in a header file
9019 that doesn't export a gstgl symbol, the main gstgl header file is
9020 included in gstvaapipluginutil.c where the symbols are used.
9021 https://bugzilla.gnome.org/show_bug.cgi?id=786597
9023 2017-08-18 18:00:24 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9025 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9026 libs: encoder: h264: fix enum namespace
9028 2017-08-17 12:26:12 +0100 Tim-Philipp Müller <tim@centricular.com>
9031 Automatic update of common submodule
9032 From 48a5d85 to 3f4aa96
9034 2017-08-17 11:03:35 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9036 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9037 libs: encoder: h264: remove spurious assignation
9039 An assigned value that is never used may represent unnecessary
9040 computation, an incorrect algorithm, or possibly the need for cleanup
9042 ip_period is assigned first to be rewritter inmediatly after. The
9043 first assignation is spurious.
9045 2017-08-15 17:36:51 +0900 Hyunjun Ko <zzoon@igalia.com>
9047 * gst/vaapi/gstvaapidecode.c:
9048 vaapidecode: fix mismatch of the return type
9049 https://bugzilla.gnome.org/show_bug.cgi?id=786307
9051 2017-08-10 13:34:21 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9053 * gst-libs/gst/vaapi/Makefile.am:
9054 * gst-libs/gst/vaapi/gstvaapidecoder_unit.h:
9055 * gst-libs/gst/vaapi/gstvaapiminiobject.h:
9056 * gst-libs/gst/vaapi/gstvaapiutils.h:
9057 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
9058 * gst-libs/gst/vaapi/gstvaapiutils_h264_priv.h:
9059 * gst-libs/gst/vaapi/gstvaapiutils_h265_priv.h:
9060 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2_priv.h:
9061 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
9062 * gst-libs/gst/vaapi/libgstvaapi_priv_check.h:
9063 * gst-libs/gst/vaapi/meson.build:
9064 libs: remove unused header
9065 Since libgstvaapi is not distributed, there is no need to check for
9066 private header inclusion. Thus removing it.
9067 https://bugzilla.gnome.org/show_bug.cgi?id=786119
9069 2017-08-10 13:27:11 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9071 * gst-libs/gst/vaapi/gstvaapiutils_h264.h:
9072 * gst-libs/gst/vaapi/gstvaapiutils_h264_priv.h:
9073 * gst-libs/gst/vaapi/gstvaapiutils_h265.h:
9074 * gst-libs/gst/vaapi/gstvaapiutils_h265_priv.h:
9075 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2.h:
9076 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2_priv.h:
9077 libs: utils: move gstvaapisurface.h to private headers
9078 Since the utils don't expose API defined in gstvaapisource.h, it is
9079 moved to their private headers where they are used.
9080 https://bugzilla.gnome.org/show_bug.cgi?id=786119
9082 2017-08-10 13:26:12 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9084 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
9085 * gst-libs/gst/vaapi/gstvaapiutils_h264.h:
9086 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
9087 * gst-libs/gst/vaapi/gstvaapiutils_h265.h:
9088 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2.c:
9089 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2.h:
9090 libs: utils: remove va.h include in header
9091 And include gstvaapicompat.h in the C files, since the VA-API is not
9092 exposed in the headers.
9093 https://bugzilla.gnome.org/show_bug.cgi?id=786119
9095 2017-08-10 13:24:06 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9097 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9098 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
9099 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
9100 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
9101 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
9102 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
9103 libs: encoder: remove va.h include
9104 Since it is already managed by gstvaapicompat.h
9105 https://bugzilla.gnome.org/show_bug.cgi?id=786119
9107 2017-08-10 13:11:04 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9109 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
9110 * gst-libs/gst/vaapi/gstvaapicompat.h:
9111 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
9112 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
9113 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
9114 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9115 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
9116 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
9117 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
9118 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
9119 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
9120 * gst-libs/gst/vaapi/gstvaapifilter.c:
9121 * gst-libs/gst/vaapi/gstvaapiutils.c:
9123 build: consolidate the VA sub API includes
9124 Include all VA sub APIs headers in a single point (gstvaapicompat.h),
9125 since they are all already included in va.h after VA-API 0.38.
9126 https://bugzilla.gnome.org/show_bug.cgi?id=786119
9128 2017-08-10 13:09:27 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9132 build: check for va_vpp.h
9133 Thus, in config.h the macro HAVE_VA_VA_VPP_H is defined. This will
9134 allow us to handle the inclusion of the header better.
9135 https://bugzilla.gnome.org/show_bug.cgi?id=786119
9137 2017-08-11 20:22:41 +0100 Tim-Philipp Müller <tim@centricular.com>
9140 meson: don't export symbols by default
9141 Only plugin entry points should be exported.
9143 2017-08-09 19:06:59 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9145 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
9146 libs: decoder: h265: remove spurious code
9148 Logically dead code: The indicated dead code may have performed some
9149 action; that action will never occur.
9150 By using pointer arithmetic is impossible to get NULL.
9152 2017-08-08 18:52:37 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9154 * gst-libs/gst/vaapi/gstvaapicontext.c:
9155 libs: context: use attribs index instead pointers
9157 Out-of-bounds write. This could cause an immediate crash or incorrect
9159 Coverity basically found that it is possible to assign more than 4
9160 attribs in the array.
9161 In my opinion this was produced because code pattern used pointer
9162 arithmetic, which is not readable nor maintainable.
9163 This patch refactors config_create() to use an array index rather than
9164 pointer arithmetic. Also a run-time check for index size was added.
9166 2017-08-08 17:38:51 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9168 * gst/vaapi/gstvaapiencode_h264.c:
9169 vaapiencode: h264: remove spurious code
9171 An unsigned value can never be negative, so this test will always
9172 evaluate the same way.
9173 As len is guint32, there is no need to check it if it is equal or
9176 2017-08-08 17:34:12 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9178 * gst/vaapi/gstvaapidecode.c:
9179 vaapidecode: initialize variable
9181 The variable will contain an arbitrary value left from earlier
9183 Variable base_only is fetched from base-only property, and it may be
9184 not assigned. It needs to be initialized.
9186 2017-08-08 17:29:54 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9188 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
9189 libs: windows: wayland: fail if cannot remove last frame
9191 If the function returns an error value, the error value may be
9192 mistaken for a normal value.
9193 If g_atomic_pointer_compare_and_exchange() fails because the frame is
9194 not the last one, the function fails. Thus, logging an info message.
9196 2017-08-08 17:21:52 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9198 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
9199 libs: utils: glx: check return value
9201 If the function returns an error value, the error value may be
9202 mistaken for a normal value.
9203 Function sscanf returns the number of assignations done. Validate this
9204 return value with the number of expected variables to match.
9206 2017-08-08 17:12:06 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9208 * gst-libs/gst/vaapi/gstvaapiobject.c:
9209 libs: vaapi: object: remove unrequired NULL check
9211 Dereference after null check: Either the check against null is
9212 unnecessary, or there may be a null pointer dereference.
9213 Variable klass has been validated as non-NULL several time before in
9214 gst_vaapi_object_new() function, so there is no need to check it
9217 2017-08-08 17:06:17 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9219 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
9220 libs: encoder: h265: remove spurious assignation
9222 An assigned value that is never used may represent unnecessary
9223 computation, an incorrect algorithm, or possibly the need for cleanup
9225 ip_period is assigned first to be rewritter inmediatly after. The
9226 first assignation is spurious.
9228 2017-08-08 16:50:39 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9230 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9231 libs: encoder: h264: fix copy & paste error
9233 The copied code will not have its intended effect.
9234 This is a bug from commit cdaf15b2, where the intention is to
9235 initialize RefPicList1 while setting RefPicList0.
9237 2017-08-08 16:33:44 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9239 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9240 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
9241 libs: encoder: h265: fix possible integer overflow
9243 Unintentional integer overflow. The expression's value may not be what
9244 the programmer intended, because the expression is evaluated using a
9245 narrow (i.e. few bits) integer type.
9246 Cast operator to guint64 before computation to avoid narrowing.
9249 2017-08-08 16:12:13 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9251 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
9252 libs: decoder: mpeg4: fail if return value is not OK
9254 An assigned value that is never used may represent unnecessary
9255 computation, an incorrect algorithm, or possibly the need for cleanup
9257 In the return value of decode_slice() or
9258 gst_mpeg4_parse_video_packet_header() are not success, thus fail
9259 decode_packet() function.
9261 2017-08-08 15:49:27 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9263 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
9264 libs: decoder: h265: check for null
9266 Dereference after null check: Either the check against null is
9267 unnecessary, or there may be a null pointer dereference.
9268 While looking for hte lowest poc, according to rest of the code, the
9269 picture in the dbp (decoded picture buffer) might be NULL, thus we
9270 could check for a NULL picture before assigned as found.
9271 Also, split a comma operator because it is considered as a bad
9272 practice because it possible side effects.
9274 2017-08-08 15:38:16 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9276 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
9277 libs: decoder: h265: untaint loop control variable
9279 Scalars (for example, integers) are not properly
9280 bounds-checked (sanitized) before being used as array or pointer
9281 indexes, loop boundaries, or function arguments are considered as
9283 In this case, num_nals were not checked before used as loop control.
9285 2017-08-08 13:46:56 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9287 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
9288 libs: decoder: h264: remove unrequired NULL check
9290 Dereference after null check: Either the check against null is
9291 unnecessary, or there may be a null pointer dereference.
9292 In the original commit for fill_picture_gaps() (commit 5abd2b90) the
9293 prev_picture could be NULL, that's why the code did a null check. But,
9294 since commit 52adebe7, the previous reference frames are tracked, thus
9295 there is no need to check null anymore.
9297 2017-08-03 23:17:44 +0300 orestisf <orestisf1993@gmail.com>
9299 * gst/vaapi/gstvaapidecode.c:
9300 vaapidecode: fix gst_caps_new_simple call
9301 https://bugzilla.gnome.org/show_bug.cgi?id=732265
9303 2017-07-25 22:25:10 +0300 orestisf <orestisf1993@gmail.com>
9305 * gst/vaapi/gstvaapidecode.c:
9306 vaapidecode: force add h264 MVC profiles in caps
9307 When vaapih264dec's base-only profile is set to TRUE, fake MVC profile
9309 https://bugzilla.gnome.org/show_bug.cgi?id=732265
9311 2017-07-25 22:54:30 +0300 orestisf <orestisf1993@gmail.com>
9313 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
9314 libs: decoder: h264: decode MVC base view only
9315 If processed SPS has mvc profile and the configuration is set to
9316 base-only, the frame is drop.
9317 https://bugzilla.gnome.org/show_bug.cgi?id=732265
9319 2017-07-25 22:06:56 +0300 orestisf <orestisf1993@gmail.com>
9321 * gst/vaapi/gstvaapidecode.c:
9322 * gst/vaapi/gstvaapidecode_props.c:
9323 vaapidecode: set h264 base-only to decoder
9324 Set the base-only value when property is set and the internal
9325 decoder is already instantiated or when the internal decoder
9327 https://bugzilla.gnome.org/show_bug.cgi?id=732265
9329 2017-07-25 22:03:34 +0300 orestisf <orestisf1993@gmail.com>
9331 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
9332 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
9333 libs: decoder: h264: add setter for base-only mode
9334 https://bugzilla.gnome.org/show_bug.cgi?id=732265
9336 2017-07-25 22:01:37 +0300 orestisf <orestisf1993@gmail.com>
9338 * gst/vaapi/gstvaapidecode_props.c:
9339 * gst/vaapi/gstvaapidecode_props.h:
9340 vaapidecode_props: h264: add base-only property
9341 https://bugzilla.gnome.org/show_bug.cgi?id=732265
9343 2017-08-01 11:11:55 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9345 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
9346 libs: encoder: h264: missing property enum documentation
9348 2017-08-02 14:54:53 +0900 Hyunjun Ko <zzoon@igalia.com>
9350 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9351 libs: encoder: h264: add multi reference support
9352 Using num_ref_frames provided and the result of the Query
9353 VAConfigAttribEncMaxRefFrames, it determines the size of reference list
9354 and perform encoding with multi reference frames as the following:
9355 1\ The num_ref_frames is being considered as the number of
9356 reference picture list0
9357 2\ Encoder adds 1 reference frame more to the reference picture list1
9358 internally if b-frame encoding.
9359 3\ If num_ref_frames is bigger than the number of refrence frames
9360 supported in the driver, it will be lowered.
9361 https://bugzilla.gnome.org/show_bug.cgi?id=783803
9363 2017-08-02 14:53:34 +0900 Hyunjun Ko <zzoon@igalia.com>
9365 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9366 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
9367 libs: encoder: h264: add refs property
9368 Users can provide the number of reference frame by this property.
9369 The value of the property will be considered as the number of
9370 reference picture list0 and will add 1 reference frame more to the
9371 reference picture list1 internally if b-frame encoding.
9372 If the value provided is bigger than the number of refrence frames
9373 supported in the driver, it will be lowered.
9374 https://bugzilla.gnome.org/show_bug.cgi?id=783803
9376 2017-07-28 15:27:20 +0900 Hyunjun Ko <zzoon@igalia.com>
9378 * gst-libs/gst/vaapi/gstvaapiencoder.c:
9379 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
9380 libs: encoder: implements gst_vaapi_encoder_ensure_max_num_ref_frames
9381 This function will query VAConfigAttribEncMaxRefFrames to get the
9382 maximum number of reference frames supported in the driver.
9383 This will be used for h264/h265 encoding.
9384 https://bugzilla.gnome.org/show_bug.cgi?id=783803
9386 2017-08-01 18:38:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9388 * gst/vaapi/gstvaapiencode_h265.c:
9389 vaapiencode: h265: compare an unsigned int if not zero
9390 An unsigned value can never be negative, so this test (greater than
9391 zero) will always evaluate the same way. Thus change it to just if
9394 2017-08-01 18:10:50 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9396 * gst/vaapi/gstvaapipluginbase.c:
9397 plugins: check gst_gl_ensure_element_data() return value
9398 Refactor gst_vaapi_plugin_base_create_gl_context() in order to check
9399 the return value of gst_gl_ensure_element_data(). The result is a code
9402 2017-08-01 17:59:38 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9404 * gst/vaapi/gstvaapipluginutil.c:
9405 plugins: avoid dead code detection
9406 By using #elif macro, the static code analysis would stop to detect
9407 these lines as dead code. Also it is inforced the mutually exclusive
9410 2017-08-01 17:39:04 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9412 * gst/vaapi/gstvaapivideobufferpool.c:
9413 vaapivideobufferpool: don't shift by negative since it's undefined
9414 The function g_bit_nth_lsf() may return -1 if the request bit position
9415 is not avaible. Thus, this patch check if the return value is not -1
9416 in order to continue.
9418 2017-08-01 17:29:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9420 * gst/vaapi/gstvaapisink.c:
9421 vaapisink: fix memory leak
9423 2017-08-01 17:23:48 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9425 * gst/vaapi/gstvaapipostprocutil.c:
9426 vaapipostproc: fix memory leaks
9428 2017-07-27 10:54:00 +0000 Tomas Rataj <rataj28@gmail.com>
9430 * gst-libs/gst/vaapi/gstvaapidisplay.c:
9431 libs: display: when appending formats change pointers to indexes
9432 Thus, it fixes an invalid read when YV12 or I420 are not supported by
9434 https://bugzilla.gnome.org/show_bug.cgi?id=785085
9436 2017-07-19 12:02:40 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
9438 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9439 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
9440 libs: encoder: h264: Add uncompliant mode reducing coded buffer size
9441 Added a new property "compliance-mode", which default is the normal
9442 strict compliant mode.
9443 The second mode, "restrict-buf-alloc", is to limit the coded buffer
9444 allocation size to improve performance in some specific Intel
9445 platforms (there is asignificant performance improvement in parallel
9446 encodings). Under this new mode, we use the MinCR field in A.3.1 for
9447 pre-calculating the coded-buffer size.
9448 https://bugzilla.gnome.org/show_bug.cgi?id=784590
9450 2017-07-05 17:13:44 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
9452 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
9453 * gst-libs/gst/vaapi/gstvaapiutils_h264_priv.h:
9454 libs: utils_h264: Extend LevelLimit table with MinCR field
9455 Add MinCR(Minimum Compression Ratio) field to GstVaapiH264LevelLimits
9457 https://bugzilla.gnome.org/show_bug.cgi?id=784590
9459 2017-07-11 17:29:13 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9461 * gst-libs/gst/vaapi/gstvaapiutils.c:
9462 libs: utils: libva 1.0 changed the logging
9463 The logging mechanism in libva has changed it's functions
9464 signatures. This patch updates that for libva versions >= 1.0
9465 https://bugzilla.gnome.org/show_bug.cgi?id=784398
9467 2017-07-11 17:27:32 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9469 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
9470 libs: decoder: h264: libva 1.0 deprecated baseline
9471 libva 1.0 deprecated H.264 baseline profile and FMO support
9473 https://bugzilla.gnome.org/show_bug.cgi?id=784398
9475 2017-07-26 20:03:35 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9479 * meson_options.txt:
9480 build: check for libva-2.0
9481 Check for libva-2.0 since libva's developers decided to increase the
9482 library's version number.
9483 https://bugzilla.gnome.org/show_bug.cgi?id=784398
9485 2017-07-11 16:55:26 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9489 build: blacklist only libva 0.99.0
9490 Intel's MSDK uses libva 0.99.0, meanwhile open source libva bumped
9491 its API version to 1.0.0. Thus we have to blacklist only the MSDK's
9493 https://bugzilla.gnome.org/show_bug.cgi?id=784398
9495 2017-07-26 20:30:37 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9497 * gst-libs/gst/vaapi/meson.build:
9498 build: meson: remove gstvaapidisplaycache.c
9499 This is a missing bit of commit ec3e10f6
9501 2017-07-26 09:53:10 -0700 U. Artie Eoff <ullysses.a.eoff@intel.com>
9504 configure: do not break configure if gtk+-3.0 devel missing
9505 Fix PKG_CHECK_MODULES rule for with_gtk=check condition to
9506 set USE_GTK=0 if gtk+-3.0 is not available.
9507 Since commit 85856c29a70d6de4aea5b708e04e9eb418190623
9508 Author: Hyunjun Ko <zzoon@igalia.com>
9509 Date: Wed Jul 5 15:59:43 2017 +0900
9510 tests: elements: add testsuite of vaapi context
9511 ...configure fails if gtk+-3.0 development files are missing.
9512 The "with_gtk" option defaults to "check" in configure.ac
9513 which implies that if it is not explicitly requested then
9514 configure will only enable it if it's available on the system.
9515 However, the PKG_CHECK_MODULES rule that get's activated on
9516 "check" condition did not provide default when gtk+-3.0 devel
9517 packages are not found on the system. Thus, it resulted in
9519 Signed-off-by: U. Artie Eoff <ullysses.a.eoff@intel.com>
9520 https://bugzilla.gnome.org/show_bug.cgi?id=785452
9522 2017-07-05 15:59:43 +0900 Hyunjun Ko <zzoon@igalia.com>
9525 * tests/elements/Makefile.am:
9526 * tests/elements/test-vaapicontext.c:
9527 tests: elements: add testsuite of vaapi context
9528 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9529 https://bugzilla.gnome.org/show_bug.cgi?id=766704
9531 2017-07-05 15:32:43 +0900 Hyunjun Ko <zzoon@igalia.com>
9533 * gst/vaapi/gstvaapisink.c:
9534 vaapisink: fail if surface display is different
9535 Replacing GstVaapiDisplay during rendering might be hiding problems
9536 at some cases, even though it's safe currently since we use cache
9538 Play safe by failing if this happens.
9539 https://bugzilla.gnome.org/show_bug.cgi?id=766704
9541 2017-07-05 15:31:55 +0900 Hyunjun Ko <zzoon@igalia.com>
9543 * gst/vaapi/gstvaapivideocontext.c:
9544 * gst/vaapi/gstvaapivideocontext.h:
9545 videocontext: support "gst.vaapi.app.Display" context
9546 Through "gst.vaapi.app.Display" context, users can set their own VADisplay
9547 and native display of their backend.
9549 - display : pointer of VADisplay
9550 - x11-display : pointer of X11 display (Display *), if they're using.
9551 This patch creates GstVaapidisplayX11 if information provided through
9552 "gst.vaapi.app.Display"
9553 https://bugzilla.gnome.org/show_bug.cgi?id=766704
9555 2017-07-05 14:33:38 +0900 Hyunjun Ko <zzoon@igalia.com>
9557 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
9558 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
9559 libs: display: x11: add gst_vaapi_display_x11_new_with_va_display()
9560 Implements new API function so that users could create GstVaapiDisplay
9561 with their own VADisplay within a native display as backend.
9562 https://bugzilla.gnome.org/show_bug.cgi?id=766704
9564 2017-07-05 14:32:35 +0900 Hyunjun Ko <zzoon@igalia.com>
9566 * gst-libs/gst/vaapi/gstvaapidisplay.c:
9567 libs: display: pass display info when foreign display
9568 When creating a GstVaapiDisplay using a foreign VADisplay, and render
9569 with that display, it also requires native display of the backend.
9570 https://bugzilla.gnome.org/show_bug.cgi?id=766704
9572 2017-06-26 21:18:25 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9574 * gst-libs/gst/vaapi/Makefile.am:
9575 * gst-libs/gst/vaapi/gstvaapidisplay.c:
9576 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
9577 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
9578 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
9579 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
9580 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
9581 * gst-libs/gst/vaapi/gstvaapidisplaycache.h:
9582 libs: display: remove cache
9583 Remove a bunch of code that handles the VADisplay cache, since the
9584 context sharing should be doing this correctly.
9585 https://bugzilla.gnome.org/show_bug.cgi?id=747946
9587 2017-07-13 10:56:18 +0900 Hyunjun Ko <zzoon@igalia.com>
9589 * tests/elements/Makefile.am:
9590 * tests/elements/test-vaapipostproc.c:
9591 tests: elements: add test for vaapipostproc
9592 https://bugzilla.gnome.org/show_bug.cgi?id=754885
9594 2017-07-12 18:25:15 +0900 Hyunjun Ko <zzoon@igalia.com>
9596 * gst/vaapi/gstvaapipostproc.c:
9597 postproc: reconfigure when width or height changes
9598 https://bugzilla.gnome.org/show_bug.cgi?id=754885
9600 2017-07-17 18:53:57 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9602 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
9603 libs: encoder: vp9: array terminated in zeros
9604 There is a crash when setting ref-pic-mode since the #GEnumValue
9605 array is not terminated with a structured with all memvers being
9607 https://bugzilla.gnome.org/show_bug.cgi?id=785032
9609 2017-07-13 16:43:34 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9611 * gst/vaapi/gstvaapiencode_h264.c:
9612 vaapiencode: h264: add plugin documentation
9613 Comment how the profile is set and other parameters.
9615 2017-05-26 15:19:00 +0000 Matt Staples <staples255@gmail.com>
9617 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
9618 libs: decoder: h264: push frames as soon as possible
9619 Push frames downstream as soon as possible instead of waiting until
9620 they are ejected from the DPB.
9621 This patch makes the decoder not comply with the H.264 specification,
9622 but it is required for some video cameras.
9623 https://bugzilla.gnome.org/show_bug.cgi?id=762509
9624 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9626 2017-07-10 19:27:57 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9628 * gst/vaapi/gstvaapidecode_props.c:
9629 vaapidecode_props: h264: set low-latency in decoder
9630 Set the low-latency property if the H264 decoder is already
9631 instantiated, thus you could change the behavior in run-time.
9632 https://bugzilla.gnome.org/show_bug.cgi?id=783588
9634 2017-07-06 20:00:15 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9636 * gst/vaapi/gstvaapidecode.c:
9637 vaapidecode: set h264 low latency to decoder
9638 https://bugzilla.gnome.org/show_bug.cgi?id=783588
9640 2017-06-14 18:30:53 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9642 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
9643 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
9644 libs: decoder: h264: add getter/setter for low latency mode
9645 https://bugzilla.gnome.org/show_bug.cgi?id=783588
9647 2017-06-14 18:31:18 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9649 * gst/vaapi/gstvaapidecode_props.c:
9650 * gst/vaapi/gstvaapidecode_props.h:
9651 vaapidecode_props: h264: add low latency property
9652 Adding support for private data.
9653 https://bugzilla.gnome.org/show_bug.cgi?id=783588
9655 2017-06-14 18:23:34 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9657 * gst/vaapi/Makefile.am:
9658 * gst/vaapi/gstvaapidecode.c:
9659 * gst/vaapi/gstvaapidecode_props.c:
9660 * gst/vaapi/gstvaapidecode_props.h:
9661 * gst/vaapi/meson.build:
9662 vaapidecode_props: add skeleton for h264 decoder properties
9663 https://bugzilla.gnome.org/show_bug.cgi?id=783588
9665 2017-06-14 17:07:30 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9667 * gst/vaapi/gstvaapidecode.c:
9668 vaapidecode: properties callback in decoders map
9669 https://bugzilla.gnome.org/show_bug.cgi?id=783588
9671 2017-07-07 12:01:59 +0100 Tim-Philipp Müller <tim@centricular.com>
9674 meson: find python3 via python3 module
9675 https://bugzilla.gnome.org/show_bug.cgi?id=783198
9677 2017-06-09 14:47:40 +0900 Hyunjun Ko <zzoon@igalia.com>
9679 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9680 libs: encoder: h264: submit sps in case of IDR picture
9681 If the picture is IDR, also submit a SPS header.
9682 This means when frame number reaches to keyframe-period or an force
9683 key unit event arrives, we insert SPS/PPS again.
9684 https://bugzilla.gnome.org/show_bug.cgi?id=776712
9686 2017-06-09 14:47:16 +0900 Hyunjun Ko <zzoon@igalia.com>
9688 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9689 libs: encoder: h264: set the frame as IDR if forced key unit
9690 GST_VIDEO_CODEC_FRAME_IS_FORCE_KEYFRAME() is a flag usually used to manage
9691 the `frame-lost` event in the case of streaming, such as RTP.
9692 In case of this event, it is needed to start new GOP rather than just
9694 https://bugzilla.gnome.org/show_bug.cgi?id=776712
9696 2017-04-05 14:48:46 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
9698 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9699 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
9700 libs: encoder: h264: insert AU delimiter
9701 Insert an AUD as the first NAL of each encoded frame.
9702 Some applications require Access Unit Delimiter for decoding the
9704 The AU delimeter insertion is done only when the aud parameter is
9705 TRUE (by default is disabled). The reason of this it is because this
9706 header is only available from Intel Gen9 and the VA intel driver
9707 should be 1.8 or superior. Otherwise, the output will be corrupted.
9708 https://bugzilla.gnome.org/show_bug.cgi?id=776712
9709 Signed-off-by: Victor Jaquez <vjaquez@igalia.com>
9711 2017-06-29 12:50:26 +0900 Hyunjun Ko <zzoon@igalia.com>
9713 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9714 libs: encoder: h264: initialize all elements of view_ids
9715 Currently when num_views is changed by multiview-mode on sink caps, it produces
9716 wrong MVC encoded stream since the array view_ids is not set properly according
9717 to changed num_views.
9718 So this patch initializes all of the array sequentially to handle this case.
9719 Side effect is not going to happen by this patch since this array is being
9720 handled by num_views.
9721 https://bugzilla.gnome.org/show_bug.cgi?id=784321
9723 2017-06-27 14:30:54 +0900 Hyunjun Ko <zzoon@igalia.com>
9725 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9726 Revert "encoder: h264: Use high profile by default"
9727 This reverts commit 4aec5bdd7207fc0e45813ef14c9c0ad5174a8f75.
9728 https://bugzilla.gnome.org/show_bug.cgi?id=757941
9730 2017-06-27 16:03:37 +0900 Hyunjun Ko <zzoon@igalia.com>
9732 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9733 libs: encoder: h264: set profile via capsfilter
9734 Until now, the encoder ignored the profile in src caps and chose one
9735 according with the given parameters. But the encoder must honor the
9736 profile specifed in src caps.
9737 This patch do that, and if the encoder needs to choose the profile,
9738 it will do it by following these rules:
9739 1\ If given parameters are not compatible with given profile, the
9740 encoder will bail out with an error.
9741 2\ The encoder will choose the higher profile indicated in the
9743 https://bugzilla.gnome.org/show_bug.cgi?id=757941
9745 2017-06-27 13:14:31 +0900 Hyunjun Ko <zzoon@igalia.com>
9747 * gst/vaapi/gstvaapiencode_h264.c:
9748 vaapiencode: h264: set profile to src caps
9749 So far vaapi encoder does not set profile to src caps. This patch makes it
9750 setting profile to src caps, which is determined by itself.
9751 In addition, if encoder chose different profile, which is not negotiated with
9752 downstream, we should set compatible profile to make negotiation working.
9753 https://bugzilla.gnome.org/show_bug.cgi?id=757941
9755 2017-06-22 09:56:49 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9757 * gst/vaapi/gstvaapiencode_h264.c:
9758 * gst/vaapi/gstvaapiencode_h264.h:
9759 vaapiencode: h264: verify if requested profile is supported
9760 Check if the requested profile in source caps, is supported by the
9761 VA driver. If it is not, an info log message is send saying that
9762 another (compatible?) profile will be used.
9763 https://bugzilla.gnome.org/show_bug.cgi?id=757941
9765 2017-06-21 21:49:27 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9767 * gst/vaapi/gstvaapiencode_h264.c:
9768 vaapiencode: h264: improve set_config() vmethod
9769 First check if downstream requests ANY caps. If so, byte-stream is
9770 used and the profile will be choose by the encoder. If dowstream
9771 requests EMPTY caps, the negotiation will fail.
9772 Lately, byte-stream and profile are looked in the allowed caps.
9773 https://bugzilla.gnome.org/show_bug.cgi?id=757941
9775 2017-06-21 19:30:55 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9777 * gst/vaapi/gstvaapiencode_h264.c:
9778 vaapiencode: h264: check for avc in set_config()
9779 The check for avc stream format was done in the vaapi encoder's
9780 vmethod get_caps(), but that is wrong since it has to be check
9781 when encoder set_format().
9782 https://bugzilla.gnome.org/show_bug.cgi?id=757941
9784 2017-06-29 12:49:24 +0900 Hyunjun Ko <zzoon@igalia.com>
9786 * gst/vaapi/gstvaapipostprocutil.c:
9787 vaapipostproc: set multivew-mode flags to src caps
9788 vaapipostproc didn't negotiate the proper multiview caps losing
9789 downstream information.
9790 This patch enables the playing of MVC encoded stream by setting
9791 the proper multiview mode/flags and views to src caps, according
9793 https://bugzilla.gnome.org/show_bug.cgi?id=784320
9795 2016-11-22 15:52:47 +0000 Julien Isorce <j.isorce@samsung.com>
9797 * gst/vaapi/gstvaapipostproc.c:
9798 vaapipostproc: add support for DMABuf caps feature
9799 https://bugzilla.gnome.org/show_bug.cgi?id=755072
9800 Signed-off-by: Julien Isorce <j.isorce@samsung.com>
9802 2017-06-01 19:42:20 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9804 * gst/vaapi/gstvaapidecode.c:
9805 vaapidecode: add support for DMABuf caps feature
9806 https://bugzilla.gnome.org/show_bug.cgi?id=755072
9807 Original-patch-by: Julien Isorce <j.isorce@samsung.com>
9809 2017-06-23 12:12:12 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9811 * gst/vaapi/gstvaapipluginbase.c:
9812 vaapipluginbase: force dmabuf allocator if DMABuf caps feature
9813 Instantiate all dmabuf allocator for src pad buffer pool if the
9814 src caps ask for memory:DMABuf feature.
9815 https://bugzilla.gnome.org/show_bug.cgi?id=755072
9817 2016-11-22 23:26:05 +0000 Julien Isorce <j.isorce@samsung.com>
9819 * gst/vaapi/gstvaapipluginutil.c:
9820 * gst/vaapi/gstvaapipluginutil.h:
9821 vaapipluginutil: add support for DMABuf caps feature
9822 https://bugzilla.gnome.org/show_bug.cgi?id=755072
9823 Signed-off-by: Julien Isorce <j.isorce@samsung.com>
9824 Signed-off-by: Victor Jaquez <vjaquez@igalia.com>
9825 vaapipluginutil: add support for DMABuf caps feature
9827 2017-06-01 19:13:52 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9829 * gst/vaapi/gstvaapipluginbase.c:
9830 vaapipluginbase: dmabuf memory map trial for raw caps
9831 Only push dmabuf-based buffers with raw caps if gst_memory_map()
9832 succeeds. Otherwise, use the the vaapi surfaces allocator.
9833 https://bugzilla.gnome.org/show_bug.cgi?id=755072
9834 https://bugzilla.gnome.org/show_bug.cgi?id=774649
9835 Original-patch-by: Julien Isorce <j.isorce@samsung.com>
9837 2016-06-08 19:11:15 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
9839 * gst/vaapi/gstvaapivideomemory.c:
9840 * gst/vaapi/gstvaapivideomemory.h:
9841 vaapivideomemory: add gst_vaapi_dmabuf_can_map()
9842 This new method checks the specified allocator can create GstMemory that can
9844 https://bugzilla.gnome.org/show_bug.cgi?id=755072
9846 2017-06-23 17:33:03 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9848 * gst/vaapi/gstvaapivideobufferpool.c:
9849 vaapivideobufferpool: fix regression with video metas
9850 There is another regression with 7a206923 when setting the video
9851 info for the video meta, it should be the one from the image's
9852 allocator rather from the allocation caps.
9854 gst-launch-1.0 filesrc location=bug766184.flv ! decodebin \
9855 ! tee ! videoconvert ! videoscale \
9856 ! video/x-raw, width=1920, height=1080 ! xvimagesink
9858 2017-06-23 14:38:10 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9860 * gst/vaapi/gstvaapipluginbase.c:
9861 * gst/vaapi/gstvaapivideobufferpool.c:
9862 plugins: update buffer size with the one reported by allocator
9863 There is a regression in 7a206923, since the buffer pool ditches all
9864 the buffers generated by them because the pool config size is
9865 different of the buffer's size.
9867 gst-launch-1.0 filesrc location=big_buck_bunny_1080p_h264.mov \
9868 ! qtdemux ! vaapih264dec ! vaapipostproc ! xvimagesink \
9869 --gst-debug=GST_PERFORMANCE:5
9870 The allocator may update the buffer size according to the VA surface
9871 properties. In order to do this, the video info is modified when the
9872 allocator is created, which reports through the allocation info the
9873 updated size, and set it to the pool config.
9875 2017-06-14 21:40:33 +0900 Hyunjun Ko <zzoon@igalia.com>
9877 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
9878 libs: decoder: h264: initialize active_sps/pps in reset
9879 Since commits in https://bugzilla.gnome.org/show_bug.cgi?id=781142 landed,
9880 they introduced regression in seek.
9881 Formerly, once seek is done, decoder drops P-frames until I-frame arrives.
9882 But since the commits landed, it doesn't drop P-frame and does try to
9883 decode it continuously because active_sps is still alive. See ensure_sps function.
9884 But there are prev_frames and prev_ref_frames reset already, then it
9886 So it's necessary to reset active_sps/pps also in reset method.
9887 https://bugzilla.gnome.org/show_bug.cgi?id=783726
9889 2017-06-15 13:24:56 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9891 * gst-libs/gst/vaapi/gstvaapiencoder.c:
9892 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9893 libs: encoder: fix compilation with old versions of libva
9894 There are some symbols that are not used when compiling with old
9895 version of libva and those generates a compilation error.
9896 Original-patch-by: Matt Staples <staples255@gmail.com>
9898 2017-06-09 14:02:20 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
9900 * gst-libs/gst/vaapi/gstvaapiencoder.c:
9901 libs: encoder: Fix the quality level clamping
9902 Change the hard-coded range of quality-level from {1-8} to {1-7},
9903 since it is the range Intel Open source driver supports.
9904 Also perform the range clamping only if the user provided
9905 quality-level is greater than the max-range suppored by the driver,
9906 because there could be non-intel drivers giving lower value than
9907 the hard-coded max value 7.
9908 https://bugzilla.gnome.org/show_bug.cgi?id=783567
9910 2017-04-06 19:35:27 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
9912 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9913 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
9914 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
9915 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
9916 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
9917 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
9918 libs: encoder: log out the name of the profile
9919 Instead of printing a number, it is more readable to log out, in
9920 case of error, the name of the failing profile.
9922 2017-05-31 12:36:17 +0900 Hyunjun Ko <zzoon@igalia.com>
9924 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
9925 libs: encoder: h264: changes raw number of profile to macro name of its
9926 Changes raw number of profile to macro name of its to improve readability.
9927 https://bugzilla.gnome.org/show_bug.cgi?id=757941
9929 2017-06-09 17:00:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9931 * gst/vaapi/gstvaapivideobufferpool.c:
9932 vaapivideobufferpool: remove allocation_vinfo private attribute
9933 There is no need to keep this attribute internally since it is
9934 already managed by the allocator.
9935 https://bugzilla.gnome.org/show_bug.cgi?id=783599
9937 2017-06-09 15:02:08 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9939 * gst/vaapi/gstvaapivideobufferpool.c:
9940 vaapivideobufferpool: refactor set_config()
9941 Refactor the set_config() virtual method considering a cleaner
9942 approach to allocator instanciation, if it it not set or if it is
9943 not valid for the pool.
9944 https://bugzilla.gnome.org/show_bug.cgi?id=783599
9946 2017-06-09 13:05:36 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9948 * gst/vaapi/gstvaapipluginbase.c:
9949 * gst/vaapi/gstvaapivideobufferpool.c:
9950 * gst/vaapi/gstvaapivideomemory.c:
9951 * gst/vaapi/gstvaapivideomemory.h:
9952 plugins: distinguish allocation and negotiation caps
9953 The vaapi video decoders might have different allocation caps from
9954 the negotiation caps, thus the GstVideoMeta shall use the negotiation
9955 caps, not the allocation caps.
9956 This was done before reusing gst_allocator_get_vaapi_video_info(),
9957 storing there the negotiation caps if they differ from the allocation
9958 ones, but this strategy felt short when the allocator had to be reset
9959 in the vaapi buffer pool, since we need both.
9960 This patch adds gst_allocator_set_vaapi_negotiated_video_info() and
9961 gst_allocator_get_vaapi_negotiated_video_info() to store the
9962 negotiated video info in the allocator, and distinguish it from
9963 the allocation video info.
9964 https://bugzilla.gnome.org/show_bug.cgi?id=783599
9966 2017-06-08 19:32:35 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9968 * gst/vaapi/gstvaapivideomemory.c:
9969 * gst/vaapi/gstvaapivideomemory.h:
9970 vaapivideomemory: rename qdata quarks and ids
9971 Also the parameter names were renamed to reflect their origin
9973 https://bugzilla.gnome.org/show_bug.cgi?id=783599
9975 2017-06-08 16:05:49 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9977 * gst/vaapi/gstvaapivideobufferpool.c:
9978 vaapivideobufferpool: rename local variables
9979 Renamed local video info structure names in set_config() vitual
9980 method. The purpose of their renaming is to clarify the origin
9981 of those structures, whether come from passed caps parameter
9982 (new_allocation_vinfo) or from the configured allocator
9984 https://bugzilla.gnome.org/show_bug.cgi?id=783599
9986 2017-06-08 15:49:05 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
9988 * gst/vaapi/gstvaapivideobufferpool.c:
9989 vaapivideobufferpool: rename video info structures
9990 Renamed private GstVideoInfo structure video_info to allocation_vinfo
9991 and alloc_info to negotiated_vinfo.
9992 The purpose of these renaming is to clarify the origin and purpose of
9993 these private variables:
9994 video_info (now allocation_vinfo) comes from the bufferpool
9995 configuration. It describes the physical video resolution to be
9996 allocated by the allocator, which may be different from the
9998 alloc_info (now vmeta_vinfo) comes from the negotiated caps in
9999 the pipeline. It represents how the frame is going to be mapped
10000 using the video meta.
10001 In Intel's VA-API backend, the allocation_vinfo resolution is
10002 bigger than the negotiated_info.
10003 https://bugzilla.gnome.org/show_bug.cgi?id=783599
10005 2017-06-08 12:51:50 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10007 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10008 libs: encoder: set framerate if bigger than 0/1
10009 Just set the framerate parameter if the framerate numerator and
10010 denominator are bigger than zero.
10011 Otherwise, in Intel Gen6 driver, a warning is raised disabling the
10013 Original-patch-by: Hyunjun Ko <zzoon@igalia.com>
10014 https://bugzilla.gnome.org/show_bug.cgi?id=783532
10016 2017-06-07 12:32:53 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10018 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10019 libs: encoder: bitrate target percentage calculation
10020 If the rate control is set to Constant Bit Rate (CBR) the target
10021 percentage is 100%, otherwise is 70%
10023 2017-06-07 12:25:24 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10025 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10026 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10027 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10028 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
10029 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
10030 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
10031 libs: encoder: h264,h265,mpeg2,vp8,vp9: refactor ratecontrol param
10032 Centralize the common configuration for the Rate Control parameter,
10033 thus can be overloaded per each specific encoder.
10035 2017-06-07 11:10:49 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10037 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10038 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10039 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10040 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
10041 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
10042 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
10043 libs: encoder: h264,h265,mpeg2,vp8,vp9: refactor framerate param
10044 Since the framerate VA parameter is calculated equally among all the
10045 encoders, it is better to handle it in the base encoder class.
10047 2016-08-09 15:53:47 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
10049 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
10050 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.h:
10051 libs: encoder: vp9: Adds CBR and VBR Encoding support
10052 https://bugzilla.gnome.org/show_bug.cgi?id=766832
10053 Signed-off-by: Hyunjun Ko <zzoon@igalia.com>
10054 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10056 2017-06-01 12:12:26 +0900 Hyunjun Ko <zzoon@igalia.com>
10058 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
10059 libs: encoder: vp8: Adds VBR Encoding support
10060 https://bugzilla.gnome.org/show_bug.cgi?id=778732
10062 2017-06-01 12:11:12 +0900 Hyunjun Ko <zzoon@igalia.com>
10064 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10065 libs: encoder: h265: Adds VBR Encoding support
10066 Enables Variable BitRate mode, which does set FrameRate and RateControl
10068 https://bugzilla.gnome.org/show_bug.cgi?id=778732
10070 2017-06-02 13:50:05 +0900 Hyunjun Ko <zzoon@igalia.com>
10072 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10073 libs: encoder: Describes more detail about the bitrate property
10074 https://bugzilla.gnome.org/show_bug.cgi?id=778732
10076 2017-06-05 20:44:22 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10078 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10079 libs: encoder: h265: add rate control parameter
10080 https://bugzilla.gnome.org/show_bug.cgi?id=783449
10082 2017-06-05 20:33:27 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10084 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10085 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10086 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
10087 libs: encoder: h264,h265,mpeg2: add framerate parameter
10088 https://bugzilla.gnome.org/show_bug.cgi?id=783449
10090 2017-06-05 20:30:07 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10092 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10093 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10094 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
10095 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
10096 libs: encoder: vp8,h264,h265,mpeg2: set misc param once
10097 Instead of recalculating the miscellaneous buffer parameters for
10098 every buffer, it is only done once, when the encoder is configured.
10099 And for every buffer, the same structures are just copied.
10100 https://bugzilla.gnome.org/show_bug.cgi?id=783449
10102 2017-06-05 17:31:10 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10104 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10105 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10106 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10107 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
10108 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
10109 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
10110 libs: encoder: vp8,h264,h265,mpeg2: refactor misc parameters
10111 This is patch pretends to decouple the assignation of the values
10112 in the parameter structures and the VA buffer's parameters setting.
10113 It may lead to some issues since HRD, framerate or controlrate may
10114 not be handled by the specific encoder, but they are set in
10115 the VA buffer's parameters.
10116 I leave as it because this patch is just a transitional patch.
10117 https://bugzilla.gnome.org/show_bug.cgi?id=783449
10119 2017-06-05 16:34:12 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10121 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
10122 libs: encoder: vp8: fix frame rate calculation
10123 According to the VA documentation:
10124 The framerate is specified as a number of frames per second,
10125 as a fraction. The denominator of the fraction is given in
10126 the top half (the high two bytes) of the framerate field, and
10127 the numerator is given in the bottom half (the low two bytes).
10128 For example, if framerate is set to (100 << 16 | 750), this is
10129 750 / 100, hence 7.5fps.
10130 If the denominator is zero (the high two bytes are both zero)
10131 then it takes the value one instead, so the framerate is just
10132 the integer in the low 2 bytes.
10133 This patch fixes the the framerate calculation in vp8 encoder
10135 https://bugzilla.gnome.org/show_bug.cgi?id=783449
10137 2017-06-02 19:46:52 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10139 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
10140 libs: encoder: vp8: refactor FrameRate parameter
10141 Move frame-rate parameter from ensure_misc_params() to
10142 ensure_contro_rate_param() since it only has meaning when the
10143 control rate is either VBR or CBR.
10144 https://bugzilla.gnome.org/show_bug.cgi?id=783449
10146 2017-06-02 19:33:36 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10148 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10149 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10150 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
10151 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
10152 libs: encoder: h264,h265,mpeg2,vp8: refactor HDR
10153 Move the Hypothetical Reference Decoder (HRD) parameter, from
10154 ensure_misc_params() to ensure_control_rate_params(), since it
10155 only shall be defined when the control rate is either VBR or CBR.
10156 https://bugzilla.gnome.org/show_bug.cgi?id=783449
10158 2017-06-02 17:21:25 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10160 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10161 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
10162 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
10163 libs: encoder: h264,vp8,mpeg2: refactor control rate
10164 Instead of filling the control rate param in ensure_misc_params(),
10165 this patch refactor it out, as a first step to merge the same code
10166 for all the encoders.
10167 https://bugzilla.gnome.org/show_bug.cgi?id=783449
10169 2017-06-02 16:28:30 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10171 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10172 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10173 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
10174 libs: encoder: h264, h265, mpeg2: remove assert
10175 Remove spurious asserts for misc parameters. If they cannot be
10176 allocated, FALSE is already returned.
10178 2017-06-05 18:19:05 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10180 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10181 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
10182 libs: encoder: use VA quality level structure
10183 Instead of using a proxy to story the buffer quality level, the
10184 encoder now uses the native VA structure, which is copied to the
10185 dynamically allocated VAEncMiscParameterBuffer.
10186 This approach is computationally less expensive.
10188 2017-05-26 11:10:34 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10190 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
10191 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
10192 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10193 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
10194 libs: standardize the FIXME comment
10195 This is a trivial patch that makes homogeneous the FIXME tag in
10197 For more info about these comment style:
10198 http://wiki.c2.com/?FixmeComment
10200 2017-05-22 17:20:45 +0200 Hyunjun Ko <zzoon@igalia.com>
10202 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
10203 libs: encoder: vp8: set quality level regardless of rate control mode
10204 https://bugzilla.gnome.org/show_bug.cgi?id=782957
10206 2017-05-15 18:38:29 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10208 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10209 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10210 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10211 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
10212 libs: encoder: check for maximum number of slices
10213 Right now, H264 and HEVC can set as a property the number of slices to
10214 process. But each driver can set a maximum number of slices, depending
10215 on the supported profile & entry point.
10216 This patch verifies the current num_slices to process against the maximum
10217 permitted by the driver and the media size.
10218 https://bugzilla.gnome.org/show_bug.cgi?id=780955
10220 2017-05-15 18:36:21 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10222 * gst-libs/gst/vaapi/gstvaapiutils_h26x_priv.h:
10223 libs: utils: mark functions as internals
10224 The functions in this header are internal to the library.
10226 2017-05-15 18:35:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10228 * gst-libs/gst/vaapi/gstvaapicontext.h:
10229 libs: context: add missing documentation
10230 Document the region-of-interest configuration variables.
10232 2017-05-12 18:46:50 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10234 * tests/elements/test-vaapisink.c:
10235 tests: elements: vaapisink: handle nav events
10236 The test app can now handle navigation events to rotate the
10239 2017-05-12 18:17:55 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10241 * tests/elements/test-vaapisink.c:
10242 tests: elements: clean up vaapisink test
10243 - Use gst_element_send_event() instead of gst_pad_push_event()
10244 - don't zero App structure
10245 - check for pipeline parsing error
10246 - only get vaapisink for property set
10248 2017-05-12 13:08:30 +0900 Hyunjun Ko <zzoon@igalia.com>
10250 * gst/vaapi/gstvaapisink.c:
10251 vaapisink: keep handle_events flag except that if user want to set
10252 When state of vaapisink is changed from PLAYING to NULL, the handle_events
10253 flag is set to FALSE, and never recovered, and then event thread is never
10255 So we should allow to set the flag only when users try it.
10256 https://bugzilla.gnome.org/show_bug.cgi?id=782543
10258 2017-05-12 13:06:24 +0900 Hyunjun Ko <zzoon@igalia.com>
10260 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
10261 libs: window: x11: fix src rect info when using vpp
10262 Since we started using VPP in VaapiWindowX11, we need to care about
10263 the case that src rect and window's size are different.
10264 So, once VPP has converted to other format, we should honor the
10265 size of the VPP's surface as source rect. Otherwise, it is cropped
10266 according the previous size of the source rect.
10267 https://bugzilla.gnome.org/show_bug.cgi?id=782542
10269 2017-04-28 15:20:01 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10271 * gst/vaapi/gstvaapipluginutil.c:
10272 plugins: remove par from caps negotiation
10273 https://bugzilla.gnome.org/show_bug.cgi?id=781759
10275 2017-03-30 17:57:42 +0900 Hyunjun Ko <zzoon@igalia.com>
10277 * tests/elements/Makefile.am:
10278 * tests/elements/test-roi.c:
10279 tests: elements: add an example for ROI
10280 This implements a pipleint to recognize difference between ROI and non-ROI.
10281 See comments in this code in detail.
10282 https://bugzilla.gnome.org/show_bug.cgi?id=768248
10283 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10285 2017-03-30 17:54:20 +0900 Hyunjun Ko <zzoon@igalia.com>
10287 * gst/vaapi/gstvaapiencode.c:
10288 * gst/vaapi/gstvaapiencode_h264.c:
10289 vaapiencode: handle custom event GstVaapiEncoderRegionOfInterest
10290 Handles new custom event GstVaapiEncoderRegionOfInterest
10291 to enable/disable a ROI region.
10292 Writes a way to use new event to document.
10293 https://bugzilla.gnome.org/show_bug.cgi?id=768248
10294 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10296 2017-02-23 18:53:18 +0900 Hyunjun Ko <zzoon@igalia.com>
10298 * tests/simple-encoder.c:
10299 tests: simple-encoder: add an option to set ROI
10300 $ simple-encoder -r inputfile.y4m
10301 And you'll got an output file in H264 with two regions of interest.
10302 https://bugzilla.gnome.org/show_bug.cgi?id=768248
10303 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10305 2017-02-23 18:52:48 +0900 Hyunjun Ko <zzoon@igalia.com>
10307 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10308 libs: encoder: h264: set ROI params during encoding
10309 Set ROI params during encoding each frame, which are set via
10310 gst_vaapi_encoder_add_roi ()
10311 https://bugzilla.gnome.org/show_bug.cgi?id=768248
10312 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10314 2017-03-28 17:41:37 +0900 Hyunjun Ko <zzoon@igalia.com>
10316 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10317 * gst-libs/gst/vaapi/gstvaapiencoder.h:
10318 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
10319 libs: encoder: add api gst_vaapi_encoder_add/del_roi
10320 Implements and exposes new api gst_vaapi_encoder_add/del_roi to set ROI regions.
10321 https://bugzilla.gnome.org/show_bug.cgi?id=768248
10323 2017-02-23 17:57:07 +0900 Hyunjun Ko <zzoon@igalia.com>
10325 * gst-libs/gst/vaapi/gstvaapicontext.c:
10326 * gst-libs/gst/vaapi/gstvaapicontext.h:
10327 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10328 libs: encoder/context: query region of interest support
10329 Queries if the driver supports "Region of Interest" (ROI) during the config
10331 This attribute conveys whether the driver supports region-of-interest (ROI)
10332 encoding, based on user provided ROI rectangles. The attribute value is
10333 partitioned into fields as defined in the VAConfigAttribValEncROI union.
10334 If ROI encoding is supported, the ROI information is passed to the driver
10335 using VAEncMiscParameterTypeROI.
10336 https://bugzilla.gnome.org/show_bug.cgi?id=768248
10337 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10339 2017-05-12 11:11:48 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10341 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
10342 libs: encoder: fix a comment
10344 2017-05-11 12:23:28 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10346 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10347 libs: encoder: guard quality level configuration
10348 The quality level appeared in VA-API 0.36. So let's guard its
10351 2017-04-19 13:04:44 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
10353 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10354 * gst-libs/gst/vaapi/gstvaapiencoder.h:
10355 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10356 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10357 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
10358 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
10359 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
10360 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
10361 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
10362 encoders: add quality level tuning
10363 This patch adds the handling of VAEncMiscParameterTypeQualityLevel,
10364 in gstreamer-vaapi encoders:
10365 The encoding quality could be set through this structure, if the
10366 implementation supports multiple quality levels. The quality level set
10367 through this structure is persistent over the entire coded sequence, or
10368 until a new structure is being sent. The quality level range can be queried
10369 through the VAConfigAttribEncQualityRange attribute. A lower value means
10370 higher quality, and a value of 1 represents the highest quality. The quality
10371 level setting is used as a trade-off between quality and speed/power
10372 consumption, with higher quality corresponds to lower speed and higher power
10374 The quality level is set by the element's parameter "quality-level" with a
10375 hard-coded range of 1 to 8.
10376 Later, when the encoder is configured in run time, just before start
10377 processing, the quality level is scaled to the codec range. If
10378 VAConfigAttribEncQualityRange is not available in the used VA backend, then
10379 the quality level is set to zero, which means "disabled".
10380 All the available codecs now process this parameter if it is available.
10381 https://bugzilla.gnome.org/show_bug.cgi?id=778733
10382 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10384 2017-05-04 18:59:31 +0300 Sebastian Dröge <sebastian@centricular.com>
10388 Back to development
10390 === release 1.12.0 ===
10392 2017-05-04 15:46:03 +0300 Sebastian Dröge <sebastian@centricular.com>
10397 * gstreamer-vaapi.doap:
10401 2017-05-04 11:49:33 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10403 * gst/vaapi/gstvaapidecodebin.c:
10404 Revert "vaapidecodebin: fix element's classification"
10405 This reverts commit 8cbe03599a4f27c2001380e2ec150c4f4267a9cf.
10407 2017-04-27 22:55:27 -0700 Scott D Phillips <scott.d.phillips@intel.com>
10411 build: Require libva < 0.99.0
10412 libva >= 0.99.0 is not currently supported by gstreamer-vaapi, so
10413 fail to configure instead of failing late in the build.
10414 This libva is bundled in msdk[1] and it is ahead in time with
10415 respect the official and open source libva[2]. GStreamer-VAAPI
10416 only supports the latter for now.
10417 1. https://software.intel.com/en-us/media-sdk/download
10418 2. https://github.com/01org/libva/
10419 https://bugzilla.gnome.org/show_bug.cgi?id=781866
10421 2017-05-02 14:08:54 +0200 Victor Toso <me@victortoso.com>
10423 * gst/vaapi/gstvaapidecodebin.c:
10424 vaapidecodebin: fix element's classification
10425 This bin should have similar classification as decodebin which is
10426 "Generic/Bin/Decoder" otherwise it will appear wrongly as video
10428 Signed-off-by: Victor Toso <victortoso@redhat.com>
10429 https://bugzilla.gnome.org/show_bug.cgi?id=782063
10431 === release 1.11.91 ===
10433 2017-04-27 17:49:52 +0300 Sebastian Dröge <sebastian@centricular.com>
10438 * gstreamer-vaapi.doap:
10442 2017-04-27 13:08:56 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10444 * gst/vaapi/gstvaapipluginutil.c:
10445 Revert "plugins: reject pixel-aspect-ratio with value 0/1"
10446 This reverts commit c0be7b1890ea8da915a81ae82bc9f504aee7cc26.
10448 2017-04-27 12:43:54 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10450 * gst/vaapi/gstvaapipluginutil.c:
10451 plugins: reject pixel-aspect-ratio with value 0/1
10452 Do not negotiate a pixel-aspect-ratio of 0/1.
10453 https://bugzilla.gnome.org/show_bug.cgi?id=781759
10455 2017-04-26 15:48:01 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10457 * gst/vaapi/gstvaapipostprocutil.c:
10458 * gst/vaapi/gstvaapisink.c:
10459 plugins: handle pixel-aspect-ratio with value 0/1
10460 When downstream negotiates a pixel-aspect-ratio of 0/1, the
10461 calculations for resizing and formatting in vaapipostproc and
10462 vaapisink, respectively, failed, and thus the pipeline.
10463 This patch handles this situation by converting p-a-r of 0/1 to
10464 1/1. This is how other sinks, such as glimagesink, work.
10465 https://bugzilla.gnome.org/show_bug.cgi?id=781759
10467 2017-04-27 14:42:55 +0900 Hyunjun Ko <zzoon@igalia.com>
10469 * gst/vaapi/gstvaapivideobufferpool.c:
10470 vaapivideobufferpool: fix leak of created allocator
10471 Since it's created by itself, it should be unref-counted
10472 after gst_buffer_pool_config_set_allocator call. Afterwards,
10473 this allocator will be ref-counted again when assigning to priv->allocator.
10474 https://bugzilla.gnome.org/show_bug.cgi?id=781577
10476 2017-04-21 19:07:18 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10478 * gst/vaapi/gstvaapivideobufferpool.c:
10479 vaapivideobufferpool: create or reconfig allocator
10480 Sometimes a video decoder could set different buffer pool
10481 configurations, because their frame size changes. In this case we
10482 did not reconfigure the allocator.
10483 This patch enables this use case, creating a new allocator inside
10484 the VAAPI buffer pool if the caps changed, if it is not dmabuf-based.
10485 If so, it is just reconfigured, since it doesn't have a surface pool.
10486 https://bugzilla.gnome.org/show_bug.cgi?id=781577
10488 2017-04-25 12:58:44 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10490 * tests/elements/Makefile.am:
10491 test: elements: fix compilation flags
10492 This issue was spotten on bug #766704
10493 Original-patch-by: Hyunjun Ko <zzoon@igalia.com>
10495 2017-04-25 16:23:08 +0900 Hyunjun Ko <zzoon@igalia.com>
10497 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
10498 libs: windows: wayland: fix leak if failure of sync
10499 Sometimes gst_vaapi_window_wayland_sync returns FALSE when poll returns EBUSY
10500 during destruction.
10501 In this case, if GstVaapiWindow is using vpp, leak of vpp surface happens.
10502 This surface is not attached to anything at this moment, so we should release
10504 https://bugzilla.gnome.org/show_bug.cgi?id=781695
10506 2017-04-24 20:30:30 +0100 Tim-Philipp Müller <tim@centricular.com>
10509 Automatic update of common submodule
10510 From 60aeef6 to 48a5d85
10512 2017-04-21 15:30:09 +0200 Hyunjun Ko <zzoon@igalia.com>
10514 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
10515 libs: window: wayland: mark frames as done
10516 When the frame listener callbacks 'done', the number of pending
10517 frames are decreased. Nonetheless, there might be occasions where
10518 the buffer listener callbacks 'release', without calling previously
10519 frame's 'done'. This leads to problem with
10520 gst_vaapi_window_wayland_sync() operation.
10521 This patch marks as done those frames which were callbacked, but if
10522 the buffer callbacks 'release' and associated frame is not marked
10523 as 'done' it is so, thus the number of pending frames keeps correct.
10524 https://bugzilla.gnome.org/show_bug.cgi?id=780442
10525 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10527 2017-04-21 14:07:44 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10529 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
10530 libs: window: wayland: don't sync at destroy()
10531 Don't call gst_vaapi_window_wayland_sync() when destroying the
10532 wayland window instance, since it might lead to a lock at
10533 gst_poll_wait() when more than one instances of vaapisink are
10534 rendering in the same pipeline, this is because they share the
10536 Since now all the frames are freed we don't need to freed the
10537 private last_frame, since its address is invalid now.
10538 https://bugzilla.gnome.org/show_bug.cgi?id=780442
10539 Signed-off-by: Hyunjun Ko <zzoon@igalia.com>
10541 2017-04-19 10:37:19 +0900 Hyunjun Ko <zzoon@igalia.com>
10543 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
10544 libs: window: wayland: null buffer at destroy()
10545 Fix leakage of the last wl buffer.
10546 VAAPI wayland sink needs to send a null buffer while destruction,
10547 it assures that all the wl buffers are released. Otherwise, the last
10548 buffer's callback might be not called, which leads to leak of
10550 This was inspired by gstwaylandsink.
10551 https://bugzilla.gnome.org/show_bug.cgi?id=774029
10552 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10554 2016-10-30 10:43:49 +0900 Jagyum Koo <koojagyum@gmail.com>
10556 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
10557 libs: window: wayland: rt event queue at destroy()
10558 The proxy object of wl_buffer for the last frame remains in the
10559 wl_map. Even though we call wl_buffer_destroy() in
10560 frame_release_callback(), the proxy object remains without being
10561 removed, since proxy object is deleted when wayland server sees the
10562 delete request and sends 'delete_id' event.
10563 We need to call roundtrip before destroying event_queue so that the
10564 proxy object is removed. Otherwise, it would be mess up as receiving
10565 'delete_id' event from previous play, when playing in the next
10566 va/wayland window with the same wl_display connection.
10567 https://bugzilla.gnome.org/show_bug.cgi?id=773689
10568 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10570 2017-04-20 20:30:52 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10572 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
10573 libs: window: wayland: cancel read at poll message
10574 Always call wl_display_cancel_read() when an errno is set, but
10575 different to EAGAIN or EINTR.
10576 https://bugzilla.gnome.org/show_bug.cgi?id=780442
10578 2017-04-21 18:05:48 +0900 Hyunjun Ko <zzoon@igalia.com>
10580 * gst/vaapi/gstvaapidecodebin.c:
10581 * gst/vaapi/gstvaapidecodebin.h:
10582 vaapidecodebin: skips configuration once it's done
10583 Skips configuration of creation of vpp/capsfilter and link them once it's done.
10584 Otherwise, it always fails when it's trying to re-start playback.
10585 https://bugzilla.gnome.org/show_bug.cgi?id=781573
10587 2017-04-20 18:44:41 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10589 * gst/vaapi/gstvaapipostprocutil.c:
10590 vaapipostproc: fixes for memory leaks
10591 The use of gst_vaapi_value_set_format() and gst_structure_*_value()
10592 requires to clear the used GValue to avoid a memory leak.
10594 2016-12-08 18:51:54 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10596 * gst/vaapi/gstvaapipluginbase.c:
10597 * gst/vaapi/gstvaapipluginbase.h:
10598 plugins: enable direct rendering with envvar
10599 Direct rendering (use vaDeriveImage rather than vaPutImage) has better
10600 performance in some Intel platforms (Haswell, for example) but in others
10601 (Skylake) is the opposite.
10602 In order to have some control, the patch enables the direct rendering
10603 through the environment variable GST_VAAPI_ENABLE_DIRECT_RENDERING.
10604 Also it seems to generating some problems with gallium/radeon backend.
10606 https://bugzilla.gnome.org/show_bug.cgi?id=775848
10608 2017-04-08 02:05:21 +1000 Jan Schmidt <jan@centricular.com>
10610 * gst/vaapi/gstvaapidecode.c:
10611 vaapidecode: Don't renegotiate on every flush
10612 If caps don't actually change, don't update the
10613 decoder and don't set the do_renego flag forcing
10614 downstream renegotiation
10615 https://bugzilla.gnome.org/show_bug.cgi?id=781142
10617 2017-04-08 01:21:23 +1000 Jan Schmidt <jan@centricular.com>
10619 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
10620 h264 decoder: Implement reset() for faster flush
10621 Implement a custom reset() function for faster flushes
10622 that just clear the reference pictures but don't reallocate
10623 the DPB or clear out SPS/PPS
10624 https://bugzilla.gnome.org/show_bug.cgi?id=781142
10626 2017-04-05 17:24:20 +1000 Jan Schmidt <jan@centricular.com>
10628 * gst-libs/gst/vaapi/gstvaapidecoder.c:
10629 * gst-libs/gst/vaapi/gstvaapidecoder.h:
10630 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
10631 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
10632 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
10633 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
10634 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
10635 * gst/vaapi/gstvaapidecode.c:
10636 Implement decoder reset on flush, rather than recreating
10637 Clear decoders out on a flush but keep the same instance,
10638 rather than completely recreating them. That avoids
10639 unecessarily freeing and recreating surface pools
10640 and contexts, which can be quite expensive
10641 https://bugzilla.gnome.org/show_bug.cgi?id=781142
10643 2017-04-11 18:50:35 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10645 * gst-libs/gst/vaapi/gstvaapiwindow.c:
10646 libs: window: don't add an unused function
10647 The macro GST_VAAPI_OBJECT_DEFINE_CLASS_WITH_CODE only defines
10648 a function that is never used, thus when compiling we might see
10649 this warning (clang):
10650 gstvaapiwindow.c:147:1: warning: unused function 'gst_vaapi_window_class' [-Wunused-function]
10651 GST_VAAPI_OBJECT_DEFINE_CLASS_WITH_CODE (GstVaapiWindow,
10653 https://bugzilla.gnome.org/show_bug.cgi?id=759533
10655 2017-04-11 18:22:00 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10657 * gst-libs/gst/vaapi/gstvaapiwindow.c:
10658 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
10659 libs: window: remove surface_format member
10660 Since we always convert to NV12, there is no need to keep a
10661 variable for that. Let us hard code it.
10662 https://bugzilla.gnome.org/show_bug.cgi?id=759533
10664 2017-04-10 17:23:26 +0900 Hyunjun Ko <zzoon@igalia.com>
10666 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
10667 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
10668 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
10669 libs: window: x11/wayland: use new api for conversion
10670 Since gst_vaapi_window_vpp_convert_internal is created,
10671 GstVaapiWindowX11/Wayland can use it for conversion.
10672 Note that once it chooses to use vpp, it's going to use vpp
10673 until the session is finished.
10674 https://bugzilla.gnome.org/show_bug.cgi?id=759533
10676 2017-04-10 11:41:29 +0900 Hyunjun Ko <zzoon@igalia.com>
10678 * gst-libs/gst/vaapi/gstvaapiwindow.c:
10679 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
10680 libs: window: add gst_vaapi_window_vpp_convert_internal()
10681 If a backend doesn't support specific format, we can use vpp for conversion
10682 and make it playing.
10683 This api is originated from GstVaapiWindowWayland and moved to GstVaapiWindow,
10684 so that GstVaapiWindowX11 could use it.
10685 https://bugzilla.gnome.org/show_bug.cgi?id=759533
10687 2017-04-03 16:45:36 +0900 Hyunjun Ko <zzoon@igalia.com>
10689 * gst-libs/gst/vaapi/gstvaapiwindow.c:
10690 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
10691 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
10692 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
10693 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
10694 libs: window: x11/wayland: chaining up to GstVaapiWindow
10695 Currently, GstVaapiWindowX11/Wayland are not descendants of GstVaapiWindow.
10696 This patch chains them up to GstVaapiWindow to handle common members in GstVaapiWindow.
10697 https://bugzilla.gnome.org/show_bug.cgi?id=759533
10699 2017-04-05 11:19:15 -0700 Scott D Phillips <scott.d.phillips@intel.com>
10701 * gst/vaapi/gstvaapipluginutil.c:
10702 plugins: Fix usage of GST_GL_HAVE_WINDOW_* defines
10703 When these definitions are false, they are undef in the
10704 preprocessor, not a defined value of 0. When they are unset the
10705 compile fails with:
10706 'GST_GL_HAVE_WINDOW_WAYLAND' undeclared (first use in this function)
10707 https://bugzilla.gnome.org/show_bug.cgi?id=780948
10709 2017-04-10 23:51:06 +0100 Tim-Philipp Müller <tim@centricular.com>
10712 Automatic update of common submodule
10713 From 39ac2f5 to 60aeef6
10715 === release 1.11.90 ===
10717 2017-04-07 16:36:21 +0300 Sebastian Dröge <sebastian@centricular.com>
10722 * gstreamer-vaapi.doap:
10726 2017-04-03 14:52:41 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10728 * gst/vaapi/gstvaapiencode_h265.c:
10729 vaapiencode: h265: add main-10 in caps template
10730 This patch adds h265's main-10 profile in encoder src caps template.
10731 https://bugzilla.gnome.org/show_bug.cgi?id=771291
10733 2017-04-03 15:34:51 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10735 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10736 libs: encoder: h265: chroma and luma with format
10737 If the profile is main-10 the bit_depth_luma_minus8, in the sequence
10738 parameter buffer, shall be the color format bit depth minus 8, 10-8
10739 which is 2. Also for bit_depth_chroma_minus8.
10740 This patch gets the negotiated sink caps format and queries its
10741 luma's depth and uses that value to fill the mentioned parameters.
10742 https://bugzilla.gnome.org/show_bug.cgi?id=771291
10744 2017-03-29 19:20:26 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10746 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10747 libs: encoder: admit YUV420_10BPP as valid chroma
10748 Accepts as supported the GST_VAAPI_CHROMA_TYPE_YUV420_10BPP chroma
10750 https://bugzilla.gnome.org/show_bug.cgi?id=771291
10752 2017-03-29 19:16:50 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10754 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10755 libs: encoder: h265: ensures profile given format
10756 Set the VA profile as GST_VAAPI_PROFILE_H265_MAIN10 if the
10757 configured color format is P010_10LE.
10758 Otherwise, keep GST_VAAPI_PROFILE_H265_MAIN
10759 https://bugzilla.gnome.org/show_bug.cgi?id=771291
10761 2017-04-06 17:21:21 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10763 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10764 libs: encode: merge all possible surface formats
10765 When the function gst_vaapi_encoder_get_surface_formats() was added
10766 it was under the assumption that any VA profile of the specific codec
10767 supported the same format colors. But it is not, for example the
10768 profiles that support 10bit formats.
10769 In other words, different VA profiles of a same codec may support
10770 different color formats in their upload surfaces.
10771 In order to expose all the possible color formats, if no profile is
10772 specified via source caps, or if the encoder doesn't have yet a
10773 context, all the possible VA profiles for the specific codec are
10774 iterated and their color formats are merged.
10775 https://bugzilla.gnome.org/show_bug.cgi?id=771291
10777 2017-04-06 16:28:12 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10779 * gst/vaapi/gstvaapiencode.c:
10780 * gst/vaapi/gstvaapiencode.h:
10781 * gst/vaapi/gstvaapiencode_h264.c:
10782 * gst/vaapi/gstvaapiencode_h265.c:
10783 vaapiencode: add get_profile() vmethod
10784 This new virtual method, get_profile(), if implemented by specific
10785 encoders, will return the VA profile potentially determined by the
10787 Also it is implemented by h264 and h265 encoders, which are the main
10788 users of this vmethod.
10789 https://bugzilla.gnome.org/show_bug.cgi?id=771291
10791 2017-04-06 12:49:24 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10793 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10794 * gst-libs/gst/vaapi/gstvaapiencoder.h:
10795 * gst/vaapi/gstvaapiencode.c:
10796 libs: encoder: pass profile to get_surface_formats()
10797 In order to get the supported surface formats within a specific
10798 profile this patch adds the GstVaapiProfile as property to
10799 gst_vaapi_encoder_get_surface_formats().
10800 Currently the extracted formats are only those related with the
10801 default profile of the element's codec.
10802 https://bugzilla.gnome.org/show_bug.cgi?id=771291
10804 2017-04-06 12:28:51 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10806 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10807 libs: encoder: dummy context for get_surface_formats()
10808 Instead of creating (if it doesn't exist, yet) the encoder's context
10809 the method gst_vaapi_encoder_get_surface_formats() now it creates
10810 dummy contexts, unless the encoder has it already created.
10811 The purpose of this is to avoid setting a encoder's context with a
10813 https://bugzilla.gnome.org/show_bug.cgi?id=771291
10815 2017-04-04 14:39:59 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10817 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10818 libs: encoder: refactor init_context_info()
10819 In order to generate vaapi contexts iterative, the function
10820 init_context_info() is refactored to pass, as parameters the
10821 GstVaapiContextInfo and the GstVaapiProfile.
10822 https://bugzilla.gnome.org/show_bug.cgi?id=771291
10824 2017-04-04 14:21:43 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10826 * gst-libs/gst/vaapi/gstvaapiencoder.c:
10827 * gst-libs/gst/vaapi/gstvaapiutils.c:
10828 * gst-libs/gst/vaapi/gstvaapiutils.h:
10829 libs: encoder: initialize chroma_type
10830 Instead of initialize the chroma_type with a undefined value, which
10831 will be converted to GST_VAAPI_CHROMA_TYPE_YUV420 by GstVaapiContext,
10832 this patch queries the VA config, given the received
10833 GstVaapiContextInfo's parameters, and gets the first response.
10834 In order to get the GstVaapiChromaType value, also it was needed to
10835 add a new utility function: to_GstVaapiChromaType(), which, given a
10836 VA_RT_FORMAT_* will return the associated GstVaapiChromaType.
10837 https://bugzilla.gnome.org/show_bug.cgi?id=771291
10839 2017-03-31 11:21:21 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10841 * gst/vaapi/gstvaapiencode.c:
10842 vaapiencode: enhance logs of negotiated caps
10843 https://bugzilla.gnome.org/show_bug.cgi?id=771291
10845 2017-04-05 11:15:41 -0700 Scott D Phillips <scott.d.phillips@intel.com>
10847 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
10848 libs: encoder: vp9: Fix initialization of ref_list
10849 gcc 7.0.1 gives a memset-elt-size warning in gst_vaapi_encoder_vp9_init:
10850 'memset' used with length equal to number of elements without
10851 multiplication by element size [-Werror=memset-elt-size]
10852 https://bugzilla.gnome.org/show_bug.cgi?id=780947
10854 2017-03-31 14:12:43 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
10856 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10857 encoder: h264: Fix Backward ReferencePicture flag setting
10858 This is a regression introduced by e829b62 which
10859 override the reference flags and caused issues with
10860 latest intel-vaapi-driver.
10862 2017-03-29 13:22:47 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10864 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10865 libs: encoder: h265: fix code style
10866 Trivial patch to remove a double ';' as end of instruction.
10868 2017-03-28 10:53:20 -0700 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
10870 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10871 encoder: h264: Fix B frame encoding artifacts
10872 The current implementation is updating the POC values only
10873 in Slice parameter Buffer.But we are not filling the
10874 picture order count and reference flags in VAPictureH264
10875 while populating VA Picture/Slice structures.The latest
10876 intel-vaapi-driver is directly accessing the above fields
10877 from VAPicutreH264 provided as RefPicLists, which resulted
10878 some wrong maths and prediction errors in driver.
10879 https://bugzilla.gnome.org/show_bug.cgi?id=780620
10881 2017-03-21 16:13:56 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10883 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10884 libs: encoder: h265: remove unused macro definition
10885 Since the h265 encoder doesn't use GValueArray, there is no need to
10886 disable the Glib deprecation warnings, thus removing the macro
10889 2017-03-20 16:45:01 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
10891 * gst/vaapi/gstvaapidecode.c:
10892 * gst/vaapi/gstvaapiencode.c:
10893 * gst/vaapi/gstvaapipostproc.c:
10894 * gst/vaapi/gstvaapisink.c:
10895 plugins: when debug disabled, default category is NULL
10896 As in gstreamer-vaapi a common base class is used, the specific
10897 default category is passed to the base-plugin initializator, thus
10898 the log messages are categorized with the used plugin.
10899 Nonetheless, when the gst-debug is disabled in compilation time,
10900 it is needed to pass NULL to the base-plugin initializator. This
10902 https://bugzilla.gnome.org/show_bug.cgi?id=780302
10904 2017-03-17 17:14:01 +0900 Hyunjun Ko <zzoon@igalia.com>
10906 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10907 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10908 * gst-libs/gst/vaapi/gstvaapiutils_h26x.c:
10909 * gst-libs/gst/vaapi/gstvaapiutils_h26x_priv.h:
10910 libs: h26x: adds gst_vaapi_utils_h26x_write_nal_unit()
10911 Implements gst_vaapi_utils_h26x_write_nal_unit(), which writes NAL
10912 unit length and data to a bitwriter.
10913 Note that this helper function applies EPB (Emulation Prevention
10914 Bytes), since otherwise produced codec_data might be broken when
10915 decoder/parser considering EPB, starts parsing.
10916 See sections 7.3 and 7.4 of the H264 and H264 specifications, which
10917 describes the emulation_prevention_three_byte.
10918 https://bugzilla.gnome.org/show_bug.cgi?id=778750
10919 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10921 2017-03-17 16:49:41 +0900 Hyunjun Ko <zzoon@igalia.com>
10923 * gst-libs/gst/vaapi/Makefile.am:
10924 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10925 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10926 * gst-libs/gst/vaapi/gstvaapiutils_h26x.c:
10927 * gst-libs/gst/vaapi/gstvaapiutils_h26x_priv.h:
10928 * gst-libs/gst/vaapi/meson.build:
10929 libs: utils: h26x: create vaapiutils_h26x
10930 Since there is duplicated code in h264/265 encoder, we could
10931 refactor it to avoid duplicated code.
10932 https://bugzilla.gnome.org/show_bug.cgi?id=778750
10933 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10935 2017-03-17 16:32:36 +0900 Hyunjun Ko <zzoon@igalia.com>
10937 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
10938 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10939 libs: encoder: h264/5: fix wrong return value
10940 https://bugzilla.gnome.org/show_bug.cgi?id=778750
10942 2017-03-13 17:29:59 +0900 Hyunjun Ko <zzoon@igalia.com>
10944 * gst/vaapi/gstvaapiencode_h264.c:
10945 * gst/vaapi/gstvaapiencode_h265.c:
10946 docs: h264/h265: put parser to the example pipeline
10947 https://bugzilla.gnome.org/show_bug.cgi?id=778749
10949 2017-03-13 16:20:59 +0900 Hyunjun Ko <zzoon@igalia.com>
10951 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
10952 libs: encoder: h265: fix reserved length of bits
10953 Fix reserved length of bits for bit_depth_luma_minus8 and bit_depth_chroma_minus8
10954 https://bugzilla.gnome.org/show_bug.cgi?id=778749
10956 2017-03-12 18:59:42 +0100 Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
10958 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
10959 * tests/test-display.c:
10960 O_CLOEXEC needs _GNU_SOURCE defined
10962 The O_CLOEXEC, O_DIRECTORY, and O_NOFOLLOW flags are not specified
10963 in POSIX.1-2001, but are specified in POSIX.1-2008. Since glibc
10964 2.12, one can obtain their definitions by defining either
10965 _POSIX_C_SOURCE with a value greater than or equal to 200809L or
10966 _XOPEN_SOURCE with a value greater than or equal to 700. In glibc
10967 2.11 and earlier, one obtains the definitions by defining
10969 And indeed, with the uClibc C library, O_CLOEXEC is not exposed if
10970 _GNU_SOURCE is not defined. Therefore, this commit fixes the build of
10971 gstreamer-vaapi with the uClibc C library.
10972 Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
10973 https://bugzilla.gnome.org/show_bug.cgi?id=779953
10975 2017-03-14 16:07:08 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10978 README: fix "Sources" section
10979 Update the URL where the release source tarballs can be downloaded.
10981 2017-03-12 21:39:53 +0100 Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
10984 README: fix "Reporting bugs" section
10985 The "Reporting bugs" section gives
10986 https://bugzilla.gnome.org/enter_bug.cgi?product=gstreamer-vaapi as the
10987 link to report a bug, but this link says "Sorry, entering a bug into the
10988 product gstreamer-vaapi has been disabled.".
10989 This commit fixes the URL to point to the proper location, and also
10990 removes the following paragraph that is no longer correct.
10991 Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
10992 https://bugzilla.gnome.org/show_bug.cgi?id=779954
10994 2017-03-03 19:55:00 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
10996 * gst/vaapi/gstvaapipluginutil.c:
10997 plugins: retry to create the VA display
10998 Particularly in GNOME Wayland, the negotiated or created GL context
10999 defines a GLX environment, but VAAPI fails to create a GLX VA
11000 display because there is no a DRI2 connection.
11001 This patch retries to create the VA display if VA cannot create one
11002 with the GL context parameters. Now using the old list of display
11004 This should also work in the case of systems with two GPU, when the
11005 non-VAAPI has the graphics environment, and the VAAPI-enabled one
11006 shall work headless.
11007 https://bugzilla.gnome.org/show_bug.cgi?id=772838
11009 2016-10-19 15:33:41 +0100 Julien Isorce <j.isorce@samsung.com>
11011 * gst/vaapi/gstvaapipostproc.c:
11012 vaapipostproc: texture upload if driver supports GL
11013 Removes GstVideoGLTextureUploadMeta caps feature if the driver
11014 doesn't support opengl.
11015 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11016 https://bugzilla.gnome.org/show_bug.cgi?id=772838
11018 2016-10-25 17:48:47 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11020 * gst/vaapi/gstvaapidecode.c:
11021 vaapidecode: texture upload if driver supports GL
11022 When the allowed source pad caps are generated, the GLTextureUpload caps are
11023 only inserted if the driver support OpenGL.
11024 https://bugzilla.gnome.org/show_bug.cgi?id=772838
11026 2017-02-22 15:02:01 -0800 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
11029 configure: Add missing compiler flags
11030 The AC_CHECK_HEADERS macro was failing to locate some headers, in
11031 particular the va_enc_* headers due to missing compiler flags.
11032 https://bugzilla.gnome.org/show_bug.cgi?id=779101
11033 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11035 2017-03-01 14:48:46 +0900 Hyunjun Ko <zzoon@igalia.com>
11037 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
11038 libs: window: wayland: handle more VAStatus to use vpp
11039 Since the commit landed https://github.com/01org/intel-vaapi-driver/pull/55,
11040 we should consider more returned VAStatus to use vpp.
11041 https://bugzilla.gnome.org/show_bug.cgi?id=779400
11043 2017-02-23 15:16:06 +0900 Hyunjun Ko <zzoon@igalia.com>
11045 * gst-libs/gst/vaapi/gstvaapiencoder.c:
11046 libs: encoder: ensure profile when context initialization
11047 We can't be sure that encoder's profile is assgined already or not
11048 at context initialization.
11049 https://bugzilla.gnome.org/show_bug.cgi?id=779120
11051 2017-02-23 15:13:59 +0900 Hyunjun Ko <zzoon@igalia.com>
11053 * gst-libs/gst/vaapi/gstvaapiencoder.c:
11054 libs: encoder: set rate control info only when query succeed
11055 Currently, it set rate control information even when query fails.
11056 In addition, it doesn't update any more since the flag
11057 got_rate_control_mask is set to TRUE.
11058 https://bugzilla.gnome.org/show_bug.cgi?id=779120
11060 2017-02-24 16:00:23 +0200 Sebastian Dröge <sebastian@centricular.com>
11063 meson: Update version
11065 2017-02-24 15:38:22 +0200 Sebastian Dröge <sebastian@centricular.com>
11068 Back to development
11070 === release 1.11.2 ===
11072 2017-02-24 15:10:21 +0200 Sebastian Dröge <sebastian@centricular.com>
11077 * gstreamer-vaapi.doap:
11080 2017-02-16 18:37:59 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11082 * gst/vaapi/gstvaapiencode.c:
11083 vaapiencode: merge tags for downstream's info
11084 Add encoder and codec name and the bitrate into the output for
11085 informational purposes. Some muxers or application use it as
11087 https://bugzilla.gnome.org/show_bug.cgi?id=778781
11089 2017-02-17 01:27:52 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11091 * gst-libs/gst/vaapi/gstvaapiencoder.c:
11092 libs: encoder: caps can change at any time
11093 The encoder should be able to change its caps even it is already
11094 processing a stream.
11095 This is suppose to happen after a flush so the codedbuf_queue should
11097 https://bugzilla.gnome.org/show_bug.cgi?id=775490
11099 2017-02-17 01:19:00 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11101 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
11102 libs: encoder: h265: bail if nal unit type fails
11103 Bail out if the NAL unit type is not recognized.
11104 https://bugzilla.gnome.org/show_bug.cgi?id=778782
11106 2017-02-16 18:11:50 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11108 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
11109 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
11110 libs: decoder: h264,h265 avoid uninitialized variable
11111 Configuring GCC to verify possible usage of uninitialized variables,
11112 shows that found_index might be used without previous assignation.
11113 This patch assigns a initial value to found_index, also avoid a
11114 branching when returning the result value.
11115 https://bugzilla.gnome.org/show_bug.cgi?id=778782
11117 2017-02-13 16:39:41 -0800 Scott D Phillips <scott.d.phillips@intel.com>
11120 * gst-libs/gst/vaapi/Makefile.am:
11121 * gst/vaapi/gstvaapidecode.c:
11122 * gst/vaapi/gstvaapidecodebin.c:
11123 build: rename USE_HEVC_DECODER to USE_H265_DECODER
11124 Rename to be consistent with H.264 and also H.265 encoder. The
11125 meson build assumed this was already consistently named, and so
11126 previously was not able to actually build the H.265 decoder.
11127 https://bugzilla.gnome.org/show_bug.cgi?id=778576
11129 2017-02-15 19:14:59 +0000 Tim-Philipp Müller <tim@centricular.com>
11132 meson: gstreamer-codecparsers is a required dep
11133 Just like in configure.ac.
11135 2017-02-15 00:26:21 +0000 Tim-Philipp Müller <tim@centricular.com>
11138 meson: dist meson build files
11139 Ship meson build files in tarballs, so people who use tarballs
11140 in their builds can start playing with meson already.
11142 2017-02-10 09:51:38 +0900 Hyunjun Ko <zzoon@igalia.com>
11144 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
11145 libs: encoder: vp8: add CBR encoding mode
11146 This patch enables the Constant BitRate encoding mode in VP8 encoder.
11147 Basically it adds the configuration parameters required by libva to
11149 Original-Patch-By: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11150 https://bugzilla.gnome.org/show_bug.cgi?id=749950
11152 2017-02-09 12:39:19 +0900 Hyunjun Ko <zzoon@igalia.com>
11154 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
11155 libs: encoder: vp8: fix bitrate calculation
11156 Base encoder's unit of bitrate is in Kbps. We should honor it so
11157 we use the value of bitrate in VA, in which is expressed in bps.
11158 https://bugzilla.gnome.org/show_bug.cgi?id=749950
11160 2017-02-09 12:49:44 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11162 * gst/vaapi/gstvaapipluginbase.c:
11163 plugins: fix build when gcc
11164 In commit a8e482f9 we added a function without parameters, but gcc
11167 2017-02-06 15:46:20 -0800 Scott D Phillips <scott.d.phillips@intel.com>
11169 * gst-libs/gst/base/meson.build:
11170 * gst-libs/gst/meson.build:
11171 * gst-libs/gst/vaapi/meson.build:
11172 * gst-libs/meson.build:
11174 * gst/vaapi/meson.build:
11176 * meson_options.txt:
11177 vaapi: add meson build
11178 https://bugzilla.gnome.org/show_bug.cgi?id=778250
11180 2017-02-08 10:17:40 -0800 Scott D Phillips <scott.d.phillips@intel.com>
11183 * gst-libs/gst/vaapi/Makefile.am:
11184 * gst-libs/gst/vaapi/gstvaapidisplay.c:
11185 * gst-libs/gst/vaapi/gstvaapiversion.h.in:
11186 make: remove gstvaapiversion.h generation
11187 https://bugzilla.gnome.org/show_bug.cgi?id=778250
11189 2016-10-19 15:47:41 +0100 Julien Isorce <j.isorce@samsung.com>
11191 * gst/vaapi/gstvaapipluginbase.c:
11192 plugins: use linear storage if not the same device
11193 When dmabuf is negotiated downstream and decoding and rendering are
11194 not done on the same device, the layout has to be linear in order for
11195 the memory to be shared accross devices, since each device has its
11196 own way to do tiling.
11197 Right now this code is rather just a to-do comment, since we are not
11198 fetching the device ids.
11199 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11201 2017-02-08 14:17:05 +0900 Hyunjun Ko <zzoon@igalia.com>
11203 * gst-libs/gst/vaapi/gstvaapiutils.c:
11204 libs: utils: add HEVC profiles representation
11205 https://bugzilla.gnome.org/show_bug.cgi?id=778318
11207 2017-02-07 16:17:39 +0900 Hyunjun Ko <zzoon@igalia.com>
11209 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
11210 libs: decoder: h264: reduce frame number of gaps
11211 Reduce frame num gaps so that we don't have to create unnecessary
11212 dummy pictures, just throw them away.
11213 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11214 https://bugzilla.gnome.org/show_bug.cgi?id=777506
11216 2016-10-16 01:04:09 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11218 * gst/vaapi/gstvaapidecode.c:
11219 vaapidecode: don't GLTextureUpload if dmabuf
11220 Do not add the meta:GstVideoGLTextureUploadMeta feature if the render
11221 element can handle dmabuf-based buffers, avoiding its negotiation.
11223 2016-10-19 16:21:21 +0100 Julien Isorce <j.isorce@samsung.com>
11225 * gst/vaapi/gstvaapidecode.c:
11226 vaapidecode: make pool to export decoder's surface
11227 Use new -base API gst_video_decoder_allocate_output_frame_full() to
11228 pass the current proxy/surface to the pool.
11229 The pool will will export thins given surface instead of exporting a
11230 brand new surface that will never be filled in with meaningfull data.
11231 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11233 2017-02-03 17:06:29 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11235 * gst/vaapi/gstvaapipluginbase.c:
11236 plugins: decoder can negotiate dmabuf downstream
11238 2016-10-19 16:07:07 +0100 Julien Isorce <j.isorce@samsung.com>
11240 * gst/vaapi/gstvaapivideobufferpool.c:
11241 vaapivideobufferpool: override acquire_buffer()
11242 Overriding the vmethod acquire_buffer() it is possible to attach the
11243 right GstMemory to the current acquired buffer.
11244 As a matter of fact, this acquired buffer may contain any instantiated
11245 GstFdmemory, since this buffer have been popped out from the buffer
11246 pool, which is a FIFO queue. So there is no garantee that this buffer
11247 matches with the current processed surface. Evenmore, the VA driver
11248 might not use a FIFO queue. Therefore, it is no way to guess on the
11250 In short, acquire_buffer on the VA driver and on the buffer pool return
11251 none matching data, we have to manually attach the right GstFdMemory to
11252 the acquired GstBuffer. The right GstMemory is the one associated with
11253 the current surface.
11254 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11256 2016-10-19 16:05:04 +0100 Julien Isorce <j.isorce@samsung.com>
11258 * gst/vaapi/gstvaapivideobufferpool.c:
11259 * gst/vaapi/gstvaapivideomemory.c:
11260 vaapivideomemory: export surface if it is provided
11261 gst_vaapi_dmabuf_memory_new() always exports a surface. Previously, it
11262 had to create that surface. Now it can also export an already provided
11263 surface. It is useful to export decoder's surfaces (from VA context).
11264 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11266 2016-10-19 15:55:27 +0100 Julien Isorce <j.isorce@samsung.com>
11268 * gst/vaapi/gstvaapivideobufferpool.h:
11269 vaapivideobufferpool: add GstVaapiVideoBufferPoolAcquireParams
11270 Useful to let the pool know the current surface proxy when calling
11271 gst_buffer_pool_alloc_buffer() / gst_buffer_pool_acquire_buffer()
11272 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11274 2016-10-19 15:09:34 +0100 Julien Isorce <j.isorce@samsung.com>
11276 * gst-libs/gst/vaapi/gstvaapisurface.c:
11277 * gst-libs/gst/vaapi/gstvaapisurface.h:
11278 libs: surface: add gst_vaapi_surface_{set,peek}_buffer_proxy()
11279 These functions are useful when a dmabuf-based memory is instantiated in
11280 order to relate the generated buffer @proxy with the processed @surface.
11281 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11283 2016-10-19 15:07:31 +0100 Julien Isorce <j.isorce@samsung.com>
11285 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
11286 * gst-libs/gst/vaapi/gstvaapibufferproxy.h:
11287 * gst-libs/gst/vaapi/gstvaapibufferproxy_priv.h:
11288 libs: bufferproxy: gst_vaapi_buffer_proxy_{set,peek}_mem()
11289 This patch adds a GstMemory as a variable member of the buffer proxy,
11290 because we will need to associate the buffer proxy with the memory
11291 which exposes it. Later, we will know which memory, in the video buffer
11292 pool, is attached to the processed surface.
11293 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11295 2016-10-19 15:33:41 +0100 Julien Isorce <j.isorce@samsung.com>
11297 * gst/vaapi/gstvaapipostproc.c:
11298 vaapipostproc: don't GLTextureUpload if dmabuf
11299 Do not add the meta:GstVideoGLTextureUploadMeta feature if the render
11300 element can handle dmabuf-based buffers, avoiding its negotiation.
11301 Similar as "vaapidecode: do not add meta:GstVideoGLTextureUploadMeta
11302 feature if can dmabuf"
11303 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11305 2016-12-16 14:12:30 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11307 * gst/vaapi/gstvaapipluginbase.c:
11308 plugins: enable DMAbuf allocator to downstream
11309 If the negotiated caps are raw caps and downstream supports the
11310 EGL_EXT_image_dma_buf_import extension, then the created allocator
11311 is the DMAbuf, configured to downstream.
11312 At this moment, the only element which can push dmabuf-based buffers
11313 to downstream, is vaapipostproc.
11315 2016-06-02 22:13:51 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11317 * gst/vaapi/gstvaapipluginbase.c:
11318 * gst/vaapi/gstvaapipluginbase.h:
11319 plugins: check if negotiate dmabuf with downstream
11320 In order to enable, in the future, dmabuf-based buffers, the vaapi base
11321 plugin needs to check if downstream can import dmabuf buffers.
11322 This patch checks if downstream can handle dmabuf, by introspecting the
11323 shared GL context. If the GL context is EGL/GLES2 and have the extension
11324 EGL_EXT_image_dma_buf_import, then dmabuf can be negotiated.
11325 Original-patch-by: Julien Isorce <j.isorce@samsung.com>
11327 2016-10-19 15:37:04 +0100 Julien Isorce <j.isorce@samsung.com>
11329 * gst/vaapi/gstvaapivideomemory.c:
11330 vaapivideomemory: release proxy's data if downstream
11331 The surface created for downstream is going to be filled by VAAPI
11332 elements. So, the driver needs write access on that surface.
11333 This patch releases the derived image held by the proxy, thus the
11334 surface is unmarked as busy.
11335 This is how it has to be done as discussed on libva mailing list.
11336 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11338 2016-10-19 15:01:04 +0100 Julien Isorce <j.isorce@samsung.com>
11340 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
11341 * gst-libs/gst/vaapi/gstvaapibufferproxy.h:
11342 libs: bufferproxy: add gst_vaapi_buffer_proxy_release_data()
11343 Adds an API to request the user's data release in the buffer proxy.
11344 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11346 2016-10-19 15:27:03 +0100 Julien Isorce <j.isorce@samsung.com>
11348 * gst/vaapi/gstvaapipluginbase.c:
11349 * gst/vaapi/gstvaapivideomemory.c:
11350 * gst/vaapi/gstvaapivideomemory.h:
11351 vaapivideomemory: add direction to dmabuf allocator
11352 Add GstPadDirection param to gst_vaapi_dmabuf_allocator_new(), thus
11353 we later could do different thing when the allocated memory is for
11354 upstream or dowstream, as required by VA-API.
11355 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11357 2016-12-15 15:59:30 +0900 Hyunjun Ko <zzoon@igalia.com>
11359 * gst-libs/gst/vaapi/gstvaapiutils_core.c:
11360 libs: utils: return NULL if failed to get surface formats
11361 Thus, when generating the allowed caps, the element will throw a
11362 warning and it will use its caps template.
11363 This behavior might be a bug in the VA driver.
11364 https://bugzilla.gnome.org/show_bug.cgi?id=775490
11366 2015-11-26 18:21:08 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11368 * gst-libs/gst/vaapi/gstvaapidisplay.c:
11369 Revert "vaapidisplay: mark X11 display as compatible with EGL"
11370 This reverts commit 200b1baabc066f8a4102f82f539655d588200ec9.
11372 2017-02-01 14:32:45 +0900 Hyunjun Ko <zzoon@igalia.com>
11374 * gst/vaapi/gstvaapipostproc.c:
11375 vaapipostproc: set GST_VAAPI_POSTPROC_FLAG_SIZE according to src caps
11376 A value of width/height property should be set to out caps,
11377 if negotiation had been going properly.
11378 So we can use srcpad_info when making decision of scaling.
11379 https://bugzilla.gnome.org/show_bug.cgi?id=778010
11381 2017-01-27 12:10:54 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11383 * gst/vaapi/gstvaapidecode.c:
11384 * gst/vaapi/gstvaapiencode.c:
11385 * gst/vaapi/gstvaapipluginutil.c:
11386 * gst/vaapi/gstvaapipluginutil.h:
11387 * gst/vaapi/gstvaapipostproc.c:
11388 * gst/vaapi/gstvaapisink.c:
11389 plugins: handle GL params through context query
11390 If the element instantiated the GL display and context, they should
11391 handle them too through the context query.
11392 https://bugzilla.gnome.org/show_bug.cgi?id=777409
11394 2017-01-26 12:02:56 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11396 * gst/vaapi/gstvaapipluginbase.c:
11397 * gst/vaapi/gstvaapipluginbase.h:
11398 * gst/vaapi/gstvaapipluginutil.c:
11399 plugins: create a GL context on certain conditions
11400 If a GstVaapiDisplay is not found in the GStreamer context sharing,
11401 then VAAPI elements look for a local GstGLContext in gst context
11402 sharing mechanism ('gst.gl.local.context').
11403 If this GstGLContext not found either then, only the VAAPI decoders
11404 and the VAAPI post-processor, will try to instantiate a new
11406 If a valid GstGLContext is received, then a new GstVaapiDisplay will
11407 be instantiated with the platform, API and windowing specified by the
11408 instantiated GstGLContext.
11409 Original-Patch-By: Matt Fischer <matt.fischer@garmin.com>
11410 https://bugzilla.gnome.org/show_bug.cgi?id=777409
11412 2016-08-02 15:48:25 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11414 * gst/vaapi/gstvaapivideocontext.c:
11415 vaapivideocontext: context type can be rejected
11416 Instead of calling g_return_val_if_fail() to check the context type, we
11417 should use a normal conditional, since it is possible that other context types
11418 can arrive and try to be assigned. Otherwise a critical log message is
11420 This happens when we use playbin3 with vaapipostproc as video-filter.
11421 https://bugzilla.gnome.org/show_bug.cgi?id=777409
11423 2017-01-20 19:57:52 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11425 * gst/vaapi/gstvaapipostprocutil.c:
11426 vaapipostproc: use sink caps par if not requested
11427 Use the sink caps pixel-aspect-ratio to fixate the src caps, if it
11428 is not already set.
11429 https://bugzilla.gnome.org/show_bug.cgi?id=777395
11431 2017-01-20 19:00:24 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11433 * gst/vaapi/gstvaapipostproc.c:
11434 * gst/vaapi/gstvaapipostprocutil.c:
11435 vaapipostproc: set interlace mode
11436 if the vaapipostproc is configured to not do deinterlacing, the
11437 interlace-mode in the src caps should be the same as the input caps.
11438 https://bugzilla.gnome.org/show_bug.cgi?id=777395
11440 2017-01-20 16:10:32 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11442 * gst/vaapi/gstvaapisink.c:
11443 vaapisink: fix gcc compiler warning
11444 warning: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement]
11446 2017-01-12 19:54:41 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11448 * gst/vaapi/gstvaapisink.c:
11449 vaapisink: don't use member variable outside lock
11450 Thus a race condition segfault is avoided.
11451 Original-patch-by: Matt Staples <staples255@gmail.com>
11452 https://bugzilla.gnome.org/show_bug.cgi?id=777146
11454 2017-01-18 17:20:21 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11456 * gst/vaapi/gstvaapipluginbase.c:
11457 * gst/vaapi/gstvaapipostproc.c:
11458 plugins: avoid log flood when activating pool
11459 Every time a new buffer is allocated, the pool is activated. This
11460 doesn't impact in performance since gst_buffer_pool_set_active()
11461 checks the current state of the pool. Nonetheless it logs out a
11462 message if the state is the same, and it floods the logging subsystem
11464 To avoid this log flooding first the pool state is checked before
11467 2017-01-13 21:26:15 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11469 * gst-libs/gst/vaapi/gstvaapidecoder.c:
11470 * gst-libs/gst/vaapi/gstvaapidecoder.h:
11471 * gst/vaapi/gstvaapidecode.c:
11472 * gst/vaapi/gstvaapidecode.h:
11473 vaapidecode: update internal decoder sink caps
11474 When a new sink caps arrive the internal decoder state is updated
11475 and, if it is, request a downstream renegotiation.
11476 Previously, when new caps arrived the whole decoder where destroyed
11477 and recreated. Now, if the caps are compatible or has the same codec,
11478 the internal decoder is kept, but a downstream renegotiation is
11480 https://bugzilla.gnome.org/show_bug.cgi?id=776979
11482 2017-01-12 16:33:13 +0200 Sebastian Dröge <sebastian@centricular.com>
11485 Back to development
11487 === release 1.11.1 ===
11489 2017-01-12 16:27:12 +0200 Sebastian Dröge <sebastian@centricular.com>
11494 * gstreamer-vaapi.doap:
11497 2017-01-12 12:49:55 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11499 * gst-libs/gst/vaapi/gstvaapiutils.c:
11500 vaapi: bump ifdef to API 0.40.0 for log redirect
11501 vaSetInfoCallback() is not released yet. It is going to appear in
11504 2017-01-12 13:45:29 +0200 Sebastian Dröge <sebastian@centricular.com>
11506 * gst-libs/gst/vaapi/gstvaapiutils.c:
11507 vaapiutils: Fix compilation with latest and previous libva releases
11508 vaSetInfoCallback() was defined after 0.39.4 / 1.7.3, so check for
11511 2017-01-11 16:04:24 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11513 * gst-libs/gst/vaapi/gstvaapidisplay.c:
11514 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
11515 * gst-libs/gst/vaapi/gstvaapiutils.c:
11516 * gst-libs/gst/vaapi/gstvaapiutils.h:
11517 libs: display: redirect logging at initialize
11518 Redirect libva's logs to GStreamer logging mechanism. This is
11519 particularly useful when VA is initialized, because it always logs
11520 out the drivers details.
11521 In order to achieve this a new helper function was added as a wrapper
11522 for the vaInitialize() function.
11523 https://bugzilla.gnome.org/show_bug.cgi?id=777115
11525 2017-01-10 15:15:31 +0900 Hyunjun Ko <zzoon@igalia.com>
11527 * gst/vaapi/gstvaapipluginbase.c:
11528 plugins: provide at least two buffers in sink pool
11529 Adds two buffers as the default value of minimum buffer.
11530 This would be used when creating and proposing vaapi bufferpool for
11531 sink pad, hence the upstream element will keep, at least, these two
11533 https://bugzilla.gnome.org/show_bug.cgi?id=775203
11534 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11536 2017-01-10 13:49:27 +0900 Hyunjun Ko <zzoon@igalia.com>
11538 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
11539 libs: decoder: h264: don't update cloned attributes
11540 If the frame is a cloned picture, its PTS comes from its parent
11541 picture. In addition, the base decoder doesn't set a valid PTS to
11542 the frame corresponding to the cloned picture.
11543 https://bugzilla.gnome.org/show_bug.cgi?id=774254
11545 2017-01-09 19:25:24 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11547 * gst/vaapi/gstvaapivideomemory.c:
11548 vaapivideomemory: check for display
11549 This patch fixes the check of display, rather than check for the
11550 meta, which it is known it exists.
11552 2017-01-09 16:23:56 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11554 * gst/vaapi/gstvaapivideomemory.c:
11555 * gst/vaapi/gstvaapivideomemory.h:
11556 vaapivideomemory: rename dmabuf allocator parameters
11557 Rename the parameters 'vip' and 'flags' to 'alloc_info' and
11558 'surface_alloc_flags' respectively. The purpose of this change is
11559 to auto-document those parameters.
11560 Also, aligned to this patch, the local 'alloc_info' variable was
11561 renamed as 'surface_info', because it stores the possible surface's
11562 video info, not the allocate one.
11564 2017-01-09 16:18:32 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11566 * gst/vaapi/gstvaapivideomemory.c:
11567 * gst/vaapi/gstvaapivideomemory.h:
11568 vaapivideomemory: rename vip to alloc_info parameter
11569 In order to auto-document the code, this patch renames the 'vip'
11570 parameter in the functions related to gst_vaapi_video_allocator_new ()
11571 to 'alloc_info', since it declares the allocation video info from
11572 the vaapi buffer pool.
11574 2017-01-09 16:08:17 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11576 * gst/vaapi/gstvaapivideomemory.c:
11577 vaapivideomemory: use GST_VIDEO_INFO_FORMAT_STRING()
11578 Use the existing local macro GST_VIDEO_INFO_FORMAT_STRING() to get
11579 the video format string.
11581 2017-01-09 12:51:11 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11583 * gst/vaapi/gstvaapivideomemory.c:
11584 vaapivideomemory: unroll gst_vaapi_surface_new_with_format()
11585 gst_vaapi_surface_new_with_format() is a wrapper for
11586 gst_vaapi_surface_new_full (). In this case, the former is simpler
11587 than the first. This patch changes that.
11589 2017-01-04 19:23:06 +0900 Hyunjun Ko <zzoon@igalia.com>
11591 * gst/vaapi/gstvaapipostproc.c:
11592 vaapipostproc: notify if src pad caps changed
11593 If src pad caps have changed, it needs to notify it downstream. In
11594 addition, do not set passthrough if they have changed.
11595 Otherwise, transform sometimes starts processing before caps change.
11596 The passthrough value will be set in fixate later in this case.
11597 https://bugzilla.gnome.org/show_bug.cgi?id=775204
11599 2016-12-14 15:51:01 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11601 * gst/vaapi/gstvaapidecodebin.c:
11602 vaapidecodebin: capsfilter to optimize negotiation
11603 Add a capsfilter forcing the caps
11604 "video/x-raw(memory:VASurface), format=(string)NV12" between the
11605 queue and the vaapipostproc so no renegotiation is required.
11606 https://bugzilla.gnome.org/show_bug.cgi?id=776175
11608 2016-12-21 17:38:07 +0900 Hyunjun Ko <zzoon@igalia.com>
11610 * gst/vaapi/gstvaapivideomemory.c:
11611 vaapivideomemory: fail surface/image configuration
11612 To detect and handle errors during allocator_configure_surface_info()
11613 and allocator_conigure_image_info().
11614 https://bugzilla.gnome.org/show_bug.cgi?id=776084
11615 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11617 2016-12-21 17:50:41 +0900 Hyunjun Ko <zzoon@igalia.com>
11619 * gst/vaapi/gstvaapisink.c:
11620 vaapisink: ensures raw caps at start()
11621 Calls gst_vaapi_plugin_base_get_allowed_raw_caps() at start() to avoid
11622 race conditions at get_caps(), especially with multiple src elements.
11623 https://bugzilla.gnome.org/show_bug.cgi?id=776303
11625 2016-12-09 14:51:52 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11627 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
11628 libs: surface: fix error handling code style
11630 2016-12-09 16:14:14 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11632 * gst/vaapi/gstvaapivideomemory.c:
11633 vaapivideomemory: add gst_video_info_update_from_surface()
11634 With this function is possible to refactor and remove duplicated code
11635 between dmabuf configuration and direct rendering/uploading
11638 2016-12-09 15:51:32 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11640 * gst/vaapi/gstvaapivideomemory.c:
11641 vaapivideomemory: no log object at initialization
11642 When an instance of GstVaapiVideoAllocator fails at initializing, the
11643 log message should not include the allocator's object, because it is
11644 going to be unrefed.
11646 2016-12-09 17:56:02 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11648 * gst/vaapi/gstvaapidecode.c:
11649 vaapidecode: lock stream when setting caps
11651 2016-12-09 17:42:42 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11653 * gst/vaapi/gstvaapidecode.c:
11654 Revert "vaapidecode: implement negotiate() vmethod"
11655 This reverts commit 3285121181295c544480fc6ba756845b16285d30.
11656 videodecode's negotiate() vmethod is also called when events arrive,
11657 but this would mean that the proper configuration of sink pad might
11658 not be complete, thus we should not update the src pad.
11659 Let's keep the old non-vmethod negotitate().
11661 2016-12-07 16:52:35 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11663 * gst/vaapi/gstvaapi.c:
11664 Revert "encoders: demote to RANK_NONE since not fit for autoplugging yet"
11665 This reverts commit f182b8be2ba05965e6d31a4d380d6563b9b53a77.
11667 2016-12-01 18:57:10 +0900 Hyunjun Ko <zzoon@igalia.com>
11669 * gst/vaapi/gstvaapiencode.c:
11670 * gst/vaapi/gstvaapiencode.h:
11671 vaapiencode: get surface formats in get_caps()
11672 Query for the supported surface formats in config at get_caps() vmethod.
11673 https://bugzilla.gnome.org/show_bug.cgi?id=769266
11674 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11676 2016-12-07 11:26:37 +0100 Hyunjun Ko <zzoon@igalia.com>
11678 * gst-libs/gst/vaapi/gstvaapiencoder.c:
11679 * gst-libs/gst/vaapi/gstvaapiencoder.h:
11680 libs: encoder: add gst_vaapi_encoder_get_surface_formats()
11681 This method will return the valid surface formats in the current
11682 config. If the are no VAConfig it is created with the information
11684 https://bugzilla.gnome.org/show_bug.cgi?id=769266
11685 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11687 2016-12-07 11:10:42 +0100 Hyunjun Ko <zzoon@igalia.com>
11689 * gst-libs/gst/vaapi/gstvaapiencoder.c:
11690 libs: encoder: split set_context_info()
11691 Split set_context_info() adding init_context_info() which only
11692 initialises the GstVaapiContextInfo structure inside GstVaapiEncoder
11693 required for VAConfig.
11694 https://bugzilla.gnome.org/show_bug.cgi?id=769266
11695 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11697 2016-12-02 09:30:52 +0900 Hyunjun Ko <zzoon@igalia.com>
11699 * gst-libs/gst/vaapi/gstvaapicontext.c:
11700 libs: context: skip VAContext if no frame size
11701 If GstVaapiContextInfo has just initial information, without frame's
11702 width and height, skip the creation of the VAContext, just keep the
11704 https://bugzilla.gnome.org/show_bug.cgi?id=769266
11705 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11707 2016-12-02 09:28:07 +0900 Hyunjun Ko <zzoon@igalia.com>
11709 * gst-libs/gst/vaapi/gstvaapicontext.c:
11710 libs: context: split context_create()
11711 Split the funcion context_create() into context_create() and
11713 Decoupling VAConfig and VAContext during context creation, we could
11714 query the VAConfig for the supported surface's formats without creating
11716 https://bugzilla.gnome.org/show_bug.cgi?id=769266
11718 2016-12-06 17:33:42 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11720 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
11721 libs: drm: find render node in hybrid system
11722 Originally the drm backend only tried to open the first render node
11723 found. But in hybrid system this first render node might not support
11724 VA-API (propietary Nvidia driver, for example).
11725 This patch tries all the available nodes until a finding one with a
11726 VA-API supported driver.
11727 https://bugzilla.gnome.org/show_bug.cgi?id=774811
11728 Original-patch-by: Stirling Westrup <swestrup@gmail.com> and
11729 Reza Razavi <reza@userful.com>
11731 2016-11-14 17:45:55 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11733 * gst/vaapi/gstvaapivideomemory.c:
11734 vaapivideomemory: refactor code for readability
11735 Added the inlined function allocator_configure_pools() moving out code
11736 from gst_vaapi_video_allocator_new() to make clear that it is a
11737 post-initalization of the object.
11739 2016-11-14 17:40:37 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11741 * gst/vaapi/gstvaapivideomemory.c:
11742 vaapivideomemory: log error if not VA image
11743 Log an error message if the test image for surface downloading
11744 cannot be allocated or mapped.
11746 2016-11-14 17:33:41 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11748 * gst/vaapi/gstvaapivideomemory.c:
11749 vaapivideomemory: remove unused macros
11750 These macros are not used. Let us remove them.
11752 2016-11-11 19:45:45 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11754 * gst/vaapi/gstvaapivideomemory.c:
11755 * gst/vaapi/gstvaapivideomemory.h:
11756 vaapivideomemory: rename video for allocation_info
11757 Since video_info stores the GstVideoInfo of the allocation caps,
11758 it is clear if we rename it as allocation_info, to distinguish it
11759 later from negotiation_info.
11761 2016-10-19 15:27:03 +0100 Julien Isorce <j.isorce@samsung.com>
11763 * gst/vaapi/gstvaapivideomemory.c:
11764 * gst/vaapi/gstvaapivideomemory.h:
11765 vaapivideomemory: add real GstVaapiDmaBufAllocator
11766 Instead of defining GstVaapiDmaBufAllocator as a hackish decorator of
11767 GstDmaBufAllocator, now, since the expose of the GstDmaBufAllocator's
11768 GType, GstVaapiDmaBufAllocator is a full feature GstAllocator inherited
11769 from GstDmaBufAllocator.
11770 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11771 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11773 2016-10-19 15:30:09 +0100 Julien Isorce <j.isorce@samsung.com>
11775 * gst/vaapi/gstvaapivideomemory.c:
11776 vaapivideomemory: errors in gst_vaapi_dmabuf_allocator_new()
11777 Add a helper function to initialize the gst_debug_vaapivideomemory,
11778 to use it either by the GstVaapiVideoAllocatorClass or
11779 GstVaapiDmabufAllocator (which is a decorator of GstDmaBufAllocator).
11780 Later, log possible errors when calling gst_vaapi_dmabuf_allocator_new ()
11781 https://bugzilla.gnome.org/show_bug.cgi?id=755072
11783 2016-11-29 15:14:32 +0900 Hyunjun Ko <zzoon@igalia.com>
11785 * gst/vaapi/gstvaapiencode.c:
11786 vaapiencode: release internal encoder at stop()
11787 As the internal encoder is created at start(), let's release it at
11788 stop() vmethod, to be consistent.
11789 gst_vaapiencode_destroy() is called since it also resets the input and
11790 output states, which is something that the base class does internally
11791 after calling stop() vmethod.
11792 https://bugzilla.gnome.org/show_bug.cgi?id=769266
11793 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11795 2016-12-03 08:20:56 +0100 Edward Hervey <bilboed@bilboed.com>
11798 Automatic update of common submodule
11799 From f49c55e to 39ac2f5
11801 2016-11-29 14:59:02 +0900 Hyunjun Ko <zzoon@igalia.com>
11803 * gst/vaapi/gstvaapiencode.c:
11804 vaapiencode: call ensure_encoder() at start()
11805 Currently, specific encoder is created during set_format(). This might
11806 lead to race condition when creating profiles with multiple encoders.
11807 This patch moves ensure_encoder() call to start() vmethod to ensure
11808 avoiding the race condition.
11809 https://bugzilla.gnome.org/show_bug.cgi?id=773546
11811 2016-11-21 19:29:22 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11813 * gst/vaapi/gstvaapidecodebin.c:
11814 vaapidecodebin: bring back dynamic configuration
11815 In commit ca0c3fd6 we remove the dynamic configuration if the bin
11816 because we assumed that the bin will be always static as it is
11818 Nonetheless we were wrong, because it is possible to request, with a
11819 property, to avoid the use of the post-processor.
11820 Since we want to add a way to disable the post-processor through
11821 environment variables, this remove feature is required again.
11822 If the environment variable GST_VAAPI_DISABLE_VPP is defined the
11823 postprocessor inside of the vaapidecodebin is disabled, then
11824 vaapidecodebin is an alias of the old vaapidecode.
11825 https://bugzilla.gnome.org/show_bug.cgi?id=775041
11827 2016-11-21 18:25:18 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11829 * gst/vaapi/gstvaapidecode.c:
11830 vaapidecode: implement negotiate() vmethod
11831 Instead of decorating the negotiate() method, let us override it,
11832 so the stream is locked while called.
11833 https://bugzilla.gnome.org/show_bug.cgi?id=775040
11835 2016-11-26 11:27:26 +0000 Tim-Philipp Müller <tim@centricular.com>
11838 common: use https protocol for common submodule
11839 https://bugzilla.gnome.org/show_bug.cgi?id=775110
11841 2016-11-24 21:17:54 +0100 Dominique Leuenberger <dimstar@opensuse.org>
11843 * gst-libs/gst/vaapi/Makefile.am:
11844 build: add LIBVA_WAYLAND_CFLAGS to libgstvaapiegl
11845 In case libva-wayland has its headers not installed in default
11846 locations (like /usr/include), the build fails to include "wayland-client.h":
11847 CC libgstvaapi_egl_la-gstvaapiutils_egl.lo
11848 In file included from gstvaapidisplay_wayland.h:27:0,
11849 from gstvaapidisplay_egl.c:35:
11850 /usr/include/va/va_wayland.h:31:28: fatal error: wayland-client.h: No such file or directory
11851 #include <wayland-client.h>
11852 As we already passed VA_CLAGS, /usr/include/va/va_wayland.h could be found, but it is
11853 our fault not to instruct the system that we ALSO care for va_wayland. We correctly query
11854 for libva-wayland.pc in configure and use this in other places as well. It is thus only
11855 correct and consequent, to do it also at this spot.
11856 https://bugzilla.gnome.org/show_bug.cgi?id=773946
11858 2015-10-28 12:55:18 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11860 * gst/vaapi/gstvaapivideocontext.c:
11861 vaapivideocontext: log a message if no bus
11862 Raise a warning if there is no bus when the element tries to post a
11865 2015-10-28 12:57:14 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11867 * gst/vaapi/gstvaapisink.c:
11868 vaapisink: create display at open()
11869 Instead of creating the VA display before setting the bus to the
11870 element, it is created when the element is opened.
11871 Basically, this commit is a revert of
11872 5e5d62cac79754ba60057fc2516135aad8d7de35
11873 That was done when the GStreamer's context sharing was not mature
11874 enough as now. There is no reason to keep this hack.
11876 2016-11-04 18:04:36 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11878 * gst/vaapi/gstvaapidecode.c:
11879 vaapidecode: don't add video crop meta
11880 Since the differentiation of negotiation caps and allocation caps,
11881 there is no need to add a video crop meta with the negotiation caps.
11882 Hence, removing it.
11883 https://bugzilla.gnome.org/show_bug.cgi?id=773948
11885 2015-03-19 21:20:26 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
11887 * gst/vaapi/gstvaapivideobuffer.c:
11888 * gst/vaapi/gstvaapivideometa.c:
11889 * gst/vaapi/gstvaapivideometa.h:
11890 remove the video converter from vaapi buffer meta
11891 Since all the video converter were deprecated in gstreamer-1.2, we don't need
11892 to handle them anymore in the vaapi's buffer meta.
11893 This patch removes its usage and the buffer meta's API for that.
11894 https://bugzilla.gnome.org/show_bug.cgi?id=745728
11896 2016-11-21 18:28:18 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11898 * gst/vaapi/gstvaapipluginbase.c:
11899 plugins: destroy source pad allocator and pool
11900 First, deactivate source pad pool when the out caps change, and if so,
11901 destroy texture map, the source pad allocator and pool only if the
11902 new caps are different from the ones already set.
11904 2016-11-21 19:17:07 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11906 * gst/vaapi/gstvaapipluginbase.c:
11907 plugins: don't destroy sink pad allocator
11908 Don't destroy sink pad allocator at _set_caps() because it will be done at
11909 ensure_sinkpad_buffer_pool() if it is required.
11911 2016-11-21 18:27:00 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11913 * gst/vaapi/gstvaapipluginbase.c:
11914 plugins: first validate the out caps
11915 When calling _set_caps() first validate the out caps before doing
11918 2016-11-21 18:42:02 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11920 * gst/vaapi/gstvaapidecode.c:
11921 vaapidecode: negotiate after destroying allocator
11922 This is related with bug 758907 when no vaapipostproc is used (no
11923 vaapidecodebin). In order to negotiate downstream we need to destroy
11924 the source pad allocator, otherwise the same allocated buffers are
11925 used, failing the mapping.
11927 2016-11-21 16:35:34 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11929 * gst/vaapi/gstvaapivideomemory.c:
11930 vaapivideomemory: remove GST_VAAPI_TYPE_VIDEO_INFO
11931 Remove redundant GST_VAAPI_TYPE_VIDEO_INFO, since it is a duplicate of
11932 GST_TYPE_VIDEO_INFO created before gstreamer 1.6, where the boxed type
11934 https://bugzilla.gnome.org/show_bug.cgi?id=774782
11936 2016-11-21 12:51:25 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11938 * gst/vaapi/gstvaapipluginbase.c:
11939 plugins: update the src pad allocator video info
11940 Update the size, stride and offset of the source pad allocator video
11941 info, so the pool could set the correct GstVideoMeta
11942 https://bugzilla.gnome.org/show_bug.cgi?id=774782
11944 2016-11-21 12:36:27 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11946 * gst/vaapi/gstvaapivideomemory.c:
11947 vaapivideomemory: add missing documentation
11948 https://bugzilla.gnome.org/show_bug.cgi?id=774782
11950 2016-11-21 12:29:26 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11952 * gst/vaapi/gstvaapipluginbase.c:
11953 plugins: use early return without goto
11954 https://bugzilla.gnome.org/show_bug.cgi?id=774782
11956 2016-11-21 11:25:21 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11958 * gst/vaapi/gstvaapipluginbase.c:
11959 * gst/vaapi/gstvaapivideomemory.c:
11960 * gst/vaapi/gstvaapivideomemory.h:
11961 plugins: add allocator to allocation query
11962 This patch adds the created allocator to the allocation query either
11963 in decide_allocation() and propose_allocation() vmehtods.
11964 With it, there's no need to set the modified allocator's size in the
11965 pool configuration.
11966 https://bugzilla.gnome.org/show_bug.cgi?id=774782
11968 2016-10-19 15:15:01 +0100 Julien Isorce <j.isorce@samsung.com>
11970 * gst/vaapi/gstvaapivideobufferpool.c:
11971 vaapivideobufferpool: set correct buffer size
11972 We should set the correct buffer size when we are configuring the pool,
11973 otherwise the buffer will be discarded when it returns to the pool.
11974 Indeed when the ref-count of a buffer reaches zero, its pool will queue
11975 it back (and ref it) if, and only if, the buffer size matches the
11976 configured buffer size on the pool.
11977 This issue can be debugged with GST_DEBUG=*PERF*:6, see gstbufferpool.c
11978 https://bugzilla.gnome.org/show_bug.cgi?id=774782
11980 2016-11-10 13:26:31 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11982 * gst/vaapi/gstvaapivideomemory.c:
11983 vaapivideomemory: check for memory allocator
11984 When calling gst_vaapi_video_memory_copy() the allocator of the memory
11985 to copy should be allocated by the vaapi allocator.
11986 This patch does this verification.
11988 2016-11-10 13:25:30 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11990 * gst/vaapi/gstvaapivideomemory.c:
11991 vaapivideomemory: code style fixes
11992 A cosmetic commit for enhance readability of the casts and method
11995 2016-11-09 19:49:22 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
11997 * gst/vaapi/gstvaapivideomemory.c:
11998 vaapivideomemory: unroll gst_vaapi_video_allocator_free()
11999 Instead of having a gst_vaapi_video_memory_free() that is only going to
12000 be called by gst_vaapi_video_allocator_free(), let's just remove the first
12001 and merged into the second.
12003 2016-11-09 19:29:12 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12005 * gst/vaapi/gstvaapivideomemory.c:
12006 vaapivideomemory: avoid virtual methods casting
12007 Use the expected virtual method signatures for readability.
12009 2016-11-09 18:58:20 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12011 * gst/vaapi/gstvaapivideomemory.c:
12012 vaapivideomemory: remove unimplemented methods
12013 Remove unimplemented method for allocator mem_share() and mem_is_span().
12015 2016-11-09 18:54:47 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12017 * gst/vaapi/gstvaapivideomemory.c:
12018 vaapivideomemory: fail if frame map can't get plane
12019 If map() vmethod in GstVideMeta cannot get the plane data, return false,
12020 thus the caller will not try to read invalid memory.
12021 https://bugzilla.gnome.org/show_bug.cgi?id=774213
12023 2016-11-09 18:39:06 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12025 * gst/vaapi/gstvaapivideomemory.c:
12026 * gst/vaapi/gstvaapivideomemory.h:
12027 vaapivideomemory: lock map and unmap operations
12028 In order to avoid race condition when two threads call map/unmap the same
12029 VA surface, this patch mutex these operations.
12030 https://bugzilla.gnome.org/show_bug.cgi?id=774213
12032 2016-11-09 17:37:06 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12034 * gst/vaapi/gstvaapivideomemory.c:
12035 vaapivideomemory: refactor vaapi memory unmapping
12036 There were duplicated code in gst_video_meta_unmap_vaapi_memory() and
12037 gst_vaapi_video_memory_unmap() when unmapping.
12038 This patch refactors both methods adding the common function
12039 unmap_vaapi_memory(). This also ensures, if direct rendering is enabled, it
12040 is correctly reset.
12041 Additionally, only when mapping flag has the WRITE bit, it set the image as
12042 current, which was done in gst_video_meta_map_vaapi_memory() but no in
12043 gst_vaapi_video_memory_map().
12044 In order to make this, the mapping flags were required, so instead of
12045 overloading mem_unmap() virtual function, mem_unmap_full() is overloaded.
12046 https://bugzilla.gnome.org/show_bug.cgi?id=774213
12048 2016-11-09 13:54:23 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12050 * gst/vaapi/gstvaapivideomemory.c:
12051 vaapivideomemory: refactor vaapi memory mapping
12052 There were duplicated code in gst_video_meta_map_vaapi_memory() and
12053 gst_vaapi_video_memory_map() when doing the READ and WRITE mapping.
12054 This patch refactors both methods adding the common function
12055 map_vaapi_memory().
12056 Additionally, only when flag has the READ bit it calls
12057 ensure_images_is_current(), which was done in
12058 gst_video_meta_map_vaapi_memory() but no in
12059 gst_vaapi_video_memory_map().
12060 https://bugzilla.gnome.org/show_bug.cgi?id=772151
12062 2016-10-27 18:22:18 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12064 * gst/vaapi/gstvaapidecode.c:
12065 vaapidecode: unref allowed_sinkpad_caps at close()
12066 The variable member allowed_sinkpad_caps is constructed querying the
12067 current VA display. Bearing that in mind, the variable shall be freed
12068 when the VA display changes or is removed.
12069 This patch moves the freeing of allowed_sinkpad_caps to close(), when
12070 the VA display is freed.
12072 2016-11-11 11:40:09 +0900 Hyunjun Ko <zzoon@igalia.com>
12074 * gst/vaapi/gstvaapisink.c:
12075 vaapisink: finish event thread at stop()
12076 The thread that handles window's events should be finished during
12077 pipeline's shutdown, otherwise it will remain alive during pipeline
12078 re-activation, leading to unexpected problems.
12079 This patch fixes failures of intensive_state_change scenario of
12081 https://bugzilla.gnome.org/show_bug.cgi?id=774241
12083 2016-11-08 09:35:00 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12085 * gst/vaapi/gstvaapipostproc.c:
12086 vaapipostproc: enhance debug message
12087 "gst_pad_push" is not a good description of the event.
12089 2016-11-08 10:05:32 +0900 Hyunjun Ko <zzoon@igalia.com>
12091 * gst/vaapi/gstvaapipostproc.c:
12092 postproc: honor gst_pad_push() return value
12093 Returning GST_FLOW_ERROR always when gst_pad_push fails might lead to
12094 deadlock during seek.
12095 This patch returns the same error of gst_pad_push() and log out the
12097 https://bugzilla.gnome.org/show_bug.cgi?id=774030
12099 2016-11-04 16:26:18 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12101 * gst/vaapi/gstvaapidecode.c:
12102 vaapidecode: guard GST_VAAPI_MAKE_GLTEXUPLOAD_CAPS
12103 In commit 6d11a00 were introduced a regression when gstreamer-vaapi is
12104 compiled with out EGL/GLX support: it shall not support
12105 GST_VAAPI_MAKE_GLTEXUPLOAD_CAPS.
12106 This patch guards the inclusion of GST_VAAPI_MAKE_GLTEXUPLOAD_CAPS in the
12107 allowed src caps for vaapedecode if EGL/GLX.
12109 2016-11-04 12:55:23 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12111 * gst/vaapi/gstvaapivideomemory.c:
12112 vaapivideomemory: increment map counter only if succeeded
12113 Previously the frame map counter increased independently if the map succeeded
12114 or not. This leaded to critical messages and crashes if the frame was unable
12115 to be mapped, but the counter increased.
12116 This patch increases the map counter only if the map operation occurred.
12117 https://bugzilla.gnome.org/show_bug.cgi?id=773939
12119 2016-11-03 17:30:46 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12121 * gst/vaapi/gstvaapipluginbase.c:
12122 plugins: set negotiation caps in src allocator
12123 When the allocator is created, it stores the allocation caps. But sometimes
12124 the "allocation caps" may be different from the "negotiation caps".
12125 In this case, the allocator should store the negotiation caps since they
12126 are the ones used for frame mapping with GstVideoMeta.
12127 When vaapispostproc is used, this is not a problem since the element is assume
12128 to resize. But when using a vaapi decoder only, with a software renderer, it
12129 fails in this case.
12130 https://bugzilla.gnome.org/show_bug.cgi?id=773323
12132 2016-07-19 16:40:10 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12134 * gst/vaapi/gstvaapidecode.c:
12135 * gst/vaapi/gstvaapidecode.h:
12136 vaapidecode: generate source pad caps
12137 Just as vaapipostproc, VA decoder's context can be queried to get the possible
12138 raw formats, so, the src caps can negotiate the exact caps that the context
12141 2016-02-25 18:57:30 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12143 * gst-libs/gst/vaapi/gstvaapidecoder.c:
12144 * gst-libs/gst/vaapi/gstvaapidecoder.h:
12145 libs: decoder: add _get_surface_formats()
12146 This function exposes the available formats of the surfaces in the the current
12147 context to the plugins.
12149 2016-02-18 19:32:58 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12151 * gst-libs/gst/vaapi/gstvaapicontext.c:
12152 * gst-libs/gst/vaapi/gstvaapicontext.h:
12153 libs: context: ensure context formats
12154 This patch ensures to get the formats, as filter does, available in the
12155 decoder / encoder context.
12156 The context fills up the array as soon it is created, otherwise the pipeline
12157 could get stalled (perhaps this is a bug in my HSW backend).
12158 https://bugzilla.gnome.org/show_bug.cgi?id=752958
12160 2016-02-18 19:20:10 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12162 * gst-libs/gst/vaapi/gstvaapifilter.c:
12163 * gst-libs/gst/vaapi/gstvaapiutils_core.c:
12164 * gst-libs/gst/vaapi/gstvaapiutils_core.h:
12165 libs: move get_surface_formats to utils_core
12166 The query of all the supported formats for a VA config were only used by the
12167 postprocessor (vaapifilter). But, in order to enable the vaapidecoder to
12168 negotiate a suitable raw format with downstream, we need to query these
12169 formats against the decoder's config.
12170 This patch is the first step: moves the code in filter's ensure_image() to a
12171 generic gst_vaapi_get_surface_formats() in vaapiutils_core, so it can be
12172 shared later by the decoder.
12173 https://bugzilla.gnome.org/show_bug.cgi?id=752958
12175 2016-07-19 18:56:15 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12177 * gst/vaapi/gstvaapipluginbase.c:
12178 plugins: remove set_sinkpad_dmabuf_allocator()
12179 Since when the sink pad allocator is created, it is decided if the required
12180 one is vaapi allocator or dmabuf allocator, there is no need to force its set
12183 2016-10-27 11:25:55 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12185 * gst/vaapi/gstvaapipluginbase.c:
12186 plugins: ensure display when getting raw caps
12187 When running gst-discoverer-1.0, in certain media, vaapipostroc is stopped
12188 meanwhile it is transforming caps. The problem is that stop() calls
12189 gst_vaapi_plugin_base_close(), which nullifies the element's va display, but
12190 the va display is used in tranform_caps() when it is extracting the possible
12191 format conversions. This display disappearing generates warning messages.
12192 This patch holds a local reference of va display at ensure_allowed_raw_caps()
12193 hence it doesn't go away meanwhile it is used, even if the
12194 gst_vaapi_plugin_base_close() is called in other thread.
12195 https://bugzilla.gnome.org/show_bug.cgi?id=773593
12197 2016-11-03 12:54:23 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12199 * gst/vaapi/gstvaapiencode_h264.c:
12200 * gst/vaapi/gstvaapiencode_h265.c:
12201 plugins: fix code style for errors
12203 2016-10-19 19:04:20 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12205 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
12206 * gst-libs/gst/vaapi/gstvaapicodedbuffer.c:
12207 * gst-libs/gst/vaapi/gstvaapicodedbufferproxy.c:
12208 * gst-libs/gst/vaapi/gstvaapicontext.c:
12209 * gst-libs/gst/vaapi/gstvaapicontext_overlay.c:
12210 * gst-libs/gst/vaapi/gstvaapidecoder.c:
12211 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
12212 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
12213 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
12214 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
12215 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
12216 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
12217 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
12218 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
12219 * gst-libs/gst/vaapi/gstvaapidisplay.c:
12220 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
12221 * gst-libs/gst/vaapi/gstvaapiencoder.c:
12222 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
12223 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
12224 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
12225 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
12226 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
12227 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
12228 * gst-libs/gst/vaapi/gstvaapifilter.c:
12229 * gst-libs/gst/vaapi/gstvaapiimage.c:
12230 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
12231 * gst-libs/gst/vaapi/gstvaapiparser_frame.c:
12232 * gst-libs/gst/vaapi/gstvaapipixmap.c:
12233 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
12234 * gst-libs/gst/vaapi/gstvaapisurface.c:
12235 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
12236 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
12237 * gst-libs/gst/vaapi/gstvaapitexture.c:
12238 * gst-libs/gst/vaapi/gstvaapiutils.c:
12239 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
12240 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
12241 * gst-libs/gst/vaapi/gstvaapiwindow.c:
12242 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
12243 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
12244 libs: fix code style for errors
12246 2016-11-03 09:31:17 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12248 * gst/vaapi/gstvaapipluginbase.c:
12249 * gst/vaapi/gstvaapivideocontext.c:
12250 plugins: update GstGL deprecated symbol
12251 GST_GL_TYPE_CONTEXT was deprecated. Now it is GST_TYPE_GL_CONTEXT.
12253 2016-10-21 11:48:54 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12255 * gst/vaapi/gstvaapipluginbase.c:
12256 plugins: direct render when raw video
12257 Enable the direct rendering with linear surfaces if the negotiated src caps
12258 are video/x-raw without features.
12259 Pass also the caps, since they are needed to know the requested caps features.
12261 2016-10-24 20:09:59 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12263 * gst/vaapi/gstvaapipluginbase.c:
12264 plugins: move src allocator error to instantiator
12265 Just as we did in ensure_sinkpad_allocator(), let's move the error message
12266 into the ensure_srcpad_allocator() from the caller,
12267 gst_vaapi_plugin_base_decide_allocation()
12269 2016-10-20 19:37:01 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12271 * gst/vaapi/gstvaapipluginbase.c:
12272 plugins: enable direct upload if raw video
12273 Enable the direct upload with linear surfaces if the negotiated sink caps are
12274 video/x-raw without features.
12276 2016-10-21 11:21:04 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12278 * gst/vaapi/gstvaapipluginutil.c:
12279 * gst/vaapi/gstvaapipluginutil.h:
12280 pluginutil: add gst_caps_is_video_raw()
12282 2016-10-24 19:25:27 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12284 * gst/vaapi/gstvaapipluginbase.c:
12285 plugins: receive caps in ensure_sinkpad_allocator()
12286 Instead of receiving the GstVideoInfo structure as parameter, get the original
12287 GstCaps from ensure_sinkpad_buffer_pool(), in this way we could decide better
12288 which allocator instantiate.
12290 2016-10-20 19:31:58 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12292 * gst/vaapi/gstvaapivideomemory.c:
12293 vaapivideomemory: destroy derived image at unmap
12294 If the allocator was configured to use direct upload or rendering, the
12295 generated derived image created at mapping needs to be destroyed after
12296 unmapping, because, in order to process the surface, it should not be marked
12297 as "busy" by the driver.
12299 2016-10-21 11:57:55 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12301 * gst/vaapi/gstvaapivideomemory.c:
12302 vaapivideomemory: enhance logs for direct modes
12303 Print, conditionally, only the enabled direct mode.
12305 2016-10-20 17:02:49 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12307 * gst/vaapi/gstvaapivideomemory.c:
12308 * gst/vaapi/gstvaapivideomemory.h:
12309 vaapivideomemory: add direct upload flag
12310 Adds the direct-upload flag in the GstVaapiVideoAllocator and
12311 GstVaapiVideoMemory.
12312 It still doesn't apply any functional change.
12314 2016-10-20 16:49:22 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12316 * gst/vaapi/gstvaapipluginbase.c:
12317 * gst/vaapi/gstvaapivideomemory.c:
12318 * gst/vaapi/gstvaapivideomemory.h:
12319 vaapivideomemory: set direct rendering at run-time
12320 The way to experiment with the direct rendering is through and internal
12321 compiler pre-processor flag.
12322 The current change set enables a way to specified at run-time, as a flag
12323 passed to the allocator at instanciation time.
12325 2016-10-20 18:09:59 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12327 * gst/vaapi/gstvaapivideomemory.c:
12328 vaapivideomemory: log in perf category when copy
12329 Log in performance category when the derive image handling fails, falling back
12332 2016-10-20 16:31:21 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12334 * gst/vaapi/gstvaapivideomemory.c:
12335 vaapivideomemory: error log is derive image fails
12336 Instead of a silently failure of the derive image, this patch log an error
12337 message according to the failure.
12339 2016-10-20 12:52:18 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12341 * gst/vaapi/gstvaapivideomemory.c:
12342 vaapivideomemory: store surface alloc flags in qdata
12343 For sake of consistency, we should add the requested surface allocation flags
12344 to the object's qdata structure.
12346 2016-10-20 12:22:06 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12348 * gst/vaapi/gstvaapivideomemory.c:
12349 vaapivideomemory: category init when object define
12350 Move the Gstreamer debug category initialize to the GObject definition.
12352 2016-11-03 08:31:16 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12354 * gst-libs/gst/vaapi/gstvaapitexturemap.c:
12355 libs: vaapitexturemap: trivial code-style fix
12357 2016-11-02 20:01:09 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12359 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
12360 libs: display: egl: avoid recreate native display
12361 Instead of passing the native descriptor of the display, just pass the received
12362 GstVaapiDisplay and reuse it.
12364 2016-11-02 15:38:52 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12366 * gst/vaapi/gstvaapipluginbase.c:
12367 plugins: log the GstVaapiDisplay name
12368 Now that GstVaapiDisplay is descendant of GstObject, it has a human-friendly
12369 name. Log it instead of the memory address.
12371 2016-11-02 18:37:00 +0900 Hyunjun Ko <zzoon@igalia.com>
12373 * gst-libs/gst/vaapi/gstvaapiwindow_egl.c:
12374 libs: window: egl: pass native va display
12375 When creating a GstVaapiWindowEGL, it also creates native window by its own
12376 native display. It should pass the native display, either X11 or Wayland.
12377 https://bugzilla.gnome.org/show_bug.cgi?id=768266
12379 2016-10-13 12:53:17 +0900 Hyunjun Ko <zzoon@igalia.com>
12381 * gst-libs/gst/vaapi/gstvaapidebug.h:
12382 * gst-libs/gst/vaapi/gstvaapidisplay.c:
12383 * gst-libs/gst/vaapi/gstvaapidisplay.h:
12384 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
12385 * gst-libs/gst/vaapi/gstvaapidisplay_drm.h:
12386 * gst-libs/gst/vaapi/gstvaapidisplay_drm_priv.h:
12387 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
12388 * gst-libs/gst/vaapi/gstvaapidisplay_egl.h:
12389 * gst-libs/gst/vaapi/gstvaapidisplay_egl_priv.h:
12390 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
12391 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
12392 * gst-libs/gst/vaapi/gstvaapidisplay_glx_priv.h:
12393 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
12394 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
12395 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.h:
12396 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
12397 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
12398 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
12399 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
12400 * gst-libs/gst/vaapi/gstvaapitexture_egl.c:
12401 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
12402 * gst-libs/gst/vaapi/gstvaapiutils_egl.h:
12403 * gst/vaapi/gstvaapivideocontext.c:
12404 libs: display: GstVaapiDisplay as GstObject descendant
12405 This patch is to change the inheritance of GstVaapiDisplay to GstObject,
12406 instead of GstVaapiMiniObject. In this way we can use all the available
12407 infrastructure for GObject/GstObject such as GstTracer, GIR, etc.
12408 In addition, a new debug category for GstVaapiDisplay is created to make it
12409 easier to trace debug messages. It is named "vaapidisplay" and it transverse
12410 all the VA display backends (DRM, GLX, EGL, Wayland, ...)
12411 This patch is a step forward to expose GstVaapiDisplay for users in a future
12413 https://bugzilla.gnome.org/show_bug.cgi?id=768266
12414 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12416 === release 1.11.0 ===
12418 2016-11-01 18:54:54 +0200 Sebastian Dröge <sebastian@centricular.com>
12421 Back to development
12423 === release 1.10.0 ===
12425 2016-11-01 18:19:32 +0200 Sebastian Dröge <sebastian@centricular.com>
12430 * gstreamer-vaapi.doap:
12433 2016-10-27 17:13:48 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12435 * gst/vaapi/gstvaapidecodebin.c:
12436 vaapidecodebin: resurrect disable-vpp property
12437 https://bugzilla.gnome.org/show_bug.cgi?id=773589
12439 2016-10-27 16:32:23 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12441 * gst/vaapi/gstvaapidecodebin.c:
12442 vaapidecodebin: name the internal queue
12443 https://bugzilla.gnome.org/show_bug.cgi?id=773589
12445 2016-10-27 16:27:45 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12447 * gst/vaapi/gstvaapidecodebin.c:
12448 * gst/vaapi/gstvaapidecodebin.h:
12449 vaapidecodebin: remove unused variables
12450 Since vaapipostproc is only registered if the driver supports it, all the
12451 support for dynamic loading were removed. Though some leftovers remained.
12452 https://bugzilla.gnome.org/show_bug.cgi?id=773589
12454 2016-10-27 12:53:54 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12456 * docs/plugins/Makefile.am:
12457 * docs/plugins/gstreamer-vaapi-plugins-docs.xml.in:
12458 * docs/plugins/running.xml:
12459 docs: document environment variables
12460 https://bugzilla.gnome.org/show_bug.cgi?id=773544
12462 2016-10-27 12:31:49 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12464 * docs/plugins/gstreamer-vaapi-plugins-docs.xml.in:
12465 * docs/plugins/gstreamer-vaapi-plugins-sections.txt:
12466 * gst/vaapi/Makefile.am:
12467 * gst/vaapi/gstvaapidecode.c:
12468 * gst/vaapi/gstvaapidecodebin.c:
12469 * gst/vaapi/gstvaapidecodedoc.c:
12470 docs: replace vaapidecode with each codec
12471 In the spirit of the codec split, this patch removes the documentation of
12472 vaapidecode and adds a page per each possible decoder.
12473 Nonetheless, only those available in the compilation system are going to be
12474 instrospected, because the rest are not registered.
12476 2016-10-27 11:06:06 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12478 * gst/vaapi/gstvaapipluginutil.c:
12479 * gst/vaapi/gstvaapivideobuffer.c:
12480 * gst/vaapi/gstvaapivideometa.c:
12481 docs: add missing long descriptions
12483 2016-10-25 14:32:44 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12485 * gst/vaapi/gstvaapipostproc.c:
12486 vaapipostproc: use GST_*_OBJECT when possible
12487 Since we can have several vaapipostproc operating in a pipeline, it is useful
12488 to know which one is generating the logging message.
12489 https://bugzilla.gnome.org/show_bug.cgi?id=773497
12491 2016-07-19 17:00:23 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12493 * gst/vaapi/gstvaapidecode.c:
12494 * gst/vaapi/gstvaapidecode.h:
12495 vaapidecode: rename member to allowed_sinkpad_caps
12496 vaapidecode has a member named allowed_caps, but this name is not enough
12497 explicit. This patch renames allowed_caps to allowed_sinkpad_caps.
12498 No functional changes were included.
12499 https://bugzilla.gnome.org/show_bug.cgi?id=773497
12501 2016-10-20 18:12:04 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12503 * gst/vaapi/gstvaapiencode_h264.c:
12504 * gst/vaapi/gstvaapiencode_h265.c:
12505 * gst/vaapi/gstvaapisink.c:
12506 * gst/vaapi/gstvaapivideomemory.c:
12507 * gst/vaapi/gstvaapivideometa.c:
12508 * gst/vaapi/gstvaapivideometa_texture.c:
12509 plugins: fix code style for errors
12510 https://bugzilla.gnome.org/show_bug.cgi?id=773497
12512 2016-10-20 17:01:57 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12514 * gst/vaapi/gstvaapivideomemory.c:
12515 vaapivideomemory: comment style
12516 https://bugzilla.gnome.org/show_bug.cgi?id=773497
12518 2016-10-20 11:19:50 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12520 * gst/vaapi/gstvaapivideomemory.c:
12521 * gst/vaapi/gstvaapivideomemory.h:
12522 vaapivideomemory: rename input parameter
12523 In order to clarify the use of flag as input parameter, it is renamed to
12524 surface_alloc_flag, since it is used when creating a VA surface with certain
12526 https://bugzilla.gnome.org/show_bug.cgi?id=773497
12528 2016-10-25 19:22:03 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12530 * gst/vaapi/gstvaapidecode.c:
12531 vaapidecode: rename element description
12532 So encoders and decoders have similar descriptions.
12533 https://bugzilla.gnome.org/show_bug.cgi?id=773497
12535 2016-08-02 11:32:19 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12537 * gst/vaapi/gstvaapiencode_h264.c:
12538 * gst/vaapi/gstvaapiencode_h265.c:
12539 vaapiencode: h264, h265: rename codec name
12540 So encoder and decoders have the same codec name.
12541 https://bugzilla.gnome.org/show_bug.cgi?id=773497
12543 2016-07-29 15:17:33 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12545 * gst/vaapi/gstvaapipluginbase.c:
12546 plugins: simplify code
12547 Merge two lines of variable declarations.
12548 https://bugzilla.gnome.org/show_bug.cgi?id=773497
12550 2016-10-07 18:46:22 +0900 Hyunjun Ko <zzoon@igalia.com>
12552 * gst-libs/gst/vaapi/gstvaapidecoder.c:
12553 * gst-libs/gst/vaapi/gstvaapipixmap_priv.h:
12554 * gst-libs/gst/vaapi/gstvaapitexture_priv.h:
12555 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
12556 libs: minor correction for logical consistency
12557 GstVaapiDecode is a descendant of GstVaapiMiniObject, so, thought we should
12558 use its methods, even though it doesn't change functionality.
12559 GstVaapiPixmap, GstVaapiTexture and GstVaapiWindow are descendant of
12560 GstVaapiObject, hence its methods shall be used.
12561 https://bugzilla.gnome.org/show_bug.cgi?id=772554
12563 2016-10-19 15:39:54 +0100 Julien Isorce <j.isorce@samsung.com>
12565 * gst/vaapi/gstvaapivideomemory.c:
12566 vaapivideomemory: add explanation about the call 'dup (dmabuf_fd)'
12567 In short GstFdMemory is configured to call close when using
12569 https://bugzilla.gnome.org/show_bug.cgi?id=755072
12571 2016-09-02 16:42:45 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12574 * gst-libs/gst/vaapi/Makefile.am:
12575 build: clean up the dlopen usage
12577 2016-10-08 14:33:59 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12579 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
12580 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
12581 encoder: h264,h265: fix regression in offset count
12582 In commit dc35dafa a bug was introduced because I assumed that
12583 GST_CLOCK_TIME_NONE is zero when is -1. This patch fixes that mistake.
12584 https://bugzilla.gnome.org/show_bug.cgi?id=772259
12586 2016-10-18 17:02:59 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12589 docs: update README
12591 2016-09-27 17:29:25 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12593 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
12594 libs: display: egl: remove unused header include
12595 The header gmodule.h is not used since the library dynamic loading for EGL
12596 display was removed.
12597 https://bugzilla.gnome.org/show_bug.cgi?id=772599
12599 === release 1.9.90 ===
12601 2016-09-30 13:05:20 +0300 Sebastian Dröge <sebastian@centricular.com>
12607 * gstreamer-vaapi.doap:
12610 2016-03-04 16:35:11 +0900 Vineeth TM <vineeth.tm@samsung.com>
12612 * gst/vaapi/gstvaapidecode.c:
12613 * gst/vaapi/gstvaapidecodebin.c:
12614 * gst/vaapi/gstvaapiencode_h264.c:
12615 * gst/vaapi/gstvaapiencode_h265.c:
12616 * gst/vaapi/gstvaapiencode_jpeg.c:
12617 * gst/vaapi/gstvaapiencode_mpeg2.c:
12618 * gst/vaapi/gstvaapiencode_vp8.c:
12619 * gst/vaapi/gstvaapipostproc.c:
12620 * gst/vaapi/gstvaapisink.c:
12621 vaapi: use new gst_element_class_add_static_pad_template()
12622 https://bugzilla.gnome.org/show_bug.cgi?id=763083
12623 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12625 2016-09-22 16:34:48 +0900 Hyunjun Ko <zzoon@igalia.com>
12627 * gst/vaapi/gstvaapipluginbase.c:
12628 plugins: reset textures at negotiation/shutdown
12629 When caps reconfiguration is called, the new downstream frame size might be
12630 different. Thus, if the downstream caps change,the display's texture map is
12632 In addition, during pipeline shutdown, textures in texture map have to be
12633 released, since each one have a reference to the GstVaapiDisplay object, which
12634 is a dangerous circular reference.
12635 https://bugzilla.gnome.org/show_bug.cgi?id=769293
12636 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12638 2016-09-22 16:34:38 +0900 Hyunjun Ko <zzoon@igalia.com>
12640 * gst-libs/gst/vaapi/gstvaapidisplay.c:
12641 * gst-libs/gst/vaapi/gstvaapidisplay.h:
12642 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
12643 * gst-libs/gst/vaapi/gstvaapidisplay_egl_priv.h:
12644 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
12645 * gst-libs/gst/vaapi/gstvaapidisplay_glx_priv.h:
12646 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
12647 * gst/vaapi/gstvaapivideometa_texture.c:
12648 libs: display{egl,glx}: cache GstVaapiTextures
12649 instances when created and reuse
12650 This patch improves performance when glimagesink uploads a GL texture.
12651 It caches the GStVaapiTexture instances in GstVaapiDisplay{GLX,EGL}, using an
12652 instance of GstVaapiTextureMap, so our internal texture structure can be found
12653 by matching the GL texture id for each frame upload process, avoiding the
12654 internal texture structure creation and its following destruction.
12655 https://bugzilla.gnome.org/show_bug.cgi?id=769293
12656 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12658 2016-09-22 16:33:06 +0900 Hyunjun Ko <zzoon@igalia.com>
12660 * gst-libs/gst/vaapi/Makefile.am:
12661 * gst-libs/gst/vaapi/gstvaapitexturemap.c:
12662 * gst-libs/gst/vaapi/gstvaapitexturemap.h:
12663 libs: vaapitexturemap: implement GstVaapiTextureMap
12664 Implement GstVaapiTextureMap object, which caches VAAPI textures, so them can be
12665 reused. Internally it is a hash table.
12666 Note that it is GstObject based rather than GstVaapiObject, as part of the future
12667 converstion to GstObject of most of the code.
12668 https://bugzilla.gnome.org/show_bug.cgi?id=769293
12669 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12671 2016-09-21 09:55:53 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
12673 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
12674 encoder: vp8: Increase the allocation size for coded buffer
12675 We are not getting enough compression for some streams and
12676 encoded frame end up with more size than allocated.
12677 Assuming a compression ratio of 4, which should be good enough
12678 for holding the frames.
12679 https://bugzilla.gnome.org/show_bug.cgi?id=771528
12681 2016-09-21 09:52:21 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
12683 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
12684 encoder: vp9: Fix refresh frame flag setting
12685 While doing the mode-1 referece picture selection,
12686 the circular buffer logic was not correctly setting the
12687 refresh frame flags as per VP9 spec.
12688 Make sure refresh_flag[0] get updated correclty after
12689 each cycle of GST_VP9_REF_FRAMES.
12690 https://bugzilla.gnome.org/show_bug.cgi?id=771507
12692 2016-09-14 18:42:09 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12694 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
12695 vaapidecode: codec_data minimal size is 7
12696 When the format of a H.264 stream is AVC3, the SPS and PPS are inside the
12697 stream, not in the codec_data, so the size of codec_data might be 7.
12698 This patch reduces the minimal size of the codec_data buffer from 8 to 7.
12699 https://bugzilla.gnome.org/show_bug.cgi?id=771441
12701 2016-09-14 16:29:01 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12703 * gst/vaapi/gstvaapidecode.c:
12704 vaapidecode: reset decoder hard when set_format()
12705 set_format() is called by upstream when the stream capabilites has changed.
12706 Before, if the new stream is compatible with the old one the VA decoder was
12707 not destroyed. Nonetheless, with this behavoir, the VA decoder ignores
12708 when the upstreamer parsers gets more details of the stream, such as the
12709 framerate. Hence, when the src caps are negotiates, the further sink caps
12710 updates are ignored.
12711 This patch forces the VA decoder destroying and recreation when set_format()
12713 https://bugzilla.gnome.org/show_bug.cgi?id=770921
12715 2016-09-14 11:31:39 +0200 Sebastian Dröge <sebastian@centricular.com>
12718 configure: Depend on gstreamer 1.9.2.1
12720 2016-09-09 12:03:37 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12722 * gst/vaapi/gstvaapivideomemory.c:
12723 Revert "vaapivideomemory: load VA Image when mapping to write"
12724 This reverts commit c67edea4aba35f16d9e97c78a0b49ad1b590b112.
12726 2016-09-10 20:52:21 +1000 Jan Schmidt <jan@centricular.com>
12729 Automatic update of common submodule
12730 From b18d820 to f980fd9
12732 2016-09-10 09:58:25 +1000 Jan Schmidt <jan@centricular.com>
12735 Automatic update of common submodule
12736 From f49c55e to b18d820
12738 2016-09-08 16:16:09 +0900 Hyunjun Ko <zzoon@igalia.com>
12740 * gst/vaapi/gstvaapipluginbase.c:
12741 plugins: set allocator's image size to sinkpad bufferpool
12742 Otherwise the buffer is always ditched by the bufferpool, losing performance.
12743 https://bugzilla.gnome.org/show_bug.cgi?id=771035
12745 2016-09-07 17:34:08 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12747 * gst-libs/gst/vaapi/gstvaapisurface.c:
12748 libs: surface: ensure composite overlay is not bigger
12749 Ensure the composition overlay rectangle (subtitles) is not bigger than
12750 the surface where it is going to be composited and rendered.
12751 https://bugzilla.gnome.org/show_bug.cgi?id=766978
12753 2016-09-07 17:51:23 +0900 Hyunjun Ko <zzoon@igalia.com>
12755 * gst/vaapi/gstvaapivideomemory.c:
12756 vaapivideomemory: load VA Image when mapping to write
12757 When calling gst_video_frame_map() with GST_MAP_WRITE flag, it doesn't call
12758 ensure_image_is_current(), which means it doesn't guarentee VAImage is valid
12760 https://bugzilla.gnome.org/show_bug.cgi?id=766978
12762 2016-09-06 12:27:45 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12764 * gst/vaapi/gstvaapidecode.c:
12765 vaapidecode: merge vc1 and wmv3 elements
12766 This patch merges vaapivc1dec and vaapiwmv3dec into a single
12767 vaapivc1dec. Also, removed the WMVA format, since it is not
12768 supported by libva.
12769 https://bugzilla.gnome.org/show_bug.cgi?id=734093
12771 2016-09-06 11:19:05 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12773 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
12774 decoder: vc1: fails only on advanced profile
12775 In commit 2eb4394 the frame coding mode was verified for progressive
12776 regardless the profile. But the FCM is only valid in the advanced
12777 profile. This patch checks for the advanced profile before verifying FCM for
12779 https://bugzilla.gnome.org/show_bug.cgi?id=769250
12781 2016-09-01 12:39:15 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12783 * gst/vaapi/gstvaapi.c:
12784 vaapi: DISPLAY envvar as dependency
12785 In a multiple video cards system, a X11 environment may have different VA
12786 capabilities. This patch tracks the DISPLAY environment variable to
12787 invalidates the GStreamer features cache. Also tracks WAYLAND_DISPLAY.
12788 https://bugzilla.gnome.org/show_bug.cgi?id=770357
12790 2016-08-26 14:55:17 -0700 Scott D Phillips <scott.d.phillips@intel.com>
12792 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
12793 decoder: vc1: Fail only on actual interlaced frames
12794 In the earlier patch:
12795 f31d9f3 decoder: vc1: Print error on interlaced content
12796 Decoding would error out if the interlace flag was set in the
12797 sequence bdu. This isn't quite right because a video can have this
12798 flag set and yet not have any interlaced pictures.
12799 Here instead we error out when either parsing a field bdu or
12800 decoding a frame bdu which has fcm set to anything other than
12802 Signed-off-by: Scott D Phillips <scott.d.phillips@intel.com>
12803 https://bugzilla.gnome.org/show_bug.cgi?id=769250
12805 2016-09-01 12:34:48 +0300 Sebastian Dröge <sebastian@centricular.com>
12808 Back to development
12810 === release 1.9.2 ===
12812 2016-09-01 12:34:38 +0300 Sebastian Dröge <sebastian@centricular.com>
12817 * gstreamer-vaapi.doap:
12820 2016-08-16 11:58:38 +0300 Scott D Phillips <scott.d.phillips@intel.com>
12822 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
12823 decoder: vc1: Print error on interlaced content
12824 Interlaced video is as yet unsupported in the vc1 element. Print
12825 an error to make that more obvious.
12826 https://bugzilla.gnome.org/show_bug.cgi?id=769250
12828 2016-08-10 13:29:45 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12830 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
12831 encoder: h264: fix C90 mixed declarations and code
12832 Commit 4259d1a introduced this compilation error. This patch fixes it.
12834 2016-07-21 17:38:40 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12836 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
12837 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
12838 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
12839 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
12840 encoder: h264,h265,mpeg2,vp8: use gst_util_uint64_scale() for bitrate
12841 Use gst_util_uint64_scale() to calculate bitrate instead of normal arithmetic
12842 to avoid overflows, underflows and loss of precision.
12843 https://bugzilla.gnome.org/show_bug.cgi?id=768458
12845 2016-07-05 20:07:15 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12847 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
12848 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
12849 vaapiencode: h264,h265: validate fps numerator
12850 Validate that fps numerator is non-zero so it can be used to calculate
12851 the duration of the B frame.
12852 Also it gst_util_uint64_scale() is used instead of normal arithmetic in
12853 order to aviod overflows, underflows and loss of precision.
12854 https://bugzilla.gnome.org/show_bug.cgi?id=768458
12856 2016-08-06 12:54:17 +0100 Tim-Philipp Müller <tim@centricular.com>
12858 * gst/vaapi/gstvaapi.c:
12859 encoders: demote to RANK_NONE since not fit for autoplugging yet
12860 Encoders claim to support a whole bunch of input formats but then
12861 just error out if the format is not actually supported, even if
12862 there's a converter in front. This means they're not fit for
12863 autoplugging in encodebin or camerabin yet and therefore should
12864 not have a rank. People can still use them in custom pipelines.
12865 https://bugzilla.gnome.org/show_bug.cgi?id=769266
12867 2016-07-19 19:24:08 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12869 * gst/vaapi/gstvaapipluginbase.c:
12870 plugins: check dmabuf-import for sink pad allocator
12871 Check earlier if upstream video source has activated the dmabuf-import
12872 io-mode (hack to disappear soon), thus we can avoid the re-assignation of a
12875 2016-07-19 20:02:54 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12877 * gst/vaapi/gstvaapipluginbase.c:
12878 plugins: reset allocators if video info changed
12879 If the frame size or format, change, the allocators are reset, so a new ones
12880 can be created with the new video info.
12882 2016-07-19 19:27:27 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12884 * gst/vaapi/gstvaapipluginbase.c:
12885 plugins: remove sink pad allocator if caps change
12886 If the negotiated sinkpad caps change, destroy the assignated allocator,
12887 because it is not valid anymore.
12889 2016-07-19 20:01:05 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12891 * gst/vaapi/gstvaapipluginutil.c:
12892 * gst/vaapi/gstvaapipluginutil.h:
12893 pluginutil: const params to gst_video_info_changed()
12894 Since they are not modified, we should mark them as const.
12896 2016-07-29 15:13:29 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12898 * gst/vaapi/gstvaapipluginbase.c:
12899 * gst/vaapi/gstvaapivideomemory.c:
12900 * gst/vaapi/gstvaapivideomemory.h:
12901 gstvaapivideomemory: allocator's image size getter
12902 Add the method gst_allocator_get_vaapi_image_size() for the
12903 GstVaapiVideoAllocator, which gets the size of the allocated images with the
12904 current video info.
12905 This method replaces the direct call to the allocator's image info when the
12906 pool is configured.
12908 2016-07-29 18:06:30 +0900 Hyunjun Ko <zzoon@igalia.com>
12910 * gst/vaapi/gstvaapipluginbase.c:
12911 plugins: update buffer pool size with new allocator's image size
12912 Depends on media, video size is sometimes updated with new allocator.
12913 It leads to dismatch between bufferpool's set size and real allocated buffer size.
12914 In this case, it causes every buffer is freed during release in bufferpool,
12915 which should be reused. This affects performance.
12916 https://bugzilla.gnome.org/show_bug.cgi?id=769248
12918 2016-07-27 19:49:59 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12921 configure: remove gmodule-2.0 EGL dependency
12922 Since commit 27429ce, EGL support doesn't depend on dynamic loading libraries,
12923 thus the dependency to gmodule-2.0 is not mandatory anymore.
12925 2016-07-27 10:09:38 -0700 Scott D Phillips <scott.d.phillips@intel.com>
12928 configure: Fix non-fatal PKG_CHECK_MODULES invocations
12929 Some invocations of PKG_CHECK_MODULES were intended to be non-fatal if
12930 the package is missing, but action-if-not-found was given as an empty
12931 string which still causes the default action to run, which halts
12933 Signed-off-by: Scott D Phillips <scott.d.phillips@intel.com>
12934 https://bugzilla.gnome.org/show_bug.cgi?id=769237
12936 2016-07-13 18:34:57 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12938 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
12939 libs: egl: remove dynamic library loading code
12940 Since the upstream of gstreamer-vaapi, the library is not a public shared
12941 object anymore. But the EGL support depended on this dynamic library, so the
12942 EGL support was broken.
12943 This patch removes the dynamic library loading code and instantiates the
12944 EGL display using either X11 or Wayland if available.
12945 https://bugzilla.gnome.org/show_bug.cgi?id=767203
12947 2016-07-12 23:47:41 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12949 * gst/vaapi/gstvaapi.c:
12950 * gst/vaapi/gstvaapidecode.c:
12951 * gst/vaapi/gstvaapidecode.h:
12952 vaapidecode: register only the available decoders
12953 In order to register only the available decoders, this patch queries the
12954 created test VA display, which uses the currently used back-end (X11, Wayland,
12955 DRM, …) on the used display device.
12956 https://bugzilla.gnome.org/show_bug.cgi?id=724352
12958 2016-06-28 11:43:15 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12960 * gst/vaapi/gstvaapi.c:
12961 * gst/vaapi/gstvaapipluginutil.c:
12962 * gst/vaapi/gstvaapipluginutil.h:
12963 vaapi: register only the available encoders
12964 In order to register only the available encoders, this patch queries the
12965 created test VA display, which uses the currently used back-end (X11,
12966 Wayland, DRM, …) on the used display device.
12967 https://bugzilla.gnome.org/show_bug.cgi?id=724352
12969 2016-06-07 16:28:07 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12971 * gst/vaapi/gstvaapidecode.c:
12972 * gst/vaapi/gstvaapidecodebin.c:
12973 vaapidecode: split all the codecs
12974 Split the vaapidecode to all the supported codecs with the format
12976 vaapidecode is stil registered as a GObject type, but not as a
12977 GStreamer feature, so it can be used internally by vaapidecodebin without
12978 changing its code too much.
12979 https://bugzilla.gnome.org/show_bug.cgi?id=734093
12981 2016-07-12 22:19:37 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12983 * gst/vaapi/gstvaapidecodebin.c:
12984 vaapidecodebin: simplify the code
12985 Since the elements dependant of the VA video processor are now only registered
12986 if it is available, vaapidecodebin code can be simplified a lot, removing all
12987 the code required to check if the VA video processor was available.
12988 https://bugzilla.gnome.org/show_bug.cgi?id=768899
12990 2016-07-12 17:54:26 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
12992 * gst/vaapi/gstvaapidecode.c:
12993 vaapidecode: delay the GstVaapiDisplay instantiating
12994 Delay the GstVaapiDisplay instantiating until when changing the state from
12995 READY to PAUSE. In this way the element has more chances to find an already
12996 created GstVaapiDisplay, or a GL context, in the pipeline.
12997 https://bugzilla.gnome.org/show_bug.cgi?id=766206
12999 2016-07-12 17:49:50 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13001 * gst/vaapi/gstvaapipluginutil.c:
13002 pluginutil: set GLX display type
13003 The function gst_vaapi_create_display_from_gl_context() cretes a
13004 GstVaapiDisplay given a GstGLContext. But it didn't created a GLX VA display
13005 when the GL platform was GLX, but a plain X11 VA display.
13006 This patch fixes that, by querying the GL platform earlier.
13007 https://bugzilla.gnome.org/show_bug.cgi?id=766206
13009 2016-06-02 19:57:08 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13011 * gst/vaapi/gstvaapipluginbase.c:
13012 * gst/vaapi/gstvaapivideocontext.c:
13013 * gst/vaapi/gstvaapivideocontext.h:
13014 plugins: add gst_vaapi_plugin_base_find_gl_context()
13015 Using the GstContext mechanism, it is possible to find if the pipeline
13016 shares a GstGLContext, even if we are not to negotiating GLTextureUpload
13017 meta. This is interesting because we could negotiate system memory caps
13018 feature, but enable DMABuf if the GstGLContext is EGL with some extensions.
13019 https://bugzilla.gnome.org/show_bug.cgi?id=766206
13021 2016-06-28 17:14:06 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13023 * gst/vaapi/gstvaapipluginbase.c:
13024 * gst/vaapi/gstvaapipluginbase.h:
13025 plugins: remove gst_vaapi_plugin_base_driver_is_whitelisted()
13026 Since nobody is calling gst_vaapi_plugin_base_driver_is_whitelisted(),
13029 2016-07-12 18:24:10 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13031 * gst/vaapi/gstvaapipluginbase.h:
13032 plugins: remove common change_state() vmethod
13033 Remove the common change_state() vmethod for all the plugins, since no one is
13036 2016-07-12 20:38:07 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13038 * gst/vaapi/gstvaapidecode.c:
13039 vaapidecode: remove change_state() vmethod
13040 Since the driver checkup is done at registering, there is no need to do it
13041 when changing the element state from NULL to READY. This patch remove this
13042 vmethod from vaapidecode.
13044 2016-07-12 20:29:12 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13046 * gst/vaapi/gstvaapi.c:
13047 vaapi: register vaapipostproc only if supported
13048 Query the GstVaapiDisplay to know if the driver supports video
13049 postprocessing. If does, then register vaapipostproc and vaapidecodebin
13051 This patch will simplify the design of vaapidecodebin.
13052 https://bugzilla.gnome.org/show_bug.cgi?id=724352
13054 2016-06-29 12:36:26 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13056 * gst/vaapi/gstvaapi.c:
13057 vaapi: don't register if VA driver is unsupported
13058 Using the test VA display, the driver name is queried, and if it is not
13059 white-listed, the plugin rejects to register any element.
13060 https://bugzilla.gnome.org/show_bug.cgi?id=724352
13062 2016-06-28 17:14:06 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13064 * gst/vaapi/gstvaapipluginbase.c:
13065 * gst/vaapi/gstvaapipluginutil.c:
13066 * gst/vaapi/gstvaapipluginutil.h:
13067 plugins: add gst_vaapi_driver_is_whitelisted()
13068 Move some of the logic in gst_vaapi_plugin_base_driver_is_whitelisted() to a
13069 new function gst_vaapi_driver_is_whitelisted(), in this way, it can be used
13070 when registering the plugin's feature set with the test VA display.
13071 https://bugzilla.gnome.org/show_bug.cgi?id=724352
13073 2016-07-12 19:56:02 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13075 * gst/vaapi/gstvaapi.c:
13076 vaapi: instantiate a VA display when registering
13077 This patch tries to instantiate a GstVaapiDisplay when registering the plugin
13078 features, if it fails, no gstreamer-vaapi element is registering.
13079 The purpose of this patch is to avoid a situation where the user has
13080 gstreamer-vaapi installed but their VA-API setup is not functional, which may
13081 lead to unexpected behavior.
13082 https://bugzilla.gnome.org/show_bug.cgi?id=724352
13084 2016-06-28 11:33:18 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13087 * gst/vaapi/gstvaapi.c:
13088 vaapi: declare external dependencies
13089 There are two main external dependencies that define the feature set of this
13090 plugin: a) the kernel and b) the VA driver
13091 This patch tracks both dependencies, if any of them change, GStreamer will
13092 re-inspect the plugin.
13093 The kernel is tracked through the device files /dev/dri/card*
13094 The VA driver is tracked through the files VA_DRIVERS_PATH/*_drv_video.so,
13095 where VA_DRIVERS_PATH is the one defined in libva package configuration. Also,
13096 the environment variables LIBVA_DRIVERS_PATH and LIBVA_DRIVER_NAME are tracked
13097 since they modify the driver lookup.
13098 Additionally, the environment variable GST_VAAPI_ALL_DRIVERS is tracked too.
13099 https://bugzilla.gnome.org/show_bug.cgi?id=724352
13101 2016-07-19 16:02:27 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13103 * gst/vaapi/gstvaapidecode.c:
13104 vaapidecode: remove unneeded initializations
13105 GObject's memory is set to zero, so there is no need to initialize to zero or
13106 NULL it's class variables.
13108 2016-07-19 18:28:28 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13110 * gst/vaapi/gstvaapipluginbase.h:
13111 plugins: remove undefined macros
13113 2016-07-19 17:43:28 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13115 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
13116 Revert "gstvaapisurface_drm: release image when done"
13117 This reverts commit 1dbcc8a0e199f2da6a0ab8e949f13341916128a3 and commit
13118 372a03a9e38acbf435eb80bf31d9a9844069e504.
13119 While the dmabuf handle is exported, the derive image must exist, otherwise
13120 the image's VA buffer is invalid, thus the dmabuf handle is never released,
13121 leading into a file descriptors leak.
13123 2016-07-21 17:38:17 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13125 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
13126 encoder: h265: fix code-style
13128 2016-07-22 16:55:59 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13130 * gst/vaapi/gstvaapipostproc.c:
13131 vaapipostproc: update filters at color balance
13132 This is a fix for a regression of previous commit, which updates the filters
13133 only when the property is set, because it is also required to update the
13134 filter when the color balance interface change its values.
13136 2016-07-22 12:10:23 +0900 Hyunjun Ko <zzoon@igalia.com>
13138 * gst/vaapi/gstvaapipostproc.c:
13139 * gst/vaapi/gstvaapipostproc.h:
13140 vaapipostproc: make it enable/disable pass-through mode
13141 In case that sink caps and src caps are same, and no filtering parameter set,
13142 pass-through mode is enabled.
13143 If new filtering parameter is set during playback, it makes it reconfiguring,
13144 so that pass-through mode is changed
13145 In addition, updating filter is performed during reconfiguration, if needed.
13146 https://bugzilla.gnome.org/show_bug.cgi?id=751876
13148 2016-07-22 11:51:26 +0900 Hyunjun Ko <zzoon@igalia.com>
13150 * gst-libs/gst/vaapi/gstvaapifilter.c:
13151 * gst-libs/gst/vaapi/gstvaapifilter.h:
13152 * gst/vaapi/gstvaapipostproc.c:
13153 vaapipostproc: checking and updating filter parameter only when it's set
13154 This patch is to avoid checking filter value at every frame.
13155 https://bugzilla.gnome.org/show_bug.cgi?id=751876
13157 2016-07-21 11:24:31 +0300 Allen Zhang <Zhang, Zhangfei>
13159 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
13160 decoder: h265: handle the SEI NAL units included in codec_data
13161 The prefix/suffix SEI nal units can appear in codec_data too
13162 which weren't handled before. Parse these SEI headers to
13164 https://bugzilla.gnome.org/show_bug.cgi?id=768544
13166 2016-07-15 16:32:26 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13168 * docs/plugins/Makefile.am:
13169 build: doc: do not redefine MAINTAINERCLEANFILES
13170 MAINTAINERCLEANFILES is defined in gtk-doc-plugins.mak, thus instead of
13171 overload it, the files should be added.
13173 2016-07-15 14:41:27 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13175 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
13176 encoder: h264: Fix MVC encode while enabling dct8x8
13177 Pack the transform_8x8_mode_flag and other necessary rbsp data
13178 in packed_pps header for MVC encode.
13179 https://bugzilla.gnome.org/show_bug.cgi?id=768647
13181 2016-07-12 23:58:55 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13183 * gst/vaapi/gstvaapisink.c:
13184 vaapisink: demote a debug message to trace
13185 Reduces noise when debugging.
13187 2016-07-13 17:21:01 +0900 Jagyum Koo <koojagyum@gmail.com>
13189 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
13190 wayland: Error check before using cached wl_display
13191 A planar(or some other) buffer allocation may fail on the driver, then
13192 the wayland connection becomes invalid, not able to send request or
13193 receive any event. So we need to set up a new wayland connection if
13194 there's an error detected on the cached wl_display.
13195 https://bugzilla.gnome.org/show_bug.cgi?id=768761
13196 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13198 2016-07-11 21:15:57 +0200 Stefan Sauer <ensonic@users.sf.net>
13201 Automatic update of common submodule
13202 From ac2f647 to f49c55e
13204 2016-07-05 18:23:22 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13206 * gst/vaapi/gstvaapiencode.c:
13207 vaapiencode: demote a log to trace level
13208 Removes noise when debugging.
13210 2016-07-06 11:17:23 +0900 Hyunjun Ko <zzoon@igalia.com>
13212 * gst/vaapi/gstvaapiencode.c:
13213 vaapiencode: implement flush() vmethod
13214 In order to handle correctly seek and other operations, vaapiencode should
13215 flush all the remaining data from the encoder without pushing it downstream.
13216 This patch implements the flush() vmethod, only after of pausing the
13217 source pad task, and restarting it again after the flush stop.
13218 https://bugzilla.gnome.org/show_bug.cgi?id=767176
13219 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13221 2016-07-11 08:43:04 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13223 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
13224 decoder: vc1: flush dpb only if opened
13225 Flush the decode picture buffer, if and only if, the decoder is
13226 started. Otherwise the dpb structure might be NULL.
13227 https://bugzilla.gnome.org/show_bug.cgi?id=742922
13229 2016-07-01 14:42:20 +0900 Hyunjun Ko <zzoon@igalia.com>
13231 * gst/vaapi/gstvaapidecode.c:
13232 * gst/vaapi/gstvaapidecode.h:
13233 vaapidecode: drop non-keyframe in reverse playback
13234 To avoid surface-exhausted situation during reverse playback,
13235 drop frames except for key frame.
13236 Also, to avoid the corruption of the parser state, flush() vmethod
13237 doesn't destroy the VA decoder when playing in reverse.
13238 https://bugzilla.gnome.org/show_bug.cgi?id=742922
13239 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13241 2016-07-10 19:33:14 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13243 * gst/vaapi/gstvaapidecode.c:
13244 vaapidecode: unref output frame earlier
13245 The queue in GstVaapiDecode adds an extra reference to the frames. This patch
13246 unref that extra reference earlier making the code simpler to follow.
13247 https://bugzilla.gnome.org/show_bug.cgi?id=768652
13249 2016-07-10 19:01:17 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13251 * gst/vaapi/gstvaapidecode.c:
13252 vaapidecode: remove gst_vaapidecode_internal_flush()
13253 As gst_vaapidecode_finish() is the only callee of
13254 gst_vaapidecode_internal_flush(), it is better to inline it.
13255 https://bugzilla.gnome.org/show_bug.cgi?id=768652
13257 2016-07-10 18:18:57 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13259 * gst/vaapi/gstvaapidecode.c:
13260 vaapidecode: flush output adapter at drain()
13261 Calling drain() vmethod means "decode any data it can at this point, but that
13262 more data may arrive after". Hence, vaapidecode should check if there is data
13263 in the output adapter and process them, without destroying the decoded picture
13265 Since this operation is done by gst_vaapidecode_internal_flush(), the operation
13266 was refactored into a new function gst_vaapidecode_flush_output_adapter().
13267 https://bugzilla.gnome.org/show_bug.cgi?id=768652
13269 2016-07-10 13:46:25 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13271 * gst/vaapi/gstvaapidecode.c:
13272 vaapidecode: call purge at flush()
13273 Calling flush() vmethod means "to flush all remaining data from the decoder
13274 without pushing it downstream".
13275 Nonetheless flush() is calling gst_vaapidecode_internal_flush(), which calls
13276 gst_video_decoder_have_frame() if there is still something in the input
13277 adapter, which may push buffers to downstream by calling handle_frame().
13278 This patch changes this behavior by calling gst_vaapidecode_purge() rather
13279 than gst_vaapidecode_internal_flush(), which does what we want: flushes the VA
13280 decoder and releases all the rest of decoded frames.
13281 https://bugzilla.gnome.org/show_bug.cgi?id=768652
13283 2016-07-06 18:38:37 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13285 * tests/elements/Makefile.am:
13286 * tests/elements/test-vaapisink.c:
13287 test: elements: remove spurious linkage
13288 Element tests only need to link against gstreamer libraries.
13290 2016-07-06 14:41:21 +0300 Sebastian Dröge <sebastian@centricular.com>
13293 configure: Require GLib >= 2.40 like everywhere else
13295 2016-07-06 13:51:21 +0300 Sebastian Dröge <sebastian@centricular.com>
13298 Back to development
13300 === release 1.9.1 ===
13302 2016-07-06 13:48:07 +0300 Sebastian Dröge <sebastian@centricular.com>
13308 * gstreamer-vaapi.doap:
13311 2016-07-05 20:59:49 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13313 * tests/elements/test-vaapisink.c:
13314 tests: elements: rotate orientation event
13316 2016-07-01 16:01:54 +0900 Hyunjun Ko <zzoon@igalia.com>
13319 * tests/Makefile.am:
13320 * tests/elements/Makefile.am:
13321 * tests/elements/test-vaapisink.c:
13322 tests: elements: Add testsuite for vaapisink
13323 https://bugzilla.gnome.org/show_bug.cgi?id=765798
13325 2016-07-01 16:00:46 +0900 Hyunjun Ko <zzoon@igalia.com>
13327 * gst-libs/gst/vaapi/gstvaapitypes.h:
13328 * gst-libs/gst/vaapi/gstvaapivalue.c:
13329 * gst/vaapi/gstvaapisink.c:
13330 * gst/vaapi/gstvaapisink.h:
13331 vaapisink: add support for GST_TAG_IMAGE_ORIENTATION
13332 https://bugzilla.gnome.org/show_bug.cgi?id=765798
13334 2016-06-29 13:57:02 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13336 * gst/vaapi/gstvaapipostproc.c:
13337 vaapipostproc: return caps template if no display
13338 This patch is a fix for my bad review of commit 6d73ca8d. The element should
13339 be able to return the available raw caps handled by the VA display, but that
13340 only should happen when there a VA display. If there's none, the element
13341 should use the caps template.
13342 https://bugzilla.gnome.org/show_bug.cgi?id=768161
13344 2016-06-29 16:42:18 +1000 Matthew Waters <matthew@centricular.com>
13346 * gst/vaapi/gstvaapipostproc.c:
13347 vaapipostproc: don't require a vaapi display for all caps queries
13348 This delays the requirement of having a GstVaapiDisplay until later
13349 https://bugzilla.gnome.org/show_bug.cgi?id=768161
13351 2016-06-28 15:48:39 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13353 * gst-libs/gst/vaapi/gstvaapiutils.c:
13354 utils: report VP9 profiles
13355 Add VP9Profile0-3 name mapping.
13357 2016-06-28 14:05:51 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13359 * gst/vaapi/gstvaapi.c:
13360 vaapi: nest includes under USE_ENCODER macro
13361 This is a missed changeset from commit 1c05c53, since also header includes
13364 2016-06-28 11:54:33 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13366 * gst/vaapi/gstvaapi.c:
13367 vaapi: nest encoders under USE_ENCODER macro
13368 Though USE_{JPEG,VP8,VP9,H265}_ENCODER macros definition depend on USE_ENCODER
13369 macro, it is clearer to nest them, showing explicitly the dependency relation.
13371 2016-06-24 12:05:24 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13373 * gst/vaapi/gstvaapivideocontext.c:
13374 vaapivideocontext: check if query context is NULL
13375 Under certain conditions the element might receive a positive context query
13376 but without a context instance. This situation will lead to a segmentation
13377 fault when traversing the context list in the pipeline.
13378 https://bugzilla.gnome.org/show_bug.cgi?id=767946
13380 2016-06-20 13:22:36 -0700 Scott D Phillips <scott.d.phillips@intel.com>
13382 * gst-libs/gst/vaapi/Makefile.am:
13383 * gst-libs/gst/vaapi/glibcompat.h:
13384 * gst-libs/gst/vaapi/sysdeps.h:
13385 remove unused glibcompat.h
13386 glibcompat.h is no longer doing anything. Remove it.
13387 Signed-off-by: Scott D Phillips <scott.d.phillips@intel.com>
13388 https://bugzilla.gnome.org/show_bug.cgi?id=767886
13390 2016-06-22 14:28:44 -0700 Scott D Phillips <scott.d.phillips@intel.com>
13392 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
13393 encoder: h264: Use high profile by default
13394 Change defaults for max-bframes, cabac, and dct8x8 to be enabled
13395 by default. This will cause the default profile to be high instead
13396 of baseline. In most situations this is the right decision, and
13397 the profile can still be lowered in the case of caps restrictions.
13398 Signed-off-by: Scott D Phillips <scott.d.phillips@intel.com>
13399 https://bugzilla.gnome.org/show_bug.cgi?id=757941
13401 2016-06-22 12:15:29 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13403 * gst/vaapi/gstvaapidecodebin.c:
13404 vaapidecodebin: element warning if missing element
13405 Raise an element warning if a required element is not available, thus the
13406 pipeline will post a warning message and the application will be informed.
13408 2016-06-22 15:11:56 +0300 Hyunjun Ko <zzoon@igalia.com>
13410 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
13411 decoder: h265: fix to release all dpb pictures
13412 Without this, all dpb pictures are not released during flush,
13413 because we used the global dpb_count variable for checking the
13414 dpb fullness which get decremented in dpb_remove_index()
13415 routine during each loop iteration.
13416 https://bugzilla.gnome.org/show_bug.cgi?id=767934
13418 2016-06-21 11:48:54 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
13421 Automatic update of common submodule
13422 From ac2f647 to f363b32
13424 2016-06-20 19:53:26 +0900 Hyunjun Ko <zzoon@igalia.com>
13426 * gst/vaapi/gstvaapidecode.c:
13427 * gst/vaapi/gstvaapipostproc.c:
13428 vaapi: fix minor leaks
13429 https://bugzilla.gnome.org/show_bug.cgi?id=767868
13431 2016-06-17 17:00:03 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13433 * gst/vaapi/gstvaapi.c:
13434 vaapi: remove an already included header
13435 gst/gst.h is already included in gstcompat.h
13437 2016-06-17 16:53:26 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13439 * gst/vaapi/gstvaapidecodebin.c:
13440 vaapidecodebin: add vp9 in sink pad template
13442 2016-06-15 20:19:27 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13444 * gst/vaapi/gstvaapisink.c:
13445 vaapisink: return caps template if no display
13446 If vaapisink received a caps query before getting a VA display, it returned
13447 only the surfaces related caps. This behavior broke the autovideosink
13449 This patch returns the pad's template caps if no VA display, otherwise the
13450 caps are crafted as before.
13451 https://bugzilla.gnome.org/show_bug.cgi?id=767699
13453 2016-06-15 11:25:23 -0700 Scott D Phillips <scott.d.phillips@intel.com>
13455 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
13456 decoder: vp9: Update comment about context resets
13457 Clarify that vaapi context resets are never needed for vp9, but
13458 that ensure_context() needs called when the size increases so that
13459 new surfaces can be allocated.
13460 Signed-off-by: Scott D Phillips <scott.d.phillips@intel.com>
13461 https://bugzilla.gnome.org/show_bug.cgi?id=767474
13463 2016-05-17 15:34:23 -0700 Scott D Phillips <scott.d.phillips@intel.com>
13465 * gst-libs/gst/vaapi/gstvaapicontext.c:
13466 * gst-libs/gst/vaapi/gstvaapicontext.h:
13467 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
13468 gstvaapicontext: control reset_on_resize with option
13469 Signed-off-by: Scott D Phillips <scott.d.phillips@intel.com>
13470 https://bugzilla.gnome.org/show_bug.cgi?id=767474
13472 2016-06-14 09:45:22 -0700 Scott D Phillips <scott.d.phillips@intel.com>
13474 * gst/vaapi/gstvaapivideobufferpool.c:
13475 vaapivideobufferpool: add video meta to config when needed
13476 In cases where we know the video meta must be present, add it to
13477 the pool configuration.
13478 Signed-off-by: Scott D Phillips <scott.d.phillips@intel.com>
13479 https://bugzilla.gnome.org/show_bug.cgi?id=766184
13481 2016-06-13 10:48:41 -0700 Scott D Phillips <scott.d.phillips@intel.com>
13483 * gst/vaapi/gstvaapipluginbase.c:
13484 plugins: retry pool config
13485 if gst_buffer_pool_set_config returns FALSE, check the modified
13486 config and retry set_config if the config is still acceptable.
13487 Signed-off-by: Scott D Phillips <scott.d.phillips@intel.com>
13488 https://bugzilla.gnome.org/show_bug.cgi?id=766184
13490 2016-06-08 18:42:43 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13492 * gst/vaapi/gstvaapivideomemory.h:
13493 vaapivideomemory: internal attributes to methods
13494 Mark as internal the functions used by VA-API dmabuf allocator.
13496 2016-06-02 15:41:22 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13498 * gst/vaapi/gstvaapipluginbase.c:
13499 plugins: remove precondition for decide_allocation()
13500 There's no need to check for the display in the plugin object when
13501 decide_allocation() vmethod is called, because the display will created or
13502 re-created along the method execution.
13504 2016-06-08 18:44:34 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13506 * gst/vaapi/gstvaapipluginbase.c:
13507 plugins: avoid possible memory leaks
13508 Get the pool config just before use it, to avoid a memory leak if the
13509 allocator cannot be instantiated. Similarly, return FALSE if the configuration
13510 cannot be set, avoid keep a not used allocator in the pool.
13512 2016-05-31 11:52:57 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13514 * gst/vaapi/gstvaapipluginbase.c:
13515 plugins: use GstParentBufferMeta
13516 Instead of using the VASurface proxy's notify, which is internal gstvaapi API,
13517 use the GStreamer's GstParentBufferMeta.
13518 https://bugzilla.gnome.org/show_bug.cgi?id=765435
13520 2016-05-31 08:26:33 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13522 * gst/vaapi/gstvaapipluginbase.c:
13523 plugins: cache VASurfaces from dmabufs
13524 This patch avoids the creation of a VASurface each time a new input buffer is
13525 processed, caching them in the input buffer itself.
13526 https://bugzilla.gnome.org/show_bug.cgi?id=765435
13528 2016-05-30 23:55:43 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13530 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
13531 * gst-libs/gst/vaapi/gstvaapisurface_drm.h:
13532 * gst/vaapi/gstvaapipluginbase.c:
13533 libs: change gst_vaapi_surface_new_with_dma_buf_handle()
13534 Instead of passing the data already in GstVideoInfo, let's just pass the
13535 GstVideoInfo structure.
13536 https://bugzilla.gnome.org/show_bug.cgi?id=765435
13538 2016-05-25 12:31:15 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13540 * gst/vaapi/gstvaapipluginbase.c:
13541 * gst/vaapi/gstvaapipluginbase.h:
13542 plugins: use an unique allocator per pad
13543 Instead of instantiating an allocator per vaapivideobufferpool, only one
13544 allocator is instantiated per element's pad and shared among future pools.
13545 If the pad's caps changes, the allocator is reset.
13546 https://bugzilla.gnome.org/show_bug.cgi?id=765435
13548 2016-05-25 10:58:01 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13550 * gst/vaapi/gstvaapipluginbase.c:
13551 * gst/vaapi/gstvaapivideobufferpool.c:
13552 * gst/vaapi/gstvaapivideobufferpool.h:
13553 vaapivideobufferpool: share options flag with pluginbase
13554 Originally, vaapivideobufferpool has a set of boolean variables for the
13555 buffer configuration options.
13556 This pach changes these boolean variables for a single bitwise, just as
13557 it is used in pluginbase. Hence, the internal enum was moved to
13558 vaapivideobufferpool header.
13559 https://bugzilla.gnome.org/show_bug.cgi?id=765435
13561 2016-05-10 15:57:06 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13563 * gst/vaapi/gstvaapipluginbase.c:
13564 plugins: add gst_vaapi_plugin_base_create_pool()
13565 This patch refactors the code in pluginbase in order to centralize the buffer
13566 pool instantiation. As the buffer pool config may have different options, these
13567 are gathered using a bitwise flag.
13568 https://bugzilla.gnome.org/show_bug.cgi?id=765435
13570 2016-05-20 18:46:14 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13572 pluginbase negotiates allocator with bufferpool
13573 Originally vaapivideobufferpool instantiates its own allocator regardless the
13574 received configuration, and it relies in custom configuration options to
13575 choose which kind of allocator instantiate.
13576 This patch transfers the responsibility of the allocator instantiate to
13577 vaapipluginbase and pass it to the vaapivideobufferpool through its
13579 * gst/vaapi/gstvaapipluginbase.c
13580 + set_dmabuf_allocator(): inserts a dmabuf allocator in the bufferpool
13581 + ensure_sinkpad_buffer_pool(): set a normal vaapi video allocator in
13582 bufferpool configuration
13583 + gst_vaapi_plugin_base_propose_allocation(): call set_dmabuf_allocator() if
13585 + gst_vaapi_plugin_base_decide_allocation(): set a normal vaapi video
13586 allocator in bufferpool configuration
13587 * gst/vaapi/gstvaapivideobufferpool.c
13588 + gst_vaapi_video_buffer_pool_set_config(): instead of instantiate the
13589 allocator, process the received one through its configuration.
13590 * gst/vaapi/gstvaapivideobufferpool.h: removed
13591 GST_BUFFER_POOL_OPTION_DMABUF_MEMORY since it is not used anymore.
13592 * gst/vaapi/gstvaapivideomemory.c
13593 + gst_vaapi_is_dmabuf_allocator(): new helper function to identify a dmabuf
13594 allocator with the vaapi qdata.
13595 https://bugzilla.gnome.org/show_bug.cgi?id=765435
13597 2016-05-20 14:39:23 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13599 * gst/vaapi/gstvaapivideobufferpool.c:
13600 vaapivideobufferpool: keep only current video info
13601 Instead of keeping old and new GstVideoInfo video structure, we only keep one,
13602 the current one, the negotiated. The old one is not needed at all.
13603 https://bugzilla.gnome.org/show_bug.cgi?id=765435
13605 2016-05-23 15:38:07 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13607 * gst/vaapi/gstvaapipluginbase.c:
13608 * gst/vaapi/gstvaapipluginutil.c:
13609 * gst/vaapi/gstvaapipluginutil.h:
13610 * gst/vaapi/gstvaapivideomemory.c:
13611 pluginutil: add gst_video_info_force_nv12_if_encoded()
13612 This lines repeat a couple times in the code, so it would be better to put it
13614 https://bugzilla.gnome.org/show_bug.cgi?id=765435
13616 2016-05-20 14:15:53 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13618 * gst/vaapi/gstvaapipluginutil.c:
13619 * gst/vaapi/gstvaapipluginutil.h:
13620 * gst/vaapi/gstvaapipostproc.c:
13621 * gst/vaapi/gstvaapivideobufferpool.c:
13622 pluginutil: add gst_video_info_changed() helper
13623 This function is shared among different elements, so let factorized it.
13624 https://bugzilla.gnome.org/show_bug.cgi?id=765435
13626 2016-06-08 10:14:16 +0900 Hyunjun Ko <zzoon@igalia.com>
13628 * gst/vaapi/gstvaapipostprocutil.c:
13629 vaapipostproc: Add colorimetry attributes to src caps
13630 https://bugzilla.gnome.org/show_bug.cgi?id=766596
13632 2016-06-08 10:17:46 +0900 Hyunjun Ko <zzoon@igalia.com>
13634 * gst/vaapi/gstvaapidecode.c:
13635 vaapidecode: remove chroma-site and colorimetry from src caps
13636 https://bugzilla.gnome.org/show_bug.cgi?id=766596
13638 2016-06-07 14:19:50 -0700 Scott D Phillips <scott.d.phillips@intel.com>
13640 * gst/vaapi/gstvaapipostproc.c:
13641 * gst/vaapi/gstvaapipostproc.h:
13642 vaapipostproc: add postproc_lock to protect data members
13643 Add a mutex to postproc to protect concurrent access to data members.
13644 Previously set_caps() could release the allowed_srcpad_caps while
13645 transform_caps was in the middle of using it.
13646 Signed-off-by: Scott D Phillips <scott.d.phillips@intel.com>
13647 https://bugzilla.gnome.org/show_bug.cgi?id=766940
13649 2016-05-30 11:30:40 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13651 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
13652 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.h:
13653 encoder: vp9: Add simple algorithms for reference picture selection
13654 Added two modes(as properties) for reference picture selection:
13655 ref-mode-0: AltRef and GoldRef pointing to the recent keyframe
13656 and LastRef is pointing to the previous frame.
13657 ref-mode-1: Previous frame (n) as LastRef , n-1 th frame as GoldRef
13658 and n-2 th frame as AltRef
13659 https://bugzilla.gnome.org/show_bug.cgi?id=766048
13661 2016-05-30 11:25:52 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13663 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
13664 encoder: vp9: Define Max frame width and height
13665 https://bugzilla.gnome.org/show_bug.cgi?id=766048
13667 2016-05-30 11:25:03 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13669 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
13670 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.h:
13671 encoder: vp9: Add more propertis for tuning encode quality
13672 Added three tuning properties:
13675 3: luma ac quant-table index
13676 https://bugzilla.gnome.org/show_bug.cgi?id=766048
13678 2016-05-30 11:24:14 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13680 * gst/vaapi/Makefile.am:
13681 * gst/vaapi/gstvaapi.c:
13682 * gst/vaapi/gstvaapiencode_vp9.c:
13683 * gst/vaapi/gstvaapiencode_vp9.h:
13684 Add vp9 encode element to "vaapi" plugin
13685 https://bugzilla.gnome.org/show_bug.cgi?id=766048
13687 2016-05-30 11:23:12 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13689 * gst-libs/gst/vaapi/Makefile.am:
13690 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.c:
13691 * gst-libs/gst/vaapi/gstvaapiencoder_vp9.h:
13692 Add vp9 encoder support in libgstvaapi
13693 https://bugzilla.gnome.org/show_bug.cgi?id=766048
13695 2016-05-30 11:22:35 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13698 build: Add check for VP9 encode API support in libva
13699 https://bugzilla.gnome.org/show_bug.cgi?id=766048
13701 2016-05-26 11:42:32 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13703 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
13704 gstvaapisurface_drm: fix internal documentation
13706 2016-05-26 11:41:25 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13708 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
13709 gstvaapisurface_drm: fix code-style
13711 2016-05-25 12:28:48 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13713 * gst/vaapi/gstvaapipluginbase.c:
13714 plugins: remove unused header
13715 Remove the include of gst/allocators/allocators.h since it is not used.
13717 2016-05-25 10:36:33 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13719 * gst/vaapi/gstvaapivideobufferpool.h:
13720 vaapivideobufferpool: remove GL_TEXTURE_UPLOAD_META
13721 Since gstreamer-vaapi is coupled with gstreamer releases, there is no need to
13722 keep compatibility definition.
13723 This patch removes the definition of
13724 GST_BUFFER_POOL_OPTION_VIDEO_GL_TEXTURE_UPLOAD_META since it is in
13725 gst-plugins-base version 1.2.2
13727 2016-05-23 22:49:11 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13729 * gst/vaapi/gstvaapipluginbase.c:
13730 plugins: add gst_vaapi_buffer_pool_caps_is_equal()
13731 This is a helper function to improve the readability of
13732 ensure_sinkpad_buffer_pool(). It makes clearer when the buffer pool needs to be
13735 2016-05-24 16:29:33 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13737 * gst/vaapi/gstvaapipluginbase.c:
13738 plugins: deactivate buffer pool before unref
13739 This buffer pool may still be processing buffers when a caps renegotiation is
13740 done. This one-liner patch deactivates the pool to drain it before it
13743 2016-05-24 16:22:24 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13745 * gst/vaapi/gstvaapipluginbase.c:
13746 plugins: no sinkpad bufferpool when decoder
13747 Right now, the decoders create a buffer pool for their sink pad which is not
13748 used at all, because the decoders have never proposed it to upstream.
13749 This patch avoids the buffer pool instantiating when the element inherits from
13750 the GstVideoDecoder class.
13752 2016-05-24 13:39:25 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13754 * gst/vaapi/gstvaapipluginbase.c:
13755 plugins: avoid to get/set pool config twice
13756 This patch is a bit of optimization, since the bufferpool configuration is get
13757 when the pool is created. Hence, we only need to request it when the pool from
13758 the allocation query is reused.
13760 2016-05-13 13:14:23 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13762 * gst/vaapi/gstvaapipluginbase.c:
13763 plugins: destroy pool earlier if non-vaapi
13764 If the offered pool in decide_allocation() vmethod doesn't have the
13765 VAAPI_VIDEO_META option, it is destroyed immediatly and the pointer cleared,
13766 so it could be created later.
13768 2016-05-23 22:30:04 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13770 * gst/vaapi/gstvaapivideobufferpool.c:
13771 vaapivideobufferpool: split caps validation
13772 When validating the caps from bufferpool config, this patch distinguishes the
13773 error from no caps received (NULL) from the invalid caps (cannot be converted
13774 into GstVideoInfo structure).
13776 2016-05-23 22:21:15 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13778 * gst/vaapi/gstvaapidecode.c:
13779 * gst/vaapi/gstvaapipluginbase.c:
13780 plugins: check for caps in query earlier
13781 Check for caps as soon gst_query_parse_allocation() returns.
13783 2016-05-23 23:13:55 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13785 * gst/vaapi/gstvaapipluginbase.c:
13786 * gst/vaapi/gstvaapipluginbase.h:
13787 plugins: remove unused variables
13788 This variables stopped to be used since commit 001a5c63, which removed the
13791 2016-05-23 18:47:46 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13793 * gst/vaapi/gstvaapipluginbase.c:
13794 plugins: fix potential memleak from commit 9159328
13795 If gst_video_info_from_caps() fails it is required to unref the instantiated
13798 2016-05-23 18:04:47 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13800 * gst/vaapi/gstvaapipostproc.c:
13801 vaapipostproc: handle if gst_video_info_from_caps() fails
13802 Return FALSE is the received caps cannot be transformed into a GstVideoInfo
13805 2016-05-23 17:55:35 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13807 * gst/vaapi/gstvaapipluginbase.c:
13808 plugins: handle if gst_video_info_from_caps() fails
13809 Let's play safe and return error if, somehow, the received caps are wrong.
13811 2016-05-23 17:47:36 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13813 * gst/vaapi/gstvaapipluginbase.c:
13814 plugins: relate errors to instance
13815 Use GST_{ERROR,WARNING}_OBJECT instead of GST_{ERROR,WARNING}, thus the logs
13816 will show the name of the vaapipluginbase instance that failed.
13817 Also, the code-style is fixed, where some error labels need to be surrounded
13820 2016-05-20 21:01:02 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13822 * gst/vaapi/gstvaapipluginbase.c:
13823 * gst/vaapi/gstvaapipluginutil.c:
13824 * gst/vaapi/gstvaapisink.c:
13825 plugins: use GstVideoInfo accessors
13826 Instead of access to GstVideInfo members directly, use their accessors
13827 macros. This patch makes more resistance to future changes in GStreamer core.
13829 2016-05-20 19:33:39 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13831 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
13832 * gst/vaapi/gstvaapipluginbase.c:
13833 * gst/vaapi/gstvaapipostproc.c:
13834 * tests/simple-encoder.c:
13835 remove spurious gst_video_info_init()
13836 gst_video_info_set_format() and gst_video_info_from_caps() call, internally,
13837 gst_video_info_init(), hence it is not required to call it before them. This
13838 patch removes these spurious calls.
13840 2016-05-20 19:15:11 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13842 * gst/vaapi/gstvaapipluginbase.c:
13843 vaapipluginbase: code-style: rename goto label
13844 The error labels have error_ prefix, but this one.
13846 2016-05-19 16:34:50 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13848 * gst/vaapi/gstvaapivideomemory.c:
13849 vaapivideomemory: use allocator custom alloc flag
13850 Instead of a dummy alloc() vmethod, the allocator instance set the flag
13851 GST_ALLOCATOR_FLAG_CUSTOM_ALLOC, which is used by the framework to avoid call
13852 gst_allocator_alloc() on the allocator.
13854 2016-05-06 13:17:47 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13856 * gst/vaapi/gstvaapivideobufferpool.c:
13857 vaapivideobufferpool: relate errors to instance
13858 Use GST_ERROR_OBJECT instead of GST_ERROR, thus the logs will show the name of
13859 the vaapivideobufferpool instance that failed.
13861 2016-05-10 16:14:48 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13863 * gst/vaapi/gstvaapipluginbase.c:
13864 plugins: remove gst_vaapi_plugin_base_set_pool_config()
13865 This function helper make sense for GStreamer 1.2, but it is not helpful for
13866 greater version since the validation is already done in the API implementation.
13867 Thus, it is removed.
13869 2016-05-18 17:05:03 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13871 * gst/vaapi/gstvaapipluginbase.c:
13872 gstvaapipluginbase: Fix typo in doc
13874 2016-05-13 11:45:20 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13876 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
13877 encoder : h264: Disable b-frame encode in low-power mode
13878 This is a workaround since vaapi-intel-driver doesn't have
13879 support for B-frame encode when utilizing low-power-enc
13881 Fixme :We should query the VAConfigAttribEncMaxRefFrames
13882 instead of blindly disabling b-frame support and set b/p frame count,
13883 buffer pool size etc based on the query result.
13884 https://bugzilla.gnome.org/show_bug.cgi?id=766050
13886 2016-05-13 11:44:57 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13888 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
13889 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
13890 encoder: h264 : Use "tune=low-power" for enabling lowpower encode
13891 Remove the duplicate property "low-power-enc" and use the
13892 tune property for enabling low power encoding mode.
13893 https://bugzilla.gnome.org/show_bug.cgi?id=766050
13895 2016-05-11 12:06:38 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13897 * gst-libs/gst/vaapi/gstvaapiencoder.c:
13898 gstvaapiencoder:Use internal api to dervie configured VAEntrypoint
13899 https://bugzilla.gnome.org/show_bug.cgi?id=766050
13901 2016-05-11 12:05:36 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13903 * gst-libs/gst/vaapi/gstvaapiencoder.c:
13904 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
13905 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
13906 encoder: h264: Add support of low power/high performance encoding mode
13907 Added a new property "low-power-enc" for enabling low power
13908 encoding mode. Certain encoding tools may not be available
13909 with the VAEntrypointEncSliceLP.
13910 https://bugzilla.gnome.org/show_bug.cgi?id=766050
13912 2016-05-11 12:04:46 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13914 * gst-libs/gst/vaapi/gstvaapitypes.h:
13915 * gst-libs/gst/vaapi/gstvaapiutils.c:
13916 * gst-libs/gst/vaapi/gstvaapivalue.c:
13917 Add mapping for Macroblock level rate control (VA_RC_MB)
13919 2016-05-11 12:03:08 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13921 * gst-libs/gst/vaapi/gstvaapidisplay.c:
13922 gstvaapidisplay: Add VAEntrypointEncSliceLP support
13923 https://bugzilla.gnome.org/show_bug.cgi?id=766050
13925 2016-05-11 11:59:59 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
13927 * gst-libs/gst/vaapi/gstvaapiprofile.c:
13928 * gst-libs/gst/vaapi/gstvaapiprofile.h:
13929 gstvaapiprofile : Add VAEntrypointEncSliceLP definitions
13930 This is for implementations that supports low_power/high_performance
13931 variant for slice level encode.
13932 https://bugzilla.gnome.org/show_bug.cgi?id=766050
13934 2016-05-05 18:23:10 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13936 * gst/vaapi/gstvaapipostprocutil.c:
13937 vaapipostproc: negotiate frame size fixation
13938 Refactor _fixate_frame_size(). Now, instead of fixating the frame size only
13939 using the sink caps, also it use the next capsfilter.
13940 This code is a shameless copy of gst_video_scale_fixate_caps() from
13941 https://cgit.freedesktop.org/gstreamer/gst-plugins-base/tree/gst/videoscale/gstvideoscale.c?id=1.8.1#n634
13942 https://bugzilla.gnome.org/show_bug.cgi?id=758548
13944 2016-05-06 10:50:10 +0200 Scott D Phillips <scott.d.phillips@intel.com>
13946 * gst/vaapi/gstvaapipostprocutil.c:
13947 vaapipostproc: don't use GstVideoInfo for src caps
13948 Instead of using gst_video_info_to_caps () to generated the fixed src caps,
13949 this patch enables the first step for caps negotiation with a possible
13950 following caps filter.
13951 _get_preferred_caps() will traverse the possible src caps looking for the one
13952 wit the preferred feature and the preferred color format. Then the color
13953 format, the frame size and the frame rate are fixated.
13954 https://bugzilla.gnome.org/show_bug.cgi?id=758548
13956 2016-05-05 15:32:36 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13958 * gst/vaapi/gstvaapipostproc.c:
13959 * gst/vaapi/gstvaapipostprocutil.c:
13960 * gst/vaapi/gstvaapipostprocutil.h:
13961 vaapipostproc: move gst_vaapipostproc_fixate_srccaps()
13962 Move gst_vaapipostproc_fixate_srccaps() to gstvaapiposptprocutil.
13963 No functional changes.
13964 https://bugzilla.gnome.org/show_bug.cgi?id=758548
13966 2016-05-05 15:19:02 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13968 * gst/vaapi/gstvaapipostproc.c:
13969 vaapipostproc: simplify code
13970 Change a convoluted snippet to find the preferred color format in the peer
13972 https://bugzilla.gnome.org/show_bug.cgi?id=758548
13974 2016-05-05 15:16:02 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13976 * gst/vaapi/gstvaapipostproc.c:
13977 vaapipostproc: use othercaps for preferred caps
13978 Instead of the allowed_srcpad_caps variable, this patch uses the othercaps
13979 from fixate_caps() vmethod to find the preferred caps feature and color
13981 https://bugzilla.gnome.org/show_bug.cgi?id=758548
13983 2016-05-05 13:46:11 +0200 Scott D Phillips <scott.d.phillips@intel.com>
13985 * gst/vaapi/gstvaapipostproc.c:
13986 vaapipostproc: add fixate_caps() vmethod
13987 Instead of fixating the srcpad caps in transform_caps() vmethod, this patch
13988 implements the fixate_caps() vmethod and moves code around.
13989 https://bugzilla.gnome.org/show_bug.cgi?id=758548
13991 2016-05-05 12:07:59 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
13993 * gst/vaapi/Makefile.am:
13994 * gst/vaapi/gstvaapipostproc.c:
13995 * gst/vaapi/gstvaapipostprocutil.c:
13996 * gst/vaapi/gstvaapipostprocutil.h:
13997 vaapipostproc: set early properties restrictions
13998 When running transform_caps() vmethod, returning the srcpad caps, the caps are
13999 early restricted to the element properties set: width, height, format and
14001 A new file was added gstvaapipostprocutil.{c,h} where the utilities functions
14003 https://bugzilla.gnome.org/show_bug.cgi?id=758548
14005 2016-04-25 13:45:04 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14007 * gst/vaapi/gstvaapipostproc.c:
14008 vaapipostproc: log the caps transformation
14009 https://bugzilla.gnome.org/show_bug.cgi?id=758548
14011 2016-04-27 21:20:32 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14013 * gst/vaapi/gstvaapipostproc.c:
14014 vaapipostproc: no GLTextureUpload in sinkpad
14015 As the vaapipostproc does not process frames with the VideoGLTextureUpload
14016 meta, the feature is removed from the sink pad template.
14017 https://bugzilla.gnome.org/show_bug.cgi?id=765931
14019 2015-09-28 08:49:39 +0100 Julien Isorce <j.isorce@samsung.com>
14021 * gst/vaapi/gstvaapivideobufferpool.c:
14022 vaapibufferpool: do not create texture upload meta if dmabuf
14023 https://bugzilla.gnome.org/show_bug.cgi?id=755072
14025 2016-05-06 12:16:26 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14028 build: update git.mk
14030 2016-04-29 13:11:48 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14032 * gst/vaapi/gstvaapipluginbase.h:
14033 plugin: fix macro processor check
14034 Instead of #ifdef it should be used #if becasuse USE_GST_GL_HELPERS is always
14035 defined in config.h, but it would be 0 or 1 depending on the configure output.
14036 https://bugzilla.gnome.org/show_bug.cgi?id=765702
14038 2016-04-29 12:53:06 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14040 * tests/test-display.c:
14041 tests: display: guard possible unused variables
14042 https://bugzilla.gnome.org/show_bug.cgi?id=765702
14044 2016-04-29 12:48:44 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14050 * tests/simple-decoder.c:
14051 * tests/test-decode.c:
14052 * tests/test-display.c:
14053 * tests/test-filter.c:
14054 * tests/test-h264.c:
14055 * tests/test-jpeg.c:
14056 * tests/test-mpeg2.c:
14057 * tests/test-mpeg4.c:
14058 * tests/test-subpicture-data.c:
14059 * tests/test-subpicture.c:
14060 * tests/test-surfaces.c:
14061 * tests/test-textures.c:
14062 * tests/test-vc1.c:
14063 * tests/test-windows.c:
14064 tests: inforce gstreamer code-style
14066 2016-04-27 17:10:26 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14068 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
14069 encoder: h265: Enable cu_qp_delta_enabled_flag for CBR
14070 It seems driver requires enablement of cu_qp_delta_enabled_flag
14071 for modifying QP values to controll the CBR mode bitrate.
14072 https://bugzilla.gnome.org/show_bug.cgi?id=749852
14074 2016-04-27 17:06:09 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14076 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
14077 encoder: h265: Add CBR Encoding support
14078 https://bugzilla.gnome.org/show_bug.cgi?id=749852
14080 2015-11-27 05:09:10 +0000 Julien Isorce <j.isorce@samsung.com>
14082 * gst-libs/gst/vaapi/gstvaapisurface.c:
14083 gstvaapisurface: explicitely clear TILING flag if dmabuf
14084 https://bugzilla.gnome.org/show_bug.cgi?id=755072
14086 2015-10-04 23:44:16 +0100 Julien Isorce <j.isorce@samsung.com>
14088 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
14089 gstvaapisurface_drm: release image when done
14090 Otherwise intel-vaapi-driver will fail to process the exported surface because
14091 it will find it is currently derived, so considered as busy.
14092 https://bugzilla.gnome.org/show_bug.cgi?id=755072
14094 2015-09-26 06:25:12 +0100 Julien Isorce <j.isorce@samsung.com>
14096 * gst/vaapi/gstvaapipostproc.c:
14097 vaapipostproc: already have a surface proxy if dmabuf
14098 https://bugzilla.gnome.org/show_bug.cgi?id=755072
14100 2016-03-11 08:58:51 +0000 Julien Isorce <j.isorce@samsung.com>
14102 * gst/vaapi/gstvaapipostproc.c:
14105 2016-04-21 15:14:47 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14107 * gst/vaapi/gstvaapidecode.c:
14108 * gst/vaapi/gstvaapipluginbase.c:
14109 * gst/vaapi/gstvaapipluginbase.h:
14110 vaapidecode: search driver in whitelist
14111 If the backend driver vendor string is not in a white-list, and the
14112 environment variable GST_VAAPI_ALL_DRIVERS is not set either, the decoder will
14113 change it state from NULL to READY, hence the auto-plug mechanism will look
14114 for another decoder.
14115 This patch assumes the GstContext has already being shared along the pipeline
14116 and the element has a valid GstVaapiDisplay instance.
14117 https://bugzilla.gnome.org/show_bug.cgi?id=764673
14119 2016-04-21 12:57:30 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14121 * gst/vaapi/gstvaapidecode.c:
14122 * gst/vaapi/gstvaapiencode.c:
14123 * gst/vaapi/gstvaapipluginbase.c:
14124 * gst/vaapi/gstvaapipluginbase.h:
14125 * gst/vaapi/gstvaapipostproc.c:
14126 * gst/vaapi/gstvaapisink.c:
14127 plugins: rework set_context() vmethod definition
14128 In bug 757598 was added the set_context() vmethod chain up in
14129 GstVaapiPluginBase. But it is buggy, since the parent_class address is
14130 assigned to the last element which called gst_vaapi_plugin_base_class_init().
14131 No error has shown up since none of the element's base classes redefined
14132 set_context() vmethod from GstElement, so always the correct function was
14133 called. Still this code is wrong and this patch make it right.
14134 Since set_context() is the same code, a macro is used to implement that code
14135 in all the gst-vaapi elements.
14136 https://bugzilla.gnome.org/show_bug.cgi?id=765368
14138 2016-04-15 17:57:25 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14140 * gst/vaapi/gstvaapipostproc.c:
14141 vaapipostproc: resize if negotiated and allocation caps are different
14142 Since commit 859a2b2, in vaapidecode, allocation query can be different from
14143 the negotiated caps.
14144 When connecting the vaapidecoder to the vaapipostprocessor, the last one will
14145 resize the frame to the negotiated, if and only if, some other parameter is
14146 activated to avoid the passthrough. If it is not, the surface won't be mapped
14147 into a image. If not, the image won't be resized and the output buffer would be
14149 This patch will break the passthrough if the allocation query is different
14150 from the negotiation caps, forcing the resizing.
14151 https://bugzilla.gnome.org/show_bug.cgi?id=765095
14153 2016-04-05 13:09:37 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14155 * gst/vaapi/gstvaapidecode.c:
14156 vaapidecode: improve code readability
14157 No functional changes.
14159 2016-04-05 13:37:40 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14161 * gst/vaapi/gstvaapidecode.c:
14162 vaapidecode: move GstCapsFeatures near to its use
14163 Move the handling of the GstCapsFeatures just after it is used, in order to
14164 avoid handling its memory.
14166 2016-04-05 13:07:14 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14168 * gst/vaapi/gstvaapidecode.c:
14169 vaapidecode: caps negotiation checks
14170 Check that GLUploadTexture is not negotatiated if gstreamer-vaapi is not
14171 compiled with GL support.
14173 2016-03-10 16:43:16 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14175 * gst/vaapi/gstvaapidecode.c:
14176 * gst/vaapi/gstvaapipluginbase.c:
14177 * gst/vaapi/gstvaapipluginbase.h:
14178 * gst/vaapi/gstvaapipostproc.c:
14179 plugins: remove param in gst_vaapi_plugin_base_decide_allocation()
14181 2016-03-10 16:42:04 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14183 * gst/vaapi/gstvaapidecode.c:
14184 vaapidecode: bail early if not caps in decide_allocation()
14186 2016-03-29 14:17:54 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14188 * gst/vaapi/gstvaapidecode.c:
14189 * gst/vaapi/gstvaapipluginbase.c:
14190 * gst/vaapi/gstvaapipluginutil.c:
14191 * gst/vaapi/gstvaapipluginutil.h:
14192 * gst/vaapi/gstvaapipostproc.c:
14193 plugin: use allowed caps filter from element
14194 Instead of using the srcpad template caps for filtering the peer caps, the
14195 function gst_vaapi_find_preferred_caps_feature(), now receives a new parameter
14196 for the element's allowed caps.
14197 With this modification, the vaapipostproc element simplifies a bit its code.
14198 https://bugzilla.gnome.org/show_bug.cgi?id=765223
14200 2016-04-18 17:28:51 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14202 * gst/vaapi/gstvaapidecode.c:
14203 * gst/vaapi/gstvaapipluginbase.c:
14204 * gst/vaapi/gstvaapipluginutil.c:
14205 * gst/vaapi/gstvaapipluginutil.h:
14206 * gst/vaapi/gstvaapipostproc.c:
14207 plugin: remove function parameter
14208 The native format parameter in gst_vaapi_find_preferred_caps_feature() can be
14209 saved if the out format is used for both: in and out. Thus the code is more
14211 https://bugzilla.gnome.org/show_bug.cgi?id=765223
14213 2016-04-18 17:17:58 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14215 * gst/vaapi/gstvaapidecode.c:
14216 vaapidecode: warns if driver will do color conversions
14217 If the downstream feature is system memory, the surface has to be mapped,
14218 hence a warning message is logged saying that the driver has to do color
14219 conversions. This might be troublesome because not all the color conversion
14220 combinations are supported by the VA-API drivers, and there is not a reliable
14221 way to know them before hand.
14222 https://bugzilla.gnome.org/show_bug.cgi?id=765223
14224 2016-03-29 13:28:27 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14226 * gst/vaapi/gstvaapipluginutil.c:
14227 plugin: honour negotiated format
14228 Instead of setting the requested format by the caller, the function
14229 gst_vaapi_find_preferred_caps_feature() now returns, in the output parameter,
14230 the negotiated format.
14231 A new helper function was added: gst_vaapi_find_preferred_format(), which,
14232 given the format list from the negotiated caps, will choose the best one, if
14233 possible, given the native format.
14234 https://bugzilla.gnome.org/show_bug.cgi?id=765223
14236 2016-03-28 19:26:02 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14238 * gst/vaapi/gstvaapipluginutil.c:
14239 plugin: simplify caps feature selection
14240 This patch simplifies the function gst_vaapi_find_preferred_caps_feature().
14241 Instead of intersecting custom caps to find the preferred feature, the peer
14242 caps are traversed in order to find the preferred feature, according to an
14243 ordered feature priority list.
14244 In the case of GLTextureUploadMeta, the colour format is computed using
14245 GstVideoInfo of the selected fixed caps.
14246 https://bugzilla.gnome.org/show_bug.cgi?id=765223
14248 2016-03-31 16:39:08 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14250 * gst/vaapi/gstvaapidecode.c:
14251 vaapidecode: use macros for GstVideoInfo
14252 Instead of accessing directly to the members of the structure, use the macros.
14254 2016-04-14 17:02:23 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14256 * gst/vaapi/gstvaapidecode.c:
14257 vaapidecode: refactor is_display_resolution_changed()
14258 Make the comparisons more readable and simple.
14259 https://bugzilla.gnome.org/show_bug.cgi?id=764316
14261 2016-04-14 16:43:07 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14263 * gst/vaapi/gstvaapidecode.c:
14264 * gst/vaapi/gstvaapidecode.h:
14265 vaapidecode: keep only display_{width,height}
14266 Instead of keeping the structure GstVideoInfo when we are using its width and
14267 height, we only keep these two guints.
14268 https://bugzilla.gnome.org/show_bug.cgi?id=764316
14270 2016-04-14 16:31:34 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14272 * gst/vaapi/gstvaapidecode.c:
14273 vaapidecode: decoded_info is valid at src caps update
14274 As decoded_info is assured to be valid when gst_vaapidecode_update_src_caps()
14275 is called, then we don't need to verify or replace it with the sinkpad info
14277 https://bugzilla.gnome.org/show_bug.cgi?id=764316
14279 2016-04-14 16:22:25 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14281 * gst/vaapi/gstvaapidecode.c:
14282 vaapidecode: always a valid format in decoded_info
14283 Always set a valid format in decoded_info class variable.
14284 https://bugzilla.gnome.org/show_bug.cgi?id=764316
14286 2016-03-25 15:31:28 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14288 * gst/vaapi/gstvaapidecode.c:
14289 vaapidecode: code style fixes
14290 No functional changes.
14291 https://bugzilla.gnome.org/show_bug.cgi?id=764316
14293 2016-04-14 16:10:02 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14295 * gst/vaapi/gstvaapidecode.c:
14296 vaapidecode: init {decoded,display}_info at open()
14297 It is required to initialize {decoded,display}_info variables when the decoder
14298 is open, not only at instance initialization.
14299 https://bugzilla.gnome.org/show_bug.cgi?id=764316
14301 2016-03-28 15:30:28 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14303 * gst/vaapi/gstvaapidecode.c:
14304 * gst/vaapi/gstvaapidecode.h:
14305 vaapidecode: remove spurious class variables
14306 active, do_pool_renego and do_outstate_renego class variables were used to
14307 indicate when negotiate downstream once, but now that each time a new surface
14308 resolution is pop out a renegotation verified, these variable are not required
14310 https://bugzilla.gnome.org/show_bug.cgi?id=764316
14312 2016-04-14 15:46:32 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14314 * gst/vaapi/gstvaapidecode.c:
14315 * gst/vaapi/gstvaapipluginbase.c:
14316 * gst/vaapi/gstvaapipluginbase.h:
14317 * gst/vaapi/gstvaapipostproc.c:
14318 remove custom allocation query
14319 When resolving bug 753914, a custom allocation query was added, overlapping
14320 the responsibilities of GstVideoDecoder.
14321 But with the merge of the patches from bug 764421 this overlapping was not
14322 required anymore. This patch restores this situation setting the
14323 allocation_caps in the GstVideoCodecState when needed.
14324 https://bugzilla.gnome.org/show_bug.cgi?id=764316
14326 2016-04-14 10:04:47 +0100 Julien Isorce <j.isorce@samsung.com>
14329 Automatic update of common submodule
14330 From 6f2d209 to ac2f647
14332 2016-04-13 15:44:20 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14334 * gst/vaapi/gstvaapidecode.c:
14335 * gst/vaapi/gstvaapidecodebin.c:
14336 * gst/vaapi/gstvaapipostproc.c:
14337 plugins: disable GL_TEXTURE_UPLOAD if no EGL/GLX
14338 The plugins should not expose the feature meta:GstVideoGLTextureUploadMeta in
14339 their caps templates if they were not compiled either with GLX or EGL support.
14341 2016-04-13 20:33:32 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14343 * gst/vaapi/gstvaapipluginutil.c:
14344 plugins: fix compilation when EGL/GLX is disabled
14345 The compiler might complain of gst_vaapi_create_display_from_handle() being
14346 unused if both EGL and GLX are disabled. This patch avoid that compilation
14349 2016-04-13 14:09:00 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
14351 * gst-libs/gst/vaapi/gstvaapicodedbufferpool.c:
14352 libs: fix deleting a GstVaapiCodedBufferPool object
14353 Call gst_vaapi_video_pool_finalize() in coded_buffer_pool_finalize().
14354 Otherwise it is not called when the pool is destroyed and all objects
14355 referenced by the GstVaapiVideoPool are never released.
14356 https://bugzilla.gnome.org/show_bug.cgi?id=764993
14358 2016-04-07 18:03:42 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14360 * gst-libs/gst/vaapi/gstvaapisurface.c:
14361 surface: destroy derived image
14362 If gst_vaapi_image_new_with_image() fails, the created derived image should be
14363 destroyed, otherwise the surface cannot be processed because is being used.
14364 https://bugzilla.gnome.org/show_bug.cgi?id=764607
14366 2016-03-18 20:00:52 -0300 Thiago Santos <thiagoss@osg.samsung.com>
14368 * gst/vaapi/gstvaapidecode.c:
14369 vaapidecode: add stop function
14370 Clear any status on the current stream:
14371 stored frames, caps and decoder configuration
14372 https://bugzilla.gnome.org/show_bug.cgi?id=763460
14374 2016-04-01 14:00:28 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14376 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
14377 decoder: vp9: Align with the ABI changes in vp9 codecparser
14378 The subsampling_x, subsampling_y, bit_depth, color_space and color_range
14379 fileds are moved from GstVp9FrameHdr to the global GstVp9Parser structure.
14380 These fields are only present in keyframe or intra-only frame, no need to
14381 duplicate them for inter-frames.
14382 https://bugzilla.gnome.org/show_bug.cgi?id=764082
14384 2016-04-01 13:59:59 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14386 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
14387 decoder: vp9 : Add 10bit decoding support (Profile2)
14388 https://bugzilla.gnome.org/show_bug.cgi?id=764082
14390 2016-04-01 13:57:45 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14392 * gst-libs/gst/vaapi/gstvaapiprofile.c:
14393 * gst-libs/gst/vaapi/gstvaapiprofile.h:
14394 gstvaapiporfile: Add more VP9 profile definitions
14395 https://bugzilla.gnome.org/show_bug.cgi?id=764082
14397 2016-02-03 20:34:49 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14401 build: possibility to disable tests
14402 The configuration option --disable-examples will disable the compilation of
14403 the sample apps in tests/ directory.
14405 2016-03-29 14:25:55 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14407 * gst/vaapi/gstvaapiencode_h264.c:
14408 * gst/vaapi/gstvaapiencode_h265.c:
14409 * gst/vaapi/gstvaapiencode_jpeg.c:
14410 * gst/vaapi/gstvaapiencode_mpeg2.c:
14411 * gst/vaapi/gstvaapiencode_vp8.c:
14412 * gst/vaapi/gstvaapipluginutil.h:
14413 * gst/vaapi/gstvaapisink.c:
14414 unify caps template for VAAPI encoders and decoders
14415 There is no difference in VAAPI surface caps between encoders and decoders.
14416 Thus, the patch makes a simplification by removing encoders specific caps and
14417 shares the same definition of VAAPI surfaces caps for all the elements.
14419 2016-03-10 17:42:55 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14421 * gst/vaapi/gstvaapidecode.c:
14422 vaapidecode: move gst_vaapidecode_negotiate() code
14423 With it we can remove a function declaration, making the code a bit
14425 https://bugzilla.gnome.org/show_bug.cgi?id=764316
14427 2016-03-29 13:50:00 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14429 * docs/plugins/gstreamer-vaapi-plugins-docs.xml.in:
14430 * docs/plugins/gstreamer-vaapi-plugins-sections.txt:
14431 * gst/vaapi/gstvaapidecode.c:
14432 docs: generate vaapijpegdec documentation
14433 https://bugzilla.gnome.org/show_bug.cgi?id=764314
14435 2016-03-30 14:37:21 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14437 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
14438 decoder: h265: Fix offset calculation when there is more than one vps/sps/pps present in codec_data
14439 The array_completeness, reserved bit and num_nal_units fields
14440 in HEVCDecoderConfigurationRecord will be present for each VPS/SPS/PPS array list,
14441 but not for each occurance of similar headers.
14442 https://bugzilla.gnome.org/show_bug.cgi?id=764274
14444 2016-03-29 15:34:38 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14446 * gst-libs/gst/vaapi/video-format.c:
14447 video-format: Keep the HW order preference while mapping to GstVideoFormats
14449 2016-03-29 15:02:46 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14451 * gst/vaapi/gstvaapidecode.c:
14452 * gst/vaapi/gstvaapipluginbase.c:
14453 * gst/vaapi/gstvaapipluginutil.c:
14454 * gst/vaapi/gstvaapipluginutil.h:
14455 * gst/vaapi/gstvaapivideomemory.c:
14456 gst/vaapi: keep precedence for NV12 over I420
14457 Use NV12 as default "assumption" format all over.
14458 NV12 is the default high priority format used my most of the
14461 2016-03-29 14:34:37 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14463 * gst/vaapi/gstvaapidecode.c:
14464 vaapidecode: Use video format derived from decoded surface as default src pad format
14465 Use the surface format derived from first decoded surface
14466 to negotiate the downstream video format capabilities.
14467 https://bugzilla.gnome.org/show_bug.cgi?id=759181
14469 2016-03-29 14:34:00 +0300 Scott D Phillips <scott.d.phillips@intel.com>
14471 * gst-libs/gst/vaapi/gstvaapiimage.c:
14472 * gst-libs/gst/vaapi/video-format.c:
14473 * gst/vaapi/gstvaapidecode.c:
14474 * gst/vaapi/gstvaapipluginutil.h:
14475 * gst/vaapi/gstvaapisink.c:
14476 Add P010 video format support
14477 The P010 video format is the native format used by the vaapi intel driver
14478 for HEVCMain10 decode . Add support for planes and images of this video format.
14479 https://bugzilla.gnome.org/show_bug.cgi?id=759181
14481 2016-03-27 09:11:00 +0000 Stephen <elfarto@elfarto.com>
14483 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
14484 libs: x11: allows 30-bit colour depth
14485 The colour depth is clamped to 24 when it is not equal {15,16,24,32}. But this
14486 fails with the NVIDIA binary driver as it doesn't advertise a TrueColor visual
14487 with a depth of 24 (only 30 and 32). Allowing the depth to be 30, lets everything
14489 https://bugzilla.gnome.org/show_bug.cgi?id=764256
14491 2016-03-28 13:13:56 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14493 * gst/vaapi/gstvaapivideometa_texture.c:
14494 gltextureupload: use an array for texture type
14495 Instead of using a single value for the texture type, use an array with 4
14496 elements, just as the GstVideoGLTextureUploadMeta, avoiding a buffer
14498 https://bugzilla.gnome.org/show_bug.cgi?id=764231
14500 2016-03-24 15:09:43 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14502 * gst/vaapi/gstvaapidecode.c:
14503 * gst/vaapi/gstvaapipluginbase.c:
14504 vaapidecode: Fix decide_allocation handling
14505 Set the already configured pool in decide_allocation query
14506 in cases where pool renegotiation is not required.
14507 https://bugzilla.gnome.org/show_bug.cgi?id=753914
14509 2016-03-24 15:09:15 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14511 * gst/vaapi/gstvaapidecode.c:
14512 vaapidecode: Derive and save the decoded surface format
14513 After the decoding of first frame, try to extract the exact
14514 decoded surface format using vaDeriveImage and keep this
14515 as the format in decoded_info.
14516 https://bugzilla.gnome.org/show_bug.cgi?id=753914
14518 2016-03-24 15:08:50 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14520 * gst/vaapi/gstvaapidecode.c:
14521 * gst/vaapi/gstvaapidecode.h:
14522 * gst/vaapi/gstvaapipluginbase.c:
14523 * gst/vaapi/gstvaapipluginbase.h:
14524 * gst/vaapi/gstvaapipostproc.c:
14525 Make vaapidecode to advertise the cropped values in srcpad, but negotiate pool only if needed
14526 -- Maintaing decoded surface resoluton and actual display resoultion separately
14527 -- Before pushing every frames downstream, check for the requirement of pool negoation and
14528 output_state negotiation: This is needed to avoid multiple issuses with cropping,
14529 multi-resoluton video handling, more complex multi resolution decode scenarios for vp9decode,
14530 possible wrong behaviour from upstream element to report uncropped values etc. Due to these reasons,
14531 We can't just reliably use the resolution change notification from libgstvaapi for pool renegotiation too.
14532 This is slight overhead, but safe enough. Optimization could be possible though.
14533 https://bugzilla.gnome.org/show_bug.cgi?id=753914
14535 2016-03-24 15:08:27 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14537 * gst/vaapi/gstvaapidecode.c:
14538 vaapidecode: Delay the output format setting until we have a decoded surface
14539 This will help to consoidate the out caps negotiation to a single place,
14540 which will make the code simpler, allows to get the exact decoded format
14541 if needed and the selected chroma type too.
14542 https://bugzilla.gnome.org/show_bug.cgi?id=753914
14544 2016-03-24 13:36:24 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14546 * gst/vaapi/gstvaapidecode.h:
14547 vaapidecode: GST_VAAPIDECODE macro is a cast
14548 This patch is the continuation of commit 1e1d3b1d because the function
14549 gst_vaapidecode_get_type() got undefined since then.
14550 Now, the macro GST_VAAPIDECODE is a simple cast to the GstVaapiDecode
14551 structure. The rest of the GObject handling macros were deleted too.
14553 2016-03-24 13:34:18 +0200 Sebastian Dröge <sebastian@centricular.com>
14556 Back to development
14558 === release 1.8.0 ===
14560 2016-03-24 13:11:05 +0200 Sebastian Dröge <sebastian@centricular.com>
14565 * gstreamer-vaapi.doap:
14568 === release 1.7.91 ===
14570 2016-03-15 12:39:20 +0200 Sebastian Dröge <sebastian@centricular.com>
14575 * gstreamer-vaapi.doap:
14578 2016-03-11 17:44:07 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14580 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
14581 decoder: h265: Fix offset calculation in codec_data parsing
14582 https://bugzilla.gnome.org/show_bug.cgi?id=762922
14584 2016-03-09 20:26:31 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14586 * gst/vaapi/gstvaapidecodebin.c:
14587 vaapidecodebin: don't handle jpeg decoding
14588 As JPEG decoder has been split and demoted, it cannot be handled by
14590 Added a fixme comment regarding the future removal of vaapidecode.
14591 https://bugzilla.gnome.org/show_bug.cgi?id=734093
14593 2016-03-09 20:25:08 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14595 * gst/vaapi/gstvaapidecode.c:
14596 vaapidecode: split out jpeg decoder
14597 Split, as a different element, the JPEG decoder.
14598 https://bugzilla.gnome.org/show_bug.cgi?id=734093
14600 2016-03-09 18:41:49 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14602 * gst/vaapi/gstvaapi.c:
14603 * gst/vaapi/gstvaapidecode.c:
14604 * gst/vaapi/gstvaapidecode.h:
14605 vaapidecode: register decoder with internal GType
14606 Don't expose the the vaapidecode GType, instead expose a function
14607 which will register element.
14608 This is the first step to split the decoder by codecs.
14609 https://bugzilla.gnome.org/show_bug.cgi?id=734093
14611 2016-03-10 12:47:49 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14613 * gst/vaapi/gstvaapidecode.c:
14614 vaapidecode: remove unused function declaration
14615 There is no need to pre-declare gst_vaapidecode_update_sink_caps(). And fixed
14616 code-style of the other pre-declared functions.
14618 2016-03-09 18:58:13 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14620 * docs/plugins/Makefile.am:
14621 build: git ignore gtkdoc generated files
14623 2016-03-09 18:55:39 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14626 build: handle git ignore in m4 directory
14628 2016-03-04 20:17:54 -0300 Thiago Santos <thiagoss@osg.samsung.com>
14630 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
14631 vaapidecoder_h265: plug leak of h265 parsing info
14632 If something goes wrong while parsing, the info object is
14634 https://bugzilla.gnome.org/show_bug.cgi?id=763121
14636 2016-03-04 20:17:20 -0300 Thiago Santos <thiagoss@osg.samsung.com>
14638 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
14639 vaapidecoder_h264: plug leak of h264 parsing info
14640 If something goes wrong while parsing, the info object is
14642 https://bugzilla.gnome.org/show_bug.cgi?id=763121
14644 2016-03-09 11:03:28 +0900 Vineeth TM <vineeth.tm@samsung.com>
14646 * gst/vaapi/gstvaapipluginutil.c:
14647 plugins: fix gstgl and vaapi memory leaks
14648 1\ Unref gl_display and gl_window as soon they are not needed.
14649 2\ Remove an unneeded display type check, since is handled by
14650 gst_vaapi_created_display_from_handle()
14651 3\ Unref vaapi's display if the display cannot be bind to a GL API.
14652 Modified-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14653 https://bugzilla.gnome.org/show_bug.cgi?id=763354
14655 2016-03-09 14:13:24 +0900 Vineeth TM <vineeth.tm@samsung.com>
14657 * gst-libs/gst/vaapi/gstvaapidisplay.c:
14658 vaapidisplay: Fix uninitialized value error for VA attribute
14659 https://bugzilla.gnome.org/show_bug.cgi?id=763362
14661 2016-03-08 10:47:56 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14663 * gst/vaapi/gstvaapidecode.c:
14664 * gst/vaapi/gstvaapiencode.c:
14665 plugins: proxy information from downstream caps
14666 Propagate to upstream the downstream information, such as fps, par, etc.
14667 This will fix several "getcaps" critical warnings in gst-validate.
14668 https://bugzilla.gnome.org/show_bug.cgi?id=763300
14670 2016-03-04 10:51:42 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14672 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
14673 decoder: vp9: Assign values for profile and bit_depth from frame header
14674 bit_depth field has added only in VA-API 0.39.0, added version check.
14676 2016-03-04 09:12:13 +0200 Sebastian Dröge <sebastian@centricular.com>
14679 build: Dist gstreamer-vaapi.doap and configure.ac/autogen.sh
14680 https://bugzilla.gnome.org/show_bug.cgi?id=763067
14682 === release 1.7.90 ===
14684 2016-03-01 19:23:51 +0200 Sebastian Dröge <sebastian@centricular.com>
14689 * gstreamer-vaapi.doap:
14692 2016-03-01 16:14:47 +0200 Sebastian Dröge <sebastian@centricular.com>
14695 configure: Use AG_GST_SET_PACKAGE_RELEASE_DATETIME_WITH_NANO to set release date
14697 2016-03-01 11:35:49 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14699 * gst/vaapi/gstvaapidecode.c:
14700 vaapidecode: intersect with filter in getcaps()
14701 In commit 6034734d I forgot to add the caps filter intersection in the
14702 getcaps() vmethod generating a regression when a capsfilter is set in the
14704 This commit adds the caps filter intersection.
14706 2016-02-29 11:55:27 +0200 Lim Siew Hoon <siew.hoon.lim@intel.com>
14708 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
14709 Add memset to initialize value for VAEncSliceParameterBufferJPEG
14710 https://bugzilla.gnome.org/show_bug.cgi?id=762850
14712 2016-02-26 12:42:46 +0200 Sebastian Dröge <sebastian@centricular.com>
14715 Automatic update of common submodule
14716 From b64f03f to 6f2d209
14718 2016-02-24 12:36:33 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14723 build: add m4 directory
14724 Instead of rely on the automatic creation of m4 directory by aclocal, we
14725 already control it. Later we could create our own m4 scripts in order to
14726 unclutter configure.ac
14727 https://bugzilla.gnome.org/show_bug.cgi?id=762528
14729 2016-02-23 10:55:02 +0200 Scott D Phillips <scott.d.phillips@intel.com>
14731 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
14732 decoder: hevc: Fill dependent slice segment headers while parsing
14733 Copy the data into the dependent slice segment header from the
14734 corresponding independent slice segment header during parsing.
14735 Previously the reference to the "previous" independent header was
14736 held through the parsing phase and then dereferenced during the
14737 decoding phase. This caused all dependent headers to be populated
14738 with the data of the AU's last independent header instead of the
14739 proper corresponding header.
14740 https://bugzilla.gnome.org/show_bug.cgi?id=762352
14742 - Reworded commit message
14744 2016-02-17 13:43:48 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14746 * gst/vaapi/gstvaapidecode.c:
14747 vaapidecode: use video decoder getcaps()
14748 The usage of getcaps() vmethod is preferred than to handle manually the sink's
14750 In order to avoid function declarations, this patch moves the class_init()
14751 method to the end of the file.
14753 2016-02-17 12:51:45 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14755 * gst/vaapi/gstvaapipluginbase.c:
14756 plugins: remove deprecated code
14757 Since we are only supporting current GStreamer version, since 1.3
14758 gst_buffer_pool_config_add_option() checks if the option to add is
14759 already set. There is no need to do it ourselves.
14761 2016-02-19 19:03:44 -0300 Thiago Santos <thiagoss@osg.samsung.com>
14763 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
14764 vaapidecoder_h265: fix parsing of NALU aligned data
14765 Don't assume the whole buffer is a single NAL, instead look for the
14766 next start code in case there are multiple NALs per buffer.
14767 https://bugzilla.gnome.org/show_bug.cgi?id=762328
14769 2016-02-19 11:10:25 -0300 Thiago Santos <thiagoss@osg.samsung.com>
14771 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
14772 vaapidecoder_h264: fix parsing of NALU aligned data
14773 Don't assume the whole buffer is a single NAL, instead look for the
14774 next start code in case there are multiple NALs per buffer.
14775 https://bugzilla.gnome.org/show_bug.cgi?id=762328
14777 2016-02-18 10:13:53 +0900 Vineeth TM <vineeth.tm@samsung.com>
14779 * gst/vaapi/gstvaapisink.c:
14780 vaapisink: Fix event,pad,structure memory leaks
14781 https://bugzilla.gnome.org/show_bug.cgi?id=762229
14783 2016-02-17 15:40:54 +0200 Lim Siew Hoon <siew.hoon.lim@intel.com>
14785 * gst/vaapi/gstvaapipluginbase.c:
14786 Add icamerasrc as dmabuf capable peer element
14787 icamerasrc is another gstreamer plugin using to capture RAW
14788 frames from camera device. It is based on libcamhal library.
14789 There are some properties available to control icamera behavior.
14790 Signed-off-by: Lim Siew Hoon <siew.hoon.lim@intel.com>
14791 Tested & Reviewed: Zhu Haiyang <haiyang.zhu@intel.com>
14792 https://bugzilla.gnome.org/show_bug.cgi?id=759481
14793 Fixme: This is the similar workaround we done for v4l2src.
14794 The workaround will be removed once we fix #755072
14796 2016-02-17 17:15:28 +0900 Vineeth TM <vineeth.tm@samsung.com>
14798 * gst/vaapi/gstvaapipluginbase.c:
14799 vaapipluginbase: Fix structure memory leak
14800 config structure is not being freed in all cases
14801 https://bugzilla.gnome.org/show_bug.cgi?id=762172
14803 2016-02-17 17:20:08 +0900 Vineeth TM <vineeth.tm@samsung.com>
14805 * gst/vaapi/gstvaapidecode.c:
14806 vaapidecode: Fix videocodec state memory leak
14807 When state is not NULL and either width/height of video info is 0, then state leaks
14808 https://bugzilla.gnome.org/show_bug.cgi?id=762173
14810 2016-02-16 15:44:48 +0000 Tim-Philipp Müller <tim@centricular.com>
14812 * gst/vaapi/gstvaapisink.c:
14813 vaapisink: post message for application for unhandled keyboard/mouse events
14814 Makes (most) keyboard shortcuts work in gst-play-1.0 when
14815 the video window has focus.
14817 2016-02-16 08:48:43 +0900 Vineeth TM <vineeth.tm@samsung.com>
14819 * gst/vaapi/gstvaapidecode.c:
14820 vaapidecode: Fix capsfeature memory leak
14821 https://bugzilla.gnome.org/show_bug.cgi?id=762116
14823 2016-02-16 08:15:40 +0900 Vineeth TM <vineeth.tm@samsung.com>
14825 * gst/vaapi/gstvaapisink.c:
14826 vaapisink: Fix capsfeature memory leak
14827 caps feature allocated is not being freeing in some cases
14828 https://bugzilla.gnome.org/show_bug.cgi?id=762111
14830 2016-02-16 15:09:01 +0200 Sebastian Dröge <sebastian@centricular.com>
14833 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
14834 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
14835 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
14836 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
14837 * gst-libs/gst/vaapi/gstvaapifilter.c:
14838 * gst/vaapi/gstvaapidecodebin.c:
14839 * gst/vaapi/gstvaapipluginutil.c:
14840 vaapi: Fix various compiler warnings and disable -Wredundant-decls for now
14842 2016-02-16 14:36:39 +0200 Sebastian Dröge <sebastian@centricular.com>
14845 configure: Fix setting of extra compiler warning flags
14847 2016-02-15 18:00:49 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14849 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
14850 libs: fix build error
14851 gst_vaapi_buffer_proxy_{acquire_handle,release_handle,finalize,class}
14852 functions are used only when libva's API version is greater than 0.36.0
14853 This patch guards those functions completely rather than just their
14854 content. The patch is a continuation of commit 38f8fea4
14855 Original-patch-by: Vineeth TM <vineeth.tm@samsung.com>
14856 https://bugzilla.gnome.org/show_bug.cgi?id=762055
14858 2016-02-15 10:01:54 +0900 Vineeth TM <vineeth.tm@samsung.com>
14860 * tests/simple-encoder.c:
14861 tests: simple-encoder: fix build error
14862 argument mismatch of gsize with 'long unsigned int'
14863 https://bugzilla.gnome.org/show_bug.cgi?id=762055
14865 2016-02-04 10:16:00 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
14867 * gst/vaapi/gstvaapisink.c:
14868 vaapisink: Fix wrong caps advertising
14869 The get_caps() should only report the supported formats.
14870 https://bugzilla.gnome.org/show_bug.cgi?id=761147
14872 2016-02-05 18:11:29 -0300 Thiago Santos <thiagoss@osg.samsung.com>
14875 Automatic update of common submodule
14876 From e97c9bb to b64f03f
14878 2016-02-03 19:07:40 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14880 * docs/plugins/gstreamer-vaapi-plugins-docs.xml.in:
14881 * docs/plugins/gstreamer-vaapi-plugins-sections.txt:
14882 * gst/vaapi/gstvaapi.c:
14883 * gst/vaapi/gstvaapiencode_h264.c:
14884 * gst/vaapi/gstvaapiencode_h265.c:
14885 * gst/vaapi/gstvaapiencode_jpeg.c:
14886 * gst/vaapi/gstvaapiencode_mpeg2.c:
14887 * gst/vaapi/gstvaapiencode_vp8.c:
14888 rename encoders to vaapi{codec}enc
14889 Trying to comply with GStreamer's element names, this patch renames the
14890 encoders using the name format vaapi{codec}enc.
14891 In this way, the plugin documentation is linked correctly.
14892 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14894 2016-02-03 18:42:36 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14897 Use new AG_GST_ARG_ENABLE_EXTRA_CHECKS #define
14898 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14900 2016-02-03 18:02:21 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14903 build: remove vp9 parser check
14904 Since the VP9 parser was added in gst-plugins-bad 1.7.1 we can remove safely
14905 the check of the parser, as we did for the others.
14906 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14908 2016-02-03 17:06:48 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14912 Back to development
14913 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14915 === release 1.6.0 ===
14917 2016-02-03 16:53:41 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14922 2016-02-03 16:45:18 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14924 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
14925 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
14926 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
14927 libs: fix compiler warnings
14928 After setting the release flags, the compiler warns about a couple
14929 initialized variables.
14930 Also marked a couple of set variables as unused, because they are only
14931 used for assertion.
14932 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14934 2016-01-29 20:41:27 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14936 * docs/plugins/gstreamer-vaapi-plugins-docs.xml.in:
14937 * docs/plugins/gstreamer-vaapi-plugins-sections.txt:
14938 * gst/vaapi/gstvaapidecode.c:
14939 * gst/vaapi/gstvaapidecodebin.c:
14940 * gst/vaapi/gstvaapiencode_h264.c:
14941 * gst/vaapi/gstvaapiencode_h265.c:
14942 * gst/vaapi/gstvaapiencode_jpeg.c:
14943 * gst/vaapi/gstvaapiencode_mpeg2.c:
14944 * gst/vaapi/gstvaapiencode_vp8.c:
14945 * gst/vaapi/gstvaapipostproc.c:
14946 * gst/vaapi/gstvaapisink.c:
14947 docs: update plugin documentation
14948 Update all the documentation of elements of the vaapi plugin.
14949 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14951 2016-01-29 15:39:09 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14955 * docs/Makefile.am:
14956 * docs/plugins/Makefile.am:
14957 * docs/plugins/gstreamer-vaapi-plugins-docs.xml.in:
14958 * docs/plugins/gstreamer-vaapi-plugins.types:
14959 * docs/reference/Makefile.am:
14960 * docs/reference/plugins/Makefile.am:
14961 * docs/reference/plugins/plugins-sections.txt:
14962 * docs/reference/plugins/plugins.types:
14963 * docs/version.entities.in:
14964 resurrect gtk-doc machinery
14965 Our auto-generated documentation has been a bit neglected. This patch replaces
14966 the 'normal' gtk-doc with the one used in GStreamer, which is adapted for
14967 plugins, elements and libraries.
14968 This patch also re-enables documentation generation.
14969 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14971 2016-01-28 19:35:24 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14977 use gst-common submodule
14978 This is 'the' big change in gstreamer-vaapi autoconf. Now it uses the official
14979 GStreamer common submodule.
14980 The documentation generation has been disable temporarily since it needs a
14981 major rework, which will be done in the following commit.
14982 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14984 2016-01-25 16:06:03 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14988 add gst-common submodule
14989 Pointing to branch 1.6
14990 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14992 2016-01-28 13:28:30 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
14994 * gstreamer-vaapi.doap:
14995 add doap descriptor
14996 DOAP (Description of a Project) is an RDF Schema and XML vocabulary to
14997 describe software projects, in particular free and open source software.
14998 The description is used in GStreamer as in many other open source projects.
14999 This patch adds the doap description of this project.
15000 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15002 2016-02-03 11:50:13 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15004 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
15005 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
15006 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
15007 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
15008 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
15009 * gst-libs/gst/vaapi/gstvaapidecoder_unit.c:
15010 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
15011 * gst-libs/gst/vaapi/gstvaapiimage.c:
15012 * gst-libs/gst/vaapi/gstvaapiparser_frame.c:
15013 * gst-libs/gst/vaapi/gstvaapipixmap.c:
15014 * gst-libs/gst/vaapi/gstvaapipixmap_x11.c:
15015 * gst-libs/gst/vaapi/gstvaapiprofile.c:
15016 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
15017 libs: humongous code style fix
15018 As part of the upstreaming process of gstreamer-vaapi into the GStreamer
15019 umbrella, we need to comply with the project's code style. This meant to
15020 change a lot of code.
15021 It was decided to use a single massive patch to update the code style.
15022 I would like to apologize with the original developers of this code because of
15023 the history breakage.
15024 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15026 2016-02-03 11:04:15 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15028 * gst-libs/gst/vaapi/gstvaapicontext.c:
15029 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15030 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
15031 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
15032 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
15033 * gst/vaapi/gstvaapivideomemory.c:
15034 libs: small refactors to enhance the code style
15035 As gst-indent generated ugly code in these cases, this patch changes the used
15036 idiomatic into other one.
15037 No functional changes were introduced.
15038 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15040 2016-02-02 17:59:57 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15042 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
15043 * gst-libs/gst/vaapi/gstvaapicodedbufferpool.c:
15044 * gst-libs/gst/vaapi/gstvaapicodedbufferproxy.c:
15045 * gst-libs/gst/vaapi/gstvaapidecoder.c:
15046 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
15047 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
15048 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
15049 * gst-libs/gst/vaapi/gstvaapidisplay.c:
15050 * gst-libs/gst/vaapi/gstvaapiencoder.c:
15051 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
15052 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
15053 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
15054 * gst-libs/gst/vaapi/gstvaapifilter.c:
15055 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
15056 * gst-libs/gst/vaapi/gstvaapisurface.c:
15057 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
15058 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
15059 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
15060 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
15061 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
15062 * gst-libs/gst/vaapi/gstvaapiwindow.c:
15063 * gst-libs/gst/vaapi/gstvaapiwindow_egl.c:
15064 libs: small code style fixes
15065 This a set of small code style fixes detected as-is by gst-indent.
15066 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15068 2016-02-02 17:50:19 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15070 * gst-libs/gst/vaapi/gstvaapicodedbufferpool.c:
15071 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
15072 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
15073 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
15074 * gst-libs/gst/vaapi/gstvaapitexture_egl.c:
15075 * gst-libs/gst/vaapi/gstvaapitexture_glx.c:
15076 * gst-libs/gst/vaapi/gstvaapiwindow_drm.c:
15077 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
15078 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
15079 libs: trivial comment style fixes
15080 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15082 2016-02-02 17:31:02 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15084 * gst-libs/gst/vaapi/gstvaapicodedbuffer.c:
15085 * gst-libs/gst/vaapi/gstvaapiwindow_egl.c:
15086 libs: avoid gst-indent mess up
15087 Guard pieces of code to avoid gst-ident to mess up the following code.
15088 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15090 2016-02-03 12:17:59 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15092 * gst/vaapi/gstvaapidecode.c:
15093 * gst/vaapi/gstvaapidecodebin.c:
15094 * gst/vaapi/gstvaapipluginutil.c:
15095 * gst/vaapi/gstvaapisink.c:
15096 * gst/vaapi/gstvaapivideomemory.c:
15097 plugins: fix code style
15098 Minor code style changes by executing gst-indent in gst/vaapi directory.
15099 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15101 2016-02-01 13:22:10 +0000 Tim-Philipp Müller <tim@centricular.com>
15103 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
15104 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
15105 Fix some more compiler warning
15106 Two (false) compiler warnings about variables potentially
15107 being used uninitialized, and one about a variable being
15109 https://bugzilla.gnome.org/show_bug.cgi?id=759192
15111 2016-02-01 13:02:13 +0000 Tim-Philipp Müller <tim@centricular.com>
15113 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
15114 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
15115 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
15116 * gst-libs/gst/vaapi/gstvaapifilter.c:
15117 * gst/vaapi/gstvaapidecode.c:
15118 * gst/vaapi/gstvaapipluginbase.c:
15119 * gst/vaapi/gstvaapipluginutil.c:
15120 * gst/vaapi/gstvaapipostproc.c:
15121 * tests/simple-encoder.c:
15122 * tests/test-filter.c:
15123 vaapi: fix 'ISO C90 forbids mixed declarations and code' compiler warnings
15124 Declare variables at the beginning of a code block, which
15125 is how it's done in GStreamer.
15126 https://bugzilla.gnome.org/show_bug.cgi?id=759192
15128 2016-01-28 14:21:04 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15130 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15131 decoder: vp9: Fix crop rectangle setting
15132 Align with software vp9dec behaviour: Add crop rectangle
15133 only if display_width/display_height is less than the
15134 frame_hdr->width/frame_hdr->height
15136 2016-01-27 08:56:45 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15138 * gst/vaapi/gstvaapidecode.c:
15139 vaapidecode: Fix renegotiation for resolution change
15140 Always renegotiate the pool if the immediate frame which going
15141 to be pushed has a different un-cropped resolution than the already
15144 2016-01-29 15:51:49 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15146 * gst/vaapi/gstvaapipluginbase.h:
15147 plugins: use the same pre-processor macro
15148 In gstvaapipluginbase.c we are using the macro USE_GST_GL_HELPERS to guard the
15149 code related with GstGL. Nonetheless, in gstvaapipluginbase.h we are using
15150 HAVE_GST_GL_GL_H macro in order to include the GstGLContext's header.
15151 We should use only one to be homogeneous. This patch sets USE_GST_GL_HELPERS
15152 in the header file.
15153 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15155 2016-01-29 18:06:29 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15157 * gst-libs/gst/vaapi/gstvaapidecoder.c:
15158 decoder: update a deprecated function
15159 Somehow this didn't show up earlier, but gst_adapter_prev_timestamp() got
15160 deprecated since GStreamer 1.0.
15161 This patch replace it with gst_adapter_prev_pts()
15162 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15164 2016-01-29 13:13:56 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15167 build: honor configure's cache
15168 The user might enable --config-cache when calling configure script. If so, our
15169 configuration variables will not be correctly calculated.
15170 This patch extracts the value of our variables either from the cache or from
15171 the operation result.
15172 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15174 2016-01-28 19:12:13 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15177 build: use common version variables
15178 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15180 2016-01-28 19:01:43 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15183 build: hard-code an unneeded macro
15184 That macro is required for EGL's dynamic module loading, but since
15185 gstreamer-vaapi doesn't creates dynamic modules, it is not required anymore.
15186 That code in gst-libs/gst/vaapi/gstvaapidisplay_egl.c should be removed.
15187 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15189 2016-01-28 17:14:51 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15192 build: refactorization of dependency tracking
15193 This patch tries to avoid branching in configure.ac using a more functional
15194 approach in macros usage.
15195 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15197 2016-01-29 12:34:30 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15200 build: check for OpenGL either GLX or EGL are requested
15201 Refactor some code in configure.ac to centralize $enable_opengl definition.
15202 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15204 2016-01-28 16:55:44 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15207 build: indent and add square braces
15208 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15210 2016-01-28 16:50:39 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15213 build: upgrade autotools version dependency
15214 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15216 2016-01-29 11:14:34 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15219 build: enhance string comparisons
15220 Add a 'x' as a prefix in string comparisons to watch out for edge cases where
15221 the string is empty or undefined.
15222 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15224 2016-01-28 14:29:16 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15227 build: remove unused variables
15228 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15230 2016-01-27 19:00:51 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15233 build: remove check for old version of gstreamer
15234 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15236 2016-01-27 17:55:02 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15239 build: remove GStreamer's parsers checks
15240 This patch removes almost all the parsers check since they are already in place,
15241 with the exception of the VP9 parser, since it was merged in Gstreamer 1.7.
15242 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15244 2016-01-29 12:11:17 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15247 * gst/vaapi/Makefile.am:
15248 build: add gstreamer-pbutils dependency
15249 This dependency was added in gstvaapidecodebin with the call
15250 gst_missing_element_message_new().
15251 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15253 2016-01-27 17:53:59 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15256 build: fix variable declaration
15257 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15259 2016-01-27 17:47:32 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15261 * gst-libs/gst/vaapi/Makefile.am:
15262 build: fix when HEVC decoder is disabled
15263 This a very pathological situation: when we have a HEVC encoder but not a HEVC
15265 The encoder needs functions that are only available when the decoder is
15267 This patch moves the utils functions into the generic sources, such as the
15269 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15271 2016-01-27 17:20:31 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15273 * gst/vaapi/Makefile.am:
15274 build: remove unused EGL specific sources
15275 These Makefile variables are not used at all.
15276 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15278 2016-01-27 17:19:32 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15280 * gst/vaapi/Makefile.am:
15281 build: remove check for GStreamer 1.2
15282 Since we are working for current stable GStreamer 1.6
15283 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15285 2016-01-26 11:49:40 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15287 * gst/vaapi/gstvaapiparse.h:
15288 Remove more video parser crufts
15289 This header is not used anymore since it declares parsers that are
15290 already in GStreamer 1.6
15291 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15293 2016-01-25 12:43:15 +0000 Tim-Philipp Müller <tim@centricular.com>
15296 * docs/reference/Makefile.am:
15297 * docs/reference/libs/Makefile.am:
15298 * docs/reference/libs/libs-docs.xml.in:
15299 * docs/reference/libs/libs-overrides.txt:
15300 * docs/reference/libs/libs-sections.txt:
15301 docs: remove library documentation which is non-public now
15302 https://bugzilla.gnome.org/show_bug.cgi?id=759192
15304 2016-01-25 12:40:49 +0000 Tim-Philipp Müller <tim@centricular.com>
15308 * debian.upstream/Makefile.am:
15309 * debian.upstream/changelog.in:
15310 * debian.upstream/compat:
15311 * debian.upstream/control.in:
15312 * debian.upstream/copyright:
15313 * debian.upstream/gstreamer-vaapi-doc.install.in:
15314 * debian.upstream/gstreamer-vaapi.install.in:
15315 * debian.upstream/rules:
15316 Remove debian.upstream packaging
15317 https://bugzilla.gnome.org/show_bug.cgi?id=759192
15319 2016-01-22 19:27:13 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15321 * gst-libs/gst/vaapi/gstvaapidecoder.c:
15322 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
15323 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
15324 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
15325 * gst-libs/gst/vaapi/gstvaapiprofile.c:
15326 * gst/vaapi/gstcompat.h:
15327 * gst/vaapi/gstvaapi.c:
15328 * gst/vaapi/gstvaapidecode.c:
15329 * gst/vaapi/gstvaapiencode.c:
15330 * gst/vaapi/gstvaapipluginbase.c:
15331 * gst/vaapi/gstvaapipluginbase.h:
15332 * gst/vaapi/gstvaapipluginutil.c:
15333 * gst/vaapi/gstvaapisink.c:
15334 Remove old gst version guards
15335 As gstreamer-vaapi now only supports from GStreamer 1.6, this patch removes
15336 all the old GStreamer version guards.
15337 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15339 2016-01-22 19:23:43 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15341 * gst/vaapi/gstvaapiparse.c:
15342 Remove video parser crufts
15343 We forgot to remove gstvaapiparse.c when we removed all the videoparser
15345 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15347 2015-12-09 19:52:33 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15349 * docs/reference/libs/Makefile.am:
15350 * docs/reference/plugins/Makefile.am:
15351 * gst-libs/gst/vaapi/Makefile.am:
15352 * gst/vaapi/Makefile.am:
15353 * tests/Makefile.am:
15354 libs: remove versioning
15355 Since we don't install libraries anymore, it makes no sense to keep
15356 versioning them according to the gstreamer's version.
15357 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15359 2015-12-09 16:59:16 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15362 * debian.upstream/Makefile.am:
15363 * debian.upstream/control.in:
15364 * debian.upstream/libgstvaapi-dev.install.in:
15365 * debian.upstream/libgstvaapi-drm.install.in:
15366 * debian.upstream/libgstvaapi-glx.install.in:
15367 * debian.upstream/libgstvaapi-wayland.install.in:
15368 * debian.upstream/libgstvaapi-x11.install.in:
15369 * debian.upstream/libgstvaapi.install.in:
15370 * gst-libs/gst/vaapi/Makefile.am:
15371 * tests/Makefile.am:
15372 libs: make libraries no installables
15373 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15375 2015-12-09 15:37:39 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15377 * debian.upstream/libgstvaapi-dev.install.in:
15378 * gst-libs/gst/vaapi/Makefile.am:
15379 Do not install libgstvaapi headers
15380 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15382 2016-01-19 10:40:54 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15386 * patches/Makefile.am:
15387 * patches/videoparsers/0001-plugins-compile-the-built-in-video-parsers-as-vaapip.patch:
15388 * patches/videoparsers/0002-h264parse-fix-build-with-older-GStreamer-1.x-stacks.patch:
15389 * patches/videoparsers/0003-h264parse-default-to-byte-stream-nalu-format-Annex-B.patch:
15390 * patches/videoparsers/0004-h264parse-Disable-3D-video-support-for-GStreamer-1.5.patch:
15391 * patches/videoparsers/0005-videoparsers-h264-Disable-passthorugh-mode-enabling.patch:
15392 * patches/videoparsers/0006-h265parse-include-gstvaapiparse.h.patch:
15393 * patches/videoparsers/0007-h265parse-fix-build-with-GStreamer-1.5.patch:
15394 * patches/videoparsers/Makefile.am:
15395 * patches/videoparsers/series.frag:
15396 Remove videoparser patches
15398 2015-12-09 15:18:11 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15402 * debian.upstream/libgstvaapi-dev.install.in:
15403 * pkgconfig/Makefile.am:
15404 * pkgconfig/gstreamer-vaapi-drm.pc.in:
15405 * pkgconfig/gstreamer-vaapi-glx.pc.in:
15406 * pkgconfig/gstreamer-vaapi-wayland.pc.in:
15407 * pkgconfig/gstreamer-vaapi-x11.pc.in:
15408 * pkgconfig/gstreamer-vaapi.pc.in:
15409 Remove pkg-config files
15410 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15412 2015-12-09 13:24:30 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15418 * ext/codecparsers:
15419 * gst-libs/gst/Makefile.am:
15420 * gst-libs/gst/codecparsers/Makefile.am:
15421 * gst-libs/gst/vaapi/Makefile.am:
15422 * gst/vaapi/Makefile.am:
15423 Remove codecparsers submodule
15424 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15426 2015-12-08 00:36:36 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15431 * debian.upstream/libgstvaapi.install.in:
15433 * ext/libvpx/Makefile.am:
15434 * ext/libvpx/gstlibvpx.c:
15435 * ext/libvpx/gstlibvpx.h:
15436 * ext/libvpx/libgstcodecparsers_vpx.vers:
15437 * ext/libvpx/sources.frag:
15438 * ext/libvpx/upstream:
15439 * gst-libs/gst/codecparsers/Makefile.am:
15440 * gst-libs/gst/codecparsers/gstvaapilibvpx.c:
15441 Remove libvpx submodule
15442 We will be using upstream codecparsers always.
15443 No more internal libvpx !
15445 2015-12-09 14:12:22 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15448 build: fix check for GstJpegParser
15449 Right now the local JPEG parser is always compiled because the check for the
15450 upstreamed version is broken: it looks for an non existent symbol:
15452 This patch changes that check for< GstJpegFrameHdr.
15453 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15455 2015-10-28 09:56:46 +0100 Michael Olbrich <m.olbrich@pengutronix.de>
15457 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
15458 wayland: free the frame in frame_release_callback()
15459 This basically reverts 62c3888b76afc69f714a020957e8c5dd9d98f561 (wayland:
15460 decouple wl_buffer from frame).
15461 Otherwise the frame may be overwritten while it is still used by the
15463 The frame done callback (frame_done_callback()) is called, when the
15464 compositor is done processing the frame and hands it to the hardware.
15465 The buffer release callback (frame_release_callback()) is called when the
15466 buffer memory is no longer used.
15467 This can be quite some time later: E.g. if weston (with the DRM backend)
15468 puts the buffer on a hardware plane, then then buffer release callback is
15469 called when the kernel is done with the buffer. This is usually when the
15470 next frame is shown, so most likely after the frame done callback for the
15472 Since 70eff01d36a2870cbf06ffb91c2a941e8cb6b804 "wayland: sync() when
15473 destroy()" the mentioned possible leak should no longer be a problem, so
15474 reverting this change should cause no leaking buffers.
15475 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15476 https://bugzilla.gnome.org/show_bug.cgi?id=758848
15477 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15479 2016-01-14 17:36:24 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15481 * gst/vaapi/gstvaapipostproc.c:
15482 vaapipostproc: check ANY caps at transform_caps()
15483 When transforming downstream caps we should check for ANY caps from peer pad,
15484 otherwise we get a segmentation fault.
15485 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15486 https://bugzilla.gnome.org/show_bug.cgi?id=759893
15488 2016-01-13 19:17:02 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15490 * gst/vaapi/gstvaapisink.c:
15491 vaapisink: ignore frame if its upload failed
15492 When gst_vaapi_plugin_base_get_input_buffer() fail to copy the input buffer
15493 into a VAAPI buffer, the return value is GST_FLOW_NOT_SUPPORTED, and it was
15494 ignored by the vaapisink, leading to a segmentation fault.
15495 This patch ignores the frame that generated the GST_FLOW_NOT_SUPPORTED
15496 returned by gst_vaapi_plugin_base_get_input_buffer(), avoiding the
15497 segmentation fault, but doing and effort to continue rendering. This is
15498 the same behavior of ximagesink.
15499 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15500 https://bugzilla.gnome.org/show_bug.cgi?id=759332
15502 2015-12-09 18:24:50 +0200 Joel Holdsworth <joel.holdsworth@vcatechnology.com>
15505 build: Don't ignore GST_PLUGIN_PATH_1_0 even if the directory doesn't exist yet
15506 https://bugzilla.gnome.org/show_bug.cgi?id=759184
15508 2015-12-08 16:14:11 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15510 * gst-libs/gst/vaapi/gstvaapicontext.c:
15511 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
15512 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
15513 * gst-libs/gst/vaapi/gstvaapiutils_h265_priv.h:
15514 Add 10 HEVC 10 bit decoding support
15515 Only supporting vaapidecode ! vaapisink combination for now.
15516 Missing dependencies:
15517 1: No support for P010 video format in GStreamer
15518 2: No support for P010 vaGetImage()/vaPutimage() in vaapi-intel-driver
15519 3: As a result of 1&2 , we have no support for Vaapi Video memory mapping
15520 through GstVideoMeta.
15521 Right now we only set chroma format (YUV420 with more than 8 bits per channel)
15522 for surface pool and keeping GST_VIDEO_FORMAT as ENCODED. The underlying format
15523 of the surfaces is implementation (driver) defined, which is P010.
15525 2001-01-01 04:59:28 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15527 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
15528 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
15529 gstvaapisurfacepool: Add new API to create surface pool based on chroma type
15530 This new API gst_vaapi_surface_pool_new_with_chroma_type() is for
15531 creating a new GstVaapiVideoPool of GstVaapiSurfaces with the specified
15532 chroam type and dimensions. The underlying format of the surfaces is
15533 implementation (driver) defined.
15535 2015-12-07 19:06:28 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15537 * gst-libs/gst/vaapi/gstvaapisurface.h:
15538 * gst-libs/gst/vaapi/gstvaapiutils.c:
15539 Add definitions for YUV420 with more than 8 bits per channel
15541 2015-12-07 17:26:24 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15543 * gst-libs/gst/vaapi/gstvaapiprofile.c:
15544 gstvaapiporfile: Fix string representation of HEVCMain10 profile
15546 2015-12-07 16:17:11 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15549 Bump version for development
15551 === release 0.7.0 ===
15553 2015-12-07 12:52:10 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15558 2015-12-07 12:49:05 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15563 2015-12-07 12:47:04 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15568 2015-12-07 12:39:23 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15573 2015-11-26 10:34:12 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15575 * gst-libs/gst/vaapi/gstvaapitexture_glx.c:
15576 texture: detect GL version and use the proper API
15577 When receiving the texture from the application or the video sink, we must
15578 know it size and border. To query the texture the API has changed according to
15579 the OpenGL version used in the GL context of the application/vsink.
15580 This patch checks the current context API type and queries the texture
15581 according to this detected API.
15582 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15583 https://bugzilla.gnome.org/show_bug.cgi?id=753099
15585 2015-11-26 10:19:32 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15587 * gst-libs/gst/vaapi/gstvaapitexture_glx.c:
15588 texture: check for expected target and format
15589 gst_vaapi_texture_glx_new_wrapped() only handles a GL_TEXTURE_2D target and
15590 formats GL_RGBA or GL_BGRA.
15591 This patch adds a debugging verification of those values.
15592 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15593 https://bugzilla.gnome.org/show_bug.cgi?id=753099
15595 2015-11-26 10:26:10 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15597 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
15598 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
15599 libs: add gl3_bind_texture_2d()
15600 Since OpenGL3.1 removed the fixed pipelines[1] enabling 2D textures is not
15601 needed. In particular, the Intel's Mesa implementation complains if it is
15603 This patch add a new binding function for 2D textures, without enabling
15604 gl3_bind_texture_2d()[2].
15605 1. https://www.opengl.org/wiki/Fixed_Function_Pipeline
15606 2. https://www.opengl.org/wiki/Common_Mistakes#OOP_and_hidden_binding
15607 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15608 https://bugzilla.gnome.org/show_bug.cgi?id=753099
15610 2015-11-26 10:14:45 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15612 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
15613 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
15614 libs: add gl_get_current_api()
15615 In order to know which OpenGL API use, we must detect the API type of current
15616 context. This patch adds the function gl_get_current_api() which returns the
15618 This function is an adaptation of gst_gl_context_get_current_gl_api() from
15620 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15621 https://bugzilla.gnome.org/show_bug.cgi?id=753099
15623 2015-11-27 12:29:11 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15626 * gst-libs/gst/vaapi/Makefile.am:
15627 build: Add gmodule dependency for libgstvaapi_egl
15628 https://bugzilla.gnome.org/show_bug.cgi?id=756259
15630 2015-11-27 14:24:55 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15632 * patches/videoparsers/0005-videoparsers-h264-Disable-passthorugh-mode-enabling.patch:
15633 * patches/videoparsers/0006-h265parse-include-gstvaapiparse.h.patch:
15634 * patches/videoparsers/0007-h265parse-fix-build-with-GStreamer-1.5.patch:
15635 * patches/videoparsers/series.frag:
15636 patches/videoparsers: h264: Disable passthorugh mode enabling
15637 This is a quick fix for regression introduced by the upstream
15638 commit e8908f5aeef952566f6bccde743c7735d3f8c6ef in h264 videoparser.
15639 The patch is disabling the passthrough mode, otherwise it will
15640 break multi-layer mvc stream parsing.
15641 https://bugzilla.gnome.org/show_bug.cgi?id=758656
15643 2015-11-25 15:12:53 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15645 * gst/vaapi/Makefile.am:
15646 * patches/videoparsers/Makefile.am:
15647 * patches/videoparsers/series.frag:
15648 build: add gsth265parse patches conditionally
15649 As gsth265parse was added in GStreamer 1.4, and gstreamer-vaapi still support
15650 GStreamer 1.2, the patching of gsth265parse must be conditional to the target
15652 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15653 https://bugzilla.gnome.org/show_bug.cgi?id=755525
15655 2015-11-25 15:11:28 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15657 * gst/vaapi/Makefile.am:
15658 build: declare correctly parse lib built files
15659 This is a continuation of commit fc8a0d12
15660 When declaring BUILT_SOURCES, those files should not be distributed. This
15661 patch avoids the distribution of the generated source code.
15662 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15663 https://bugzilla.gnome.org/show_bug.cgi?id=755525
15665 2015-11-24 17:14:20 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15667 * ext/libvpx/sources.frag:
15668 build: libvpx: Add missing source file
15670 2015-11-23 17:21:23 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15672 * gst/vaapi/gstvaapipostproc.c:
15673 vaapipostproc: Correctly detect the caps change
15674 This is a quick fix for regression introuduced by the
15675 commit 757833230bc73b8e3b4e31649e4618ba802bea51
15676 With out this, the gst_vaapipostproc_create() will
15678 https://bugzilla.gnome.org/show_bug.cgi?id=758543
15680 2015-11-18 20:48:30 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15682 * ext/libvpx/Makefile.am:
15683 * ext/libvpx/sources.frag:
15684 build: libvpx: update the sources lists
15685 `make dist` broke since commit f06798 (libvpx: Update the submodule to
15686 libvpx-1.4.0) because the sources.frag does not contain all the module
15688 This patch updates thoroughly the sources.
15689 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15690 https://bugzilla.gnome.org/show_bug.cgi?id=755525
15692 2015-11-16 17:49:01 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15694 * gst/vaapi/gstvaapipostproc.c:
15695 vaapipostproc: don't set caps change at first set
15696 When the source caps change, the filter is destroyed and recreated.
15697 Nonetheless, this happens every time the vaapipostproc starts, since the caps
15698 change detection algorithm does not take in consideration when the caps are
15700 This patch intents to be an optimization, to avoid a useless filter
15701 destroy-creation cycle when the sources caps are set for first time.
15702 The new helper function video_info_update() is a refactorization to avoid
15704 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15705 https://bugzilla.gnome.org/show_bug.cgi?id=758007
15707 2015-11-12 16:13:25 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15709 * gst/vaapi/gstvaapipostproc.c:
15710 vaapipostproc: params video_info_changed() callers
15711 The signature is video_info_changed(old_vip, new_vip). Nonetheless the callers
15712 swapped the the order. This didn't raise problems since the comparison of both
15713 structures were not affected by its semantics.
15714 But still it would be better to fix this to keep the coherence of the code.
15715 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15716 https://bugzilla.gnome.org/show_bug.cgi?id=758007
15718 2015-09-24 10:35:44 +0000 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15720 * gst/vaapi/Makefile.am:
15721 build: declare real built files
15722 When runnig the `make dist` target from a clean tree, it fails because
15723 if could not find the copied files from codecparsers submodule.
15724 They weren't copied because they weren't declared as built sources.
15725 This patch removes the stamp mechanism and use the actual file list to copy
15726 as the built sources. Also it fixes the duplication of the parser files.
15727 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15728 https://bugzilla.gnome.org/show_bug.cgi?id=755525
15730 2015-11-17 19:37:07 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15732 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15733 decoder: vp9: Fix last/golden/altref frame index setting
15734 Always fill VADecPictureParameterBufferVP9 last/golden/altref indices
15735 based on what ever reference frame indices encoded in frame header.
15737 2015-08-25 16:01:51 +0000 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15739 * debian.upstream/rules:
15740 debian: remove custom parallel compilation
15741 In order to build a debian package with upstream source, the user should
15744 cp -a debian.upstream debian
15745 debuild -eDEB_BUILD_OPTIONS="parallel=8" -us -uc -b
15746 The environment variable DEB_BUILD_OPTIONS="parallel=8" is the canonical
15747 way to make a parallel build (-j8 in this case).
15748 This commit removes the script in debian/rules that detects the number of
15749 cpus, requested by the environment variable DEBIAN_BUILD_NCPUS, which is not
15750 official in debian.
15751 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15752 https://bugzilla.gnome.org/show_bug.cgi?id=754087
15754 2015-11-16 18:22:55 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15756 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15757 decoder: vp9: Fix PTS calculation of cloned frames
15759 2015-11-16 18:22:33 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15761 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15762 decoder: vp9: Avoid unnecessary show_frame flag checking while doing picture output
15763 We always set GST_VAAPI_PICTURE_FLAG_SKIPPED for DECODE_ONLY frames and the
15764 gstvaapidecoder base calss is reponsible for handling those frames later on.
15765 No need for explicit verification of frame header's show_frame in order to
15766 do picture outputing.
15768 2015-11-16 18:22:14 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15770 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15771 decoder: vp9: Fix ref picture update while doing repeat frame
15772 Don't try to do frame decoding and reference picture update
15773 while receiving a vp9 frame having show_existing_frame flag
15776 2015-11-16 18:21:56 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15778 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15779 decoder: vp9: Add repeat-frame display handling
15780 If vp9 frame header come up with show_existing_frame flag set,
15781 we should duplicate the existing decoded frame as current frame to
15784 2015-11-12 11:07:38 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15786 * gst/vaapi/gstvaapidecodebin.c:
15787 vaapidecodebin: add me as element co-author
15788 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15789 https://bugzilla.gnome.org/show_bug.cgi?id=757957
15791 2015-11-12 12:47:01 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15793 * gst/vaapi/gstvaapidecodebin.c:
15794 vaapidecodebin: try to get display from decoder
15795 Rather than create a dummy display, if none has propagated as a context, we
15796 should try to get the one from vaapidecode.
15797 As the bin is already in READY state, the vaapidecode should be also in that
15798 state. That means that the contexts have been negotiated, and it should have
15800 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15801 https://bugzilla.gnome.org/show_bug.cgi?id=757957
15803 2015-11-11 19:04:25 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15805 * gst/vaapi/gstvaapidecodebin.c:
15806 * gst/vaapi/gstvaapidecodebin.h:
15807 vaapidecodebin: add postprocessor dynamically
15808 The former approach to left the bin unfinished has some problems: the context
15809 cannot be shared because the vaapidecode is unlinked in many cases, leading to
15810 creating a VADisplay twice.
15811 Initially the bin is fully functional, constructed as
15812 (-----------------------------------)
15814 | (-------------) (-------) |
15815 |<--| vaapidecode |--->| queue |--->|
15816 | (-------------) (-------) |
15817 (-----------------------------------)
15818 When the context is shared and the VADisplay has VPP capabilities, before
15819 changing to READY state, the bin is reconfigured dynamically, adding the
15820 vaapipostproc element afeter the queue:
15821 (--------------------------------------------------------)
15823 | (-------------) (-------) (---------------) |
15824 |<--| vaapidecode |--->| queue |--->| vaapipostproc |--->|
15825 | (-------------) (-------) (---------------) |
15826 (--------------------------------------------------------)
15827 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15828 https://bugzilla.gnome.org/show_bug.cgi?id=757957
15830 2015-11-11 16:33:24 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15832 * gst/vaapi/gstvaapidecodebin.c:
15833 vaapidecodebin: delay the bin configuration
15834 Delay the bin configuration until changing to READY state. This is because we
15835 should add the vaapipostproc element until the vaapidecode has emitted the
15836 HAVE_CONTEXT message, so de gst_bin_add() could set the context set to
15838 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15839 https://bugzilla.gnome.org/show_bug.cgi?id=757957
15841 2015-11-13 19:39:56 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15843 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15844 decoder: vp9: Add crop rectangle support.
15845 Set crop rectange if:
15846 There is display_width and display_height which is different from actual width/height
15848 The changed resolution is less than the actual configured dimension of surfaces
15850 2015-11-13 19:23:05 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15852 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15853 decoder: vp9: Fix the context and surface pool reset for multi resolution video
15854 Unlike other decoders, vp9 decoder doesn't need to reset the
15855 whole context and surfaces for each resolution change. Context
15856 reset only needed if resolution of any frame is greater than
15857 what actullay configured. There are streams where a bigger
15858 resolution set in ivf header or webm header but actual resolution
15859 of all frames are less. Also it is possible to have inter-prediction
15860 between these multi resolution frames.
15862 2015-11-13 18:58:33 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15864 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15865 decoder: vp9: Fill the VADecPictureParameterBufferVP9 width/height from frame header
15866 Always fill width/height of VADecPictureParameterBufferVP9 from frame header.
15867 Preliminary fix for supproting multi resolution video decode.
15869 2015-11-13 18:51:27 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15871 * gst/vaapi/gstvaapidecode.c:
15872 vaapidecode: Add comments for corner case fixes and fix couple of indentations.
15874 2015-11-13 18:41:53 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15876 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15877 decoder: vp9: Set lossless flag from frame header
15879 2015-11-13 18:40:52 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15881 * ext/codecparsers:
15882 codecparsers: Update to gst-vaapi-branch d9f25
15883 d9f2527: codecparsers: vp9: Set lossless flag in frame header
15885 2015-11-11 19:16:16 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15887 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
15888 libs: vp9: remove unused symbols
15889 clang complains about a couple variables and one label which were not
15890 used. This patch removes them.
15891 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15892 https://bugzilla.gnome.org/show_bug.cgi?id=757958
15894 2015-11-10 19:00:22 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
15896 * ext/codecparsers:
15897 codecparsers: Update to gst-vaapi-branch da251bb
15898 da251bb: codecparsers: vp9: Optimize the memory allocation
15899 f5759f4: codecparsers: vp9: Fix the wrong memcpy of probability arrays
15901 2015-11-05 12:58:52 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15903 * gst/vaapi/gstvaapipluginbase.c:
15904 * gst/vaapi/gstvaapipluginbase.h:
15905 plugin: guard pointers to pad query functions
15906 Since gstreamer 1.4 is not required to have pad query functions if the query
15908 This patch guards out the pad query functions for gstreamer < 1.4
15909 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15910 https://bugzilla.gnome.org/show_bug.cgi?id=757629
15912 2015-11-05 12:39:55 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15914 * gst/vaapi/gstvaapiencode.c:
15915 vaapiencode: use pad query vmethods
15916 GstVideoEncoder, the base class of vaapiencode, added support for pad queries
15917 as virtual methods since gstreamer 1.4. This patch enables those vmethods,
15918 while keeps support for previous versions of gstreamer.
15919 This patch is relevant since GstVideoEncoder takes care of other queries that
15920 we are currently ignoring.
15921 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15922 https://bugzilla.gnome.org/show_bug.cgi?id=757629
15924 2015-10-28 13:01:04 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15926 * gst/vaapi/gstvaapidecode.c:
15927 vaapidecode: return pad's template caps if no display
15928 A caps query can occur before the element has a display. In that case, the
15929 element can return its pad's template. But when the element already has a
15930 display, and the caps probe fails, the element shall return an empty caps, so
15931 the auto-plug could try with another decoder.
15932 If the element has a display and the caps probe works, then the computed caps
15933 should be returned.
15934 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15935 https://bugzilla.gnome.org/show_bug.cgi?id=757598
15937 2015-10-28 12:59:02 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15939 * gst/vaapi/gstvaapidecode.c:
15940 * gst/vaapi/gstvaapisink.c:
15941 plugins: don't create display at caps query
15942 Caps query can happen before the element has a bus. The display creation should
15943 be should occur on the context negotiation, when the bus is already configured.
15944 Then at caps query no display should be created.
15945 Instead of force the display creation, we graciously fail the allowed_caps()
15947 This change only applies for vaapidecode and vaapisink. The vaapipostroc, as a
15948 basetransform descendant, seems to be not affected by this, nor the encoders.
15949 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15950 https://bugzilla.gnome.org/show_bug.cgi?id=757598
15952 2015-11-04 21:38:42 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15954 * gst/vaapi/gstvaapidecode.c:
15955 * gst/vaapi/gstvaapiencode.c:
15956 * gst/vaapi/gstvaapipluginutil.c:
15957 * gst/vaapi/gstvaapipluginutil.h:
15958 * gst/vaapi/gstvaapipostproc.c:
15959 * gst/vaapi/gstvaapisink.c:
15960 plugins: fix context query handling
15961 The current context query handling design is flawed: the function
15962 gst_vaapi_reply_to_query() returns FALSE either if the query is not a
15963 GST_CONTEXT_QUERY of if the query could not be handled correctly. But the
15964 pad query function should handle differently each case.
15965 This patch changes the gst_vaapi_reply_to_query() for
15966 gst_vaapi_handle_context_query() and changes it usage in all the vaapi plugins
15967 to match the correct context query handling.
15968 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15969 https://bugzilla.gnome.org/show_bug.cgi?id=757598
15971 2015-11-04 20:37:05 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15973 * gst/vaapi/gstvaapipluginutil.c:
15974 plugin: don't lose previous context at query
15975 When processing the GST_CONTEXT_QUERY we should not lose the previous
15976 context in the query, we should only add our display structure.
15977 This patch copies the old context, if it is there, and stamp our display on
15978 it. Otherwise, a new context is created.
15979 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15980 https://bugzilla.gnome.org/show_bug.cgi?id=757598
15982 2015-11-04 20:29:03 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15984 * gst/vaapi/gstvaapivideocontext.c:
15985 * gst/vaapi/gstvaapivideocontext.h:
15986 vaapivideocontext: add gst_vaapi_video_context_set_display()
15987 This function set the display to an already created context. This function is
15988 going to be used later.
15989 Also, gst_vaapi_video_context_new_with_display() now uses this function.
15990 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15991 https://bugzilla.gnome.org/show_bug.cgi?id=757598
15993 2015-10-30 12:27:16 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
15995 * gst/vaapi/gstvaapipluginutil.c:
15996 * gst/vaapi/gstvaapivideocontext.c:
15997 * gst/vaapi/gstvaapivideocontext.h:
15998 plugins: check if display is set in sync
15999 Since the context messages are sync'ed, the display assignation happens in the
16000 same thread, hence we can know if the display was found or not as soon we call
16002 In order to take advantage of it, gst_vaapi_video_context_prepare() receives,
16003 as a new parameter, the address of the plugin's display, and reports back if
16004 the display was found and set.
16005 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16006 https://bugzilla.gnome.org/show_bug.cgi?id=757598
16008 2015-10-30 12:33:48 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16010 * gst/vaapi/gstvaapipluginutil.c:
16011 * gst/vaapi/gstvaapivideocontext.c:
16012 plugins: set display through context
16013 Instead of setting the display to the plugin directly after its creation, do
16014 it through the gstreamer's context mechanism, avoiding double assignations.
16015 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16016 https://bugzilla.gnome.org/show_bug.cgi?id=757598
16018 2015-11-02 18:20:07 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16020 * gst/vaapi/gstvaapivideocontext.c:
16021 vaapivideocontext: rename context structure
16022 The context structure is named "display" which is too generic. The contrary
16023 happens, for example, with GstGL, what uses the same name as the context, and
16024 its logs make more sense.
16025 This patch renames the context structure with the same name as the
16026 context, thus GST_PTR_FORMAT can pretty print it.
16027 https://bugzilla.gnome.org/show_bug.cgi?id=757598
16029 2015-11-04 19:02:34 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16031 * gst/vaapi/gstvaapivideocontext.c:
16032 vaapivideocontext: refactor gst_vaapi_video_context_prepare()
16033 First, refactorized run_context_query() into _gst_context_run_query(), adding
16034 a new parameter: the pad direction, in order to simplify the code.
16035 Second, added a new helper function: _gst_context_query(), which is a generic
16036 context query function. It isolates the operation of running the query and
16037 sets the context if found, also it enhances the logs.
16038 _gst_context_query() is similar to the one used in GstGL. Perhaps, in the
16039 future this helper function will be merged into the core libraries of
16041 Finally, gst_vaapi_video_context_prepare() was rewritten to use
16042 _gst_context_query().
16043 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16044 https://bugzilla.gnome.org/show_bug.cgi?id=757598
16046 2015-10-30 11:18:47 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16048 * gst/vaapi/gstvaapivideocontext.c:
16049 vaapivideocontext: refactor context category debug
16050 Refactor the extraction GST_CAT_CONTEXT logging using a only once
16051 initializator, so we could get the debug category from different code
16053 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16054 https://bugzilla.gnome.org/show_bug.cgi?id=757598
16056 2015-10-23 11:17:01 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16058 * gst/vaapi/gstvaapivideocontext.c:
16059 gstvaapivideocontext: fix indentation
16060 gst-indent does not handle correctly some expression like function
16061 declaration with attributes, breaking the following expressions.
16062 This patch makes gst-indent to ignore the attributed function
16063 declartion so the followed function definition is not mangled, such
16064 as happened in commit b4154a
16065 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16066 https://bugzilla.gnome.org/show_bug.cgi?id=757598
16068 2015-11-02 16:48:27 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16070 * gst/vaapi/gstvaapipluginbase.c:
16071 plugin: chain up set_context() vmethod
16072 Since Gstreamer 1.7, set_context() vmethod needs to be chained up with
16073 the parent class in order to broadcast all its contexts when the element
16074 is added into a bin:
16075 http://cgit.freedesktop.org/gstreamer/gstreamer/commit/?id=d5ded1588920c4471eefe055d09095d9e5e989b5
16076 There is no need to guard the call, because before GStreamer 1.7, the
16077 set_context() vmethod was NULL in the element class, hence the conditional
16079 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16080 https://bugzilla.gnome.org/show_bug.cgi?id=757598
16082 2015-11-06 10:20:34 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16084 * tests/simple-encoder.c:
16085 tests: simple-encoder: remove dead code
16086 The caps creation for codec state configuration is not used. Let's remove it.
16087 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16089 2015-11-02 19:05:07 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16091 * gst/vaapi/gstvaapidecodebin.c:
16092 vaapidecodebin: fix a leaked display instance
16093 The display returned by gst_vaapi_video_context_get_display() increments the
16094 references. Thus, we have to unref the returned display.
16095 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16096 https://bugzilla.gnome.org/show_bug.cgi?id=757595
16098 2015-11-04 16:50:44 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16100 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
16101 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
16102 libs: remove unneeded headers
16103 Since gstvaapidisplay_glx.h do not expose gl.h/glx.h structures, it is not
16104 required to include them in the header. It is not also required to include
16105 them in gstvaapidisplay_glx.c, since gstvaapiutils_glx.h includes them and
16106 exposes their structures (e.g. GLXPixmap).
16107 Nonetheless, glext.h neither glxext.h are required to include, they are
16108 already included conditionally by gl.h and glx.h, respectively.
16109 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16110 https://bugzilla.gnome.org/show_bug.cgi?id=757577
16112 2015-11-06 19:18:54 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16114 * ext/codecparsers:
16115 codecparsers: Update to gst-vaapi-branch 0ea6792
16116 0ea6792: codecparsers: vp9: Add header comments
16117 347ffc7: codecparsers: vp9: Use g_slice_free() for releasing memory allocated from the slice allocator
16119 2015-11-06 15:19:38 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16121 * gst/vaapi/gstvaapidecode.c:
16122 VP9: plugins: Add VP9 decoder
16124 2015-11-06 15:12:51 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16126 * gst-libs/gst/vaapi/Makefile.am:
16127 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.c:
16128 * gst-libs/gst/vaapi/gstvaapidecoder_vp9.h:
16129 VP9: libgstvaapi: Add VP9 decoder
16131 2015-11-06 14:57:00 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16133 * gst-libs/gst/vaapi/gstvaapiprofile.c:
16134 * gst-libs/gst/vaapi/gstvaapiprofile.h:
16135 VP9: gstvaapiprofile: Add profile definitions
16137 2015-11-06 14:39:22 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16140 VP9: build: Check availability of vp9 decoder APIs
16142 2015-11-06 14:24:08 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16146 * gst-libs/gst/codecparsers/Makefile.am:
16147 VP9: Allow building vp9 codecparser internally
16149 2015-11-06 12:38:46 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16151 * ext/codecparsers:
16152 codecparsers: Update to gst-vaapi-branch commit ac5dc1a
16153 ac5dc1a: codecparsers: vp9: Add vp9 codec parser
16154 e7d9217: codecparser: h264: initialize parsing structures
16155 403d400: codecparser: h265: initialize parsing structures
16157 2015-11-04 15:37:34 +0100 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16160 configure.ac: don't use an undefined variable
16161 If the environment lacks of gstreamer development packages, this error will
16162 be reported to the user: "gstreamer- was not found"
16163 This is because we are using an undefined variable in the printed message. The
16164 fix simple changes the variable for the hard-coded string "1.0".
16165 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16166 https://bugzilla.gnome.org/show_bug.cgi?id=757283
16168 2015-10-16 15:55:40 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16170 * gst/vaapi/gstvaapidecode.c:
16171 vaapidecode: relax guards for memory:VASurface capsfeature
16172 Though caps features are supported since GStreamer 1.2, there are some
16173 issues with the features caps negotiation in that version. Nonetheless,
16174 those issues are fixed in GStreamer 1.4. So, the memoy:VASurface caps
16175 feature negotiation is relaxed for GStreamer 1.4.
16176 The guard is the same as in vaapisink's caps template.
16177 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16178 https://bugzilla.gnome.org/show_bug.cgi?id=756686
16180 2015-10-15 18:18:36 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16182 * gst/vaapi/gstvaapidecode.c:
16183 vaapidecode: decide allocation doesn't update srccaps
16184 The received caps query will bring the already negotiated caps, so they are
16185 not expected to change.
16186 This patch removes this verification which is dead code path.
16187 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16188 https://bugzilla.gnome.org/show_bug.cgi?id=756686
16190 2015-10-14 20:30:30 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16192 * gst/vaapi/gstvaapidecode.c:
16193 * gst/vaapi/gstvaapipluginutil.c:
16194 * gst/vaapi/gstvaapipluginutil.h:
16195 vaapidecode: use caps to check the features
16196 Instead of calling gst_vaapi_find_preferred_caps_feature(), which is
16197 expensive, we check the caps from the allocation query, to check the
16198 negotiated feature.
16199 In order to do this verification a new utility function has been implemented:
16200 gst_vaapi_caps_feature_contains().
16201 As this new function shared its logic with gst_caps_has_vaapi_surface(), both
16202 have been refactorized.
16203 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16204 https://bugzilla.gnome.org/show_bug.cgi?id=756686
16206 2015-10-14 20:22:43 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16208 * gst/vaapi/gstvaapidecode.c:
16209 vaapidecode: set format before decide allocation
16210 There is a regression from commit 3d8e5e. It was expected the buffer pool
16211 allocation occur before the caps negotiation, but it is not.
16212 This patch fixes this regression: the caps negotiation is done regardless the
16213 allocation query from downstream.
16214 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16215 https://bugzilla.gnome.org/show_bug.cgi?id=756686
16217 2015-10-16 20:21:50 +0800 Lim Siew Hoon <siew.hoon.lim@intel.com>
16220 build: check for patch and fix yasm check
16221 Add configure checking for GNU patch tools and fixed configure
16222 checking YASM to correct sequence.
16223 Signed-off-by: Lim Siew Hoon <siew.hoon.lim@intel.com>
16224 https://bugzilla.gnome.org/show_bug.cgi?id=756690
16226 2015-10-15 19:20:21 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16228 * ext/libvpx/Makefile.am:
16229 build: Remove disable-md5 option for libvpx build
16230 The configure option --disable-md5 was provided in libvpx-1.3.0 which
16231 has been removed in 1.4.0.
16233 2015-10-15 19:00:26 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16235 * ext/libvpx/upstream:
16236 libvpx: Update the submodule to libvpx-1.4.0
16237 libvpx git commit: c74bf6d889992c3cabe017ec353ca85c323107cd
16239 2015-10-15 10:59:08 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16242 configure: mark support for GStreamer 1.2 as obsolete.
16243 Support for GStreamer 1.2 is obsolete. i.e. it is no longer supported.
16244 Our goal is to support the last two stable versions of GStreamer which
16245 are 1.4 and 1.6 at the moment.
16246 We still keep the 1.2 specific codes until the next gstreamer-vaapi-0.7
16247 release and will get rid of those in 0.8.
16249 2015-10-12 14:13:03 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16251 * gst/vaapi/gstvaapidecode.c:
16252 vaapidecode: Fix buffer copy assertion
16253 Don't try to copy the NULL buffer-codec_data.
16255 2015-09-28 14:57:33 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16258 build: allow builds against GStreamer 1.7.x
16259 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16261 2015-09-23 16:02:46 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16263 * gst/vaapi/gstcompat.h:
16264 gstcompat: add gst_buffer_copy_deep() if gst < 1.5
16265 gst_buffer_copy_deep() was added in GStreamer 1.5. If want to use it we should
16266 add an implementation if gstreamer-vaapi is linked to previous versions.
16267 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16269 2015-09-23 12:13:41 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16271 * gst/vaapi/gstvaapidecode.c:
16272 vaapidecode: simplify copy of GstVideoCodecState
16273 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16275 2015-09-14 19:21:08 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16277 * patches/videoparsers/0006-h265parse-fix-build-with-GStreamer-1.5.patch:
16278 * patches/videoparsers/series.frag:
16279 patches/videoparsers: h265parser: more API fences
16280 Add more API fences according with its version and refresh the patch.
16281 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16282 https://bugzilla.gnome.org/show_bug.cgi?id=754845
16284 2015-09-14 19:19:56 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16286 * patches/videoparsers/0005-h265parse-include-gstvaapiparse.h.patch:
16287 * patches/videoparsers/series.frag:
16288 patches/videoparsers: h265parser: rename patch keeping number
16289 Refresh the patch and rename it in order to keep the patch number.
16290 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16291 https://bugzilla.gnome.org/show_bug.cgi?id=754845
16293 2015-09-14 19:18:33 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16295 * patches/videoparsers/0002-h264parse-fix-build-with-older-GStreamer-1.x-stacks.patch:
16296 patches/videoparsers: h264parser: more API fences and refresh
16297 Add more API fences according with its version and refresh the patch.
16298 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16299 https://bugzilla.gnome.org/show_bug.cgi?id=754845
16301 2015-09-14 19:16:51 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16303 * patches/videoparsers/0004-h264parse-Disable-3D-video-support-for-GStreamer-1.5.patch:
16304 * patches/videoparsers/series.frag:
16305 patches/videoparsers: h264parser: fix description and refresh
16306 Fix a typo in the patch description and refresh it in order to avoid the
16307 creation of .orig files and break the distcheck target.
16308 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16309 https://bugzilla.gnome.org/show_bug.cgi?id=754845
16311 2015-09-14 19:15:18 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16313 * patches/videoparsers/0001-plugins-compile-the-built-in-video-parsers-as-vaapip.patch:
16314 * patches/videoparsers/0003-h264parse-default-to-byte-stream-nalu-format-Annex-B.patch:
16315 patches/videoparsers: h264parser: refresh patches
16316 In order to avoid the creation of .orig files and break the distcheck target.
16317 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16318 https://bugzilla.gnome.org/show_bug.cgi?id=754845
16320 2015-09-15 16:53:31 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16322 * gst/vaapi/Makefile.am:
16323 build: link libgstvaapi_parse against codec parser
16324 GST_CODEC_PARSER_* variables are defined if builtin codec parsers are disabled
16325 when running configure.
16326 Right now, libgstcodecparsers links only to libgstvaapi, but libgstvaapi_parse
16327 need it if builtin codec parsers are disabled.
16328 This patch adds GST_CODEC_PARSER_* variables to libgstvaapi_parse
16329 compilation. If builtin codec parsers are enable, this variable is null, so it
16330 should work using libgstvaapi, as normal.
16331 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16332 https://bugzilla.gnome.org/show_bug.cgi?id=754845
16334 2015-09-11 16:35:30 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16337 build: verify for H264 MVC and H265 SPS
16338 Currently the H264 and H265 parsers look for MVC and SPS respectively, and
16339 the required symbols for those were added in GStreamer 1.5
16340 If we try to compile in GStreamer < 1.4, without enabling the builtin codec
16341 parsers, the compilation fails, because the lack of those symbols.
16342 This patch verifies if the installed H264 and H265 parsers have those symbols. If
16343 they do not, the specific built in codec parsers are enabled and used.
16344 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16345 https://bugzilla.gnome.org/show_bug.cgi?id=754845
16347 2015-09-11 16:49:16 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16349 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
16350 decoder: h264: initialize PPS's slice_group_id
16351 When the GstVaapiParserInfoH264 is allocated, the memory is not initialized,
16352 so it contains random data.
16353 When gst_h264_parser_parse_pps() fails, the PPS structure keeps slice_group_id
16354 pointer uninitialized, leading to a segmentation fault when the memory is
16356 This patch prevents this by initializing the slice_group_id before the PPS
16358 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16359 https://bugzilla.gnome.org/show_bug.cgi?id=754845
16361 2015-09-15 11:01:29 +0300 Mark Nauwelaerts <mnauw@users.sourceforge.net>
16363 * gst/vaapi/gstvaapidecode.c:
16364 vaapidecode: proper numerator and denominator for forced latency framerate
16365 https://bugzilla.gnome.org/show_bug.cgi?id=755040
16367 2015-09-11 20:51:42 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16369 * ext/codecparsers:
16370 codecparsers: Update to gst-vaapi-branch commit f9e284b
16371 dae1a84: h264parse/h265parse: Fix negotiation crash
16372 45a9f8a: codecparsers: h265 : Fix default scaling list values
16373 28eaaf5: codecparsers: h265: Fix the selection of Active Ref Pic Set
16374 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16376 2015-09-04 22:19:55 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16378 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16379 decoder: hevc: Don't flush dpb for EOS/EOB nal
16380 Explicit flushing of dpb for EOS and EOB nal decoding is wrong,
16381 the dpb_add() itself will handle the flusing(if needed) of dpb
16382 for end of sequence and end of bitstream.
16383 https://bugzilla.gnome.org/show_bug.cgi?id=754010
16385 2015-09-04 22:11:10 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16387 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16388 decoder: hevc: Fix the dpb_add() based on C.5.2.3
16389 Follow the spec as it is in C.5.2.3, add the decoded frame to dpb
16390 just after the PicLatencyCnt setting of existing dpb frames.
16391 https://bugzilla.gnome.org/show_bug.cgi?id=754010
16393 2015-09-04 22:02:55 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16395 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16396 decoder: hevc: Fix the picture addition in dpb() based on spec H265 v3 (04/2015)
16397 This fix is based on the V3 vesion of spec which was missing in older versions.
16398 When the current picture has PicOutputFlag equal to 1, for each picture in the
16399 DPB that is marked as "needed for output" and follows the current picture in output order,
16400 the associated variable PicLatencyCount is set equal to PicLatencyCount + 1 (C.5.2.3).
16401 https://bugzilla.gnome.org/show_bug.cgi?id=754010
16403 2015-09-04 22:00:36 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16405 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16406 decoder: h265: Fix indentation
16408 2015-06-13 01:39:31 +1000 Jan Schmidt <jan@centricular.com>
16410 * gst-libs/gst/vaapi/gstvaapidecoder.c:
16411 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
16412 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
16413 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
16414 * gst/vaapi/gstvaapidecode.c:
16415 * gst/vaapi/gstvaapipluginutil.c:
16416 multiview: initial attempt at stereo/multiview support
16417 Add support for marking caps and buffers for multiview or
16418 stereoscopic output.
16419 https://bugzilla.gnome.org/show_bug.cgi?id=750835
16421 2015-08-28 17:12:12 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16423 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16424 decoder: hevc: remove unused functions
16425 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16426 https://bugzilla.gnome.org/show_bug.cgi?id=754250
16428 2015-08-31 13:11:54 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16430 * gst/vaapi/gstvaapidecode.c:
16431 vaapidecode: remove (another) unused variable
16432 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16434 2015-08-28 17:10:40 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16436 * gst/vaapi/gstvaapidecode.c:
16437 vaapidecode: remove unused variable
16438 Thus silence the compilation warnings.
16440 2015-08-28 16:06:08 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16442 * gst/vaapi/gstvaapidecode.c:
16443 vaapidecode: compilation fix
16444 gst_vaapi_decoder_state_changed() returns void. This patch fixes the
16445 compilation where the toolchain uses restrictive flags as clang.
16446 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16448 2015-08-29 00:27:05 +0300 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16450 * gst/vaapi/gstvaapidecode.c:
16451 vaapidecode: renegotiate if caps are not equal
16452 The use of gst_caps_is_always_compatible() for this optimization may lead to
16453 false positives. It is better to stick to gst_caps_is_strictly_equal() to know
16454 if it is required a re-negotiation.
16455 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16456 https://bugzilla.gnome.org/show_bug.cgi?id=750835
16458 2015-08-29 00:18:57 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16460 * gst/vaapi/gstvaapidecode.c:
16461 * gst/vaapi/gstvaapidecode.h:
16462 vaapidecode: Rework the re-negotiation code to handle multi resoultion videos
16463 Delaying the pool re-negotiation untill we push all decoded (and queued)
16464 frames downstream. Otherwise for the multi-resolution videos, the
16465 GstVideoVideoMemory will be having wrong resolution and which leads
16466 to nasty behaviours, especially when using software renderers.
16467 sample media file: RAP_B_Bossen_1.bin
16469 The first SPS Nal will report resoultion of 448x256 and having crop rectangles to
16470 get the final resoultion 416x240.
16471 Starting from 25 th frame, the resolution will change to 416x240. But parser
16472 elements won't report this since the effective croped resolution is same in
16473 both cases. Here the core libgstvaapi will detect this through it's internal
16474 parsing and do all context/pool destory/reset stuffs. Also it will notify this
16475 change to plugins in advance. But if the plugin try to do re-negotiaion of pool
16476 immediately, this will not sync with the resolution of already decoded and queued
16477 frames and which will lead to failure in gst_video_frame_map() in downstream(if we use the
16478 software renderer). So we have to delay the pool renegotiation in vaapidecode,
16479 untill we push all decoded frames downstream.
16480 https://bugzilla.gnome.org/show_bug.cgi?id=753914
16482 2015-08-28 23:43:47 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16484 * gst/vaapi/gstvaapidecode.c:
16485 vaapidecode: Always keep a copy of input codec state
16486 Currently we are sharing the input GstVideoCodecState with
16487 GstVaapiDecoder(gst-libs/gst/vaapi) by just doing ref and unref for
16488 each caps change. This is troublesome in many cases, for eg: if
16489 resoultion changes with in a singe stream. Because, when ever there
16490 is a resolution change, GstVideoDecoder will first change the Codec_state->caps
16491 fields with new resolution, but since we are using the same codecstate (ref)
16492 in gstvaapidecode.c, the caps check for input caps change will always fail.
16493 https://bugzilla.gnome.org/show_bug.cgi?id=753914
16495 2015-08-26 07:25:03 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16497 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16498 decoder: hevc: Fix the scaling list scan order
16499 The default scan order of scaling lists are up-right-diagonal
16500 as per hevc specification. Use the newly implemented
16501 uprightdiagonal_to_raster conversion codecparser APIs to
16502 get the the scaling_list values in raster order, which is
16503 what the VA intel driver requires.
16505 2015-08-26 07:20:09 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16507 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16508 decoder: hevc: sync with the codecparser changes
16509 The Tile Scanning Conversion process (spec 6-3 and 6-4) is implemented
16510 in codecparsers now. Remove the duplication from gstvaapidecoder_h265
16512 2015-08-26 07:04:22 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16514 * patches/videoparsers/0002-h264parse-fix-build-with-older-GStreamer-1.x-stacks.patch:
16515 patches/Videoparsers: update patch to fix build with older GStreamer 1.2 stacks
16517 2015-08-26 06:57:36 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16519 * ext/codecparsers:
16520 codecparsers: Update to gst-vaapi-branch commit 69550f1
16521 c207c6d: codecparsers: h265: Fix tile row and column parsing
16522 47074c5: codecparsers: h265: Add APIs for up-right-diagonal/raster scan conversion
16523 cd28b18: codecparsers: h265: Fix the range of delta_chroma_log2_weight_denom
16524 1746bbe: videoparsers: Use gst_base_parse_merge_tags()
16525 2f0932b: h264parse: Clear SPS info after processing
16526 f57d6b0: videoparsers: enable accept-template flag
16528 2015-08-25 15:38:42 +0000 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16530 * debian.upstream/control.in:
16531 debian: add yasm as build dependency
16532 As the compilation of libvpx (for vp8 parser) is enabled by default,
16533 yasm is required by default too.
16534 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16536 2015-08-14 19:21:04 +0800 Lim Siew Hoon <siew.hoon.lim@intel.com>
16538 * debian.upstream/rules:
16539 debian: remove --with-gstreamer-api option
16540 It is no longer valid in gstreamer-vaapi.
16541 Signed-off-by: Lim Siew Hoon <siew.hoon.lim@intel.com>
16542 [removed unused GST_API_VERSION variable]
16543 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16544 https://bugzilla.gnome.org/show_bug.cgi?id=753618
16546 2015-08-24 19:22:14 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16548 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
16549 wayland: rename is_cancelled to sync_failed
16550 Since commit 065a18a3, the semantics of the variable is_cancelled did not make
16551 sense. This commit renames this variable to sync_failed.
16553 2015-08-13 15:12:44 -0400 Olivier Crete <olivier.crete@collabora.com>
16555 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
16556 wayland: Don't return GST_FLOW_ERROR on flushing
16557 Setting the sink to flushing causes gst_vaapi_window_wayland_sync() to
16558 return FALSE which makes gst_vaapi_window_wayland_render() return
16559 FALSE which ends up posting an ERROR message in
16560 gst_vaapisink_show_frame_unlocked(). Solution is to just return TRUE
16562 https://bugzilla.gnome.org/show_bug.cgi?id=753598
16564 2015-08-06 12:28:51 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16566 * gst/vaapi/gstvaapi.c:
16567 Revert "Marking rank of vaapidecodebin as GST_RANK_MARGINAL for now."
16568 This reverts commit 3ccb198b513dc6ad287fe44117d03bec4d6a966a.
16570 2015-07-06 20:22:57 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16572 * gst/vaapi/gstvaapidecodebin.c:
16573 vaapidecodebin: check for postproc instance
16574 If the VPP's deinterlace-method is set, first we should check if the postproc
16575 is already instanced to set it. Otherwise we just store it until the VPP is
16576 added into the bin.
16577 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16578 https://bugzilla.gnome.org/show_bug.cgi?id=749554
16580 2015-08-06 18:48:13 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16582 * gst/vaapi/gstvaapidecodebin.c:
16583 * gst/vaapi/gstvaapipluginutil.c:
16584 * gst/vaapi/gstvaapipluginutil.h:
16585 vaapidecodebin: ensure VPP before going to READY
16586 There are sometimes that the VA-API display context is not shared among the
16587 pipeline, but it is important to know it before going to READY state (when the
16588 pipeline is already linked).
16589 One instance of this case is this:
16590 gst-launch-1.0 filesrc location=media ! decodebin ! vaapipostproc ! vaapisink
16591 This patch adds a new function in gstvaapipluginutil called
16592 gst_vaapi_create_test_display(). Its purpose is to create a disposable VA-API
16593 display, which only will be used for verify if the VAEntrypointVideoProc is
16594 available by the hardware. Afterwards, it should be unrefed.
16595 If the vaapidecodebin is going to READY state, and the element still doesn't
16596 know if VPP is available, the last resort is to create a new instance of the
16597 VA-API display and test for it.
16598 https://bugzilla.gnome.org/show_bug.cgi?id=749554
16600 2015-08-06 12:39:52 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16602 * gst/vaapi/gstvaapidecodebin.c:
16603 vaapidecodebin: post an error message if fails
16604 If the construction of the bin fails, post an error message in the bus.
16605 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16606 https://bugzilla.gnome.org/show_bug.cgi?id=749554
16608 2015-08-06 12:36:07 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16610 * gst/vaapi/gstvaapidecodebin.c:
16611 * gst/vaapi/gstvaapidecodebin.h:
16612 vaapidecodebin: has_vpp as a tri-state variable
16613 has_vpp can be UNKNOWN while the context message hasn't being received.
16614 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16615 https://bugzilla.gnome.org/show_bug.cgi?id=749554
16617 2015-08-03 16:33:02 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16619 * gst/vaapi/gstvaapivideomemory.c:
16620 gstvaapivideomemory: native format with no derived image
16621 If USE_NATIVE_FORMATS is defined we bail out before configuring the surface
16622 info based on the derived image configuration.
16623 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16624 https://bugzilla.gnome.org/show_bug.cgi?id=744042
16626 2015-07-23 20:07:59 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
16628 * gst-libs/gst/vaapi/video-format.c:
16629 * gst-libs/gst/vaapi/video-format.h:
16630 * gst/vaapi/gstvaapivideomemory.c:
16631 surface pool config based on video info
16632 First added the function gst_vaapi_video_format_get_best_native(), which
16633 returns the best native format that matches a particular chroma type:
16634 YUV 4:2:0 -> NV12, YUV 4:2:2 -> YUY2, YUV 4:0:0 -> Y800
16635 RGB32 chroma and encoded format map to NV12 too.
16636 That format is used to configure, initially, the surface's pool for the
16638 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16639 https://bugzilla.gnome.org/show_bug.cgi?id=744042
16641 2015-07-23 16:03:43 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
16643 * gst/vaapi/gstvaapivideomemory.c:
16644 gstvaapivideomemory: refactor gst_vaapi_video_allocator_new()
16645 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16646 https://bugzilla.gnome.org/show_bug.cgi?id=744042
16648 2015-06-19 15:51:07 +0200 Victor Jaquez <vjaquez@igalia.com>
16650 * gst-libs/gst/vaapi/gstvaapiencoder.c:
16651 gstvaapiencoder: validate chroma according to the VA's RT format
16652 Before, only YUV420 color space where supported. With this patch, the
16653 encoder is queried to know the supported formats and admits YUV422
16654 color space if its available.
16655 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16656 https://bugzilla.gnome.org/show_bug.cgi?id=744042
16658 2015-08-13 05:07:52 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16660 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16661 decoder: hevc: Add calculation of WpOffsetHalfRangeC
16662 This is necessary for finding ChromaOffsetL0/ChromaOffsetL1
16663 prediction weight table values with out using any hard coding.
16664 Fixme: We don't have parser API for sps_range_extension, so
16665 assumed zero value for high_precision_offsets_enabled_flag.
16666 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16668 2015-08-13 04:09:44 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16670 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16671 decoder: hevc: Fix ChromaOffsetL0/ChromaOffsetL1 calculation
16672 Based on ITU-T rec H265(4/2015): 7-56
16673 This was a wrong equation in rec H265 (4/2013): 7-44...
16674 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16676 2015-08-13 04:08:03 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16678 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16679 decoder: hevc: Fix default value assignment of pred_weight_table
16681 2015-08-13 03:48:43 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16683 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16684 decoder: hevc: Fix the value assigning for delta_chroma_log2_weight_denom
16685 Assign only if ChromaArrayType != 0..
16686 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16688 2015-08-13 03:06:32 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16690 * ext/codecparsers:
16691 codecparsers: Update to gst-vaapi-branch commit 1c70432
16692 8e98b41: codecparsers: h265: Fix the range of delta_chroma_log2_weight_denom
16693 839c5bc: codecparsers: h265: Fix the parsing of ref_pic_lists_modification
16694 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16696 2015-08-11 08:09:10 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16698 * ext/codecparsers:
16699 codecparsers: Update to gst-vaapi-branch commit c18b8ad
16700 8a03e67: videoparsers: h265: Avoid skipping of EOS and EOB nals
16701 a033083: videoparsers: h265: Fix the frame start detection code
16702 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16704 2015-08-10 05:50:50 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16706 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16707 decoder: hevc: Add SEI parsing
16708 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16710 2015-08-07 08:43:44 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16712 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16713 decoder: hevc: Workaround to recognize wrongly encoded main profile streams
16714 HACK: This is a work-around to identify some main profile streams having wrong profile_idc.
16715 There are some wrongly encoded main profile streams(eg: ENTP_C_LG_3.bin) which doesn't
16716 have any of the profile_idc values mentioned in Annex-A, instead general_profile_idc
16717 has been set as zero and having general_profile_compatibility_flag[general_profile_idc]
16718 is TRUE. Assuming them as MAIN profile for now.
16719 https://bugzilla.gnome.org/show_bug.cgi?id=753226
16720 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16722 2015-08-07 08:41:57 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16724 * patches/videoparsers/0002-h264parse-fix-build-with-older-GStreamer-1.x-stacks.patch:
16725 patches/videoparsers: Fix the wrong source file path
16726 This is something wrongly typed in commit 6d7b631
16728 2015-08-07 08:34:55 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16730 * ext/codecparsers:
16731 codecparsers: Update to gst-vaapi-branch commit b8d8be4
16732 ee7e81b: h264parse: Don't discard first AU delimiter
16733 3690fb9: h264parse: Add more NAL types for debugging output
16734 108d368: h265parse: Avoid checking for Non Mandatory VPS NAL
16735 ace61048: h265parse: expose compatible profiles to downstream
16736 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16738 2015-08-06 13:07:53 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16740 * gst/vaapi/gstvaapipluginutil.c:
16741 * gst/vaapi/gstvaapivideocontext.c:
16742 * gst/vaapi/gstvaapivideocontext.h:
16743 gstvaapivideocontext: remove unused parameter
16744 gst_vaapi_video_context_prepare() received an unused parameter. This patch
16745 removes it and the structure passed by the caller.
16746 This a left over of "Removal of gstreamer-1.0 support" (commit 8b36e25f).
16747 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16749 2015-08-06 04:01:24 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16751 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16752 decoder: hevc: Fix decoding of stream when it has temporal sublayers
16753 We are calculating the dpb size based on max_dec_pic_buffering.
16754 But if there are more than one temporal sublayers, we are supposed
16755 to use the max_dec_pic_buffering[max_sub_layers_minus] for dpb
16756 size calculation (Assuming HighestTid as max_sub_layers_minus).
16757 Sample streams: TSCL_A_VIDYO_5.bin, TSCL_B_VIDYO_4.bin
16758 https://bugzilla.gnome.org/show_bug.cgi?id=753226
16759 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16761 2015-08-05 14:11:12 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16763 * gst/vaapi/gstvaapidecode.c:
16764 * gst/vaapi/gstvaapidecodebin.c:
16765 * gst/vaapi/gstvaapiencode_h264.c:
16766 * gst/vaapi/gstvaapiencode_h265.c:
16767 * gst/vaapi/gstvaapiencode_jpeg.c:
16768 * gst/vaapi/gstvaapiencode_mpeg2.c:
16769 * gst/vaapi/gstvaapiencode_vp8.c:
16770 * gst/vaapi/gstvaapipluginutil.h:
16771 * gst/vaapi/gstvaapipostproc.c:
16772 * gst/vaapi/gstvaapisink.c:
16773 refactor vaapi caps strings for pad templates
16774 Refactor the main vaapi caps strings into three macros:
16775 GST_VAAPI_MAKE_SURFACE_CAPS, GST_VAAPI_MAKE_ENC_SURFACE_CAPS and
16776 GST_VAAPI_MAKE_GLTEXUPLOAD_CAPS.
16777 Those are in gstvaapipluginutil.h so all the elements could use them, instead
16778 of re-declaring them every time.
16779 No functional changes.
16780 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16782 2015-08-05 14:15:07 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16784 * gst/vaapi/gstvaapipostproc.c:
16785 gstvaapipostproc: fix code style
16786 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16788 2015-08-05 05:23:20 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16790 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
16791 decoder: hevc: Fix the decoding of dependent slice segment
16792 Decoding process for reference picture list construction needs to be
16793 invoked only for P and B slice and the value for slice_type of dependent slice
16794 segment should be taken from the previous independent slice segment header
16796 https://bugzilla.gnome.org/show_bug.cgi?id=753226
16797 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16799 2015-06-22 17:38:41 +0200 Victor Jaquez <vjaquez@igalia.com>
16801 * gst/vaapi/gstvaapipluginbase.c:
16802 plugins: reduce the noise of warnings
16803 Those messagse should be attached to the object, also the lack of
16804 caps is not an error, in particular in the case of JPEG encoding.
16805 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16806 https://bugzilla.gnome.org/show_bug.cgi?id=744042
16808 2015-06-23 17:49:51 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
16810 * gst-libs/gst/vaapi/gstvaapicontext.c:
16811 gstvaapicontext: fix the JPEG encoder attribs value
16812 When we query for the VAConfigAttribEncJPEG, we get a value which packs the
16813 VAConfigAttribValEncJPEG structure, but we did not assign it. This patch
16814 assigns the returned value to the attribute.
16815 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16816 https://bugzilla.gnome.org/show_bug.cgi?id=744042
16818 2015-06-18 17:37:46 +0200 Victor Jaquez <vjaquez@igalia.com>
16820 * gst-libs/gst/vaapi/gstvaapiencoder.c:
16821 gstvaapiencoder: framerate 0/1 is valid too
16822 Framerate 0/1 is valid, and it is particularly useful for picture
16823 encoding, such as jpeg. This patch makes the encoder to admit that
16825 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16826 https://bugzilla.gnome.org/show_bug.cgi?id=744042
16828 2015-07-03 09:35:16 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16830 * gst/vaapi/gstvaapipostproc.c:
16831 vaapipostroc: GLTextureUploadMeta in sink template
16832 Advertise GLTextureUploadMeta in sink caps template.
16833 https://bugzilla.gnome.org/show_bug.cgi?id=752130
16835 2015-07-23 13:11:40 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
16837 * gst/vaapi/Makefile.am:
16838 * gst/vaapi/gstvaapiencode.c:
16839 * gst/vaapi/gstvaapipluginbase.c:
16840 * gst/vaapi/gstvaapipluginbase.h:
16841 * gst/vaapi/gstvaapipostproc.c:
16842 * gst/vaapi/gstvaapisink.c:
16843 * gst/vaapi/gstvaapiuploader.c:
16844 * gst/vaapi/gstvaapiuploader.h:
16845 remove gstvaapiuploader
16846 Working on bug #744042 I realized that the gstvaapiuploader is practically not
16848 This patch removes the gstvaapiuploader and add the method
16849 gst_vaapi_plugin_base_get_allowed_raw_caps () that returns the raw caps that
16850 the system can handle, which is used by vaapisink and vaapipostproc.
16851 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16852 https://bugzilla.gnome.org/show_bug.cgi?id=752777
16854 2015-07-27 18:49:13 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16856 * gst/vaapi/gstvaapidecode.c:
16857 vaapidecode: don't lost GLTextureUpload on seek
16858 When seeking, the decoder is reset, but the buffer pool is not
16859 re-negotiated, but in reset_full() the code forgets if the negotiated buffer
16860 pool has the GLTextureUpload meta.
16861 The decoder knows that GLTextureUpload meta was negotiated in
16862 decide_allocation(), but this method is not called when seeking.
16863 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16864 https://bugzilla.gnome.org/show_bug.cgi?id=752929
16866 2015-07-21 18:45:56 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16868 * gst/vaapi/gstvaapipluginutil.c:
16869 plugins: don't use gst_pad_get_allowed_caps()
16870 gst_pad_get_allowed_caps() query the pad and the peer pad. In the case
16871 decoders, that is OK, but in the case of the postproc might lead loops,
16872 since the gst_base_transform_query_caps() forwards the query upstream
16874 Instead of gst_pad_get_allowed_caps() we only query the peer with
16875 gst_pad_peer_query_caps() using the pad's template as filter.
16876 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
16877 https://bugzilla.gnome.org/show_bug.cgi?id=752558
16879 2015-07-22 12:40:19 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16884 2015-07-22 09:45:26 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16886 * patches/videoparsers/0004-h264parse-Disable-3D-video-support-for-GStremaer-1.5.patch:
16887 * patches/videoparsers/series.frag:
16888 patches/videoparsers: h264parse: Disable 3D video support for GStremaer < 1.5
16889 All API/ABI changes for S3D/MVC are added in 1.5, backporting
16890 them to older verison is not recommended.
16891 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16893 2015-07-22 09:41:34 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16895 * patches/videoparsers/0002-h264parse-fix-build-with-older-GStreamer-1.x-stacks.patch:
16896 patches/Videoparsers: update patch to fix build with older GStreamer 1.2 stacks
16898 2015-07-22 09:38:42 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16900 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
16901 decoder: jpeg: Align with new API/ABI changes in codecparser
16902 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16904 2015-07-22 09:31:02 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16906 * ext/codecparsers:
16907 codecparsers: Update to gst-vaapi-branch commit 800bdb2
16908 ed13220: mpegvideometa: add meta transform function
16909 18d5efd: codecparsers: jpeg: add some padding to ScanHdr struct
16910 7a51722: codecparsers: jpeg: fix docs for table parsing functions
16911 06b8ded: codecparsers: jpeg: fix validity checking of data parsed
16912 387a39d: codecparsers: jpeg: fix up API
16913 db9d6a9: codecparsers: jpeg: tweak API a little
16914 bb6951e: codecparsers: jpeg: hide gst_jpeg_scan_for_marker_code()
16915 f33c30c: codecparsers: jpeg: fix and optimize scan for next marker code
16916 4658c30: codecparsers: jpeg: fix calculation of segment size
16917 759bcb9: codecparsers: jpeg: fix default Huffman tables generation
16918 b4811ee: codecparsers: jpeg: add JPEG bitstream parser
16919 9422464: h264parse: fix typo in log message
16920 9e793a0: h264parse: Move PAR calcs, and use them for stereoscopic half-aspect
16921 77704ce: nalutils: trivial patch to check if
16922 8bb9249: codecparsers: mpeg4: actually return full number of bits of resync marker
16923 7862f95: Revert "codecparsers: remove ignored increment of return"
16924 54017b1: h264parse: Add support for passing stereoscopic/multiview info
16925 8667ee4: h264parse: Don't switch to passthrough on set_caps()
16926 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16928 2015-06-29 14:27:56 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
16930 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
16931 decoder: h264: track previous reference frames.
16932 Improve closure of gaps in frame_num by strictly following and trying
16933 to fill them with previous reference frames. So, they are now tracked
16934 thus avoiding insertion of dummy ("greenish") frames.
16936 2015-06-29 13:16:09 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
16938 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
16939 decoder: h264: fix integration of second field into the DPB.
16940 If the new picture to be added to the DPB is not a first field, then
16941 it shall be the second field of the previous picture that was added
16943 This removes the need for dpb_find_picture() now that we track the
16944 immediately preceding decoded picture, in decode order.
16946 2015-07-06 14:38:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
16948 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
16949 decoder: h264: fix closure of "other-field" gap.
16950 When a dummy "other-field" is inserted, it is assumed to inherit the
16951 reference flags from the first field, and the sliding window decoded
16952 reference picture marking process is also executed so that corrupted
16953 frames are moved out as early as possible.
16954 While doing so, we also try to output frames that now contain a single
16955 valid field picture, prior to inserting any other picture into the DPB.
16956 Note: this may be superfluous currently based on the fact that dpb_add()
16957 combines the two most recent pairable fields, but this process would be
16958 further simplified later on.
16960 2015-06-24 13:58:17 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
16962 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
16963 decoder: h264: track corrupted frames.
16964 Mark the picture as "corrupted" if it is reconstructed from corrupted
16965 references or if those references are fake, e.g. resulting from lost
16967 This is useful for notifying the upper layer, or downstream elements,
16968 that the decoded frame may contain artefacts.
16969 https://bugzilla.gnome.org/show_bug.cgi?id=703921
16971 2015-06-24 13:48:46 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
16973 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
16974 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
16975 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
16976 * gst/vaapi/gstvaapidecode.c:
16977 decoder: add initial infrastructure for marking corrupted output.
16978 Add initial infrastructure in core codec library and vaapidecode to mark
16979 corrupted frames as such. A corrupted frame is such a frame that was
16980 reconstructed from invalid references for instance.
16981 https://bugzilla.gnome.org/show_bug.cgi?id=751434
16982 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
16984 2015-07-15 18:18:49 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16987 Bump version for development
16989 === release 0.6.0 ===
16991 2015-07-15 15:49:38 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
16996 2015-07-15 15:49:19 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17001 2015-07-14 19:39:20 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17003 * gst-libs/gst/vaapi/gstvaapiprofile.c:
17004 vaapidecoder: no wmv profiles gstreamer 1.4/1.2
17005 This patch fix the auto-plugging problem in gstreamer 1.2 and gstreamer 1.4
17006 Right now there is not a primary ranked parser for vc1 and the demuxers
17007 delivers caps without specifying the profile. This situation is not an issue
17008 for avdec_vc1 but for vaapidecode it is, which refuses to negotiate without a
17009 explicit profile defined in the negotiated caps.
17010 Nonetheless, in gstreamer 1.5 it seems not to be a problem since the
17011 negotiation admits caps subsets try outs.
17012 This patch solves the issue ignoring the profile negotiation in the caps. For
17013 gstreamer < 1.5 the profile string is not handled, so the auto-plugging get
17014 done without the vc1 parser, such as happens in gstreamer 1.5.
17015 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17017 2015-07-07 20:57:20 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17019 * gst/vaapi/gstvaapi.c:
17020 Marking rank of vaapidecodebin as GST_RANK_MARGINAL for now.
17021 Unfortunately vaapidecodebin element is not seems to be stable
17022 enough for autoplugging ahead of vaapidecode.
17023 Lowering the rank for now (cosidering the immediate 0.6 release).
17024 See this: https://bugzilla.gnome.org/show_bug.cgi?id=749554
17025 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17027 2015-07-07 13:32:18 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17029 * gst-libs/gst/vaapi/Makefile.am:
17030 build: Add missing CFLAGS to Makefile.am
17032 2015-07-03 15:07:02 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17034 * gst-libs/gst/vaapi/gstvaapifilter.c:
17035 gstvaapifilter: Only register STE property if it supported by corresponding VA library
17036 Fix the regression introduced in commit eb465fb.
17037 VAProcFilterSkinToneEnhancement is avaialbe from VA >= 0.36.
17038 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17040 2015-07-02 17:49:25 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17042 * gst/vaapi/gstvaapipostproc.c:
17043 vaapipostproc: no format convert on GL tex upload meta
17044 When GL texture upload meta is negotiated, vaapipostproc shall not modify the
17045 color format of the buffer.
17046 https://bugzilla.gnome.org/show_bug.cgi?id=748184
17048 2015-07-03 12:42:09 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17050 * gst-libs/gst/vaapi/gstvaapifilter.c:
17051 gstvaapifilter: Add guard for VAProcFilterSkinToneEnhancement
17052 VAProcFilterSkinToneEnhancement is avaialbe from VA >= 0.36.
17053 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17055 2015-07-02 21:57:38 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17057 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
17058 encoder:h265: Fix the check for packed-header support
17059 Use VA_ENC_PACKED_HEADER_* definition for checking.
17060 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17062 2015-07-02 21:37:56 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17064 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
17065 encoder:h264: Fix the check for packed-header support
17066 Use VA_ENC_PACKED_HEADER_* definition for checking.
17067 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17069 2015-07-02 21:00:14 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17071 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
17072 encoder: h264: submit SEI buffering_period() and picture_timing() messages for CBR mode
17073 One buffering_period() SEI message shall be present in every IDR access unit
17074 when NalHrdBpPresentFlag is inferred to be equal to 1. This is the case when we
17075 use a non-CQP mode, e.g. CBR. In other words, when
17076 nal_hrd_parameters_present_flag is set to 1.
17077 One picture_timing() SEI messages shall be present in every access unit
17078 if CpbDpbDelaysPresentFlag is equal to 1 or pic_struct_present_flag is equal to 1
17079 https://bugzilla.gnome.org/show_bug.cgi?id=722734
17080 https://bugzilla.gnome.org/show_bug.cgi?id=751831
17081 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17083 2015-07-01 14:16:50 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17085 * gst/vaapi/gstvaapidecodebin.c:
17086 vaapidecodebin: notify if vpp is disabled
17087 When the system is aware that VPP is not available by the VA driver,
17088 it would be useful to notify to the user that the disable-vpp property
17090 https://bugzilla.gnome.org/show_bug.cgi?id=749554
17092 2015-07-01 14:17:17 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17094 * gst/vaapi/gstvaapidecodebin.c:
17095 * gst/vaapi/gstvaapidecodebin.h:
17096 vaapidecodebin: enable vpp if it is available
17097 Instead of creating and adding VPP into the bin at setup, we wait until
17098 we are sure the VA driver supports it. We know that when the VA video
17099 context is received by the bin. Afterwards, it is decided to instanciate
17100 and link the VPP or not.
17101 This is more efficient and safer than waiting the VPP to fail and then
17103 https://bugzilla.gnome.org/show_bug.cgi?id=749554
17105 2015-07-02 12:29:32 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17107 * tests/test-display.c:
17108 * tests/test-windows.c:
17109 tests: Fix compilation while enabling egl as the only renderer in build
17110 Include missing header files gstvaapidisplay_egl.h and gstvaapiwindow_egl.h.
17112 2015-07-02 10:45:50 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17115 configure: fix the build while enabling egl as the only renderer
17117 2015-07-02 10:25:25 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17120 libs: Bump library major version
17122 2015-06-30 09:44:18 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17124 * gst/vaapi/gstvaapi.c:
17125 * gst/vaapi/gstvaapidecodebin.c:
17126 * gst/vaapi/gstvaapiencode.c:
17127 * gst/vaapi/gstvaapiencode_jpeg.c:
17128 * gst/vaapi/gstvaapipluginbase.c:
17129 * gst/vaapi/gstvaapipluginutil.c:
17130 * gst/vaapi/gstvaapipostproc.c:
17131 * gst/vaapi/gstvaapivideobufferpool.c:
17132 * gst/vaapi/gstvaapivideocontext.c:
17133 * gst/vaapi/gstvaapivideomemory.c:
17134 * gst/vaapi/gstvaapivideometa.c:
17135 gst/vaapi: Switch to upstreram like indentation.
17136 gst-indent for all gst/vaapi/*.c source files
17138 2015-06-30 09:35:37 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17140 * gst/vaapi/gstvaapidecodebin.c:
17141 * gst/vaapi/gstvaapidecodebin.h:
17142 vaapidecodebin: Add property to disable VPP
17143 Adding a new propery "disable-vpp", enabling it will prevent
17144 the insertion of vaapipostproc child element.
17145 This is helpful in debugging, specifically to narrow-down the
17146 vaapidecodebin/vaapipostproc related negotiation issues.
17147 No support for run-time disabling for now.
17148 https://bugzilla.gnome.org/show_bug.cgi?id=745901
17150 2015-06-29 13:35:59 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17152 * gst/vaapi/gstvaapipostproc.c:
17153 vaapipostproc: Fix wrong selection of passthrough mode.
17154 The Current code path is falling back to passthorugh mode if there is no
17155 vpp property set by the user explictily. But we should not use the
17156 passthrough mode if the negotiated src pad caps have a differnt color space
17157 format than sink pad caps (Even though the user didn't set the format property
17159 https://bugzilla.gnome.org/show_bug.cgi?id=748184
17160 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17162 2015-06-29 13:20:28 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17164 * gst/vaapi/gstvaapipostproc.c:
17165 vaapipostproc: prevent advanced-deinterlacing of non-native video formats.
17166 This is a workaround to deal with the va-intel-driver for non-native
17167 formats while doing advanced deinterlacing. The format of reference surfaces must
17168 be same as the format used by the driver internally for motion adaptive
17169 deinterlacing and motion compensated deinterlacing.
17170 A permanent solution could be to do the color space conversion internally
17171 for reference surfaces.
17172 https://bugzilla.gnome.org/show_bug.cgi?id=730925
17173 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17175 2015-06-29 13:06:30 +0300 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
17177 * gst/vaapi/gstvaapisink.c:
17178 Work around ABBA deadlock between vaapisink and vaapipostproc
17179 vaapisink takes the display lock, then does a gst_buffer_replace which can
17180 take the lock on the gst_vaapi_video_pool.
17181 vaapipostproc asks the gst_vaapi_video_pool for a new surface. This takes
17182 the lock on the gst_vaapi_video_pool; if you're unlucky, there are no free
17183 surfaces, which means that gst_vaapi_surface_create is
17184 called. gst_vaapi_surface_create takes the display lock.
17185 If vaapisink and vaapipostproc are in different threads, and this happens,
17186 you get a deadlock. vaapisink holds the display lock, and wants the
17187 gst_vaapi_video_pool lock. vaapipostproc holds the gst_vaapi_video_pool lock
17188 and wants the display lock.
17189 Work around this by releasing the display lock in vaapisink around the
17190 gst_buffer_replace.
17191 https://bugzilla.gnome.org/show_bug.cgi?id=738249
17192 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
17193 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17195 2015-04-29 16:34:07 +0200 Jacobo Aragunde Pérez <jaragunde@igalia.com>
17197 * gst/vaapi/gstvaapidecodebin.c:
17198 * gst/vaapi/gstvaapidecodebin.h:
17199 vaapidecodebin: expose deinterlace-method property from inner vaapipostproc
17200 https://bugzilla.gnome.org/show_bug.cgi?id=745901
17202 2015-05-19 11:24:10 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17204 * gst/vaapi/gstvaapipostproc.c:
17205 vaapipostproc: log negotiated caps
17207 2015-05-18 14:30:22 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17209 * gst/vaapi/gstvaapipostproc.c:
17210 vaapipostproc: remove useless debug message
17212 2015-02-12 12:31:57 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
17214 * gst/vaapi/gstvaapidecode.c:
17215 vaapidecode: log negotiated src/sink caps
17217 2015-05-07 15:57:26 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17219 * gst/vaapi/gstvaapisink.c:
17220 vaapisink: error handling if rendering fails
17221 This patch enhance the code path when an error is found when rendering a
17223 If the video meta doesn't contain a surface proxy or a surface, a warning
17224 message is printed.
17225 If the rendering backend fails, a error message is posted in the bus.
17226 https://bugzilla.gnome.org/show_bug.cgi?id=749382
17228 2015-06-18 14:55:12 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17230 * gst/vaapi/gstvaapisink.c:
17231 vaapisink: Fix the conditional pad template creation.
17233 2015-06-18 13:19:26 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17235 * tests/Makefile.am:
17236 build: Don't build simple-encoder test program if there is no VA Encoding support
17237 This will fix the build error against older VA-APIs <= 0.32
17239 2015-06-18 12:20:37 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17241 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
17242 * gst-libs/gst/vaapi/gstvaapicompat.h:
17243 Fix build error for older VA-API versions
17244 Provide guards for VA_SURFACE_ATTRIB_MEM_TYPE_KERNEL_DRM and
17245 VA_SURFACE_ATTRIB_MEM_TYPE_DRM_PRIME which are only availble from
17248 2015-06-17 14:20:37 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17250 * gst/vaapi/gstvaapisink.c:
17251 vaapisink: Fix the capsfeature advertisement in padtemplate
17252 This fixes the regression introduced in 64acc74.
17253 If a pad supports multiple set of capsfeatures, it needs to add
17254 multiple equal structures with different feature sets to the caps.
17255 Because caps structures with the same name but with a non-equal
17256 set of caps features are not compatible.
17257 Without this patch, playbin will autoplug xvimagesink instead of vaapisink.
17258 https://bugzilla.gnome.org/show_bug.cgi?id=750095
17260 2015-06-17 12:41:28 +0300 Adrian Cox <adrian@humboldt.co.uk>
17262 * gst/vaapi/gstvaapisink.c:
17263 vaapisink: Expose the overlay capability for compatibility with dvbsuboverlay.
17264 https://bugzilla.gnome.org/show_bug.cgi?id=750095
17265 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17266 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17268 2015-06-17 09:53:29 +0300 Olivier Crete <olivier.crete@collabora.com>
17270 * gst/vaapi/gstvaapipluginbase.c:
17271 vaapipluginbase: Override downstream allocation reply if no pool
17272 If the downstream replied without a pool, then override it.
17273 https://bugzilla.gnome.org/show_bug.cgi?id=748559
17275 2015-06-09 15:15:31 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17277 * gst/vaapi/gstvaapipostproc.c:
17278 * gst/vaapi/gstvaapipostproc.h:
17279 vaapipostproc: add color balance interface
17280 https://bugzilla.gnome.org/show_bug.cgi?id=720376
17282 2015-05-22 18:13:25 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17284 * gst-libs/gst/vaapi/gstvaapifilter.c:
17285 * gst-libs/gst/vaapi/gstvaapifilter.h:
17286 * gst/vaapi/gstvaapipostproc.c:
17287 * gst/vaapi/gstvaapipostproc.h:
17288 vaapipostproc: add skin tone enhancement
17289 Added the 'skin-tone-enhancement' property to vaapostproc.
17290 https://bugzilla.gnome.org/show_bug.cgi?id=744088
17292 2015-05-20 18:02:37 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17294 * docs/reference/libs/libs-docs.xml.in:
17295 doc: add VA-API reference in freedesktop
17297 2015-06-04 19:03:44 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17299 * patches/videoparsers/0001-plugins-compile-the-built-in-video-parsers-as-vaapip.patch:
17300 * patches/videoparsers/0002-h264parse-fix-build-with-older-GStreamer-1.x-stacks.patch:
17301 * patches/videoparsers/0003-h264parse-default-to-byte-stream-nalu-format-Annex-B.patch:
17302 patches/videoparsers: rebase all the h264parse patches
17303 In order to avoid the creation of .orig files and break the distcheck target
17305 2015-06-04 18:29:15 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17307 * ext/libvpx/Makefile.am:
17308 build: don't build in parallel libvpx
17309 This fixes the distcheck -j XX target.
17311 2015-06-02 08:52:53 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17313 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
17314 encoder: jpeg: Fix the packed header generation
17315 This is a work-around to satisfy the va-intel-driver.
17316 Normalize the quality factor and scale QM values (only for packed header
17317 generation) similar to what VA-Intel driver is doing . Otherwise the
17318 generated packed headers will be wrong, since the driver itself is
17319 scaling the QM values using the normalized quality factor.
17320 https://bugzilla.gnome.org/show_bug.cgi?id=748335
17321 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17323 2015-06-02 11:46:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17325 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
17326 decoder: h264: fix uninitialized variables in avcC mode.
17327 Fix uninitialized variables when decoding SPS and PPS NAL units from
17328 "codec-data" buffers. This is particularly important when seeking ops
17329 are involved, and the new persistent states are used more often.
17330 https://bugzilla.gnome.org/show_bug.cgi?id=750094
17332 2015-06-01 18:39:18 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17334 * gst/vaapi/gstvaapidecode.c:
17335 vaapidecode: remove unneeded casting
17336 And a code-style fix
17338 2015-05-21 19:38:33 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17340 * gst/vaapi/gstvaapidecode.c:
17341 vaapidecode: calculate decoding latency
17342 This is a naïve approach to the calculation of the VA-API decoding latency. It
17343 takes into consideration when the frame-rate has some insane value.
17344 https://bugzilla.gnome.org/show_bug.cgi?id=740419
17346 2015-05-21 23:16:14 +1000 Jan Schmidt <jan@centricular.com>
17349 configure: Compiling against libgstgl requires libgstvideo
17350 Fix detection of the GstGL helper headers in uninstalled
17353 2015-05-28 10:52:48 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17355 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
17356 encoder: hevc: Fix the size over-flow for encoded buffer.
17357 The approximation of 6 times compression ratio migh not
17358 work in all cases. Especially when enabling I frames.
17359 Provide large enough size for coded-buffer creation.
17361 2015-05-28 10:43:20 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17363 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
17364 encoder: vp8: Fix the size over-flow for encoded buffer.
17365 The approximation of 4 times compression ratio will not
17366 work in all cases. Especially when enabling I frames.
17367 Provide large enough size for coded-buffer creation.
17369 2015-05-28 05:43:49 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17371 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
17372 encoder: hevc: fix bug in multi slice encoding.
17373 This is a work-around for satisfying the VA-Intel driver.
17374 The driver only support slices begin from CTU row start address.
17375 Multi-Slice encoding also requires a fix in va-intel-driver:
17376 http://lists.freedesktop.org/archives/libva/2015-May/003351.html
17377 https://bugzilla.gnome.org/show_bug.cgi?id=749854
17378 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17380 2015-03-12 22:57:22 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17382 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
17383 decoder: h264: add initial support for loss of pictures.
17384 Implement decoding process for gaps in frame_num (8.5.2). This
17385 also somewhat supports unintentional loss of pictures.
17386 https://bugzilla.gnome.org/show_bug.cgi?id=745048
17387 https://bugzilla.gnome.org/show_bug.cgi?id=703921
17388 Original-patch-by: Wind Yuan <feng.yuan@intel.com>
17389 [fixed derivation of POC, ensured clone is valid for reference,
17390 actually fixed detection of gaps in FrameNum by PrevRefFrameNum]
17391 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17393 2015-05-22 11:42:52 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17395 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
17396 decoder: h264: add support for missing first field.
17397 Try to identify missing first fields too, thus disregarding any
17398 intermediate gaps in frames. We also assume that we keep the same
17399 field sequence, i.e. if previous frames were in top-field-first
17400 (TFF) order, then so are subsequent frames.
17401 Note that insertion of dummy first fields need to operate in two
17402 steps: (i) create the original first field that the current field
17403 will inherit from, and (ii) submit that field into the DPB prior
17404 to initializing the current (other) field POC values but after any
17405 reference flag was set. i.e. copy reference flags from the child
17406 (other field) to the parent (first field).
17407 https://bugzilla.gnome.org/show_bug.cgi?id=745048
17409 2015-05-07 14:00:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17411 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
17412 decoder: h264: add support for missing second field.
17413 Interlaced H.264 video frames always have two fields to decode and
17414 display. However, in some cases, e.g. packet loss, one of the field
17415 can be missing. This perturbs the reference picture marking process,
17416 whereby the number of references available in DPB no longer matches
17417 the expected value.
17418 This patch adds initial support for missing field within a decoded
17419 frame. The current strategy taken is to find out the nearest field,
17420 by POC value, and with the same parity.
17421 https://bugzilla.gnome.org/show_bug.cgi?id=745048
17423 2015-05-22 17:06:11 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17425 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
17426 decoder: h264: improve tracking of "top-field-first" flag.
17427 Try to maintain a "top-field-first" (TFF) flag, even if the H.264 standard
17428 does not mandate it. This will be useful for tracking missing fields, and
17429 also for more correct _split_fields() implementation for frames in the DPB.
17431 2015-05-05 11:56:11 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17433 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
17434 decoder: h264: skip all pictures prior the first I-frame.
17435 Don't try to decode pictures until the first I-frame is received within
17436 the currently active sequence. There is no point is decoding and then
17437 displaying frames with artifacts.
17439 2015-05-12 15:36:10 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17441 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
17442 decoder: h264: fix processing of EOSEQ NAL.
17443 Fix decoding of end_of_seq() NAL unit so that to not submit the current
17444 picture for decoding again. This is pretty vintage code that dates back
17445 before the existing of the whole decoder units machinery.
17446 One issue that could be arising if that code was kept is that we could
17447 have submitted a picture, and subsequently a GstVideoCodec frame, twice.
17448 Once without the decode_only flag set, and once with that flag set. The
17449 end result is that the GstVideoDecoder would release the codec frame
17450 twice, thus releasing stale data.
17451 In short, the piece of code that is removed by this patch is for once
17452 completely obsolete for a while, and secondly error-prone in corner
17455 2013-02-28 15:26:36 +0800 Wind Yuan <feng.yuan@intel.com>
17457 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
17458 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
17459 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
17460 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
17461 decoder: add utility function to clone picture objects.
17462 https://bugzilla.gnome.org/show_bug.cgi?id=703921
17463 Signed-off-by: Wind Yuan <feng.yuan@intel.com>
17464 [added cosmetic changes, fixed propagation of "one-field" flag to
17465 children, fixed per-codec clone modes (h264)]
17466 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17468 2015-05-27 23:49:18 +0300 Alban Browaeys <prahal@yahoo.com>
17470 * gst/vaapi/Makefile.am:
17471 build: don't compile HEVC encoder if not supported
17473 (gst-plugin-scanner:16681): GStreamer-WARNING **: Failed to load plugin '/usr/lib/x86_64-linux-gnu/gstreamer-1.0/libgstvaapi.so': /usr/lib/x86_64-linux-gnu/gstreamer-1.0/libgstvaapi.so: undefined symbol: gst_vaapi_encoder_h265_get_default_properties
17474 https://bugzilla.gnome.org/show_bug.cgi?id=749954
17475 Signed-off-by: Alban Browaeys <prahal@yahoo.com>
17476 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17478 2015-05-27 23:43:16 +0300 Alban Browaeys <prahal@yahoo.com>
17480 * gst-libs/gst/vaapi/gstvaapicompat.h:
17481 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
17482 HEVC: decode: add missing va_dec_hevc header
17483 Signed-off-by: Alban Browaeys <prahal@yahoo.com>
17484 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17485 https://bugzilla.gnome.org/show_bug.cgi?id=749953
17487 2015-05-26 13:28:32 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17489 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
17490 mpeg2: fix PTS cache for GOP start.
17491 If the GOP temporal sequence number (TSN) is interpolated from a valid
17492 PTS, then we need to compensate that PTS corresponding to the start of
17493 GOP with the next picture to be decoded, which shall be an I-frame,
17494 based on its sequence number.
17495 https://bugzilla.gnome.org/show_bug.cgi?id=748676
17497 2015-05-27 10:49:56 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17499 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
17500 mpeg2: avoid crash when seeking with debug logs
17501 Move down the debug message when the state of the decoder is verified
17502 so the slice header is not NULL.
17504 2014-12-17 00:41:10 +1100 Jan Schmidt <jan@centricular.com>
17506 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
17507 mpeg2: Avoid crashes and warnings on re-opened decoder after a seek
17508 Reset state and add some checks for safe state to avoid a crash and
17509 a warning after the decoder is destroyed/recreated during a seek.
17511 2015-05-26 10:21:59 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17513 * patches/videoparsers/0003-h264parse-default-to-byte-stream-nalu-format-Annex-B.patch:
17514 * patches/videoparsers/series.frag:
17515 patches/videoparsers: Rebase the patch on top of gst-vaapi-branch commit 20ee952
17516 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17518 2015-05-26 10:03:20 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17520 * ext/codecparsers:
17521 codecparsers: Update to gst-vaapi-branch commit 20ee952
17522 b7dded3: h264parse: don't consider unknown stream-format as avc
17523 5110ad9: h264parse: fix up handling of input caps corner cases
17524 e51db3e: h264parse: Remove dead code
17525 3d739d0: codecparser: h265: Fix the number of tile rows/columns parsing
17526 8482957: h265parse: Fix profile, tier and level setting in caps
17527 4649acb: h265parse: Fix the memory freeing of stored VPS nals
17528 f2beeb7: h265parse: Fix source caps to report cropped dimensions
17529 6886a31: h264parse: Fix profile and level setting in caps
17530 5286c1a: h264parse: Consider SEI NALU as "HEADER" packets
17531 eb97854: videoparsers: h264: bit-exact sync with upstream, minor changes here and there
17532 53074fc: build: Upgrade GStreamer dependency to 1.0
17533 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17535 2015-05-26 06:01:10 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17537 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
17538 HEVC: decode: Replace clip3 implementation with glib CLAMP macro
17539 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17541 2015-05-26 05:33:33 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17543 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
17544 HEVC: decode: Update Cropping Rectangle
17545 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17547 2015-05-25 11:58:20 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17549 * gst/vaapi/Makefile.am:
17550 * gst/vaapi/gstvaapi.c:
17551 * gst/vaapi/gstvaapiencode_h265.c:
17552 * gst/vaapi/gstvaapiencode_h265.h:
17553 HEVC_Encode: Add HEVC(h265) Encoder plugin
17554 https://bugzilla.gnome.org/show_bug.cgi?id=748874
17555 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17557 2015-05-25 11:38:34 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17559 * gst-libs/gst/vaapi/Makefile.am:
17560 * gst-libs/gst/vaapi/gstvaapiencoder_h265.c:
17561 * gst-libs/gst/vaapi/gstvaapiencoder_h265.h:
17562 HEVC_Encode: Add HEVC(h265) encoder to core libgstvaapi
17563 https://bugzilla.gnome.org/show_bug.cgi?id=748874
17564 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17566 2015-05-25 11:26:14 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17569 HEVC_Encode: build: Check availability of VA APIs for H265 encoding.
17570 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17571 https://bugzilla.gnome.org/show_bug.cgi?id=748874
17573 2015-05-25 10:58:52 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17575 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
17576 * gst-libs/gst/vaapi/gstvaapiutils_h265.h:
17577 gstvaapiutils_h265: Add H265 Tier specific utility functions
17578 -- New API: gst_vaapi_utils_h265_get_tier_from_string()
17579 -- New API: gst_vaapi_utils_h265_get_tier_string()
17580 https://bugzilla.gnome.org/show_bug.cgi?id=748874
17581 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17583 2015-05-19 10:57:42 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17585 * docs/reference/plugins/Makefile.am:
17586 doc: conditional linking for scanner
17587 Add x11 library only if it is enabled.
17588 https://bugzilla.gnome.org/show_bug.cgi?id=749018
17590 2015-05-19 10:37:13 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17592 * docs/reference/plugins/plugins.types:
17593 doc: fix scanner compilation warning
17594 https://bugzilla.gnome.org/show_bug.cgi?id=749018
17596 2015-05-06 16:19:23 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17598 * docs/reference/libs/libs-docs.xml.in:
17599 * docs/reference/libs/libs-sections.txt:
17600 doc: update sections and symbols
17601 https://bugzilla.gnome.org/show_bug.cgi?id=749018
17603 2015-05-13 10:38:24 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17607 * debian.upstream/Makefile.am:
17608 * docs/Makefile.am:
17609 * docs/reference/Makefile.am:
17610 * docs/reference/libs/Makefile.am:
17611 * docs/reference/plugins/Makefile.am:
17613 * ext/libvpx/Makefile.am:
17615 * gst-libs/Makefile.am:
17616 * gst-libs/gst/Makefile.am:
17617 * gst-libs/gst/base/Makefile.am:
17618 * gst-libs/gst/codecparsers/Makefile.am:
17619 * gst-libs/gst/vaapi/Makefile.am:
17621 * gst/vaapi/Makefile.am:
17622 * patches/Makefile.am:
17623 * patches/videoparsers/Makefile.am:
17624 * pkgconfig/Makefile.am:
17625 * tests/Makefile.am:
17627 This patch handles dinamically the gitignore files with git.mk[1].
17628 Removed the automake variable MAINTAINERCLANFILES in most of the
17629 Makefile.am files since now it is handled by the top one.
17630 1. https://github.com/behdad/git.mk/blob/master/git.mk
17631 https://bugzilla.gnome.org/show_bug.cgi?id=749321
17633 2015-05-07 11:28:15 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17635 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
17636 wayland: sync() when destroy()
17637 Before pushing a the new frame, the render() method calls sync() to flush the
17638 pending frames. Nonetheless, the last pushed frame never gets rendered, leading
17639 to a memory leak too.
17640 This patch calls sync() in the destroy() to flush the pending frames before
17641 destroying the window.
17642 Also a is_cancelled flag is added. This flag tells to not flush the event
17643 queue again since the method failed previously or were cancelled by the user.
17644 https://bugzilla.gnome.org/show_bug.cgi?id=749078
17646 2015-05-07 15:55:40 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
17648 * gst-libs/gst/vaapi/gstvaapiwindow.c:
17649 * gst-libs/gst/vaapi/gstvaapiwindow.h:
17650 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
17651 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
17652 * gst/vaapi/gstvaapisink.c:
17653 vaapisink: implement unlock/unlock_stop for wayland
17654 Otherwise wl_display_dispatch_queue() might prevent the pipeline from
17655 shutting down. This can happen e.g. if the wayland compositor exits while
17656 the pipeline is running.
17658 * renamed unlock()/unlock_stop() to unblock()/unblock_cancel() in gstvaapiwindow
17659 * splitted the patch removing wl_display_dispatch_queue()
17660 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17661 https://bugzilla.gnome.org/show_bug.cgi?id=747492
17662 https://bugzilla.gnome.org/show_bug.cgi?id=749078
17664 2015-05-07 12:33:34 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17667 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
17668 wayland: wl_display_dispatch_queue() can block forever.
17669 wl_display_dispatch_queue() might prevent the pipeline from shutting
17670 down. This can happen e.g. if the wayland compositor exits while the
17671 pipeline is running.
17672 This patch replaces it with these steps:
17673 - With wl_display_prepare_read() all threads announce their intention
17675 - wl_display_read_events() is thread save. On threads reads, the other
17676 wait for it to finish.
17677 - With wl_display_dispatch_queue_pending() each thread dispatches its
17679 wl_display_dispatch_queue_pending() was defined since wayland 1.0.2
17680 Original-patch-by: Michael Olbrich <m.olbrich@pengutronix.de>
17681 * stripped out the unlock() unlock_stop() logic
17682 * stripped out the poll handling
17683 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17684 https://bugzilla.gnome.org/show_bug.cgi?id=749078
17685 https://bugzilla.gnome.org/show_bug.cgi?id=747492
17687 2015-05-07 18:30:33 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17689 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
17690 wayland: rename frame for last_frame
17691 Since frame in the private data means the last frame sent, it would
17692 semantically better use last_frame.
17693 Also, this patch makes use of g_atomic_pointer_{compare_and_exchange, set}()
17695 https://bugzilla.gnome.org/show_bug.cgi?id=749078
17697 2015-05-07 11:18:12 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17699 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
17700 wayland: use a counter as sync flag
17701 Wayland window has a pointer to the last pushed frame and use it to set the
17702 flag for stopping the queue dispatch loop. This may lead to memory leaks,
17703 since we are not keeping track of all the queued frames structures.
17704 This patch removes the last pushed frame pointer and change the binary flag
17705 for an atomic counter, keeping track of number of queued frames and use it for
17706 the queue dispatch loop.
17707 https://bugzilla.gnome.org/show_bug.cgi?id=749078
17709 2015-05-07 10:36:17 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17711 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
17712 wayland: decouple wl_buffer from frame
17713 This patch takes out the wayland's buffer from the the frame structure. The
17714 buffer is queued to wayland and destroyed in the "release" callback. The
17715 frame is freed in the surface's "done" callback.
17716 In this way a buffer may be leaked but not the whole frame structure.
17717 - surface 'done' callback is used to throttle the rendering operation and to
17718 unallocate the frame, but not the buffer.
17719 - buffer 'release' callback is used to destroy wl_buffer.
17720 Original-patch-by: Zhao Halley <halley.zhao@intel.com>
17722 * kept the the event_queue for buffer's proxy
17723 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17724 https://bugzilla.gnome.org/show_bug.cgi?id=749078
17726 2015-05-14 16:22:36 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17728 * gst/vaapi/gstvaapisink.c:
17729 vaapisink: fix indentation
17731 2015-05-13 11:54:01 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17733 * debian.upstream/Makefile.am:
17734 * ext/libvpx/Makefile.am:
17735 * gst-libs/gst/vaapi/Makefile.am:
17736 * gst/vaapi/Makefile.am:
17737 * tests/Makefile.am:
17738 build: fix make distcheck
17739 This patch fixes several issues found when running the `make distcheck`
17741 - In commit c561b8da, the update of gstcompat.h in Makefile.am was
17743 - In commit c5756a91 add the simple_encoder_source_h in EXTRA_DIST was
17745 - vpx.build.stamp is not generated at all, only vpx.configure.stamp.
17746 - The make target distcleancheck failed because some autogenerated files
17747 were not handled with the DISTCLEANFILES variable.
17748 Note: `make distcheck -jXX` is not currently supported.
17750 2015-05-13 13:28:17 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17752 * patches/videoparsers/0001-plugins-compile-the-built-in-video-parsers-as-vaapip.patch:
17753 * patches/videoparsers/0002-h264parse-fix-build-with-older-GStreamer-1.x-stacks.patch:
17754 * patches/videoparsers/0004-h264parse-default-to-byte-stream-nalu-format-Annex-B.patch:
17755 h264parse: update patches with upstream
17756 These patches didn't applied cleanly, breaking the `make distcleancheck`
17757 target. Re-sync'ed the patches against the current git's submodule.
17759 2015-05-12 16:04:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17761 * tests/simple-encoder.c:
17762 tests: simple-encoder: fix build warnings on 64-bit platforms.
17763 Add a cosmetic change to replace VAAPI buffer with VA buffer and most
17764 importantly fix warnings spitted out during build on 64-bit platforms.
17765 ../../tests/simple-encoder.c:211:5: warning: format ‘%d’ expects argument of type ‘int’, but argument 4 has type ‘gssize’ [-Wformat=]
17766 g_warning ("Invalid VAAPI buffer size (%d)", size);
17768 ../../tests/simple-encoder.c:217:5: warning: format ‘%d’ expects argument of type ‘int’, but argument 4 has type ‘gssize’ [-Wformat=]
17769 g_warning ("Failed to create output buffer of size %d", size);
17772 2015-05-08 15:54:09 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17774 * gst/vaapi/gstvaapipluginbase.c:
17775 * gst/vaapi/gstvaapipluginutil.c:
17776 * gst/vaapi/gstvaapipluginutil.h:
17777 * gst/vaapi/gstvaapivideocontext.h:
17778 plugins: remove gstreamer-0.10 crumbs
17779 GstVideoContext was used in gstreamer-0.10, which is not supported anymore.
17780 Still, its definition was still in the code. This patch removes it.
17781 https://bugzilla.gnome.org/show_bug.cgi?id=749113
17783 2015-05-05 13:08:25 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17785 * tests/Makefile.am:
17786 * tests/simple-encoder.c:
17787 * tests/y4mreader.c:
17788 * tests/y4mreader.h:
17789 tests: add simple-encoder program
17790 This patch adds a simple-encoder test program that uses libgstvaapi for video
17791 encoding to elementary (raw) streams. Input stream is raw YUV in the Y4M
17792 format. That can be from a regular file or standard input when the input
17794 Usage: simple-encoder [options]* <source>
17796 --output|-o output file name
17797 --codec|-c codec to use for video encoding
17798 --bitrate|-b desired bitrate (kbps)
17799 By default, and as an initial patch, the encoded stream shall conform to the
17800 minimally supported profile. That is "Constrained Baseline Profile" for H.264
17801 and "Simple Profile" for MPEG-2. Though, those are the defaults to be
17802 generated by libgstvaapi.
17803 You can find Y4M sample files here http://samples.mplayerhq.hu/yuv4mpeg2/
17804 Original-patch-by: Changzhi Wei <changzhix.wei@intel.com>
17805 * general code clean-up
17806 * removed the yuv reader thread
17807 * re-wrote the y4m file parser
17808 * updated used API fixed some wrong usage
17809 * fixed a lot of memory leaks
17810 * added the bitrate setting
17811 * keep fps' numerator and denominator
17812 * simplified the thread control
17813 * removed custom logging and use glib
17814 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17815 https://bugzilla.gnome.org/show_bug.cgi?id=719528
17817 2015-05-05 13:02:19 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17819 * gst-libs/gst/vaapi/gstvaapiencoder.h:
17820 libs: trivial documentation fix
17821 GST_VAAPI_ENCODER_STATUS_NO_SURFACE and GST_VAAPI_ENCODER_STATUS_NO_BUFFER
17822 are not errors, so they do not have the ERROR namespace.
17823 This patch fixes this typo in documentation.
17825 2015-02-15 15:01:03 +0000 Simon Farnsworth <simon@farnz.org.uk>
17827 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
17828 window: Correct prototype to match implementation
17829 On s390x, guintptr and GstVaapiID are not compatible types. The
17830 implementation of gst_vaapi_window_new_internal() and all its callers
17831 seem to assume that its third argument is a GstVaapiID, while the
17832 header gives it guintptr type.
17833 https://bugzilla.gnome.org/show_bug.cgi?id=744559
17835 2015-05-04 14:24:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
17837 * gst/vaapi/gstvaapidecode.c:
17838 * gst/vaapi/gstvaapidecodebin.c:
17839 vaapidecode: add guards for disabled codecs.
17840 Fix link when building plugin elements without HEVC support. e.g. don't
17841 try to call into gst_vaapi_decoder_h265_set_alignment() if there is no
17842 support HEVC enabled in libgstvaapi.
17843 Also, drop disabled codecs from static template caps. Add the missing
17844 HEVC static template caps into vaapidecodebin too.
17846 2015-04-30 13:29:48 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17849 * gst-libs/gst/vaapi/glibcompat.h:
17850 build: upgrade glib dependency to 2.32
17851 Since bug #745728 was fixed the oldest supported version of GStreamer is
17852 1.2. That GStreamer release requires glib 2.32, so we can upgrade our
17854 This patch changes the required version of glib in configure.ac and removes
17855 the hacks in glibcompat.h
17856 https://bugzilla.gnome.org/show_bug.cgi?id=748698
17858 2015-04-30 13:21:08 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17860 * gst/vaapi/gstvaapipluginbase.c:
17861 plugins: check if the pool config is already set
17862 In commit 97b768, a regression for GStreamer 1.2 was introduced:
17863 GStreamer 1.2 doesn't check, in gst_buffer_pool_set_config() if the
17864 config option is already set. This patch adds an inline function to
17865 first verify if the option is not in the pool config berfore add it.
17867 2015-04-29 12:39:50 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17869 * gst/vaapi/gstvaapipostproc.c:
17870 vaapipostproc: tune up a couple of log messages
17871 In order to reduce the noise, the query type log was downgrade from INFO to
17872 DEBUG, and the shared display address log message is assigned to the object.
17874 2015-04-29 12:27:43 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17876 * gst/vaapi/gstvaapipluginbase.c:
17877 plugins: check gst_buffer_pool_set_config()
17878 Check the return value of gst_buffer_pool_set_config(). If it fails an error
17879 message is posted in the bus.
17881 2015-04-29 12:24:52 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17883 * gst/vaapi/gstvaapipluginbase.c:
17884 plugins: more specific log message
17885 Be more specific in the log message about the reason of creating a new pool.
17887 2015-04-29 12:22:29 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17889 * gst/vaapi/gstvaapipluginbase.c:
17890 plugins: delete unused variable
17891 need_pool is a boolean variable extracted from the allocation query, but it is
17892 not used afterwards.
17894 2015-04-27 19:21:12 -0400 Olivier Crete <olivier.crete@collabora.com>
17896 * gst/vaapi/gstvaapipluginbase.c:
17897 vaapipluginbase: Update the pool if there was no pool in the downstream reply
17898 Fix regression introduced by bd866479, the query after decide_allocation()
17899 always needs a pool in the first slot.
17900 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
17901 https://bugzilla.gnome.org/show_bug.cgi?id=748559
17903 2015-04-27 20:50:19 -0400 Olivier Crete <olivier.crete@collabora.com>
17905 * gst/vaapi/gstvaapivideobufferpool.c:
17906 videopool: Free members before chaining up finalize
17907 The finalize function in GObject frees the object memory, so
17908 everything else needs to have been freed before.
17909 https://bugzilla.gnome.org/show_bug.cgi?id=748563
17911 2015-04-27 20:31:50 -0400 Olivier Crete <olivier.crete@collabora.com>
17913 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
17914 wayland: don't leak the registry proxy
17915 Release the registry proxy when closing the display.
17916 https://bugzilla.gnome.org/show_bug.cgi?id=748564
17918 2015-04-21 17:17:06 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
17920 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
17921 wayland: refactor _sync() method and rename callback
17922 This patch only intends to improve readability: in the method
17923 gst_vaapi_window_wayland_sync() the if/do instructions are squashed into a
17925 Also renames the frame_redraw_callback() callback into frame_done_callback(),
17926 which is a bit more aligned to Wayland API.
17928 2015-02-03 16:52:06 +0100 Michael Olbrich <m.olbrich@pengutronix.de>
17930 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
17931 wayland: free frame in buffer release callback
17932 The Wayland compositor may still use the buffer when the frame done
17933 callback is called.
17934 This patch destroys the frame (which contains the buffer) until the
17935 release callback is called. The draw termination callback only controls
17936 the display queue dispatching.
17937 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
17938 https://bugzilla.gnome.org/show_bug.cgi?id=747492
17940 2015-04-21 10:00:36 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
17942 * gst/vaapi/gstvaapidecode.c:
17943 vaapidecode: refactor gst_vaapidecode_internal_flush()
17944 This a cosmetic refactor: gst_vaapidecode_internal_flush() removes its only
17945 label; gst_vaapidecode_finish() is more readable and gst_vaapidecode_purge()
17946 shares the same error message of gst_vaapidecode_internal_flush() when flush
17949 2015-04-20 13:27:27 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
17951 * gst/vaapi/gstvaapidecode.c:
17952 vaapidecode: refactor gst_vaapidecode_destroy()
17953 Add the method gst_vaapidecode_purge(). This method releases the
17954 flushed frames from the decoder.
17955 This new method add more readablity to gst_vaapidecode_destroy()
17957 2015-04-16 12:53:18 -0400 Olivier Crete <olivier.crete@collabora.com>
17959 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
17960 * gst/vaapi/gstvaapidecode.c:
17961 vaapidecode: Tell the base class about released frames on close
17962 The base class needs to be informed about frames that were still queued
17963 in the decoder on release, otherwise they are leaked.
17964 https://bugzilla.gnome.org/show_bug.cgi?id=747999
17966 2015-04-19 11:19:03 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
17968 * gst/vaapi/gstvaapidecode.c:
17969 vaapidecode: reduce logging noise
17970 When a frame is rejected by downstream, the message is logged twice. This
17971 patch removes one of those logging messages.
17972 Also, the reject of a frame doesn't mean an alarming error. This patch demotes
17973 the log message from error to info.
17975 2015-04-16 20:18:13 -0400 Olivier Crete <olivier.crete@collabora.com>
17977 * gst/vaapi/gstvaapidecode.c:
17978 vaapidecode: Use the GstVideoDecoder error reporting function
17979 This way, the decoder won't stop on the first decoding error,
17980 in most cases it can recover after some glitchiness.
17981 https://bugzilla.gnome.org/show_bug.cgi?id=744620
17983 2015-04-17 19:10:35 +0000 Olivier Crete <olivier.crete@collabora.com>
17985 * gst/vaapi/gstvaapipluginbase.c:
17986 vaapipluginbase: The allocation query can return without a pool
17987 It is possible to return the min/max/size without actually providing
17988 a pool. This way the source knows how many buffers downstream needs.
17989 https://bugzilla.gnome.org/show_bug.cgi?id=748076
17991 2015-04-17 16:45:22 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
17994 * gst/vaapi/Makefile.am:
17995 * gst/vaapi/gstvaapiparse.c:
17996 * gst/vaapi/gstvaapiparse.h:
17997 * patches/videoparsers/0001-h265parse-include-gstvaapiparse.h.patch:
17998 * patches/videoparsers/series.frag:
17999 plugins: Add h265 videoparser element "vaapiparse_h265"
18000 This is a mirror of h265parse element in upstream gst-plugins-bad.
18001 There could be additional patches but all should go to upstream.
18002 This is for making development faster.
18003 Note: vaapiparse_h265 will get build only for GStreamer version >= 1.4
18005 2015-04-17 15:44:04 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18007 * ext/codecparsers:
18008 codecparsers: Update to gst-vaapi-branch commit 43a0368
18009 45f1c28: codecparser: h265: Fix nal unit size checking
18010 f25987b: codecparser: h265: Calculate crop rectangle dimensions
18011 639573a: codecparser: h265: Fix parsing multiple SEI messages in a single SEI Nal
18012 4c8ec41: Add h265 videoparser plugin source files
18014 2015-04-17 10:10:10 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18017 autogen: drop videoutils submodule.
18019 2015-04-17 10:36:25 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18021 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
18022 decoder: hevc: cosmetics.
18023 Mostly coding style updates. Avoid integer signess inconsistencies.
18024 Optimize dpb_find_lowest_poc() to align with original h264's decoder.
18026 2015-04-16 14:13:59 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18028 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
18029 decoder: hevc: Add Support for tiled video decoding
18030 Based up on the value of uniform_spacing_flag in Picture Parameter Set,
18031 the tile column width and tile row height should be calculated.
18032 Equations: 6-1, 6-2
18033 Tiled video Descriptions: 7.3.2.3, 7.4.3.3
18035 2015-04-16 14:13:21 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18037 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
18038 decoder: hevc: Fix decoding when there are RASL pictures present.
18039 -- Set NoRaslOutputFlag based on EOS and EOB Nal units
18040 -- Fix PicOutputFlag setting for RASL picture
18041 -- Fix prev_poc_lsb/prev_poc_msb calculation
18042 -- Drop the RASL frames if NoRaslOutputFlag is TRUE for the associated IRAP picture
18043 -- Fixed couple of crashes and added cosmetics
18045 2015-04-14 10:54:54 +0100 Martin Sherburn <martin.sherburn@datapath.co.uk>
18047 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
18048 display: drm: fix race condition setting device type
18049 There is a race condition where g_drm_device_type can be left set to
18050 DRM_DEVICE_RENDERNODES when it shouldn't.
18051 If thread 1 comes in and falls into the last else statement setting up both
18052 RENDERNODES and LEGACY types. And begins to process the first type (RENDERNODES),
18053 it sets g_drm_device_type = RENDERNODES.
18054 Now when thread 2 comes in and sees g_drm_device_type is RENDERNODES, it queues
18055 up that type to be tried but then encounters the lock and has to wait until the
18056 first thread finishes. Once the lock is acquired it will then proceed to ONLY try
18057 RENDERNODES and fail it. But it doesn't try LEGACY. And from then on, all future
18058 attempts will only try RENDERNODES.
18059 So to avoid this situation I have simply moved the acquisition of the lock higher
18060 up in the attached patch.
18061 https://bugzilla.gnome.org/show_bug.cgi?id=747914
18063 2015-04-15 15:26:12 -0400 Olivier Crete <olivier.crete@collabora.com>
18065 * gst/vaapi/gstvaapipostproc.c:
18066 vaapipostproc: Don't create filter on caps query
18067 The problem with this is that creating the filter causes the display to
18068 be selected, and the caps query happens while linking the element. So,
18069 if the downstream or upstream element is using a specific display
18070 object, it won't be propagated correctly to the postproc as it already
18071 has a display at this point.
18072 https://bugzilla.gnome.org/show_bug.cgi?id=747945
18074 2015-04-15 15:20:17 -0400 Olivier Crete <olivier.crete@collabora.com>
18076 * gst-libs/gst/vaapi/gstvaapivideopool.c:
18077 videopool: Release lock while allocating new object
18078 The video pool can be accessed with the display lock held, for example,
18079 when releasing a buffer from inside vaapisink_render, but allocating
18080 a new object can may also take the display lock. Which means a possible
18082 https://bugzilla.gnome.org/show_bug.cgi?id=747944
18084 2015-04-15 17:26:43 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18086 * gst/vaapi/gstvaapisink.c:
18087 vaapisink: use GstVideoSink vmethod show_frame()
18088 vaapisink inherits from GstVideoSink, in order to use its functionality (such
18089 as ::show-preroll-frame property), we should use its vmethod show_frame(),
18090 rather than call ourselves render() and preroll().
18092 2015-04-15 18:16:47 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18094 * gst/vaapi/gstvaapisink.c:
18095 * gst/vaapi/gstvaapisink.h:
18096 vaapisink: add 'handoff' signal
18097 This patch adds the signal ::handoff and the property signal-handoffs. If the
18098 property is set TRUE, the signal ::handoff is emitted just after the buffer is
18100 Based on Zhao Halley <halley.zhao@intel.com>
18101 https://bugzilla.gnome.org/show_bug.cgi?id=747905
18103 2015-04-14 10:17:16 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18105 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
18106 HEVC: silence the compiler
18107 Fixed a couple of clang complains.
18109 2015-02-02 16:42:43 +0100 Michael Olbrich <m.olbrich@pengutronix.de>
18111 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
18112 wayland: destroy vpp buffer pool on resize
18113 Otherwise the old buffers with the old size are used.
18114 https://bugzilla.gnome.org/show_bug.cgi?id=747491
18116 2015-04-14 10:08:47 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18119 * gst-libs/gst/vaapi/Makefile.am:
18120 build: don't compile HEVC if not supported
18121 HEVC decoding was added recently libva-1.5.
18122 This patch avoids HEVC decoding support in libgstvaapi if it is not available
18123 in the installed libva.
18124 https://bugzilla.gnome.org/show_bug.cgi?id=747831
18126 2015-04-13 16:04:59 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18128 * gst/vaapi/gstvaapidecode.c:
18129 vaapidecode: Update Author name in plugin metadata
18131 2015-04-13 15:43:30 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18133 * gst/vaapi/gstvaapidecode.c:
18134 plugins: Add HEVC decoder
18135 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18137 2015-04-13 15:41:45 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18139 * gst-libs/gst/vaapi/Makefile.am:
18140 * gst-libs/gst/vaapi/gstvaapidecoder_h265.c:
18141 * gst-libs/gst/vaapi/gstvaapidecoder_h265.h:
18142 HEVC: Add HEVC(h265) decoder to core libgstvaapi
18143 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18145 2015-04-13 14:53:46 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18147 * gst-libs/gst/vaapi/Makefile.am:
18148 * gst-libs/gst/vaapi/gstvaapiutils_h265.c:
18149 * gst-libs/gst/vaapi/gstvaapiutils_h265.h:
18150 * gst-libs/gst/vaapi/gstvaapiutils_h265_priv.h:
18151 HEVC: Add codec utility methods to core libgstvaapi
18152 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18154 2015-04-13 14:52:53 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18156 * gst-libs/gst/vaapi/gstvaapiprofile.c:
18157 * gst-libs/gst/vaapi/gstvaapiprofile.h:
18158 HEVC: gstvaapiprofile: Add profile definitions
18159 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18161 2015-04-13 14:52:14 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18164 HEVC: build: Check availability of h265 decoder APIs
18165 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18167 2015-04-13 14:51:51 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18171 * gst-libs/gst/codecparsers/Makefile.am:
18172 HEVC: Allow to build h265 codecparser internally
18173 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18175 2015-04-08 18:05:20 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18177 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
18178 * gst-libs/gst/vaapi/gstvaapibufferproxy_priv.h:
18179 * gst-libs/gst/vaapi/gstvaapisurface.c:
18180 guard buffer export API if not available
18181 The support for buffer exports in VA-API was added in version 0.36. These
18182 interfaces are for interop with EGL, OpenCL, etc.
18183 GStreamer-VAAPI uses it for a dmabuf memory allocator. Though, gstreamer-vaapi
18184 has to support VA-API versions ranging from 0.30.4, which doesn't support it.
18185 This patch guards all the buffer exports handling (and dmabuf allocator) if
18186 the detected VA-API version is below 0.36.
18187 https://bugzilla.gnome.org/show_bug.cgi?id=746405
18189 2015-04-13 11:29:35 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18191 * ext/codecparsers:
18192 codecparsers: Update to gst-vaapi-branch commit 9bc72b0
18193 767bf22: codecparsers: h265: add helpers to convert quantization matrices
18194 71c8e93: codecparser: h265: skip byte alignment bits while parsing slice header
18195 3bf0355: codecparsre: h265: Fix the NumDeltaPocs calculation
18196 10e2087: codecparser: h265: Fix the NumPocTotalCurr calculatio
18197 2d753b8: codecparser: h265: Fix nal size calculation for EOS and EOB
18199 2014-12-11 12:02:38 +0100 Michael Olbrich <m.olbrich@pengutronix.de>
18201 * gst/vaapi/gstvaapidecode.c:
18202 vaapidecode: unref video codec frame twice
18203 We get one reference when the frame is passed to decode_handle_frame()
18204 and create another one in gst_vaapi_decoder_push_frame().
18205 Usually the frame is handled in gst_vaapidecode_push_decoded_frame().
18206 Here the frame is always released twice:
18207 gst_video_decoder_finish_frame() + gst_video_codec_frame_unref() or
18208 gst_video_decoder_drop_frame() + gst_video_codec_frame_unref().
18209 In gst_vaapidecode_reset_full() both references to the frame must be
18211 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18212 https://bugzilla.gnome.org/show_bug.cgi?id=743226
18214 2015-04-08 18:20:34 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18216 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
18217 libs: remove unused variables
18218 clang reports these unused variables. Let's get rid of them.
18219 This patch is a missing part of commit c82e5173
18220 https://bugzilla.gnome.org/show_bug.cgi?id=747312
18222 2015-04-03 20:38:07 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18224 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
18225 decoder: mpeg4: remove an spurious comparison
18226 The member size in GstMpeg4Packet is gsize which is unsigned, which cannot be
18227 less than zero. Hence this pre-condition test is a no-op. This patch removes
18229 https://bugzilla.gnome.org/show_bug.cgi?id=747312
18231 2015-04-03 20:33:44 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18233 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
18234 encoder: h264: casts slice_param->slice_type
18235 slice_type in slice_param is defined as (char *), but it is compared against a
18236 signed integer. clang complains about this comparison.
18237 This patch casts the variable.
18238 https://bugzilla.gnome.org/show_bug.cgi?id=747312
18240 2015-04-03 20:31:47 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18242 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
18243 encoder: avoid GstVaapiCodedBuffer redefinition
18244 The symbol GstVaapiCodedBuffer is already defined in
18245 gst-libs/gst/vaapi/gstvaapicodedbuffer.h which is loaded, at the end, by
18246 gstvaapiencoder_objects.h. Clang complains about the symbol re-definition.
18247 This patch removes that redefinition.
18248 https://bugzilla.gnome.org/show_bug.cgi?id=747312
18250 2015-04-03 20:28:22 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18252 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
18253 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
18254 libs: remove unused variables
18255 clang reports these unused variables. Let's get rid of them.
18256 https://bugzilla.gnome.org/show_bug.cgi?id=747312
18258 2015-04-03 20:27:24 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18260 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
18261 encoder: mpeg2: use fabsf() instead of abs()
18262 The member value in frame_rate_tab is float, the result of the abs() function
18263 should be float too. But abs() only manages integers.
18264 This patch replaces abs() with fabsf() to handle correctly the possible floats
18266 https://bugzilla.gnome.org/show_bug.cgi?id=747312
18268 2015-04-03 20:02:29 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18270 * gst-libs/gst/vaapi/gstvaapidecoder.c:
18271 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
18272 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
18273 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
18274 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
18275 decoder: cast GST_VAAPI_DECODER_STATUS_DROP_FRAME
18276 Since GST_VAAPI_DECODER_STATUS_DROP_FRAME is not part of the enum
18277 GstVaapiDecoderStatus, we need to cast it to avoid compiler complains.
18278 https://bugzilla.gnome.org/show_bug.cgi?id=747312
18280 2015-04-04 00:40:29 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18285 2015-04-04 00:06:56 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18289 Changing source code download links from https://gitorious to https://github
18290 -- gitmodules: Change gstreamer-codecparsers submodule source download link
18291 -- README: Change the gstreamer-vaapi webpage link
18293 2015-04-03 23:30:24 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18295 * ext/codecparsers:
18296 codecparsers: update to gst-vaapi-branch commit 1f792e4
18297 87f4a7e: bytereader: add gst_byte_reader_peek_sub_reader() and _get_sub_reader()
18298 7d8ba7a: bytereader: use unchecked inline variant for get_remaining in more places
18299 2528ea6: bytereader: add gst_byte_reader_masked_scan_uint32_peek
18300 2b92a67: h264parse: reset the parser information when caps changes
18301 05eee86: codecparsers: Indent file
18302 e27a38b: codecparsers: Add READ_UE_MAX macro
18303 2036471: Constify some static arrays everywhere
18305 2015-04-03 17:45:08 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18307 * gst/vaapi/gstvaapivideoconverter_glx.c:
18308 * gst/vaapi/gstvaapivideoconverter_x11.c:
18309 Remove the gstvaapivideoconverter_*.c source files missed in commit 51b1e4a
18311 2015-04-03 17:09:08 +0300 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18313 * gst/vaapi/gstvaapidecode.c:
18314 * gst/vaapi/gstvaapidecodebin.c:
18315 * gst/vaapi/gstvaapiencode_h264.c:
18316 * gst/vaapi/gstvaapiencode_jpeg.c:
18317 * gst/vaapi/gstvaapiencode_mpeg2.c:
18318 * gst/vaapi/gstvaapiencode_vp8.c:
18319 * gst/vaapi/gstvaapipluginbase.c:
18320 * gst/vaapi/gstvaapipluginutil.c:
18321 * gst/vaapi/gstvaapipluginutil.h:
18322 * gst/vaapi/gstvaapipostproc.c:
18323 * gst/vaapi/gstvaapisink.c:
18324 * gst/vaapi/gstvaapivideobuffer.c:
18325 * gst/vaapi/gstvaapivideobufferpool.c:
18326 * gst/vaapi/gstvaapivideocontext.c:
18327 * gst/vaapi/gstvaapivideocontext.h:
18328 * gst/vaapi/gstvaapivideoconverter_glx.h:
18329 * gst/vaapi/gstvaapivideoconverter_x11.h:
18330 * gst/vaapi/gstvaapivideomemory.h:
18331 * gst/vaapi/gstvaapivideometa_texture.c:
18332 Removal of gstreamer-1.0 support
18333 The support for GStreamer 1.0 has been obsoleted in 0.5.10 release.
18334 GStreamer 1.2 is the a minimal requirement for building the gstreamer-vaapi.
18335 This patch removes all the pre-processor conditional code compilation guarded
18337 Thus, all the video converters were removed too.
18338 https://bugzilla.gnome.org/show_bug.cgi?id=745728
18339 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18340 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18342 2015-04-03 17:08:30 +0300 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18344 * gst-libs/gst/vaapi/gstcompat.h:
18345 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
18346 * gst-libs/gst/vaapi/gstvaapiprofile.c:
18347 * gst-libs/gst/vaapi/sysdeps.h:
18348 * gst/vaapi/gstcompat.h:
18349 * gst/vaapi/gstvaapi.c:
18350 * gst/vaapi/gstvaapidecode.c:
18351 * gst/vaapi/gstvaapidecodebin.c:
18352 * gst/vaapi/gstvaapiencode.c:
18353 * gst/vaapi/gstvaapiencode_h264.c:
18354 * gst/vaapi/gstvaapiencode_jpeg.c:
18355 * gst/vaapi/gstvaapiencode_mpeg2.c:
18356 * gst/vaapi/gstvaapiencode_vp8.c:
18357 * gst/vaapi/gstvaapiparse.c:
18358 * gst/vaapi/gstvaapipluginbase.c:
18359 * gst/vaapi/gstvaapipluginutil.c:
18360 * gst/vaapi/gstvaapipostproc.c:
18361 * gst/vaapi/gstvaapisink.c:
18362 * gst/vaapi/gstvaapiuploader.c:
18363 * gst/vaapi/gstvaapivideobuffer.c:
18364 * gst/vaapi/gstvaapivideobufferpool.c:
18365 * gst/vaapi/gstvaapivideocontext.c:
18366 * gst/vaapi/gstvaapivideoconverter_glx.c:
18367 * gst/vaapi/gstvaapivideoconverter_x11.c:
18368 * gst/vaapi/gstvaapivideomemory.c:
18369 * gst/vaapi/gstvaapivideometa.c:
18370 * gst/vaapi/gstvaapivideometa_texture.c:
18372 update and move gstcompat.h
18373 The purpose of gstcompat.h is to couple the API differences among
18374 gstreamer-1.0 and gstreamer-0.10. Since gstreamer-0.10 is obsolete, the code
18375 in this compatibility layer shall be removed.
18376 Nevertheless, the gstcompat.h header should be kept, if new incompatibilites
18377 appear in the future, but it shall live in gst/vaapi, not in gst-libs.
18378 This patch removes the crumbs defined gstcompat.h and moves it to gst/vaapi.
18379 In order to avoid layer violations, gstcompat.h includes sysdeps.h and all
18380 the includes in gst/vaapi of sysdeps.h are replaced with gstcompat.h
18381 https://bugzilla.gnome.org/show_bug.cgi?id=745728
18382 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18383 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18385 2015-04-03 17:05:45 +0300 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18388 * gst/vaapi/Makefile.am:
18389 autotools: remove gstreamer-1.0 support
18390 This patch only removes the support of gstreamer-1.0 in the autotools
18391 scripts. No other files are touched.
18392 In the automake file all the converters were deprecated.
18393 https://bugzilla.gnome.org/show_bug.cgi?id=745728
18394 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18395 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18397 2015-04-03 17:03:38 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18402 Remove the gstreamer-videoutils submodule
18404 2015-04-03 17:01:45 +0300 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18407 * gst-libs/gst/Makefile.am:
18408 * gst-libs/gst/vaapi/Makefile.am:
18409 * gst-libs/gst/video/Makefile.am:
18410 * gst/vaapi/Makefile.am:
18411 * tests/Makefile.am:
18412 Remove libgstvaapi-videoutils.so
18413 This library was intended to add the base classes for video decoders which
18414 where not included in gstreamer-0.10.
18415 Since the support of gstreamer-0.10 is deprecated those classes are not
18416 required, thus the whole library is removed.
18417 https://bugzilla.gnome.org/show_bug.cgi?id=745728
18418 https://bugzilla.gnome.org/show_bug.cgi?id=732666
18419 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18421 2015-04-03 16:55:43 +0300 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18424 * gst-libs/gst/vaapi/gstvaapicontext_overlay.c:
18425 * gst-libs/gst/vaapi/gstvaapiutils.c:
18426 * tests/test-subpicture.c:
18427 Remove HAVE_GST_VIDEO_OVERLAY_HWCAPS macro
18428 This macro guarded the use of HAVE_GST_VIDEO_OVERLAY_HWCAPS, which was not
18429 defined before gstreamer 0.10.35. Since the support of gstreamer-0.10 is
18430 deprecated these guards are not required.
18431 https://bugzilla.gnome.org/show_bug.cgi?id=745728
18432 https://bugzilla.gnome.org/show_bug.cgi?id=732666
18433 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18435 2015-04-03 16:55:27 +0300 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18437 * gst-libs/gst/vaapi/gstcompat.h:
18438 * gst-libs/gst/vaapi/gstvaapicontext_overlay.c:
18439 * gst-libs/gst/vaapi/gstvaapidecoder.c:
18440 * gst-libs/gst/vaapi/gstvaapiimage.c:
18441 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
18442 * gst-libs/gst/vaapi/video-format.c:
18443 * gst-libs/gst/vaapi/video-format.h:
18444 * gst/vaapi/gstvaapi.c:
18445 * gst/vaapi/gstvaapidecode.c:
18446 * gst/vaapi/gstvaapidecodebin.c:
18447 * gst/vaapi/gstvaapidownload.c:
18448 * gst/vaapi/gstvaapidownload.h:
18449 * gst/vaapi/gstvaapiencode.c:
18450 * gst/vaapi/gstvaapiencode_h264.c:
18451 * gst/vaapi/gstvaapiencode_jpeg.c:
18452 * gst/vaapi/gstvaapiencode_mpeg2.c:
18453 * gst/vaapi/gstvaapiencode_vp8.c:
18454 * gst/vaapi/gstvaapiparse.c:
18455 * gst/vaapi/gstvaapipluginbase.c:
18456 * gst/vaapi/gstvaapipluginbase.h:
18457 * gst/vaapi/gstvaapipluginutil.c:
18458 * gst/vaapi/gstvaapipluginutil.h:
18459 * gst/vaapi/gstvaapipostproc.c:
18460 * gst/vaapi/gstvaapisink.c:
18461 * gst/vaapi/gstvaapiupload.c:
18462 * gst/vaapi/gstvaapiupload.h:
18463 * gst/vaapi/gstvaapiuploader.c:
18464 * gst/vaapi/gstvaapivideobuffer.c:
18465 * gst/vaapi/gstvaapivideoconverter_glx.c:
18466 * gst/vaapi/gstvaapivideoconverter_x11.c:
18467 * gst/vaapi/gstvaapivideometa.c:
18468 * gst/vaapi/gstvaapivideometa.h:
18469 * tests/test-filter.c:
18470 * tests/test-subpicture.c:
18471 Removal of gstreamer-0.10 support
18472 This patch removes all the pre-processor conditional code compilation guarded
18473 for gstreamer-0.10.
18474 https://bugzilla.gnome.org/show_bug.cgi?id=745728
18475 https://bugzilla.gnome.org/show_bug.cgi?id=732666
18476 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18478 2015-04-03 16:54:54 +0300 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18481 * debian.upstream/control.in:
18482 * gst-libs/gst/video/Makefile.am:
18483 * gst/vaapi/Makefile.am:
18484 autotools: remove gstreamer-0.10 support
18485 This patch only removes the support of gstreamer-0.10 in the autotools
18486 scripts. No other files are touched.
18487 The configuration parameter --gstreamer-api was deleted since now it is always
18489 The verification of vmethod query in GstBaseSinkClass was removed since it was
18490 added in gstreamer 0.10.35. The same case for GstVideoOverlayComposition and
18492 The precious variable GST_PLUGIN_PATH was removed, while GST_PLUGIN_PATH_1_0
18494 The automake files were changed accordingly.
18495 Removed, in debian/control, the vaapiupload and vaapidownload descriptions.
18496 https://bugzilla.gnome.org/show_bug.cgi?id=732666
18497 https://bugzilla.gnome.org/show_bug.cgi?id=745728
18498 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18500 2015-03-16 23:38:18 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18502 * gst/vaapi/gstvaapidecode.c:
18503 vaapidecode: add drain() vmethod
18504 In GStremer v1.6 a new vmethod drain() was added in GstVideoDecoder
18505 class. This patch implements this new method.
18506 https://bugzilla.gnome.org/show_bug.cgi?id=742922
18507 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18508 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18510 2015-03-16 23:37:29 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18512 * gst/vaapi/gstvaapidecode.c:
18513 vaapidecode: remove vmethod reset()
18514 Since in bug #745728 the support for GStreamer 1.0 is going to be dropped,
18515 this patch removes the method reset() which was deprecated in GStreamer 1.2.
18516 https://bugzilla.gnome.org/show_bug.cgi?id=742922
18517 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18518 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18520 2015-03-16 23:36:33 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18522 * gst/vaapi/gstvaapidecode.c:
18523 vaapidecode: handle flush() vmethod
18524 Since GStreamer 1.2 the vmethod reset() in GstVideoDecoderClass was deprecated
18525 and flush() was added.
18526 This patch set the vmethod flush() if the installed GStreamer version is 1.2 or
18527 superior. Otherwise, reset() is set.
18528 v2: 1) In order to avoid symbol collision, the old method gst_vaapidecode_flush()
18529 was renamed to gst_vaapidecode_internal_flush().
18530 2) The new vmethod flush() always do a hard full reset.
18531 v3: 1) Call gst_vaapidecode_internal_flush() first in flush() vmethod, in order to
18532 gather all collected data with gst_video_decoder_have_frame()
18533 https://bugzilla.gnome.org/show_bug.cgi?id=742922
18534 Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18535 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18537 2015-03-16 23:10:53 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18539 * gst/vaapi/gstvaapidecode.c:
18540 vaapidecode: call the correct query function
18541 In commit 2f8c115 (vaapidecode: use the query virtual methods in 1.4)
18542 a bug was introduced: when calling the parent's query function of the
18543 src pad, the one of the sink pad is called instead. This patch fixes
18545 https://bugzilla.gnome.org/show_bug.cgi?id=746248
18547 2015-03-15 00:36:45 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18550 gitmodules: Use https:// url instead of git:// for submodules.
18551 Gitorious is failing to clone repositories over git:// url.
18553 2015-03-14 22:12:19 +0200 Julien Isorce <j.isorce@samsung.com>
18555 * gst-libs/gst/vaapi/gstvaapidisplay.c:
18556 vaapidisplay: mark X11 display as compatible with EGL
18557 GST_GL_WINDOW=x11 GST_GL_API=gles2 GST_GL_PLATFORM=egl
18558 gst-launch-1.0 ... ! vaapidecode ! glimagesink
18559 https://bugzilla.gnome.org/show_bug.cgi?id=745902
18560 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
18562 2015-03-06 15:20:01 +0200 Olivier Crete <olivier.crete@collabora.com>
18564 * gst/vaapi/gstvaapidecode.c:
18565 vaapidecode: Don't crash if a buffer outlives the decoder
18566 Sometimes, for example, when switching video streams but keeping
18567 the same sink, the surface will be released after the decoder is
18568 stopped and replaced. This caused a crash because the release
18569 callback was called on an invalid pointer.
18570 The patch adding an additional reference to the decoder object in the buffer.
18571 https://bugzilla.gnome.org/show_bug.cgi?id=745189
18572 Signed-off-by: Olivier Crete <olivier.crete@collabora.com>
18573 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18575 2015-03-06 14:31:21 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18577 * gst/vaapi/gstvaapidecode.c:
18578 vaapidecode: clean-ups (indentation, drop unused variables)
18580 2015-03-06 14:09:22 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18582 * gst/vaapi/gstvaapidecode.c:
18583 vaapidecode: use the query virtual methods in 1.4
18584 GstVideoDecoder, the base class of vaapidecode, added support for
18585 pad queries as virtual methods. This patch enables the use of that
18586 support, while keeping support for lower versions of gstreamer.
18587 This patch is important because GstVideoDecoder takes care of other
18588 queries that might be important in the pipeline managing.
18589 v2: 1) rebase to current master
18590 2) fix indentation with gst-indent
18591 3) simplify the patch layout
18592 4) fix the context query
18593 5) initialise the filter to NULL
18594 6) improve the query log message for gst-1.2
18595 https://bugzilla.gnome.org/show_bug.cgi?id=744406
18597 2015-03-06 12:16:17 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18599 * gst/vaapi/gstvaapipostproc.c:
18600 * gst/vaapi/gstvaapipostproc.h:
18601 vaapipostproc: always activate buffer pool
18602 The vaapipostproc has a proxy flag to know if the the buffer pool is
18603 already active. But this fails in some situations where it is needed
18604 to renegotiate the buffer pool.
18605 This patch removes that flag so the renegotiation is done whenever is
18607 https://bugzilla.gnome.org/show_bug.cgi?id=745535
18609 2015-03-02 17:04:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18611 * gst/vaapi/gstvaapisink.c:
18612 vaapisink: fix minor memory leak in debug mode.
18613 The gst_video_colorimetry_to_string() function returns a newly created
18614 string that represents the GstVideoColorimetry value. So, that needs
18615 to be released after usage, in e.g. GST_DEBUG().
18617 2015-03-03 12:37:41 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18619 * gst/vaapi/gstvaapidecodebin.h:
18620 vaapidecodebin: Avoid usage of "__" prefix in macro names
18621 Avoiding "__" prefix usage in Header File Guards as per
18622 C standard recommendation.
18624 2015-03-03 12:31:11 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18626 * gst/vaapi/gstvaapi.c:
18627 plugins: Disable vaapidecodebin for GStreamer < 1.4
18628 There are autoplugging issues in GStreamer-1.2.
18629 Lets disable vaapidecodebin untill we get some workarounds for this.
18631 2015-03-02 15:19:40 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18633 * gst/vaapi/Makefile.am:
18634 * gst/vaapi/gstvaapi.c:
18635 * gst/vaapi/gstvaapidecodebin.c:
18636 * gst/vaapi/gstvaapidecodebin.h:
18637 plugins: Add a vaapidecodebin element
18638 Add a "vaapidecodebin" element to vaapi plugins.
18639 Child Elements: "vaapidecode ! queue ! vaapipostproc"
18640 The Reasons for implementing a new bin element:
18641 -- Help to Autoplug Hardware Accelerated Video Postprocessing element in playbin
18642 with out any dependency to upstream gstreamer.
18643 This is to overcome the *unacceptable* delay in upstream gstreamer to get new
18644 features in. Eg: https://bugzilla.gnome.org/show_bug.cgi?id=687182.
18645 Also customers using older gstreamer versions (1.2 and 1.4) will get the
18646 benefit of autoplugging, hardware accelerated deinterlacing support etc.
18647 -- Help to maintain a single thread implementation in vaapidecode.
18648 This will result a dead-lock free vaapidecode in most of the cases.
18649 More details here: https://bugzilla.gnome.org/show_bug.cgi?id=742605
18650 https://bugzilla.gnome.org/show_bug.cgi?id=745216
18652 2015-03-02 14:59:16 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18654 * gst/vaapi/gstvaapidecode.c:
18655 vaapidecode: re-indent (gst-indent) gstvaapidecode.c
18657 2015-03-02 14:46:38 +0200 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
18659 * gst/vaapi/gstvaapidecode.c:
18660 * gst/vaapi/gstvaapidecode.h:
18661 vaapidecode: Switch back to Single thread implementation
18662 Because the decoder uses the thread from handle_frame() to decode a frame,
18663 the src pad task creates an unsolveable AB-BA deadlock between
18664 handle_frame() waiting for a free surface and decode_loop() pushing
18665 decoded frames out.
18666 Instead, have handle_frame() take responsibility for pushing surfaces,
18667 and remove the deadlock completely. If you need a separate thread
18668 downstream, you can insert a queue between vaapidecode and its downstream
18670 Another justification for the single thread implementation is,
18671 there are two many point of locking in gstreamer-vaapi's current
18672 implementation which can lead to deadlocks.
18673 https://bugzilla.gnome.org/show_bug.cgi?id=742605
18674 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
18675 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
18676 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18678 2015-03-02 13:28:41 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18680 * gst/vaapi/gstvaapipluginbase.c:
18681 plugins: fix detection of upstream v4l2src element.
18682 Improve check for upstream element that requires DMABUF buffer pool,
18683 e.g. v4l2src element. In particular, make sure to traverse through
18684 any additional capsfilter for instance.
18685 Note: the traversal to the top-most upstream element could be made
18686 more generic, but we are insofar only interested in supporting pipes
18687 similar to v4l2src or v4l2src ! capsfilter, e.g. with an explicit
18688 specification for a desired video camera format, or resolution.
18690 2015-03-02 11:12:53 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18692 * gst/vaapi/gstvaapivideomemory.c:
18693 plugins: fix allocation of DMABUF memory.
18694 The dmabuf allocator would close the DMABUF handle passed in the init
18695 function gst_dmabuf_allocator_alloc(). So, we need to dup() it so that
18696 to avoid a double close, ultimately in the underlying driver that owns
18699 2015-02-26 12:28:02 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
18701 * gst/vaapi/gstvaapidecode.c:
18702 vaapidecode: keep src caps and output state in sync
18703 vaapidecode keeps an output state that use the format
18704 GST_VIDEO_FORMAT_ENCODED, while it crafts a different src caps
18705 for a correct negotiation.
18706 I don't see the rational behind this decoupling, it looks like
18707 unnecessary complexity. This patch simplify this logic keeping
18708 in sync the output state and the src caps.
18709 This patch improves the readability of the function
18710 gst_vaapidecode_update_src_caps() and simplify its logic. Also,
18711 the patch validates if the buffer pool has the configuration for
18712 the GL texture upload meta, in order to set the caps feature
18713 meta:GLTextureUpload. Otherwise, the I420 format is set back.
18714 https://bugzilla.gnome.org/show_bug.cgi?id=744618
18715 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
18716 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18718 2015-02-26 12:26:54 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
18720 * gst/vaapi/gstvaapidecode.c:
18721 vaapidecode: upload meta only if feature and allocation
18722 When vaapidecode finishes the decoding of a frame and pushes it,
18723 if, in the decide_allocation() method, it is determined if the
18724 next element supports the GL texture upload meta feature, the
18725 decoder adds the buffer's meta.
18726 Nonetheless, in the same spirit of the commit 71d3ce4d, the
18727 determination if the next element supports the GL texture upload
18728 meta needs to check both the preferred caps feature *and* if the
18729 allocation query request the API type.
18730 This patch, first removes the unused variable need_pool, and
18731 determines the attribute has_texture_upload_meta using the
18732 preferred caps feature *and* the allocation query.
18733 Also, the feature passed to GstVaapPluginBase is not longer
18734 determined by has_texture_upload_meta, but by the computed
18736 https://bugzilla.gnome.org/show_bug.cgi?id=744618
18737 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
18738 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18740 2015-02-26 12:24:55 +0200 Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
18742 * gst/vaapi/gstvaapidecode.c:
18743 * gst/vaapi/gstvaapidecode.h:
18744 * gst/vaapi/gstvaapipluginutil.c:
18745 * gst/vaapi/gstvaapipluginutil.h:
18746 vaapidecode: delayed src caps negotiation
18747 Currently the src caps are set immediately after the sink caps are set, but in
18748 that moment the pipeline might not fully constructed and the video sink has
18749 not negotiated its supported caps and features. As a consequence, in many cases
18750 of playback, the least optimized caps feature is forced. This is partially the
18751 responsible of bug #744039.
18752 Also, vaapidecode doesn't attend the reconfigure events from downstream,
18753 which is a problem too, since the video sink can be changed with different
18755 This patch delays the src caps, setting them until the first frame arrives to
18756 the decoder, assuming until that very moment the whole pipeline is already
18757 negotiated. Particularly, it checks if the src pad needs to be reconfigured,
18758 as a consequence of a reconfiguration event from downstream.
18759 A key part of this patch is the new GstVaapiCapsFeature
18760 GST_VAAPI_CAPS_FEATURE_NOT_NEGOTIATED, which is returned when the src pad
18761 doesn't have a peer yet. Also, for a better report of the caps allowed
18762 through the src pad and its peer, this patch uses gst_pad_get_allowed_caps()
18763 instead of gst_pad_peer_query_caps() when looking for the preferred feature.
18764 v3: move the input_state unref to close(), since videodecoder resets at
18765 some events such as navigation.
18766 v4: a) the state_changed() callback replaces the input_state if the media
18767 changed, so this case is also handled.
18768 b) since the parameter ref_state in gst_vaapidecode_update_src_caps() is
18769 always the input_state, the parameter were removed.
18770 c) there were a lot of repeated code handling the input_state, so I
18771 refactored it with the function gst_vaapi_decode_input_state_replace().
18772 https://bugzilla.gnome.org/show_bug.cgi?id=744618
18773 Signed-off-by: Víctor Manuel Jáquez Leal <victorx.jaquez@intel.com>
18774 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18776 2015-02-24 17:14:33 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
18778 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
18779 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
18780 encoder: h264: add support for more than 2 views
18781 Add support for H.264 MVC Multiview High profile encoding with
18782 more than 2 views. All views within the same accesss unit are
18783 provided in increasing order of view order index (VOIdx).
18784 Upto 10 view are supported for now.
18785 A new property "view-ids" has been provided for the plugins to
18786 set the view ids (which is an array of guint values) to be used
18788 https://bugzilla.gnome.org/show_bug.cgi?id=732453
18790 2015-02-23 16:55:36 +0100 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
18792 * gst/vaapi/gstvaapipluginbase.c:
18793 plugins: upload meta only if feature and allocation
18794 Working on bug #743687, I realized that vaapidecode always adds to its buffer
18795 pool the config option GST_BUFFER_POOL_OPTION_VIDEO_GL_TEXTURE_UPLOAD_META if
18796 the decide_allocation()'s query has GST_VIDEO_GL_TEXTURE_UPLOAD_META_API_TYPE.
18797 Nevertheless, there are occasions where the query has the API type, but the
18798 last negotiated caps don't have the feature meta:GstVideoGLTextureUploadMeta.
18799 Under this contradiction, vaapidecode adds the GLTextureUploadMeta API to its
18800 buffer pool configuration, and adds its buffer's meta to each output buffer,
18801 even if the negotiated caps feature is memory:SystemMemory with I420 color
18803 This kind of output buffers chokes ClutterAutoVideosSink, since it uses a map
18804 that relates caps <-> GL upload method. If it receives a buffer with color
18805 format I420, it assumes that it doesn't have a texture upload meta, because
18806 only those with RGB color format has it. Our buffers, with I420 format, say
18807 that they have the upload meta too. In that case the mapped method is a dummy
18808 one which does nothing. I reported this issue in bug #744039 (the patch,
18809 obviously, was rejected).
18810 This patch workarounds the problem: the buffer pool's configuration option
18811 GST_BUFFER_POOL_OPTION_VIDEO_GL_TEXTURE_UPLOAD_META is set if and only if the
18812 query has the GST_VIDEO_GL_TEXTURE_UPLOAD_META_API_TYPE *and* the negotiated
18813 caps feature is meta:GstVideoGLTextureUploadMeta.
18814 I have tested these patches with gst-master (1.5), gst-1.4 and gst-1.2 and
18815 in all they seem to work correctly.
18816 https://bugzilla.gnome.org/show_bug.cgi?id=744618
18817 [adapted to fit current EGL changes]
18818 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18820 2015-02-20 15:13:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18822 * gst-libs/gst/vaapi/gstvaapitexture.c:
18823 * gst-libs/gst/vaapi/gstvaapitexture.h:
18824 * gst-libs/gst/vaapi/gstvaapitexture_glx.c:
18825 * gst-libs/gst/vaapi/gstvaapitexture_priv.h:
18826 * gst/vaapi/gstvaapivideometa_texture.c:
18827 plugins: add support for GstVideoGLTextureOrientation.
18828 Add support for GstVideoGLTextureOrientation modes. In particular,
18829 add orientation flags to the GstVaapiTexture wrapper and the GLX
18830 implementations. Default mode is that texture memory is laid out
18831 with top lines first, left row first. Flags indicate whether the
18832 X or Y axis need to be inverted.
18834 2015-02-09 21:09:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18836 * gst/vaapi/gstvaapidecode.c:
18837 * gst/vaapi/gstvaapipluginbase.c:
18838 * gst/vaapi/gstvaapipluginutil.c:
18839 * gst/vaapi/gstvaapipluginutil.h:
18840 * gst/vaapi/gstvaapipostproc.c:
18841 * gst/vaapi/gstvaapivideometa_texture.c:
18842 plugins: add support for BGRA textures.
18843 Some frameworks (EFL) expect BGRA textures for storage. However,
18844 adding support for that broadly into GStreamer framework implies
18845 two kinds of hacks: (i) libgstgl helpers currently do not support
18846 BGRA textures correctly, (ii) we need to better parse downstream
18847 suggested caps and intersect them with what the VA plugin elements
18848 can offer to them for GL texturing.
18850 2015-01-23 09:31:57 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18852 * gst/vaapi/gstvaapipluginutil.c:
18853 plugins: fix support for Wayland/EGL running alongside X11.
18854 When multiple display servers are available, the glimagesink element
18855 (from GStreamer 1.4) may not be able to derive a global display in
18856 Wayland. Rather, a "window"-specific display is created. In this case,
18857 the GstGLDisplay handle available through GstGLContext is invalid.
18858 So, try to improve heuristics for display server characterisation in
18859 those particular situations.
18861 2015-02-20 15:29:17 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18864 * gst/vaapi/Makefile.am:
18865 * gst/vaapi/gstvaapidecode.c:
18866 * gst/vaapi/gstvaapipluginbase.c:
18867 * gst/vaapi/gstvaapipluginutil.c:
18868 * gst/vaapi/gstvaapivideobufferpool.c:
18869 * gst/vaapi/gstvaapivideometa_texture.c:
18870 plugins: add initial support for EGL.
18871 Add initial support for EGL through GstVideoGLTextureUploadMeta.
18872 Fix gst_vaapi_ensure_display() to allocate a GstVaapiDisplay off the
18873 downstream supplied GstGLContext configuration, i.e. use its native
18874 display handle to create a GstVaapiDisplay of type X11 or Wayland ;
18875 and use the desired OpenGL API to allocate the GstVaapiDisplayEGL
18877 https://bugzilla.gnome.org/show_bug.cgi?id=741079
18879 2014-12-09 11:46:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18881 * gst/vaapi/gstvaapivideometa_texture.c:
18882 plugins: track video texture size changes.
18883 Sync video texture sizes to GstVideoGLTextureUploadMeta private date,
18884 i.e. GstVaapiVideoMetaTexture, on a regular basis. In particular, we
18885 now update the texture size from the GstVideoMeta, if any, or reset
18886 to some defaults otherwise.
18888 2014-12-03 15:45:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18890 * gst/vaapi/gstvaapipluginbase.c:
18891 * gst/vaapi/gstvaapipluginbase.h:
18892 * gst/vaapi/gstvaapipluginutil.c:
18893 plugins: ensure VA display matches GL context expectations.
18894 If a GstGLContext is supplied by the downstream element, then make
18895 sure that the VA plugin element gets a compatible display to what
18896 is requested by the GL context. e.g. re-allocate a VA/GLX display
18897 when a GLX context is provided by the downstream element.
18899 2014-12-03 14:14:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18902 * gst/vaapi/Makefile.am:
18903 * gst/vaapi/gstvaapipluginbase.c:
18904 * gst/vaapi/gstvaapipluginbase.h:
18905 plugins: record downstream GstGLContext.
18906 Record GL context supplied by downstream elements. This can be useful,
18907 and further needed, to enforce run-time check that the GL context is
18908 compatible for use by libgstvaapi. e.g. check that we don't create a
18909 VA/GLX display for EGL/X11 contexts.
18910 https://bugzilla.gnome.org/show_bug.cgi?id=725643
18911 Original-path-by: Matthew Waters <ystreet00@gmail.com>
18913 2014-12-01 14:52:39 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18915 * tests/Makefile.am:
18918 Add initial support for EGL to tests. The new EGL backend can be selected
18919 through the --egl command line option. The OpenGL|ES version can further
18920 be selected with the --gles-version command line option, where the default
18921 of 0 means "desktop" OpenGL.
18923 2015-01-27 16:21:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18925 * gst-libs/gst/vaapi/Makefile.am:
18926 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
18927 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
18928 * gst-libs/gst/vaapi/gstvaapiutils_egl.h:
18929 * gst-libs/gst/vaapi/gstvaapiwindow_egl.c:
18930 * gst-libs/gst/vaapi/gstvaapiwindow_egl.h:
18931 egl: add windowing support.
18932 This provides for some basic EGL window abstraction.
18934 2015-01-24 08:29:57 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18936 * gst-libs/gst/vaapi/Makefile.am:
18937 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
18938 * gst-libs/gst/vaapi/gstvaapitexture_egl.c:
18939 * gst-libs/gst/vaapi/gstvaapitexture_egl.h:
18940 egl: add texture abstraction.
18941 Add GstVaapiTextureEGL abstraction that can create its own GL texture,
18942 or import a foreign allocated one, while still allowing updates from a
18945 2014-12-09 18:14:56 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18947 * gst-libs/gst/vaapi/Makefile.am:
18948 * gst-libs/gst/vaapi/gstvaapisurface_egl.c:
18949 * gst-libs/gst/vaapi/gstvaapisurface_egl.h:
18950 egl: allow for EGLImage imports into VA Surfaces.
18951 Add helpers to import EGLImage objects into VA surfaces. There are
18952 two operational modes: (i) gst_vaapi_surface_new_from_egl_image(),
18953 which allows for implicit conversion from EGLImage to a VA surface
18954 in native video format, and (ii) gst_vaapi_surface_new_with_egl_image(),
18955 which exactly wraps the source EGLImage, typically in RGBA format
18956 with linear storage.
18957 Note: in case of (i), the EGLImage can be disposed right after the
18958 VA surface creation call, unlike in (ii) where the user shall ensure
18959 that the EGLImage is live until the associated VA surface is no longer
18961 https://bugzilla.gnome.org/show_bug.cgi?id=743847
18963 2015-02-20 15:27:53 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18966 * gst-libs/gst/vaapi/Makefile.am:
18967 * gst-libs/gst/vaapi/egl_compat.h:
18968 * gst-libs/gst/vaapi/egl_vtable.h:
18969 * gst-libs/gst/vaapi/gstvaapidisplay.c:
18970 * gst-libs/gst/vaapi/gstvaapidisplay.h:
18971 * gst-libs/gst/vaapi/gstvaapidisplay_egl.c:
18972 * gst-libs/gst/vaapi/gstvaapidisplay_egl.h:
18973 * gst-libs/gst/vaapi/gstvaapidisplay_egl_priv.h:
18974 * gst-libs/gst/vaapi/gstvaapiutils_egl.c:
18975 * gst-libs/gst/vaapi/gstvaapiutils_egl.h:
18976 * gst-libs/gst/vaapi/ogl_compat.h:
18977 Add initial support for EGL.
18978 Add initial support for EGL to libgstvaapi core library. The target
18979 display server and the desired OpenGL API can be programmatically
18980 selected at run-time.
18981 A comprehensive set of EGL utilities are provided to support those
18982 dynamic selection needs, but also most importantly to ensure that
18983 the GL command stream is executed from within a single thread.
18984 https://bugzilla.gnome.org/show_bug.cgi?id=743846
18986 2015-01-30 21:38:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18988 * gst-libs/gst/vaapi/gstvaapivalue.c:
18989 libs: initialize GValues in a thread-safe manner.
18991 2015-01-30 21:35:10 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18993 * gst-libs/gst/vaapi/gstvaapivalue.c:
18994 libs: re-indent all GValue related source code.
18996 2015-01-22 22:45:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
18998 * gst/vaapi/gstvaapidecode.c:
18999 vaapidecode: partially revert 0777f35.
19000 Reset the VA decoder after updating the base plugin caps, and most
19001 importantly, after GstVideoDecoder negotiation. The reason behind
19002 this is that the negotiation could trigger a last decide_allocation()
19003 where we could actually derive a new GstVaapiDisplay to use from the
19004 downstream element. e.g. GLX backend.
19006 2015-02-19 13:37:09 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19008 * gst/vaapi/gstvaapidecode.c:
19009 vaapidecode: Caps query should return the list of all supported caps.
19010 Query caps filtering should be always done on top of allowed caps instead
19011 of existing fixed caps on a particular pad.
19012 This fixes the mvc stream decoding when there is a base view(high profile)
19013 and non-base view(stereo-high profile).
19015 2015-02-18 13:36:16 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
19017 * gst/vaapi/gstvaapidecode.c:
19018 vaapidecode: intersect filter from query caps
19019 According to documentation[1] when receiving a GST_QUERY_CAPS
19020 the return value should be all formats that this elements supports,
19021 taking into account limitations of peer elements further downstream
19022 or upstream, sorted by order of preference, highest preference first.
19023 This patch add those limitations intersecting with the received
19024 filter in the query. Also takes into account the already negotiated
19025 caps. Also adds the processing of the query on the SRC pad.
19026 1. http://gstreamer.freedesktop.org/data/doc/gstreamer/head/pwg/html/section-nego-getcaps.html
19027 https://bugzilla.gnome.org/show_bug.cgi?id=744406
19029 2015-02-18 11:46:11 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
19031 * gst-libs/gst/vaapi/Makefile.am:
19032 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
19033 * gst-libs/gst/vaapi/gstvaapidecoder.c:
19034 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
19035 * gst-libs/gst/vaapi/gstvaapidisplay.c:
19036 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
19037 * gst/vaapi/gstvaapivideomemory.c:
19038 Fix compiler warnings
19039 This patch fixes some warnings that gcc 4.9 reports.
19040 https://bugzilla.gnome.org/show_bug.cgi?id=744411
19042 2015-02-18 11:22:21 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
19044 * gst/vaapi/gstvaapidecode.c:
19045 vaapidecode: Use GST_DEBUG_FUNCPTR for gst_vaapidecode_query()
19046 Hence the function name is shown in the gst-inspect-1.0 information
19047 rather than the memory address.
19048 https://bugzilla.gnome.org/show_bug.cgi?id=744330
19050 2015-02-18 11:21:35 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
19052 * gst/vaapi/gstvaapidecode.c:
19053 vaapidecode: log flow error name
19054 https://bugzilla.gnome.org/show_bug.cgi?id=744387
19056 2015-02-18 11:20:42 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
19058 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
19059 VC1: decoder: Ignore VC1 user BDU's
19060 Don't return error if the processed BDU is a user one, just ignore them.
19061 https://bugzilla.gnome.org/show_bug.cgi?id=741237
19062 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19064 2015-02-18 11:19:26 +0200 Olivier Crete <olivier.crete@collabora.com>
19066 * gst/vaapi/gstvaapidecode.c:
19067 * gst/vaapi/gstvaapipluginbase.c:
19068 vaapidecode: Emit error GstMessage when returning a GST_FLOW_ERROR
19069 This is required in GStreamer, elements should never return
19070 GST_FLOW_ERROR without posting an ERROR message on the bus.
19071 https://bugzilla.gnome.org/show_bug.cgi?id=744620
19073 2015-02-13 13:45:32 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19075 * gst/vaapi/Makefile.am:
19076 * gst/vaapi/gstvaapi.c:
19077 * gst/vaapi/gstvaapiencode_vp8.c:
19078 * gst/vaapi/gstvaapiencode_vp8.h:
19079 plugins: Add VP8 Encoder
19081 2015-02-13 13:42:04 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19083 * gst-libs/gst/vaapi/Makefile.am:
19084 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.c:
19085 * gst-libs/gst/vaapi/gstvaapiencoder_vp8.h:
19086 Add VP8 Encoder to core libgstvaapi.
19088 2015-02-13 13:40:19 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19091 configure: Add Check for VP8 Encoding API
19093 2015-02-10 11:40:16 +0200 Lim Siew Hoon <siew.hoon.lim@intel.com>
19095 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
19096 decoder: vc1: Rounding control handling for VC1 simple and Main profile
19097 Added rounding control handling for VC1 simple and Main profile
19098 based on VC1 standard spec: section 8.3.7
19099 https://bugzilla.gnome.org/show_bug.cgi?id=743958
19100 Signed-off-by: Lim Siew Hoon <siew.hoon.lim@intel.com>
19101 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19103 2015-02-06 12:10:51 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19105 * gst/vaapi/Makefile.am:
19106 build: fix make dist when certain conditionals not met.
19107 Fix typo which was preventing the inclusion of jpeg encoder
19108 source files from make dist (when there is no jpeg encoder
19109 API support in libva).
19111 2015-02-05 13:08:53 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19113 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
19114 encoder: jpeg: Fix the sampling factor calculation for ENCODED format.
19115 If the incoming raw video format is GST_VIDEO_FORMAT_ENCODED,
19116 use native YUV420 format (which is i420) as default.
19118 2015-02-05 12:13:38 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19120 * gst-libs/gst/vaapi/gstvaapiencoder.c:
19121 encoder: Only support YUV420 native format as input for now
19122 Practically we should be able to support more formats, for eg:
19123 JPEG Encoder can support YUV422, RGBA and all.
19124 But this is causing more issues which need proper fix here and there.
19126 2015-02-04 18:34:59 +0200 Olivier Crete <olivier.crete@collabora.com>
19128 * gst-libs/gst/vaapi/gstvaapidecoder.h:
19129 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
19130 * gst/vaapi/gstvaapidecode.c:
19131 vaapidecode: Check the condition after taking the lock
19132 Otherwise the condition could become true before the lock
19133 is taken and the g_cond_signal() could be called
19134 before the g_cond_wait(), so the g_cond_wait() is never
19136 https://bugzilla.gnome.org/show_bug.cgi?id=740645
19138 2015-02-04 11:18:29 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19140 * gst/vaapi/Makefile.am:
19141 * gst/vaapi/gstvaapi.c:
19142 * gst/vaapi/gstvaapiencode_jpeg.c:
19143 * gst/vaapi/gstvaapiencode_jpeg.h:
19144 plugins: Add JPEG encoder element
19146 2015-02-04 11:17:58 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19148 * gst-libs/gst/vaapi/gstvaapicontext.c:
19149 gstvaapicontext: Add VAConfigAttribValEncJPEG to the attribute list using for VAConfig creation.
19151 2015-02-04 11:17:27 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19153 * gst-libs/gst/vaapi/gstvaapicontext.c:
19154 gstvaapicontext: Don't use the unsupported Ratecontrol attributes for vaCreateConfig
19155 Don't add the VAConfigAttribRateControl to the attribute list using
19156 for the vaCreateConfig if it is not supported by the driver.
19158 2015-02-04 11:17:06 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19160 * gst-libs/gst/vaapi/gstvaapiencoder.c:
19161 gstvaapiencoder: Use hardcoded packed_raw_data flag for JPEG Encoding
19163 2015-02-04 11:16:37 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19165 * gst-libs/gst/vaapi/gstvaapiencoder.c:
19166 gstvaapiencoder: Fix crash when handling rate control mask
19167 Having a ratecontrol_mask equal to zero is not a bug, but the driver
19168 might not be supporting any kind of rate control mechanisms.
19171 2015-02-04 11:16:05 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19173 * gst-libs/gst/vaapi/Makefile.am:
19174 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.c:
19175 * gst-libs/gst/vaapi/gstvaapiencoder_jpeg.h:
19176 encoder: Add JPEG Encoder
19178 2015-02-04 11:15:38 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19180 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
19181 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
19182 encoder_objects: Add QuantizationMatrix and JPEGHuffmanTable
19184 2015-02-04 11:15:00 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19186 * gst-libs/gst/vaapi/gstvaapidisplay.c:
19187 * gst-libs/gst/vaapi/gstvaapiencoder.c:
19188 * gst-libs/gst/vaapi/gstvaapiprofile.c:
19189 * gst-libs/gst/vaapi/gstvaapiprofile.h:
19190 Encode: Add support for Picture level Entrypoint
19191 This is useful for JPEG encoding which is utilizing picture level
19192 entrypoint instead of slice level entrypoint like h264,mpeg2 etc.
19194 2015-02-04 11:14:15 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19197 configure: Add Check for JPEG encoding API
19199 2015-02-03 13:08:53 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19204 2015-02-03 13:08:01 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19207 Bump version for development.
19209 === release 0.5.10 ===
19211 2015-02-03 10:00:42 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19216 2015-02-03 10:00:23 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19221 2015-02-02 11:43:58 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19223 * gst/vaapi/gstvaapivideometa_texture.c:
19224 Fix compilation error if there is no GL/gl.h header file installed
19226 2015-01-28 18:09:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19228 * gst/vaapi/gstvaapivideomemory.h:
19229 plugins: drop leftover declaration.
19230 GstVaapiVideoMemory quark is not needed any more, and the actual
19231 implementation was already removed bfore the merge. i.e. this is
19232 an oversight for a hunk that was not meant to be pushed.
19234 2015-01-26 18:30:47 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19236 * gst/vaapi/gstvaapipluginbase.c:
19237 * gst/vaapi/gstvaapivideobufferpool.c:
19238 * gst/vaapi/gstvaapivideobufferpool.h:
19239 * gst/vaapi/gstvaapivideomemory.c:
19240 * gst/vaapi/gstvaapivideomemory.h:
19241 plugins: add support for dma_buf exports (v4l2src).
19242 Allow v4l2src element to connected to vaapipostproc or vaapisink when
19243 "io-mode" is set to "dmabuf-import". In practice, this is a more likely
19244 operational mode with uvcvideo. Supporting v4lsrc with "io-mode" set
19245 to "dmabuf" could work, but with more demanding driver or kernel reqs.
19246 Note: with GStreamer 1.4, v4l2src (gst-plugins-good) needs to be built
19247 with --without-libv4l2.
19248 https://bugzilla.gnome.org/show_bug.cgi?id=743635
19250 2014-01-23 05:00:09 -0500 Wind Yuan <feng.yuan@intel.com>
19253 * gst/vaapi/Makefile.am:
19254 * gst/vaapi/gstvaapipluginbase.c:
19255 plugins: add support for dma_buf imports.
19256 Allow imports of v4l2 buffers into VA surfaces for further operation
19257 with vaapi plugins, e.g. vaapipostproc or vaapiencode_* elements.
19258 https://bugzilla.gnome.org/show_bug.cgi?id=735362
19259 [fixed memory leaks, ported to new dma_buf infrastructure, cleanups]
19260 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19262 2014-09-15 15:27:50 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19264 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
19265 * gst-libs/gst/vaapi/gstvaapisurface_drm.h:
19266 surface: add support for GEM buffer imports.
19267 Add support for GEM buffer imports. This is useful for VA/EGL interop
19268 with legacy Mesa implementations, or when it is desired or required to
19269 support outbound textures for instance.
19270 https://bugzilla.gnome.org/show_bug.cgi?id=736718
19272 2014-09-15 15:25:09 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19274 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
19275 * gst-libs/gst/vaapi/gstvaapisurface_drm.h:
19276 surface: add support for dma_buf imports.
19277 Add new gst_vaapi_surface_new_with_dma_buf_handle() helper function
19278 to allow for creating VA surfaces from a foreign DRM PRIME fd. The
19279 resulting VA surface owns the supplied buffer handle.
19280 https://bugzilla.gnome.org/show_bug.cgi?id=735362
19282 2015-01-27 11:19:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19284 * gst-libs/gst/vaapi/gstvaapisurface.c:
19285 * gst-libs/gst/vaapi/gstvaapisurface.h:
19286 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
19287 surface: add initial support for foreign buffer imports.
19288 Add gst_vaapi_surface_new_from_buffer_proxy() helper function to
19289 create a VA surface from an external buffer provided throug the
19290 new GstVaapiBufferProxy object.
19292 2014-09-15 13:47:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19294 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
19295 * gst-libs/gst/vaapi/gstvaapibufferproxy.h:
19296 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
19297 * gst-libs/gst/vaapi/gstvaapisurface_drm.h:
19298 surface: add support for GEM buffer exports.
19299 Add support for GEM buffer exports. This will only work with VA drivers
19300 based off libdrm, e.g. the Intel HD Graphics VA driver. This is needed
19301 to support interop with EGL and the "Desktop" GL specification. Indeed,
19302 the EXT_image_dma_buf_import extension is not going to be supported in
19303 Desktop GL, due to the lack of support for GL_TEXTURE_EXTERNAL_OES targets
19305 This is useful for implementing VA/EGL interop with legacy Mesa stacks,
19306 in Desktop OpenGL context.
19307 https://bugzilla.gnome.org/show_bug.cgi?id=736717
19309 2014-09-15 11:48:05 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19311 * gst-libs/gst/vaapi/Makefile.am:
19312 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
19313 * gst-libs/gst/vaapi/gstvaapibufferproxy.h:
19314 * gst-libs/gst/vaapi/gstvaapisurface_drm.c:
19315 * gst-libs/gst/vaapi/gstvaapisurface_drm.h:
19316 surface: add support for dma_buf exports.
19317 Use the new VA buffer export APIs to allow for a VA surface to be
19318 exposed as a plain PRIME fd. This is in view to simplifying interop
19319 with EGL or OpenCL for instance.
19320 https://bugzilla.gnome.org/show_bug.cgi?id=735364
19322 2014-09-15 10:58:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19324 * gst-libs/gst/vaapi/Makefile.am:
19325 * gst-libs/gst/vaapi/gstvaapibufferproxy.c:
19326 * gst-libs/gst/vaapi/gstvaapibufferproxy.h:
19327 * gst-libs/gst/vaapi/gstvaapibufferproxy_priv.h:
19328 Add abstraction for exported VA buffers.
19329 The VA buffer export APIs work for a particular lifetime starting from
19330 vaAcquireBufferHandle() and ending with vaReleaseBufferHandle(). As such,
19331 it could be much more convenient to support implicit releases by simply
19332 having a refcount reaching zero.
19333 https://bugzilla.gnome.org/show_bug.cgi?id=736721
19335 2015-01-28 18:25:09 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19337 * gst-libs/gst/vaapi/Makefile.am:
19338 Add missing header file to Makefile
19339 Add gstvaapitexture_glx.h to Makefile.am
19341 2015-01-27 11:44:12 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19343 * gst-libs/gst/vaapi/gstvaapicontext.c:
19344 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
19345 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
19346 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
19347 * gst/vaapi/gstvaapipostproc.c:
19348 * gst/vaapi/gstvaapiuploader.c:
19349 * gst/vaapi/gstvaapivideobufferpool.c:
19350 * gst/vaapi/gstvaapivideomemory.c:
19351 * gst/vaapi/gstvaapivideomemory.h:
19352 * tests/test-surfaces.c:
19353 videopool: add optional flags for surface pool allocation.
19354 Reword surface pool allocation helpers so that to allow for a simple
19355 form, e.g. gst_vaapi_surface_pool_new(format, width, height); and a
19356 somewhat more elaborated/flexible form with optional allocation flags
19357 and precise GstVideoInfo specification.
19358 This is an API/ABI change, and SONAME version needs to be bumped.
19360 2015-01-26 23:21:56 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19362 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
19363 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
19364 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
19365 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
19366 * gst-libs/gst/vaapi/gstvaapivideopool.c:
19367 * gst-libs/gst/vaapi/gstvaapivideopool.h:
19368 * gst-libs/gst/vaapi/gstvaapivideopool_priv.h:
19369 videopool: re-indent all GstVaapiVideoPool related source code.
19371 2014-12-10 20:13:21 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19373 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
19374 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
19375 window: add toplevel display indirection for visualid and colormap.
19376 Add GstVaapiDisplay::get_{visual_id,colormap}() helpers to help determine
19377 the best suitable window visual id and colormap. This is an indirection in
19378 view to supporting EGL and custom/generic replacements.
19380 2014-12-10 19:58:10 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19382 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
19383 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
19384 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
19385 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
19386 window: add toplevel API to determine the colormap.
19387 Add GstVaapiWindowClass::get_colormap() hook to help determine the
19388 currently active colormap bound to the supplied window, or actually
19389 create it if it does not already exist yet.
19391 2014-12-10 19:36:12 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19393 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
19394 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
19395 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
19396 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
19397 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
19398 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
19399 window: add toplevel API to determine a visual id.
19400 Add GstVaapiWindowClass::get_visual_id() function hook to help find
19401 the best suitable visual id for the supplied window. While doing so,
19402 also simplify the process by which an X11 window is created with a
19403 desired Visual, i.e. now use a visual id instead of a Visual object.
19405 2014-12-10 18:12:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19407 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
19408 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
19409 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
19410 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
19411 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
19412 * gst-libs/gst/vaapi/gstvaapiwindow.c:
19413 * gst-libs/gst/vaapi/gstvaapiwindow.h:
19414 * gst-libs/gst/vaapi/gstvaapiwindow_drm.c:
19415 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
19416 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
19417 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
19418 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
19419 window: add generic helper to create windows.
19420 Add a new generic helper function gst_vaapi_window_new() to create
19421 a window without having the caller to check for the display type
19422 himself. i.e. internally, there is now a GstVaapiDisplayClass hook
19423 to create windows, and the actual backend implementation fills it in.
19424 Add new generic helper functions gst_vaapi_texture_new_wrapped()
19425 This is a simplification in view to supporting EGL.
19427 2014-12-03 11:39:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19429 * gst-libs/gst/vaapi/gstvaapidisplay.c:
19430 * gst-libs/gst/vaapi/gstvaapidisplay.h:
19431 * gst-libs/gst/vaapi/gstvaapitexture.c:
19432 * gst/vaapi/gstvaapivideometa_texture.c:
19433 display: add utility function to check for OpenGL rendering.
19434 Add gst_vaapi_display_has_opengl() helper function to help determining
19435 whether the display can support OpenGL context to be bound to it, i.e.
19436 if the class is of type GST_VAAPI_DISPLAY_TYPE_GLX.
19438 2014-12-10 18:02:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19440 * gst-libs/gst/vaapi/gstvaapidisplay.c:
19441 * gst-libs/gst/vaapi/gstvaapidisplay.h:
19442 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
19443 * gst-libs/gst/vaapi/gstvaapidisplay_drm_priv.h:
19444 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
19445 * gst-libs/gst/vaapi/gstvaapidisplay_glx_priv.h:
19446 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
19447 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
19448 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
19449 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
19450 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
19451 * gst/vaapi/gstvaapipluginutil.c:
19452 * gst/vaapi/gstvaapivideobuffer.c:
19453 * gst/vaapi/gstvaapivideometa_texture.c:
19454 display: refine the meaning of display type.
19455 Make gst_vaapi_display_get_display_type() return the actual VA display
19456 type. Conversely, add a gst_vaapi_display_get_class_type() function to
19457 return the type of the GstVaapiDisplay instance. The former is used to
19458 identify the display server onto which the application is running, and
19459 the latter to identify the original object class.
19461 2014-12-02 11:23:29 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19463 * gst-libs/gst/vaapi/gstvaapidisplay.c:
19464 * gst-libs/gst/vaapi/gstvaapidisplay.h:
19465 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
19466 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
19467 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
19468 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
19469 display: cosmetics (helper macros, new internal API names).
19470 Add more helper macros to the top-level GstVaapiDisplay interfaces.
19471 Rename a few others used internally for improved consistency.
19473 2014-12-01 17:08:29 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19475 * gst-libs/gst/vaapi/gstvaapidisplay.c:
19476 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
19477 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
19478 * gst-libs/gst/vaapi/gstvaapipixmap_x11.c:
19479 * gst-libs/gst/vaapi/gstvaapitexture_glx.c:
19480 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
19481 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
19482 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
19483 display: record native display object.
19484 Record the underlying native display instance into the toplevel
19485 GstVaapiDisplay object. This is useful for fast lookups to the
19486 underlying native display, e.g. for creating an EGL display.
19488 2014-12-01 16:54:32 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19490 * gst-libs/gst/vaapi/gstvaapidisplay.c:
19491 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
19492 * gst-libs/gst/vaapi/gstvaapidisplaycache.h:
19493 display: use a recursive mutex for the display cache.
19494 Use a recursive mutex for the display cache so that a 3rdparty display
19495 object could be initialized during the initialization of the parent
19498 2014-12-04 14:36:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19500 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
19501 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
19502 * gst-libs/gst/vaapi/gstvaapitexture.c:
19503 * gst-libs/gst/vaapi/gstvaapitexture.h:
19504 * gst-libs/gst/vaapi/gstvaapitexture_glx.c:
19505 * gst-libs/gst/vaapi/gstvaapitexture_priv.h:
19506 texture: add generic helper to create textures.
19507 Add new generic helper functions gst_vaapi_texture_new_wrapped()
19508 and gst_vaapi_texture_new() to create a texture without having
19509 the caller to uselessly check for the display type himself. i.e.
19510 internally, there is now a GstVaapiDisplayClass hook to create
19511 textures, and the actual backend implementation fills it in.
19512 This is a simplification in view to supporting EGL.
19514 2014-10-23 17:44:23 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19516 * gst-libs/gst/vaapi/Makefile.am:
19517 * gst-libs/gst/vaapi/gstvaapitexture.c:
19518 * gst-libs/gst/vaapi/gstvaapitexture.h:
19519 * gst-libs/gst/vaapi/gstvaapitexture_glx.c:
19520 * gst-libs/gst/vaapi/gstvaapitexture_glx.h:
19521 * gst-libs/gst/vaapi/gstvaapitexture_priv.h:
19522 * gst/vaapi/gstvaapivideoconverter_glx.c:
19523 * gst/vaapi/gstvaapivideometa_texture.c:
19524 * tests/test-textures.c:
19525 texture: move to core libgstvaapi base library.
19526 GstVaapiTexture is a generic abstraction that could be moved to the
19527 core libgstvaapi library. While doing this, no extra dependency needs
19528 to be added. This means that a GstVaapitextureClass is now available
19529 for any specific code that needs to be added, e.g. creation of the
19530 underlying GL texture objects, or backend dependent ways to upload
19531 a surface to the texture object.
19532 Generic OpenGL data types (GLuint, GLenum) are also replaced with a
19534 https://bugzilla.gnome.org/show_bug.cgi?id=736715
19536 2014-10-23 13:11:54 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19539 * gst-libs/gst/vaapi/Makefile.am:
19540 * gst-libs/gst/vaapi/gstvaapicompat.h:
19541 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
19542 * gst-libs/gst/vaapi/gstvaapitexture.c:
19543 texture: drop support for VA/GLX interfaces.
19544 The VA/GLX interfaces are obsolete. They used to exist for XvBA, and
19545 ease of use, but they had other caveats to deal with. It's now better
19546 to move on to legacy mode, whereby VA/GLX interop is two be provided
19547 through (i) X11 Pixmap, and (ii) other modern means of buffer sharing.
19548 https://bugzilla.gnome.org/show_bug.cgi?id=736711
19550 2014-10-23 11:56:31 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19552 * gst-libs/gst/vaapi/gstvaapitexture.c:
19553 * gst-libs/gst/vaapi/gstvaapitexture.h:
19554 * gst/vaapi/gstvaapivideoconverter_glx.c:
19555 * gst/vaapi/gstvaapivideometa_texture.c:
19556 * tests/test-textures.c:
19557 texture: add support for cropping rectangle during transfer.
19558 The gst_vaapi_texture_put_surface() function is missing a crop_rect
19559 argument that would be used during transfer for cropping the source
19560 surface to the desired dimensions.
19561 Note: from a user point-of-view, he should create the GstVaapiTexture
19562 object with the cropped size. That's the default behaviour in software
19563 decoding pipelines that we need to cope with.
19564 This is an API/ABI change, and SONAME version needs to be bumped.
19565 https://bugzilla.gnome.org/show_bug.cgi?id=736712
19567 2014-10-23 11:22:10 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19569 * gst-libs/gst/vaapi/gstvaapitexture.c:
19570 * gst-libs/gst/vaapi/gstvaapitexture.h:
19571 texture: re-indent all GstVaapiTexture related source code.
19573 2015-01-27 11:16:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19575 * gst-libs/gst/vaapi/gstvaapisurface.c:
19576 * gst-libs/gst/vaapi/gstvaapisurface.h:
19577 surface: add more fine-grained allocation helper.
19578 Add new gst_vaapi_surface_new_full() helper function that allocates
19579 VA surface from a GstVideoInfo template in argument. Additional flags
19580 may include ways to
19581 - allocate linear storage (GST_VAAPI_SURFACE_ALLOC_FLAG_LINEAR_STORAGE) ;
19582 - allocate with fixed strides (GST_VAPI_SURFACE_ALLOC_FLAG_FIXED_STRIDES) ;
19583 - allocate with fixed offsets (GST_VAAPI_SURFACE_ALLOC_FLAG_FIXED_OFFSETS).
19585 2014-09-15 14:57:57 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19587 * gst-libs/gst/vaapi/gstvaapisurface.c:
19588 * gst-libs/gst/vaapi/gstvaapisurface.h:
19589 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
19590 surface: re-indent all GstVaapiSurface related source code.
19592 2015-01-23 16:44:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19594 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
19595 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
19596 surfaceproxy: add helper to create a wrapped surface object.
19597 Add new gst_vaapi_surface_proxy_new() helper to wrap a surface into
19598 a proxy. The main use case for that is to convey additional information
19599 at the proxy level that would not be suitable to the plain surface.
19601 2015-01-23 16:37:06 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19603 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
19604 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
19605 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
19606 surfaceproxy: re-indent all GstVaapiSurfaceProxy related source code.
19608 2015-01-27 18:02:56 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19611 libs: bump library major version.
19613 2014-12-04 14:36:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19615 * gst-libs/gst/vaapi/gstvaapitypes.h:
19616 libs: re-introduce a GST_VAAPI_ID_INVALID value.
19617 Re-introduce a GST_VAAPI_ID_INVALID value that represents
19618 a non-zero and invalid id. This is useful to have a value
19619 that is still invalid for cases where zero could actually
19622 2014-12-02 16:51:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19624 * gst-libs/gst/vaapi/gstvaapiminiobject.c:
19625 * gst-libs/gst/vaapi/gstvaapiminiobject.h:
19626 libs: expose GstVaapiMiniObject APIs to all backends.
19627 Make it possible to have all libgstvaapi backends (libs) access to a
19628 common GstVaapiMiniObject API and implementation. This is a minor step
19629 towards full exposure when needed, but restrict it to libgstvaapi at
19632 2014-12-02 14:15:49 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19634 * gst-libs/gst/vaapi/gstvaapiminiobject.c:
19635 * gst-libs/gst/vaapi/gstvaapiminiobject.h:
19636 * gst-libs/gst/vaapi/gstvaapiobject.c:
19637 * gst-libs/gst/vaapi/gstvaapiobject.h:
19638 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
19639 libs: re-indent all GstVaapiObject related source code.
19640 Re-indent and provide additional minor cosmetical changes to the
19641 GstVaapiMiniObject and GstVaapiObject source files.
19643 2015-01-27 16:25:21 +0200 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
19645 * gst/vaapi/gstvaapipluginutil.c:
19646 pluginutil: Fix clearing of subtitle overlay
19647 dvbsuboverlay signals no subtitles present by not setting
19648 GstVideoOverlayCompositionMeta on a buffer.
19649 Detect this, and remove subtitles whenever we have no overlay composition to
19651 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
19653 2015-01-27 16:06:02 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
19655 * gst/vaapi/gstvaapipostproc.c:
19656 vaapipostproc: clear state on stop
19657 Otherwise restarting may fail because the state of vaapipluginbase and
19658 vaapipostproc don't match. e.g. gst_vaapipostproc_set_caps() will skip
19659 initailization and not call gst_vaapi_plugin_base_set_caps()
19661 2015-01-27 14:50:12 +0200 Michael Olbrich <m.olbrich@pengutronix.de>
19663 * gst/vaapi/gstvaapidecode.c:
19664 vaapidecode: don't print an error message for GST_FLOW_FLUSHING
19666 2015-01-27 12:38:45 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19668 * gst-libs/gst/vaapi/gstvaapiencoder.c:
19669 gstvaapiencoder: Fix the negotiation issue with _ENCODED format handling
19670 Don't error out for the video format GST_VIDEO_FORMAT_ENCODED with in gstvaapiencoder,
19671 since the vaaapi context creation (gstvaapicontext.c) can still use the
19672 default chroma type which is YUV420.
19673 https://bugzilla.gnome.org/show_bug.cgi?id=743567
19674 https://bugzilla.gnome.org/show_bug.cgi?id=743035
19676 2015-01-21 18:31:22 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19678 * patches/videoparsers/0003-h264parse-add-initial-support-for-MVC-NAL-units.patch:
19679 * patches/videoparsers/series.frag:
19680 h264parse: drop patches merged upstream.
19681 0003-h264parse-add-initial-support-for-MVC-NAL-units.patch
19683 2015-01-21 18:26:12 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19685 * ext/codecparsers:
19686 codecparsers: update to gst-vaapi-branch commit d3b5c1b
19687 8194cac: h264parse: parse SPS subset
19688 64b7f52: h264parse: expose stereo-high profile
19689 774360a: h264parse: add initial support for MVC NAL units
19690 258478f: h264parser: fix stack smashing
19692 2015-01-19 11:30:12 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19694 * gst/vaapi/gstvaapivideometa_texture.c:
19695 Fix compilation error if there is no GL/gl.h header file installed
19697 2015-01-15 16:23:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19699 * gst/vaapi/gstvaapidecode.c:
19700 vaapidecode: commit updated srcpad caps to base plugin.
19701 Make sure that the GstVaapiPluginBase instance receives the new src
19702 pad caps whenever they get updated from within the GstVaapiDecoder
19704 This also ensures that downstream elements receive correctly sized
19705 SW decoded buffers if needed.
19706 https://bugs.tizen.org/jira/browse/TC-114
19708 2015-01-15 16:19:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19710 * gst/vaapi/gstvaapidecode.c:
19711 vaapidecode: always reset decoder on ::set_format().
19712 Split GstVideoDecoder::set_format() handler to first update the sink
19713 pad caps and reset the active VA decoder instance based on those, and
19714 then update the src pad caps whenever possible, e.g. when the caps
19715 specify a valid video resolution.
19717 2015-01-15 16:14:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19719 * gst/vaapi/gstvaapivideomemory.c:
19720 vaapivideomemory: don't crash when trying to allocate 0x0 images.
19721 In some occasions, a buffer pool is created for pre-initialization
19722 purposes regardless of whether a valid image size is available or
19723 not. However, during actual decode stage, the vaapidecode element
19724 is expected to update the srcpad caps with the new dimensions, thus
19725 also triggering a reset of the underlying bufferpool.
19727 2015-01-15 00:00:16 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19729 * ext/codecparsers:
19730 codecparsers: update to gst-vaapi-branch commit 191cb2f
19731 347605a: h264parse: expose compatible profiles to downstream
19732 d1ea97e: h264parse: Fix periodic SPS/PPS sending work after a seek
19733 24a3126: Revert "h264parse: expose compatible profiles to downstream"
19734 8661740: h264parse: expose compatible profiles to downstream
19735 8b7ef3f: codecparsers: fix some compiler warnings
19737 2014-11-27 12:11:03 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19739 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
19740 encoder: h264: Fix the period between I/P frames
19741 If the key-frame period is set as one, then ip_period shuld be zero
19742 https://bugzilla.gnome.org/show_bug.cgi?id=734992
19744 2014-11-27 11:21:03 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19746 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
19747 encoder: h264: Provide intra_idr_period value for VAEncSequenceParameterBufferH264
19748 https://bugzilla.gnome.org/show_bug.cgi?id=734993
19750 2014-11-27 11:14:50 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19752 * gst/vaapi/gstvaapisink.c:
19753 vaapisink: Protect the X11 API invokations with proper locking
19754 https://bugzilla.gnome.org/show_bug.cgi?id=739808
19756 2014-11-27 11:13:20 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19758 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
19759 encoder: h264: fix pixel-aspect-ratio in encoded stream.
19760 Really report sample aspect ratio (SAR) as present, and make it match
19761 what we have obtained from the user as pixel-aspect-ratio (PAR). i.e.
19762 really make sure VUI parameter aspect_ratio_info_present_flag is set
19763 to TRUE and that the indication from aspect_ratio_idc is Extended_SAR.
19764 This is a leftover from git commit a12662f.
19765 https://bugzilla.gnome.org/show_bug.cgi?id=740360
19767 2014-11-25 11:46:12 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19769 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
19770 decoder: mpeg4: fix uninitialized variables.
19771 Fix gst_vaapi_decoder_mpeg4_parse() to initialize the packet type to
19772 GST_MPEG4_USER_DATA so that a parse error would result in skipping
19773 that packet. Also fix gst_vaapi_decoder_mpeg4_decode_codec_data() to
19774 initialize status to GST_VAAPI_DECODER_STATUS_SUCCESS.
19776 2014-11-25 11:41:49 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19778 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
19779 encoder: h264: fix profile limits.
19780 Fix ensure_profile_limits() to lower profile to the desired limits,
19781 only if the latter are actually known and the profile needed to be
19784 2014-11-24 15:14:37 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19786 * gst/vaapi/gstvaapipostproc.c:
19787 vaapipostproc: fix out caps for GLMemory.
19788 If the best downstream capsfeature turns out to be GLMemory, then make
19789 sure to propagate RGBA video format in caps to that element. This fixes
19790 the following pipeline: ... ! vaapipostproc ! glimagesink.
19792 2014-11-24 14:25:33 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19794 * gst/vaapi/gstvaapipostproc.c:
19795 vaapipostproc: fix check for compatible src pad capsfilters.
19796 When an explicit output video format is selected, from an src pad
19797 capsfilter, make sure that the downstream element actually supports
19798 that format. In particular, fix crash with the following pipelines:
19799 ... ! vaapipostproc ! video/x-raw,format=XXX ! xvimagesink ; where
19800 XXX is a format not supported by xvimagesink.
19801 While doing so, also reduce the set of src pad filter caps to the
19802 actual set of allowed src pad caps.
19804 2014-11-24 14:10:11 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19806 * gst/vaapi/gstvaapipostproc.c:
19807 * gst/vaapi/gstvaapipostproc.h:
19808 * gst/vaapi/gstvaapiuploader.c:
19809 * gst/vaapi/gstvaapiuploader.h:
19810 plugins: re-indent all video processing related source code.
19812 2014-11-24 13:20:33 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19814 * gst/vaapi/gstvaapivideomemory.c:
19815 plugins: further fixes to the new "current" storage tracker.
19816 The ensure_surface() and ensure_image() functions shall only relate
19817 to the underlying backing store. The actual current flags are to be
19818 updated only through ensure_{surface,image}_is_current() or very other
19819 particular cases in GstMemory hooks.
19821 2014-11-21 15:43:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19823 * gst/vaapi/gstvaapivideomemory.c:
19824 plugins: fix "current" video memory flags.
19825 If the surface proxy is updated into the GstVaapiVideoMemory, then
19826 it is assumed it is the most current representation of the current
19827 video frame. Likewise, make a few more arrangements to have the
19828 "current " flags set more consistently.
19830 2014-11-21 15:23:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19832 * gst/vaapi/gstvaapivideomemory.c:
19833 * gst/vaapi/gstvaapivideomemory.h:
19834 * gst/vaapi/gstvaapivideometa.c:
19835 plugins: ensure VA surface is current prior to using it.
19836 When interacting with SW elements, the buffers and underlying video
19837 memory could be mapped as read/write. However, we need to use those
19838 buffers again as plain VA surfaces, we have to make sure the VA image
19839 is thus committed back to VA surface memory.
19840 This fixes pipelines involving avdec_* and vaapi{postproc,sink}.
19842 2013-07-12 06:34:15 -0400 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
19844 * gst/vaapi/gstvaapivideomemory.c:
19845 plugins: enable memory maps for read & write.
19846 Hence vaapisink can display buffers decoded by gst-libav, or HW decoded
19847 buffers can be further processed in-place, e.g. with a textoverlay.
19848 https://bugzilla.gnome.org/show_bug.cgi?id=704078
19849 [ported to current git master branch, amended commit message]
19850 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19852 2014-11-18 14:57:02 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19854 * gst/vaapi/gstvaapidecode.c:
19855 * gst/vaapi/gstvaapipluginutil.c:
19856 * gst/vaapi/gstvaapipluginutil.h:
19857 * gst/vaapi/gstvaapipostproc.c:
19858 plugins: preserve framerate when updating src caps video format.
19859 In the current implementation, gst_video_info_set_format() would reset
19860 the whole GstVideoInfo structure first, prior to setting video format
19861 and size. So, coleteral information like framerate or pixel-aspect-
19863 Provide and use a unique gst_video_info_change_format() for overcome
19864 this issue, i.e. only have it change the format and video size, and
19865 copy over the rest of the fields.
19866 https://bugzilla.gnome.org/show_bug.cgi?id=734665
19868 2014-11-18 14:07:57 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19870 * gst/vaapi/gstvaapi.c:
19871 vaapidecode: increase the rank to GST_RANK_PRIMARY + 1
19872 This is for helping decodebin to autoplug the vaapidecode element.
19873 Decodebin is selecting decoder elements only based on rank and caps.
19874 Without overriding the autoplug-* signals there is no way to autoplug
19875 HW decoders inside decodebin. An easier soulution is to raise the
19876 rank of vaapidecode, so that it gets selected first.
19877 https://bugzilla.gnome.org/show_bug.cgi?id=739332
19879 2014-11-12 07:46:53 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19881 * gst-libs/gst/vaapi/gstvaapiprofile.c:
19882 * gst/vaapi/gstvaapidecode.c:
19883 vaapidecode: only expose supported profiles when needed.
19884 JPEG and VP8 codecs do not really support the concept of "profile". So,
19885 don't try to expose any set that wouldn't be supported by jpegparse, or
19886 ivfparse for instance.
19887 https://bugzilla.gnome.org/show_bug.cgi?id=739713
19888 https://bugzilla.gnome.org/show_bug.cgi?id=739714
19890 2014-11-13 15:13:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19892 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
19893 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
19894 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
19895 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
19896 decoder: h264: add initial support for repeat-first-field (RFF) flag.
19897 Use the SEI pic_timing() message to track and propagate down the repeat
19898 first field (RFF) flag. This is only initial support as there is one
19899 other condition that could induce the RFF flag, which is not handled
19902 2014-11-13 15:05:19 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19904 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
19905 decoder: h264: fix picture ordering count type 0 with previous MMCO5.
19906 Fix the decoding process for picture order count type 0 when the previous
19907 picture had a memory_management_control_operation = 5. In particular, fix
19908 the actual variable type for prev_pic_structure to hold the full bits of
19909 the picture structure.
19910 In practice, this used to work though, due to the underlying type used to
19911 express a gboolean.
19913 2014-11-13 15:00:21 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19915 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
19916 decoder: h264: fix detection of top-field-first (TFF) flag.
19917 Use the SEI pic_timing() message to track the pic_struct variable when
19918 present, or infer it from the regular slice header flags field_pic_flag
19919 and bottom_field_flag. This fixes temporal sequence ordering when the
19920 output pictures are to be displayed.
19921 https://bugzilla.gnome.org/show_bug.cgi?id=739291
19923 2014-11-14 09:54:02 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19925 * ext/codecparsers:
19926 codecparsers: update to gst-vaapi-branch commit 3d05d9f.
19927 1241840 h264: fix derivation of MaxPicNum variable
19928 3bd718e h264: fix GstH264ParserResult documentation typo
19929 b021609 h264parse: set the HEADER flag on buffers containing SPS or PPS
19930 b08e4be h264parse: don't unnecesarily set src_caps
19932 2014-11-03 19:20:43 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19934 * gst/vaapi/gstvaapipostproc.c:
19935 vaapipostproc: Add string representation of VPP functions to ElementFactoy Klass
19936 Added the same Klass specifications used in other upstream
19937 video postprocessing elements like videoconvert, videoscale,
19938 videobalance and deinterlace.
19939 An example use case is for this is to help the playsink
19940 to autoplug the hardware accelerated deinterlacer.
19942 2014-11-03 19:19:20 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19944 * gst/vaapi/gstvaapipostproc.c:
19945 vaapipostproc: Tweak the output video format setting to enable the autoplugging
19946 This is a workaround until auto-plugging is fixed when
19947 format=ENCODED + memory:VASurface caps feature are provided.
19948 Use the downstream negotiated video format as the output video format
19949 if the user didn't ask for the colorspace conversion explicitly.
19950 Usecase: This will help to connect elements like videoscale, videorate etc
19952 https://bugzilla.gnome.org/show_bug.cgi?id=739443
19954 2014-10-29 17:30:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19956 * gst/vaapi/gstvaapipostproc.c:
19957 * gst/vaapi/gstvaapipostproc.h:
19958 vaapipostproc: allow user defined scaling mode.
19959 Add new "scale-method" property to expose the scaling mode to use during
19960 video processing. Note that this is only a hint, and the actual behaviour
19961 may differ from implementation (VA driver) to implementation.
19963 2014-10-29 16:57:12 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19965 * gst-libs/gst/vaapi/gstvaapifilter.c:
19966 * gst-libs/gst/vaapi/gstvaapifilter.h:
19967 * gst-libs/gst/vaapi/gstvaapiutils.c:
19968 * gst-libs/gst/vaapi/gstvaapiutils.h:
19969 filter: add initial support for high quality scaling.
19970 Add support for video scaling options in VPP pipelines. Only the
19971 DEFAULT mode is bound to exist. Others might be folded into that
19974 2014-10-29 16:35:46 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19976 * gst-libs/gst/vaapi/gstvaapifilter.c:
19977 * gst-libs/gst/vaapi/gstvaapifilter.h:
19978 filter: re-indent all GstVaapiFilter related source code.
19980 2014-10-29 15:45:50 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
19982 * ext/codecparsers:
19983 codecparsers: update to gst-vaapi-branch commit f9d3bde.
19984 2218b02 h264parse: expose parsed profile and level to downstream
19985 3dbfab4 h264parse: return flushing if we get chained while being set to READY
19986 d40fa8b h264: fix frame packing SEI parsing
19987 32d40be h264: Use proper bit_reader api while parsing buffering_period SEI
19988 b3e022e h264: initialize some fields of pic_timing structure
19989 a70661d vc1: fix expected level in sequence-layer parsing unit test
19990 6cee88d vc1: fix level values for simple/main profile
19991 356c189 vc1: add unit test for sequence-layer parsing
19992 ab9f641 vc1: take care of endianness when parsing sequence-layer
19993 8dc8e35 mpeg4: fix vlc table used for sprite trajectory
19995 2014-10-29 15:46:47 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
19997 * gst/vaapi/gstvaapidecode.c:
19998 vaapidecode: Expose the supported profiles as caps to upstream
19999 This will allows the playbin to fallback to Software Decoder
20000 if the Hardware Decoder does not support a particular profile.
20001 https://bugzilla.gnome.org/show_bug.cgi?id=730997
20003 2014-10-29 15:46:12 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
20005 * gst/vaapi/gstvaapiencode.c:
20006 * gst/vaapi/gstvaapiencode.h:
20007 * gst/vaapi/gstvaapiencode_h264.c:
20008 encode: Attach the codec-data to out caps only based on negotiated caps
20009 Attach the codec_data to out_caps only if downstream needed.
20010 For eg: h264 encoder doesn't need to stuff codec_data to the
20011 src caps if the negotiated caps has a stream format of byte-stream.
20012 https://bugzilla.gnome.org/show_bug.cgi?id=734902
20014 2014-10-29 15:45:44 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
20017 configure: echoing installation prefix path
20019 2014-09-24 10:14:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20021 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
20022 display: add support for DRM Render-Nodes.
20023 Add support for DRM Render-Nodes. This is a new feature that appeared
20024 in kernel 3.12 for experimentation purposes, but was later declared
20025 stable enough in kernel 3.15 for getting enabled by default.
20026 This allows headless usages without authentication at all, i.e. usages
20027 through plain ssh connections is possible.
20029 2014-09-24 13:44:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20031 * gst/vaapi/gstvaapisink.c:
20032 vaapisink: clean-ups (indentation, drop unused variables).
20034 2014-09-24 13:39:55 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20037 * gst/vaapi/gstvaapisink.c:
20038 vaapisink: fix GstNavigation "key-press" / "key-release" events.
20039 Fix arguments to XkbKeycodeToKeysym() for converting an X11 keycode
20040 to a KeySym. In particular, there is no such Window argument. Also
20041 make sure to check for, and use, the correct <X11/XKBlib.h> header
20042 where that new function is defined. Otherwise, default to the older
20043 XKeycodeToKeysym() function.
20045 2014-09-24 13:23:17 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20047 * gst/vaapi/gstvaapisink.c:
20048 vaapisink: fix GstNavigation "mouse-move" event.
20049 Really use the motion event coordinates to propagate the "mouse-move"
20050 event to upper layer, instead of those from a button event. Those are
20051 technically the same though.
20053 2014-09-16 14:25:40 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
20055 * gst/vaapi/gstvaapisink.c:
20056 vaapisink: implement the GstNavigation interface
20057 This is useful for things like DVD menus, where key/mouse events
20058 would need to be forwarded from the upstream sink element.
20059 https://bugzilla.gnome.org/show_bug.cgi?id=711479
20061 2014-03-13 18:38:33 +0000 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
20063 * gst/vaapi/gstvaapipluginbase.c:
20064 vaapipostproc: fix deinterlacing from non VA memory buffers.
20065 When we copy a buffer because we're moving it into VA-API memory, we
20066 need to copy flags. Otherwise, interlaced YUV buffers from a capture
20067 source (e.g. V4L2) don't get flagged as interlaced.
20068 https://bugzilla.gnome.org/show_bug.cgi?id=726270
20069 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
20070 [reversed order of gst_buffer_copy_into() flags to match <1.0 code]
20071 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20073 2014-08-22 15:22:32 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20075 * gst/vaapi/gstvaapipostproc.c:
20076 vaapipostproc: add support for GLTextureUploadMeta output.
20077 This allows for vaapipostproc to be chained to the glimagesink element
20079 https://bugzilla.gnome.org/show_bug.cgi?id=735231
20081 2014-08-22 15:22:32 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20083 * gst/vaapi/gstvaapipluginutil.c:
20084 * gst/vaapi/gstvaapipluginutil.h:
20085 * gst/vaapi/gstvaapipostproc.c:
20086 vaapipostproc: add support for "download" capability.
20087 Allow implicit conversions to raw video formats, while still keeping
20088 VA surfaces underneath. This allows for chaining the vaapipostproc
20089 element to a software-only element that takes care of maps/unmaps.
20091 https://bugzilla.gnome.org/show_bug.cgi?id=720174
20093 2014-08-22 18:10:54 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20095 * gst/vaapi/gstvaapipostproc.c:
20096 * gst/vaapi/gstvaapipostproc.h:
20097 vaapipostproc: use pooled vaapi video meta.
20098 Use pooled GstVaapiVideoMeta information, i.e. always allocate that on
20099 video buffer allocation. Also optimize copy of additional metadata info
20100 into the resulting video buffer: only copy the video cropping info and
20101 the source surface proxy.
20102 https://bugzilla.gnome.org/show_bug.cgi?id=720311
20103 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
20104 [fixed proxy leak, fixed double free on error, optimized meta copy]
20105 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20107 2014-08-22 15:17:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20109 * gst/vaapi/gstvaapipostproc.c:
20110 vaapipostproc: fix passthrough mode.
20111 If no explicit output surface format is supplied try to keep the one
20112 supplied through the sink pad caps. This avoids a useless copy, even
20113 if things are kept in GPU memory.
20114 This is a performance regression from git commit dfa70b9.
20116 2014-07-05 21:00:34 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
20118 * gst/vaapi/gstvaapipostproc.c:
20119 * gst/vaapi/gstvaapipostproc.h:
20120 vaapipostproc: fix output buffer to have a GstVaapiVideoMemory.
20121 https://bugzilla.gnome.org/show_bug.cgi?id=720311
20122 [used new infrastructure through base decide_allocation() impl]
20123 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20125 2014-08-22 11:13:36 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20127 * gst/vaapi/gstvaapipostproc.c:
20128 vaapipostproc: enable advanced deinterlacing with same format.
20129 If only advanced deinterlacing is requested, i.e. deinterlacing is
20130 the only active algorithm to apply with source and output surface
20131 formats being the same, then make sure to enable VPP processing.
20132 Otherwise, allow fallback to bob-deinterlacing with simple rendering
20135 2014-08-21 15:04:20 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
20137 * gst/vaapi/gstvaapivideomemory.c:
20138 vaapivideomemory: avoid NULL pointer unref if GstVaapiImage creation failed.
20139 https://bugzilla.gnome.org/show_bug.cgi?id=735156
20141 2014-08-22 13:25:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20143 * gst/vaapi/gstvaapipluginbase.c:
20144 * gst/vaapi/gstvaapipluginutil.c:
20145 plugins: fix memory leaks.
20147 2014-08-21 14:10:36 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20149 * gst/vaapi/gstvaapidecode.c:
20150 * gst/vaapi/gstvaapipluginbase.c:
20151 * gst/vaapi/gstvaapipluginbase.h:
20152 plugins: factor out decide_allocation() hook.
20153 Add a default decide_allocation() hook to GstVaapiPluginBase. The caps
20154 feature argument can be used to force a bufferpool with a specific kind
20157 2014-08-21 11:12:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20159 * gst/vaapi/gstvaapivideobufferpool.c:
20160 * gst/vaapi/gstvaapivideobufferpool.h:
20161 * gst/vaapi/gstvaapivideomemory.c:
20162 plugins: allow bufferpool to not allocate vaapi video meta.
20163 Add GST_VAAPI_VIDEO_BUFFER_POOL_ACQUIRE_FLAG_NO_ALLOC params flag that
20164 can be used to disable early allocations of vaapi video metas on buffers,
20165 thus delagating that to the bufferpool user.
20167 2014-08-21 10:45:31 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20169 * gst/vaapi/gstvaapivideobuffer.c:
20170 * gst/vaapi/gstvaapivideobuffer.h:
20171 * gst/vaapi/gstvaapivideobufferpool.c:
20172 * gst/vaapi/gstvaapivideobufferpool.h:
20173 * gst/vaapi/gstvaapivideocontext.c:
20174 * gst/vaapi/gstvaapivideocontext.h:
20175 * gst/vaapi/gstvaapivideoconverter_glx.c:
20176 * gst/vaapi/gstvaapivideoconverter_glx.h:
20177 * gst/vaapi/gstvaapivideoconverter_x11.c:
20178 * gst/vaapi/gstvaapivideoconverter_x11.h:
20179 * gst/vaapi/gstvaapivideomemory.c:
20180 * gst/vaapi/gstvaapivideomemory.h:
20181 * gst/vaapi/gstvaapivideometa.c:
20182 * gst/vaapi/gstvaapivideometa.h:
20183 * gst/vaapi/gstvaapivideometa_texture.c:
20184 * gst/vaapi/gstvaapivideometa_texture.h:
20185 plugins: re-indent all GstVaapiVideo* related source code.
20187 2014-08-22 15:12:46 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20189 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
20190 surfaceproxy: fix copy to propagate view_id.
20191 Fix gst_vaapi_surface_proxy_copy() to copy the view-id element, thus
20192 fixing random frames skipped when vaapipostproc element is used in
20193 passthrough mode. In that mode, GstMemory is copied, thus including
20194 the underlying GstVaapiVideoMeta and associated GstVaapiSurfaceProxy.
20196 2014-08-20 16:38:45 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20198 * gst/vaapi/gstvaapipostproc.c:
20199 vaapipostproc: fix detection of output surface format changes.
20200 Default to I420 format for output surfaces so that to match the usual
20201 GStreamer pipelines. Though, internally, we could still opt for NV12
20202 surface formats, i.e. default format=ENCODED is a hint for that, thus
20203 delegating the decision to the VA driver.
20205 2014-08-20 10:59:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20207 * gst/vaapi/gstvaapipostproc.c:
20208 vaapipostproc: fix construction of allowed sink pad caps.
20209 Fix construction of the set of caps allowed on the sink pad to filter
20210 out unsupported raw video caps with GStreamer >= 1.2.
20212 2014-08-20 10:37:02 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20214 * gst/vaapi/gstvaapipostproc.c:
20215 * gst/vaapi/gstvaapipostproc.h:
20216 vaapipostproc: improve heuristics for detecting native VA surfaces.
20217 Use the new gst_caps_has_vaapi_surface() helper function to detect
20218 whether the sink pad caps contain native VA surfaces, or not, i.e.
20220 Also rename is_raw_yuv to get_va_surfaces to make the variable more
20221 explicit as we just want a way to differentiate raw video caps from
20222 VA surfaces actually.
20224 2014-08-12 18:33:25 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20226 * gst/vaapi/gstvaapipostproc.c:
20227 vaapipostproc: disable discontinuity detection code.
20228 The "discontinuity" tracking code, whereby lost frames are tentatively
20229 detected, is inoperant if the sink pad buffer timestamps are not right
20231 This is a temporary workaround until the following bug is fixed:
20232 https://bugzilla.gnome.org/show_bug.cgi?id=734386
20234 2014-08-07 14:57:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20236 * gst/vaapi/gstvaapipostproc.c:
20237 vaapipostproc: fix heuristic for detecting discontinuity.
20238 In order to make the discontinuity detection code useful, we need to
20239 detect the lost frames in the history as early as the previous frame.
20240 This is because some VA implementations only support one reference
20241 frame for advanced deinterlacing.
20242 In practice, turn the condition for detecting new frame that is beyond
20243 the previous frame from field_duration*2 to field_duration*3, i.e.
20244 nothing received for the past frame and a half because of possible
20245 rounding errors when calculating the field-duration either in this
20246 element (vaapipostproc), or from the upstream element (parser element).
20247 This is a regression introduced with commit faefd62.
20248 https://bugzilla.gnome.org/show_bug.cgi?id=734135
20250 2014-08-20 11:43:08 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20252 * gst/vaapi/gstvaapipluginbase.c:
20253 * gst/vaapi/gstvaapipluginbase.h:
20254 plugins: fix detection of raw video caps.
20255 Use the new gst_caps_has_vaapi_surface() helper function to better
20256 detect raw video caps, and in particular those from RGB colorspace.
20257 https://bugzilla.gnome.org/show_bug.cgi?id=734665
20259 2014-08-20 11:30:41 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20261 * gst/vaapi/gstvaapipluginutil.c:
20262 * gst/vaapi/gstvaapipluginutil.h:
20263 plugins: add helper for detecting VA surfaces in caps.
20264 Introduce new gst_caps_has_vaapi_surface() helper function to detect
20265 whether the supplied caps has VA surfaces. With GStreamer >= 1.2, this
20266 implies a check for memory:VASurface caps features, and format=ENCODED
20267 for earlier versions of GStreamer.
20269 2014-08-12 13:01:57 +0300 Sebastian Dröge <sebastian@centricular.com>
20271 * gst/vaapi/gstvaapidecode.c:
20272 vaapidecode: start the decoder task again after finishing
20273 This allows the element to accept data again after draining without
20274 a hard reset or caps change happening in between.
20275 https://bugzilla.gnome.org/show_bug.cgi?id=734616
20277 2014-08-12 13:00:03 +0300 Sebastian Dröge <sebastian@centricular.com>
20279 * gst/vaapi/gstvaapidecode.c:
20280 vaapidecode: unlock condition variables before shutting down the element
20281 Otherwise threads might wait for them, causing the shutdown of the element
20282 to deadlock on the streaming thread.
20283 https://bugzilla.gnome.org/show_bug.cgi?id=734616
20285 2014-08-11 17:15:24 +0300 Sebastian Dröge <sebastian@centricular.com>
20287 * gst/vaapi/gstvaapidecode.c:
20288 vaapidecode: reset decoder_finish variable after stopping the decoder thread
20289 Otherwise the element is not usable again after draining/EOS.
20290 https://bugzilla.gnome.org/show_bug.cgi?id=734616
20292 2014-08-11 17:14:53 +0300 Sebastian Dröge <sebastian@centricular.com>
20294 * gst/vaapi/gstvaapidecode.c:
20295 vaapidecode: don't try flushing the decoder instance if we didn't create one yet
20296 This otherwise results in unnecessary error messages.
20297 https://bugzilla.gnome.org/show_bug.cgi?id=734616
20299 2014-08-01 06:32:32 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20301 * gst-libs/gst/vaapi/Makefile.am:
20302 * gst/vaapi/Makefile.am:
20303 build: fix with --no-undefined linker flags.
20304 https://bugzilla.gnome.org/show_bug.cgi?id=729352
20306 2014-01-23 15:44:09 +0000 Changzhi Wei <changzhix.wei@intel.com>
20308 * gst/vaapi/gstvaapisink.c:
20309 * gst/vaapi/gstvaapisink.h:
20310 vaapisink: add support for GstColorBalance interface.
20311 https://bugzilla.gnome.org/show_bug.cgi?id=722390
20312 [fixed channel names, simplified range factor, fixed memory leak]
20313 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20315 2014-01-23 15:23:00 +0000 Changzhi Wei <changzhix.wei@intel.com>
20317 * gst/vaapi/gstvaapisink.c:
20318 * gst/vaapi/gstvaapisink.h:
20319 vaapisink: add support for colorbalance adjustment.
20320 https://bugzilla.gnome.org/show_bug.cgi?id=722390
20321 [fixed and simplified tracking of colorbalance value changes]
20322 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20324 2014-07-31 13:18:21 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20326 * gst/vaapi/gstvaapisink.c:
20327 vaapisink: improve installation of properties.
20328 Simplify the creation and installation process of properties, by first
20329 accumulating them into a g_properties[] array, and next calling into
20330 g_object_class_install_properties().
20331 Also add missing docs and flags to some properties.
20333 2014-07-31 10:48:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20335 * gst/vaapi/gstvaapisink.c:
20336 * gst/vaapi/gstvaapisink.h:
20337 vaapisink: code clean-ups.
20338 Move code around in a more logical way. Introduce GST_VAAPISINK_CAST()
20339 helper macro and use it wherever we know the object is a GstBaseSink or
20340 any base class. Drop explicit initializers for values that have defaults
20343 2014-07-31 10:37:57 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20345 * gst/vaapi/gstvaapisink.c:
20346 * gst/vaapi/gstvaapisink.h:
20347 vaapisink: re-indent all GstVaapiSink related source code.
20349 2014-07-30 17:27:50 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20351 * gst/vaapi/gstvaapisink.c:
20352 * gst/vaapi/gstvaapisink.h:
20353 vaapisink: introduce separate backends.
20354 Introduce new backends vtable so that to have clean separation between
20355 display dependent code and common base code. That's a "soft" separation,
20356 we don't really need dedicated objects.
20357 https://bugzilla.gnome.org/show_bug.cgi?id=722248
20359 2014-07-30 16:47:20 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20361 * gst/vaapi/gstvaapisink.c:
20362 * gst/vaapi/gstvaapisink.h:
20363 vaapisink: drop unused "synchronous" mode.
20364 Support for X11 "synchronous" mode was never implemented, and was only
20365 to be useful for debugging. Drop that altogether, that's not going to
20366 be useful in practice.
20367 https://bugzilla.gnome.org/show_bug.cgi?id=733985
20369 2014-07-30 17:32:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20371 * gst/vaapi/gstvaapisink.c:
20372 * gst/vaapi/gstvaapisink.h:
20373 vaapisink: drop unused variables.
20374 Drop obsolete, and now unused, video_buffer_pool and video_buffer_size
20375 variables. They got merged into the GstVaapiPluginBase object.
20377 2014-07-30 16:35:32 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20379 * gst/vaapi/gstvaapisink.c:
20380 * gst/vaapi/gstvaapisink.h:
20381 vaapisink: drop GLX rendering and fancy effects.
20382 Rendering with GLX in vaapisink is kind of useless nowadays, including
20383 OpenGL related fancy effects. Plain VA/GLX interfaces are also getting
20384 deprecated in favor of EGL, or more direct buffer sharing with actual
20386 Should testing of interop with GLX be needed, one could still be using
20387 the modern cluttersink or glimagesink elements.
20388 https://bugzilla.gnome.org/show_bug.cgi?id=733984
20390 2013-11-05 14:01:11 +0100 Holger Kaelberer <hk@getslash.de>
20392 * gst/vaapi/gstvaapisink.c:
20393 * gst/vaapi/gstvaapisink.h:
20394 vaapisink: listen to window size changes on X11.
20395 Allow dynamic changes to the window, e.g. performed by the user, and
20396 make sure to refresh its contents, while preserving aspect ratio.
20397 In practice, Expose and ConfigureNotify events are tracked in X11
20398 display mode by default. This occurs in a separte event thread, and
20399 this is similar to what xvimagesink does. Any of those events will
20400 trigger a reconfiguration of the window "soft" size, subsequently
20401 the render-rect when necessary, and finally _expose() the result.
20402 The default of handle_events=true can be changed programatically via
20403 gst_x_overlay_handle_events().
20404 Thanks to Fabrice Bellet for rebasing the patch.
20405 https://bugzilla.gnome.org/show_bug.cgi?id=711478
20406 [dropped XInitThreads(), cleaned up the code a little]
20407 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20409 2014-07-29 15:47:46 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20411 * gst/vaapi/gstvaapidecode.c:
20412 vaapidecode: decode and output all pending frames on normal EOS.
20413 The gst_vaapidecode_decode_loop() function is called within a separate
20414 task to fetch and output all frames that were decoded so far. So, if
20415 the decoder_loop_status is forcibly set to EOS when _finish() is called,
20416 then we are bound to exist the task without submitting the pending
20418 If the downstream element error'ed out, then the gst_pad_push() would
20419 propagate up an error and so we will get it right for cutting off
20420 _finish() early in that case.
20421 This is a regression from 6003596.
20422 https://bugzilla.gnome.org/show_bug.cgi?id=733897
20424 2014-07-29 13:24:52 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20427 Bump version for development.
20429 2014-07-29 10:31:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20434 2014-07-29 10:31:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20439 2014-07-29 16:22:01 +1000 Matthew Waters <ystreet00@gmail.com>
20441 * gst/vaapi/gstvaapidecode.c:
20442 vaapidecode: properly return from decode loop on downstream errors.
20443 Fixes a hang/race on shutdown where _decode_loop() had already completed
20444 its execution and _finish() was waiting on a GCond for decode_loop()
20445 to complete. Also fixes the possible race where _finish() is called
20446 but _decode_loop() endlessly returns before signalling completion
20447 iff the decoder instance returns GST_FLOW_OK.
20448 Found with: ... ! vaapidecode ! {glimagesink,cluttersink}
20449 https://bugzilla.gnome.org/show_bug.cgi?id=733897
20450 [factored out GST_VIDEO_DECODER_STREAM_UNLOCK() call]
20451 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20453 2014-07-28 18:45:49 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20459 2014-07-28 18:31:09 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20461 * gst/vaapi/gstvaapisink.c:
20462 vaapisink: fix GstVideoOverlay::expose() implementation.
20463 Now that we always track the currently active video buffer, it is
20464 not necessary to automatically increase its reference since this is
20465 implicitly performed in ::show_frame() through the get_input_buffer()
20466 helper from GstVaapiPluginBase class.
20467 This is a regression from a26df80.
20469 2014-07-28 18:00:19 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20471 * gst/vaapi/gstvaapidecode.c:
20472 * gst/vaapi/gstvaapipluginutil.c:
20473 vaapidecode: simplify bufferpool configuration.
20474 Rework the logics behind the configuration of an adequate bufferpool,
20475 especially when OpenGL meta or additional capsfeatures are needed.
20476 Besides, for GStreamer >= 1.4, the first capsfeatures that gets matched,
20477 and that is not system memory, is now selected by default.
20479 2014-07-28 16:43:47 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20481 * gst/vaapi/gstvaapidecode.c:
20482 vaapidecode: fix auto-plugging of vaapisink element.
20483 Make sure to propagate memory:VASurface capsfeature to srcpad caps
20484 only for GStreamer >= 1.5 as the plug-in elements in GStreamer 1.4
20485 core currently miss additional patches available in 1.5-git (1.6).
20486 This is a temporary workaround.
20488 2014-07-28 15:54:46 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
20491 configure: allow builds against GStreamer git (1.5.x).
20492 https://bugzilla.gnome.org/show_bug.cgi?id=733688
20494 2014-07-28 14:20:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20497 configure: fix build with GStreamer 1.4.0 release.
20499 2014-03-04 19:40:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20502 * debian.upstream/Makefile.am:
20503 * debian.upstream/control.in:
20504 debian: fix packaging for new naming scheme.
20506 2014-07-28 11:52:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20511 2014-07-28 11:39:11 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20514 configure: mark support for GStreamer < 1.2 as deprecated.
20515 Supporting anything thing below GStreamer 1.2 is asking for trouble
20516 for keeping up with the required facilities to make efficient pipelines.
20517 Users are invited to upgrade to the very latest GStreamer 1.2.x release,
20520 2014-07-28 11:35:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20523 configure: mark support for GStreamer 0.10 as obsolete.
20524 Support for GStreamer 0.10 is obsolete. i.e. it is no longer supported
20525 and may actually be removed altogether for a future release. There is
20526 no real point to maintain a build for such an ancient GStreamer version
20527 that is not even supported upstream.
20529 2014-07-28 10:25:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20531 * gst/vaapi/gstvaapisink.c:
20532 * gst/vaapi/gstvaapisink.h:
20533 vaapisink: allow a specific view component to be displayed.
20534 If a multiview stream is decoded, multiple view components are submitted
20535 as is downstream. It is the responsibility of the sink element to display
20536 the required view components. By default, always select the frame buffer
20537 that matches the view-id of the very first frame to be displayed.
20538 However, introduce a "view-id" property to allow the selection of a
20539 specific view component of interest to display.
20541 2014-07-28 10:09:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20543 * gst/vaapi/gstvaapisink.c:
20544 vaapisink: always keep the last displayed buffer around.
20545 Always record the VA surface that is currently being rendered, no matter
20546 the fact we are using texturedblit or overlay. That's because in some
20547 occasions, we need to refresh or resize the displayed contents based on
20548 new events. e.g. user-resized window.
20549 Besides, it's simpler to track the last video buffer in GstVaapiSink than
20550 through the base sink "last-sample".
20552 2014-07-25 11:13:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20554 * gst/vaapi/gstvaapipluginbase.c:
20555 * gst/vaapi/gstvaapipluginbase.h:
20556 * gst/vaapi/gstvaapipluginutil.c:
20557 * gst/vaapi/gstvaapisink.c:
20558 vaapisink: add support for "display-name" property.
20559 Add a "display-name" property to vaapisink so that the end user could
20560 select the desired output. Keep "display-name" in-line with the existing
20561 "display" (GstVaapiDisplayXXX type).
20562 So, for X11 or GLX, the "display-name" is the usual display name as we
20563 know for XOpenDisplay(); for Wayland, the "display-name" is the name used
20564 for wl_display_connect(); and for DRM, the "display-name" is actually the
20566 https://bugzilla.gnome.org/show_bug.cgi?id=722247
20568 2014-07-25 17:29:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20570 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
20571 window: make gst_vaapi_window_reconfigure() thread-safe.
20572 Ensure the X11 implementation for GstVaapiWindow::get_geometry() is
20573 thread-safe by default, so that upper layer users don't need to handle
20576 2013-07-29 09:28:28 +0200 Holger Kaelberer <hk@getslash.de>
20578 * gst-libs/gst/vaapi/gstvaapiwindow.c:
20579 * gst-libs/gst/vaapi/gstvaapiwindow.h:
20580 window: allow for updating size from current geometry.
20581 Add gst_vaapi_window_reconfigure() interface to force an update of
20582 the GstVaapiWindow "soft" size, based on the current geometry of the
20583 underlying native window.
20584 This can be useful for instance to synchronize the window size when
20585 the user changed it.
20586 Thanks to Fabrice Bellet for rebasing the patch.
20587 [changed interface to gst_vaapi_window_reconfigure()]
20588 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20590 2014-07-25 16:53:41 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20592 * gst-libs/gst/vaapi/gstvaapiwindow.c:
20593 * gst-libs/gst/vaapi/gstvaapiwindow.h:
20594 * gst-libs/gst/vaapi/gstvaapiwindow_drm.c:
20595 * gst-libs/gst/vaapi/gstvaapiwindow_drm.h:
20596 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
20597 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
20598 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
20599 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
20600 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.h:
20601 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
20602 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
20603 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
20604 window: re-indent all GstVaapiWindow related source code.
20606 2014-07-25 11:24:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20608 * gst-libs/gst/vaapi/gstvaapidisplay.c:
20609 * gst-libs/gst/vaapi/gstvaapidisplay.h:
20610 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
20611 display: add interface to retrieve the display name.
20612 Add gst_vaapi_display_get_display_name() helper function to determine
20613 the name associated with the underlying native display. Note that for
20614 raw DRM backends, the display name is actually the device path.
20616 2014-07-25 10:55:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20618 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
20619 display: fix comparison of X11 display names.
20620 Make sure to not only compare display host names, but also the actual
20621 display number. The screen number does not need to be checked at this
20624 2014-07-25 15:52:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20626 * gst/vaapi/gstvaapidecode.c:
20627 * gst/vaapi/gstvaapivideobufferpool.c:
20628 vaapidecode: add support for VideoAlignment bufferpool option.
20629 Always add VideoAlignment bufferpool option if the downstream element
20630 expects its own pool to be used but does not offer it through a proper
20631 propose_allocation() implementation for instance, and that the ALLOCATION
20632 query does not expose the availability of the Video Meta API.
20633 This fixes propagation of video buffer stride information to Firefox.
20635 2014-07-25 15:44:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20637 * gst/vaapi/gstvaapivideomemory.c:
20638 vaapivideomemory: always prefer native VA surface formats.
20639 Make sure to always prefer native internal formats for the VA surfaces
20640 that get allocated. Also disable "direct-rendering" mode in this case.
20641 This is needed so that to make sure that anything that gets out of the
20642 decoder, or anything that gets into the encoder, is in native format
20643 for the hardware, and thus the driver doesn't need to perform implicit
20644 conversions in there. Interop with SW elements is still available with
20645 fast implementations of VA imaging APIs.
20647 2014-07-24 11:58:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20649 * gst/vaapi/gstvaapivideomemory.c:
20650 vaapivideomemory: disallow memory shares across buffers, use a copy.
20651 Forbid shares of GstMemory instances, and rather make copy of it. This
20652 effectively copies the GstMemory structure and enclosed metadata, but
20653 this does not copy the VA surface contents itself. It should though.
20654 This fixes preroll and makes sure to not download garbage for the first
20655 frame when a SW rendering sink is used.
20657 2014-07-24 06:46:22 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20659 * gst/vaapi/gstvaapivideomemory.c:
20660 * gst/vaapi/gstvaapivideomemory.h:
20661 vaapivideomemory: use an image pool to cache objects.
20662 Use an image pool to hold VA images to be used for downloads/uploads
20663 of contents for the associated surface.
20664 This is an optmization for size. So, instead of creating as many VA
20665 images as there are buffers (then VA surfaces) allocated, we only
20666 maintain a minimal set of live VA images, thus preserving memory
20669 2014-07-24 00:14:04 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20671 * gst/vaapi/gstvaapivideomemory.c:
20672 vaapivideomemory: forbid R/W mappings if non direct-rendering mode.
20673 Disable read-write mappings if "direct-rendering" is not supported.
20674 Since the ordering of read and write operations is not specified,
20675 this would require to always download the VA surface on _map(), then
20676 commit the temporary VA image back to the VA surface on _unmap().
20677 Some SW decoding plug-in elements still use R/W mappings though.
20678 https://bugzilla.gnome.org/show_bug.cgi?id=733242
20680 2014-07-23 23:49:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20682 * gst/vaapi/gstvaapivideomemory.c:
20683 vaapivideomemory: minor code clean-ups.
20684 Fix error messages introduced in the previous commit for the _map()
20685 imaplementation. Also use the new get_image_data() helper function
20686 to determine the base pixels data buffer from a GstVaapiImage when
20687 updating the video info structure from it.
20689 2014-07-23 18:54:13 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20691 * gst/vaapi/gstvaapivideomemory.c:
20692 * gst/vaapi/gstvaapivideomemory.h:
20693 vaapivideomemory: add support for raw pixels mappings.
20694 Allow raw pixels of the whole frame to be mapped read-only. i.e. in
20695 cases where the buffer pool is allocated without VideoMeta API, thus
20696 individual planes cannot be mapped.
20697 This is initial support for Firefox >= 30.
20698 https://bugzilla.gnome.org/show_bug.cgi?id=731886
20700 2014-07-03 18:41:11 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
20702 * gst/vaapi/gstvaapivideomemory.c:
20703 vaapivideomemory: fix determination of the surface pool format.
20704 While creating the vaapi video allocator, make sure the associated
20705 surface pool has correct format instead of defaulting to NV12 video
20706 format even though there is no direct rendering support.
20707 https://bugzilla.gnome.org/show_bug.cgi?id=732691
20709 2014-07-23 18:01:21 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20711 * gst/vaapi/gstvaapivideomemory.c:
20712 vaapivideomemory: fix association of surface to proxy.
20713 Make sure to always update the VA surface pointer whenever the proxy
20714 changes. This used to only work when the VA surface is written to, in
20715 interop with SW element ("upload" feature), and this now fixes cases
20716 when the VA surface is needed for reading, in interop with SW element
20717 ("download" feature).
20719 2014-07-23 10:23:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20721 * gst/vaapi/gstvaapidecode.c:
20722 * gst/vaapi/gstvaapipluginutil.c:
20723 * gst/vaapi/gstvaapivideomemory.c:
20724 plugins: expose I420 format for interop with SW elements.
20725 Always expose I420 format by default when the VA surface could be
20726 mapped for interoperability with non harware accelerated elements.
20727 However, the default behaviour remains the auto-plugging of vaapi
20728 elements, down to the sink.
20729 Side effect: "direct-rendering" mode is also disabled most of the
20730 times as plain memcpy() from uncached speculative write combining
20731 memory is not going to be efficient enough.
20733 2014-07-22 18:54:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20735 * gst/vaapi/gstvaapidecode.c:
20736 * gst/vaapi/gstvaapivideomemory.c:
20737 plugins: allow download capability to vaapidecode element.
20738 Fix support for VA surface download capability in vaapidecode element
20739 for GStreamer >= 1.2. This is a fix to supporting libva-vdpau-driver,
20740 but also the libva-intel-driver while performing hardware accelerated
20741 conversions from the native VA surface format (NV12) to the desired
20742 output VA image format.
20743 For instance, this fixes pipelines involving vaapidecode ! xvimagesink.
20744 https://bugzilla.gnome.org/show_bug.cgi?id=733243
20746 2014-07-17 01:51:36 +0200 Fabrice Bellet <fabrice@bellet.info>
20748 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
20749 decoder: mpeg4: fix picture decoder return value for skipped frames.
20750 The picture decoder should return GST_VAAPI_DECODER_STATUS_DROP_FRAME
20751 when a frame should be skipped, so the stream processing is not stalled.
20752 https://bugzilla.gnome.org/show_bug.cgi?id=733324
20754 2014-07-04 15:13:32 +1000 Jan Schmidt <jan@centricular.com>
20756 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
20757 decoder: mpeg2: respect any input PTS provided for a frame.
20758 The timestamp generator in gstvaapidecoder_mpeg2.c always interpolated
20759 frame timestamps within a GOP, even when it's been fed input PTS for
20761 That leads to incorrect output timestamps in some situations - for example
20762 live playback where input timestamps have been scaled based on arrival time
20763 from the network and don't exactly match the framerate.
20764 https://bugzilla.gnome.org/show_bug.cgi?id=732719
20766 2014-01-22 08:20:59 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
20768 * gst-libs/gst/vaapi/gstvaapiobject.c:
20769 GstVaapiObject: make gst_vaapi_object_new() more robust.
20770 Forbid GstVaapiObject to be created without an associated klass spec.
20771 It is mandatory that the subclass implements an adequate .finalize()
20772 hook, so it shall provide a valid GstVaapiObjectClass.
20773 https://bugzilla.gnome.org/show_bug.cgi?id=722757
20774 [made non-NULL klass argument to gst_vaapi_object_new() a requirement]
20775 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20777 2014-01-21 15:43:57 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
20779 * gst-libs/gst/vaapi/gstvaapiobject.c:
20780 GstVaapiObject: initialize the derived object data with init() hook.
20781 Call the subclass .init() function in gst_vaapi_object_new(), if
20782 needed. The default behaviour is to zero initialize the subclass
20783 object data, then the .init() function can be used to initialize
20784 fields to non-default values, e.g. VA object ids to VA_INVALID_ID.
20785 Also fix the gst_vaapi_object_new() description, which was merely
20786 copied from GstVaapiMiniObject.
20787 https://bugzilla.gnome.org/show_bug.cgi?id=722757
20788 [changed to always zero initialize the subclass]
20789 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20791 2014-06-13 21:45:04 +0100 Lionel Landwerlin <llandwerlin@gmail.com>
20793 * gst/vaapi/gstvaapidecode.c:
20794 vaapidecode: make decoder work with playbin
20795 When playbin/decodebin builds the pipeline, it puts decoders and sinks
20796 into different bins and forwards the queries from bins to bins. So in
20797 the initials steps the pipeline is built iteratively by playbin and
20800 [filesrc] -> [typefind]
20801 [filesrc] -> [typefind] -> [demuxer]
20802 [filesrc] -> [typefind] -> [demuxer] -> [decoder]
20803 At this point the decoder is asked for its SRC caps and it will make a
20804 choice based on what gst_pad_peer_query_caps() returns. The problem is
20805 that the caps returns at that point includes caps features like ANY,
20806 essentially because playbin can plug in additional elements like
20807 videoscale, videoconv or deinterlace.
20808 This patch adds a another call to
20809 gst_vaapi_find_preferred_caps_feature() when the decoder decides its
20810 allocation, to make sure we asks the downstream elements when the
20811 entire pipeline has been built.
20812 https://bugzilla.gnome.org/show_bug.cgi?id=731645
20814 2014-06-27 11:57:11 +0100 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
20816 * gst/vaapi/gstvaapipostproc.c:
20817 vaapipostproc: don't let tmp_rect go out of scope.
20818 A compiler change showed me that tmp_rect went out of scope before
20819 it was used. Move it to the beginning of the function instead.
20820 https://bugzilla.gnome.org/show_bug.cgi?id=726363
20821 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
20822 [added guards for GStreamer 0.10 builds]
20823 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20825 2014-07-03 22:44:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20830 2014-07-03 22:34:35 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20834 Drop references to deprecated plugins (vaapiupload, vaapidownload),
20835 mention that support for GStreamer 0.10 is deprecated, make overview
20836 more descriptive in certain aspects.
20838 2014-07-03 22:21:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20843 2014-07-03 19:42:02 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20847 build: mention that support for GStreamer 0.10 is deprecated.
20849 2014-07-03 17:17:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20852 * gst/vaapi/gstvaapidecode.c:
20853 build: fix for GStreamer 0.10.
20855 2014-07-03 18:01:09 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20857 * patches/videoparsers/0002-h264parse-fix-build-with-GStreamer-1.2.patch:
20858 * patches/videoparsers/0002-h264parse-fix-build-with-older-GStreamer-1.x-stacks.patch:
20859 * patches/videoparsers/series.frag:
20860 build: fix for GStreamer 1.0.x.
20862 2014-07-03 13:48:48 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20864 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
20865 decoder: h264: detect incorrectly paired fields in frames.
20866 When a DPB flush is required, e.g. at a natural and of stream or issued
20867 explicitly through an IDR, try to detect any frame left in the DPB that
20868 is interlaced but does not contain two decoded fields. In that case, mark
20869 the picture as having a single field only.
20870 This avoids a hang while decoding tv_cut.mkv.
20872 2014-07-03 11:13:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20874 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
20875 decoder: h264: simplify the DPB output process.
20876 Simplify the dpb_output() function to exclusively rely on the frame store
20877 buffer to output, since this is now always provided. Besides, also fix
20878 cases where split fields would not be displayed.
20879 This is a regression from f48b1e0.
20881 2014-07-01 17:20:44 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20883 * patches/videoparsers/0003-h264parse-fix-and-optimize-NAL-collection-function.patch:
20884 * patches/videoparsers/0005-h264parse-introduce-new-state-tracking-variables.patch:
20885 * patches/videoparsers/0006-h264parse-improve-conditions-for-skipping-NAL-units.patch:
20886 * patches/videoparsers/0007-h264parse-fix-collection-of-access-units-to-preserve.patch:
20887 * patches/videoparsers/series.frag:
20888 h264parse: drop patches merged upstream.
20889 0003-h264parse-fix-and-optimize-NAL-collection-function.patch
20890 0005-h264parse-introduce-new-state-tracking-variables.patch
20891 0006-h264parse-improve-conditions-for-skipping-NAL-units.patch
20892 0007-h264parse-fix-collection-of-access-units-to-preserve.patch
20894 2014-07-01 17:18:08 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20896 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
20897 decoder: h264: fix memory leak in PPS.
20898 Cope with latest changes from codecparsers/h264. It is now required
20899 to explicitly clear the GstH264PPS structure as it could contain
20900 additional allocations (slice_group_ids).
20902 2014-07-01 17:13:56 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20904 * ext/codecparsers:
20905 codecparsers: update to gst-vaapi-branch commit 2d53b69.
20906 c4ace00 h264parse: fix collection of access units to preserve config headers
20907 0f9f7c9 h264parse: improve conditions for skipping NAL units
20908 9ffb25c h264parse: introduce new state tracking variables
20909 64955d3 h264parse: fix and optimize NAL collection function
20910 13cd2a3 h264: clarifications and documentation fixes
20911 53e7dd1 h264: fix identification of EOSEQ and EOS NALs
20912 18f0de0 h264: fix memory leak in GstH264PPS
20913 fdcb54c h264: fix typo in GstH264VUIParams description
20914 fd4dae9 vp8: move up built-in range decoder private data
20916 2014-06-30 19:01:35 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20918 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
20919 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
20920 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
20921 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
20922 decoder: propagate MVC metadata ("view-id", head of multiview set).
20923 Add new GstVaapiSurfaceProxy flag FFB, which means "first frame in
20924 bundle", and really expresses the first view component of a multi
20925 view coded frame. e.g. in H.264 MVC, the surface proxy has flag FFB
20927 Likewise, new API is exposed to retrieve the associated "view-id".
20929 2014-06-30 18:46:45 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20931 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
20932 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
20933 decoder: propagate "one-field" flags.
20934 Allow decoders to set the "one-field" attribute when the decoded frame
20935 genuinely has a single field, or if the second field was mis-decoded but
20936 we still want to display the first field.
20938 2014-06-30 18:34:45 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20940 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
20941 decoder: output decoded frames only once.
20942 Make sure to output the decoded picture, and push the associated
20943 GstVideoCodecFrame, only once. The frame fully represents what needs
20944 to be output, included for interlaced streams. Otherwise, the base
20945 GstVideoDecoder class would release the frame twice.
20946 Anyway, the general process is to output decoded frames only when
20947 they are complete. By complete, we mean a full frame was decoded or
20948 both fields of a frame were decoded.
20950 2014-06-30 16:12:52 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20952 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
20953 decoder: h264: decode current picture earlier.
20954 Slightly optimize decoding process by submitting the current VA surface
20955 for decoding earlier to the hardware, and perform the reference picture
20956 marking process and DPB update process afterwards.
20957 This is a minor optimization to let the video decode engine kick in work
20958 earlier, thus improving parallel resources utilization.
20960 2014-06-30 16:09:17 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20962 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
20963 decoder: h264: fix output of second field when first field is not in DPB.
20964 Fix decoding of interlaced streams where a first field (e.g. B-slice)
20965 was immediately output and the current decoded field is to be paired
20966 with that former frame, which is no longer in DPB.
20967 https://bugzilla.gnome.org/show_bug.cgi?id=701340
20969 2014-06-30 11:06:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20971 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
20972 decoder: h264: slightly optimize the process to detect new pictures.
20973 Optimize the process to detect new pictures or start of new access
20974 units by checking if the previous NAL unit was the end of a picture,
20975 or the end of the previous access unit.
20977 2014-06-13 15:42:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20979 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
20980 decoder: h264: handle access unit ("au") optimization.
20981 Optimize parsing when buffers are supplied with access unit alignment.
20982 This helps determining faster when the end of an access unit is reached.
20984 2014-06-28 07:25:35 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20986 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
20987 decoder: h264: fix tracking of DPB size changes.
20988 Add support for MVC streams with multiple SPS and subset SPS headers
20989 emitted regularly, e.g. at around every I-frame. Track the maximum
20990 number of views in ensure_context() and really reset the DPB size to
20991 the expected value, always. i.e. even if it decreased. dpb_reset()
20992 only cares of ensuring the DPB allocation.
20994 2014-06-27 20:44:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
20996 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
20997 decoder: h264: fix the DPB compaction process.
20998 Fix the compaction process when the DPB is cleared for a specific
20999 view, i.e. fix the process of filling in the holes resulting from
21000 removing frame buffers matching the current picture.
21002 2014-06-27 16:38:03 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21004 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21005 encoder: h264: generate new SPS only when codec config changed.
21006 It is not necessary to periodically send SPS or subset SPS headers.
21007 This is up to the upper layer (e.g. transport layer) to decide on
21008 if/how to periodically submit those. For now, only generate new SPS
21009 or subset SPS headers when the codec config changed.
21010 Note: the upper layer could readily determine the config headers
21011 (SPS/PPS) through the gst_vaapi_encoder_h264_get_codec_data() function.
21012 https://bugzilla.gnome.org/show_bug.cgi?id=732083
21013 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21015 2014-06-27 18:43:27 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21017 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21018 encoder: h264: track encoder config changes.
21019 Track and report when encoder configuration changed. For now, this covers
21020 resolution, profile/level and bitrate changes.
21022 2014-06-27 13:15:13 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21024 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21025 encoder: h264: add pixel-aspect-ratio to VUI parameters.
21026 Report sample aspect ratio (SAR) as present, and make it match what
21027 we have obtained from the user as pixel-aspect-ratio (PAR). i.e. the
21028 VUI parameter aspect_ratio_info_present_flag now defaults to TRUE.
21030 2014-06-27 00:49:34 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21032 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21033 encoder: h264: fix number of anchor and non-anchor reference pictures.
21034 Set the value of num_anchor_refs_l0, num_anchor_refs_l1, num_non_anchor_refs_l0,
21035 and num_non_anchor_refs_l1 to zero since the inter-view prediction is not yet
21038 2014-06-27 10:37:38 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21040 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21041 encoder: h264: fix timing_info_present_flag value in subset SPS.
21042 When the seq_parameter_set_data() syntax structure is present in a subset
21043 sequence parameter set and vui_parameters_present_flag is equal to 1, then
21044 timing_info_present_flag shall be equal to 0 (H.7.4.2.1.1).
21046 2014-06-26 14:39:52 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21048 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21049 encoder: h264: add cpbBrNalFactor values for MVC profiles.
21051 2014-06-26 14:51:32 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21053 * patches/videoparsers/0003-h264parse-add-initial-support-for-MVC-NAL-units.patch:
21054 h264parse: fix detection of access unit boundaries for MVC.
21055 The gst_h264_parse_collect_nal() function is a misnomer. In reality,
21056 this function is used to determine access unit boundaries, i.e. that
21057 is the key function for alignment=au output format generation.
21059 2014-06-26 14:49:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21061 * patches/videoparsers/0007-h264parse-fix-collection-of-access-units-to-preserve.patch:
21062 * patches/videoparsers/series.frag:
21063 h264parse: fix collection of access units to preserve config headers.
21064 Always use a GstAdapter when collecting access units (alignment="au")
21065 in either byte-stream or avcC format. This is required to properly
21066 preserve config headers like SPS and PPS when invalid or broken NAL
21067 units are subsequently parsed.
21068 More precisely, this fixes scenario like:
21069 <SPS> <PPS> <invalid-NAL> <slice>
21070 where we used to reset the output frame buffer when an invalid or
21071 broken NAL is parsed, i.e. SPS and PPS NAL units were lost, thus
21072 preventing the next slice unit to be decoded, should this also
21073 represent any valid data.
21074 https://bugzilla.gnome.org/show_bug.cgi?id=732203
21076 2014-06-26 14:48:08 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21078 * patches/videoparsers/0003-h264parse-add-initial-support-for-MVC-NAL-units.patch:
21079 * patches/videoparsers/0006-h264parse-improve-conditions-for-skipping-NAL-units.patch:
21080 * patches/videoparsers/series.frag:
21081 h264parse: improve conditions for skipping NAL units.
21082 Carefully track cases when skipping broken or invalid NAL units is
21083 necessary. In particular, always allow NAL units to be processed
21084 and let that gst_h264_parse_process_nal() function decide on whether
21085 the current NAL needs to be dropped or not.
21086 This fixes parsing of streams with SEI NAL buffering_period() message
21087 inserted between SPS and PPS, or SPS-Ext NAL following a traditional
21088 SPS NAL unit, among other cases too.
21089 Practical examples from the H.264 AVC conformance suite include
21090 alphaconformanceG, CVSE2_Sony_B, CVSE3_Sony_H, CVSEFDFT3_Sony_E
21091 when parsing in stream-format=byte-stream,alignment=au mode.
21092 https://bugzilla.gnome.org/show_bug.cgi?id=732203
21094 2014-06-26 14:45:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21096 * patches/videoparsers/0003-h264parse-add-initial-support-for-MVC-NAL-units.patch:
21097 * patches/videoparsers/0005-h264parse-introduce-new-state-tracking-variables.patch:
21098 * patches/videoparsers/series.frag:
21099 h264parse: introduce new state tracking variables.
21100 Improve parser state tracking by introducing new flags reflecting
21101 it: "got-sps", "got-pps" and "got-slice". This is an addition for
21102 robustness purposes.
21103 Older have_sps and have_pps variables are kept because they have
21104 a different meaning. i.e. they are used for deciding on when to
21105 submit updated caps or not, and rather mean "have new SPS/PPS to
21108 2014-06-26 14:39:30 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21110 * patches/videoparsers/0004-h264parse-default-to-byte-stream-nalu-format-Annex-B.patch:
21111 * patches/videoparsers/series.frag:
21112 h264parse: default to byte-stream/nalu format (Annex B).
21113 Always default to stream-format=byte-stream,alignment=nalu if avcC
21114 format was not detected. This is the natural stream format specified
21115 in the standard (Annex.B): a series of NAL units prefixed with the
21117 https://bugzilla.gnome.org/show_bug.cgi?id=732167
21119 2014-06-26 14:37:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21121 * patches/videoparsers/0003-h264parse-fix-and-optimize-NAL-collection-function.patch:
21122 * patches/videoparsers/series.frag:
21123 h264parse: fix and optimize NAL collection function.
21124 Use gst_h264_parser_identify_nalu_unchecked() to identify the next
21125 NAL unit. We don't want to parse the full NAL unit, but only the
21126 header bytes and possibly the first RBSP byte for identifying the
21127 first_mb_in_slice syntax element.
21128 Also fix check for failure when returning from that function. The
21129 only success condition for that is GST_H264_PARSER_OK, so use it.
21130 https://bugzilla.gnome.org/show_bug.cgi?id=732154
21132 2014-06-26 11:39:38 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21134 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21135 encoder: h264: fix NAL unit types in packed headers.
21136 Submit Prefix NAL headers (nal_unit_type = 14) before every packed
21137 slice header (nal_unit_type = 1 or 5) only for the base view. In non
21138 base views, a Coded Slice Extension NAL header (nal_unit_type = 20)
21139 is required, with an appropriate nal_unit_header_mvc_extension() in
21140 the NAL header bytes.
21141 https://bugzilla.gnome.org/show_bug.cgi?id=732083
21143 2014-06-25 22:05:52 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21145 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21146 encoder: h264: add missing field in packed Subset SPS header.
21147 Write the missing num_level_values_signalled_minus1 syntax element
21148 into the packed header for subset sequence parameter set.
21149 https://bugzilla.gnome.org/show_bug.cgi?id=732083
21151 2014-06-25 22:26:32 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21153 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21154 decoder: h264: fix marking of non-reference picture into DPB.
21155 Fix search for a picture in the DPB that has a lower POC value than
21156 the current picture. The dpb_find_lowest_poc() function will return
21157 a picture with the lowest POC in DPB and that is marked as "needed
21158 for output", but an additional check against the actual POC value
21159 of the current picture is needed.
21160 This is a regression from 1c46990.
21161 https://bugzilla.gnome.org/show_bug.cgi?id=732130
21163 2014-06-19 17:08:47 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21165 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21166 decoder: h264: fix DPB clear when no decoding actually started.
21167 Fix dpb_clear() to clear previous frame buffers only if they actually
21168 exist to begin with. If the decoder bailed out early, e.g. when it
21169 does not support a specific profile, that array of previous frames
21170 might not be allocated beforehand.
21172 2014-02-06 08:30:10 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21174 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21175 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
21176 * gst/vaapi/gstvaapidecode.c:
21177 decoder: h264: add support for NALU "alignment" optimization.
21178 We can avoid scanning for start codes again if the bitstream is fed
21179 in NALU chunks. Currently, we always scan for start codes, and keep
21180 track of remaining bits in a GstAdapter, even if, in practice, we
21181 are likely receiving one GstBuffer per NAL unit. i.e. h264parse with
21183 https://bugzilla.gnome.org/show_bug.cgi?id=723284
21184 [use gst_adapter_available_fast() to determine the top buffer size]
21185 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21187 2014-06-18 18:53:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21189 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21190 decoder: h264: fix caps to report interlace-mode accordingly.
21191 The `vaapipostproc' element could never determine if the H.264 stream
21192 was interlaced, and thus always assumed it to be progressive. Fix the
21193 H.264 decoder to report interlace-mode accordingly, thus allowing the
21194 vaapipostproc element to automatically enable deinterlacing.
21196 2014-06-19 13:35:23 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21198 * gst/vaapi/gstvaapipostproc.c:
21199 vaapipostproc: don't crash with dynamic framerate (0/1).
21200 Avoid reaching an assert if dynamic framerates (0/1) are used. One
21201 way to solve this problem is to just stick field_duration to zero.
21202 However, this means that, in presence of interlaced streams, the
21203 very first field will never be displayed if precise presentation
21204 timestamps are honoured.
21205 https://bugzilla.gnome.org/show_bug.cgi?id=729604
21207 2014-02-07 12:27:50 +0000 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
21209 * gst/vaapi/gstvaapipostproc.c:
21210 vaapipostproc: create filter surface pool if it does not exist yet.
21211 ensure_srcpad_buffer_pool() tries to avoid unnecessarily deleting and
21212 recreating filter_pool. Unfortunately, this also meant it didn't create
21213 it if it did not exist.
21214 Fix it to always create the buffer pool if it does not exist.
21215 https://bugzilla.gnome.org/show_bug.cgi?id=723834
21216 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
21218 2013-12-12 10:01:13 +0800 Zhao, Halley <halley.zhao@intel.com>
21220 * gst/vaapi/gstvaapipostproc.c:
21221 vaapipostproc: reset deinterlacer state when there is a discontinuity.
21222 Reset deinterlacer state, i.e. past reference frames used for advanced
21223 deinterlacing, when there is some discontinuity detected in the course
21224 of processing source buffers.
21225 This fixes support for advanced deinterlacing when a seek occurred.
21226 https://bugzilla.gnome.org/show_bug.cgi?id=720375
21227 [fixed type of pts_diff variable, fetch previous buffer PTS from the
21228 history buffer, reduce heuristic for detecting discontinuity]
21229 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21231 2014-06-18 16:16:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21233 * gst/vaapi/gstvaapipostproc.c:
21234 vaapipostproc: add support for crop regions in VPP mode.
21235 Apply video cropping regions stored in GstVideoCropMeta, or in older
21236 GstVaapiSurfaceProxy representation, to VPP pipelines. In non-VPP modes,
21237 the crop meta are already propagated to the output buffers.
21238 https://bugzilla.gnome.org/show_bug.cgi?id=720730
21240 2014-03-14 17:49:40 +0000 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
21242 * gst/vaapi/gstvaapipostproc.c:
21243 vaapipostproc: make deinterlace-mode behave as expected.
21244 deinterlace-mode didn't behave in the way you'd expect if you have
21245 past experience of the deinterlace element. There were two bugs:
21246 1. "auto" mode wouldn't deinterlace "interleaved" buffers, only "mixed".
21247 2. "force" mode wouldn't deinterlace "mixed" buffers flagged as progressive.
21248 Fix these up, and add assertions and error messages to detect cases that
21250 https://bugzilla.gnome.org/show_bug.cgi?id=726361
21251 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
21252 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21254 2014-01-15 16:36:29 +0000 Matthieu Bouron <matthieu.bouron@collabora.com>
21256 * gst/vaapi/gstvaapidecode.c:
21257 vaapidecode: do not discard video info props when the format changed.
21258 gst_video_info_set_format() does not preserve video info properties. In
21259 order to keep important information in the caps such as interlace mode,
21260 framerate, pixel aspect ratio, ... we need to manually copy back those
21261 properties after setting the new video format.
21262 https://bugzilla.gnome.org/show_bug.cgi?id=722276
21264 2014-02-23 01:43:39 +1100 Matthew Waters <ystreet00@gmail.com>
21266 * gst/vaapi/gstvaapidecode.c:
21267 vaapidecode: plug a memory leak.
21268 It can happen that there is a pool provided that does not advertise
21269 the vappivideometa. We should unref that pool before using our own.
21270 Discovered with vaapidecode ! {glimagesink,cluttersink}
21271 https://bugzilla.gnome.org/show_bug.cgi?id=724957
21272 [fixed compilation by adding the missing semi-colon]
21273 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21275 2014-06-18 13:47:36 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21277 * gst/vaapi/gstvaapidecode.c:
21278 vaapidecode: parse source data until a frame is obtained.
21279 Parse any pending data until a complete frame is obtained. This is a
21280 memory optimization to avoid expansion of video packets stuffed into
21281 the GstAdapter, and a fix to EOS condition to detect there is actually
21282 pending data that needs to be decoded, and subsequently output.
21283 https://bugzilla.gnome.org/show_bug.cgi?id=731831
21285 2014-06-05 15:32:29 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21287 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21288 encoder: h264: fix multiple slices support in packed headers mode.
21289 Handle packedi slice headers and packed raw data on a per-slice basis,
21290 which is necessary for multi slice encoding.
21292 2014-06-05 15:30:38 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21294 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
21295 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
21296 encoder: add infrastructure for per-slice handling of packed headers.
21297 The packed slice header and packed raw data need to be paired with
21298 the submission of VAEncSliceHeaderParameterBuffer. So handle them
21299 on a per-slice basis insted of a per-picture basis.
21300 [removed useless initializer]
21301 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21303 2014-03-07 17:40:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21305 * gst/vaapi/gstvaapisink.c:
21306 vaapisink: fix initialization with "drm" display type.
21307 Force early initializatin of the GstVaapiDisplay so that to make sure
21308 that the sink element display object is presented first to upstream
21309 elements, as it will be correctly featuring the requested display type
21311 Otherwise, we might end up in situations where a VA/X11 display is
21312 initialized in vaapidecode, then we try VA/DRM display in vaapisink
21313 (as requested by the "display" property), but this would cause a failure
21314 because we cannot acquire a DRM display that was previously acquired
21315 through another backend (e.g. VA/X11).
21317 2014-03-07 17:38:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21319 * gst/vaapi/gstvaapipluginbase.c:
21320 plugins: fix initialization with foreign context.
21321 When a new display is settled through GstElement::set_context() (>= 1.2),
21322 or GstVideoContext::set_context() (<= 1.0), then we shall also update the
21323 associated display type.
21325 2014-04-28 17:44:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21330 * gst/vaapi/Makefile.am:
21331 * gst/vaapi/gstvaapiparse.c:
21332 * gst/vaapi/gstvaapiparse.h:
21333 * patches/Makefile.am:
21334 * patches/videoparsers/0001-plugins-compile-the-built-in-video-parsers-as-vaapip.patch:
21335 * patches/videoparsers/0002-h264parse-fix-build-with-GStreamer-1.2.patch:
21336 * patches/videoparsers/0003-h264parse-add-initial-support-for-MVC-NAL-units.patch:
21337 * patches/videoparsers/Makefile.am:
21338 * patches/videoparsers/series.frag:
21339 plugins: add built-in video parsers as "vaapiparse" element.
21340 The built-in video parsers elements are built into a single DSO named
21341 libgstvaapi_parse.so. The various video parsers could be accessed as
21343 For now, this only includes a modified version of h264parse so that to
21344 support H.264 MVC encoded streams.
21346 2014-06-13 11:36:56 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21348 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21349 decoder: h264: cope with new gst_h264_quant_matrix_*() interfaces.
21350 New gst_h264_quant_matrix_*_get_raster_from_zigzag() were renamed
21351 from gst_h264_video_quant_matrix_*_get_raster_from_zigzag().
21353 2014-06-13 11:34:07 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21355 * ext/codecparsers:
21356 codecparsers: update to gst-vaapi-branch commit d6325ac.
21357 7d8d045 h264parse: use new gst_h264_video_calculate_framerate()
21358 d2f965a h264parse: set field_pic_flag when parsing a slice header
21359 24c15b8 Import h264parse
21360 a9283e5 bytereader: Use concistant derefence method
21361 a8252c6 bytereader: Use pointer instead of index access
21362 b1bebfc Import GstBitReader and GstByteReader
21363 2f58788 h264: recognize SVC NAL units
21364 4335da5 h264: fix SPS copy code for MVC
21365 cf9b6dc h264: fix quantization matrix conversion routine names
21366 b11ce2a h264: add gst_h264_video_calculate_framerate()
21367 126dc6f add C++ guards for MPEG-4 and VP8 parsers
21369 2014-06-10 18:30:21 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21371 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21372 decoder: h264: factor out DPB pruning for MVC.
21373 Factor out the removal process of unused inter-view only reference
21374 pictures from the DPB, prior to the possible insertion of the current
21376 Ideally, the compiler could still opt for generating two loops. But
21377 at least, the code is now clearer for maintenance.
21379 2014-06-10 17:42:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21381 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21382 decoder: h264: clean-ups.
21383 Fix GST_VAAPI_PICTURE_IS_{INTER_VIEW,ANCHOR}() definitions to use
21384 the base GST_VAAPI_PICTURE_FLAG_IS_SET() macro.
21386 2014-06-10 16:07:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21388 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21389 decoder: h264: improve pruning of unused MVC inter-view frames.
21390 Improve process for the removal of pictures from DPB before possible
21391 insertion of the current picture (C.4.4) for H.264 MVC inter-view only
21392 reference components. In particular, handle cases where picture to be
21393 inserted is not the last one of the access unit and if it was already
21394 output and is no longer marked as used for reference, including for
21395 decoding next view components within the same access unit.
21397 2014-06-03 17:36:38 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21399 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21400 decoder: h264: improve DPB bumping process for MVC.
21401 While invoking the DPB bumping process in presence of many views,
21402 it could be necessary to output previous pictures that are ready,
21403 in a whole. i.e. emitting all view components from the very first
21404 view order index zero to the very last one in its original access
21405 unit; and not starting from the view order index of the picture
21406 that caused the DPB bumping process to be invoked.
21407 As a reminder, the maximum number of frames in DPB for MultiView
21408 High profile with more than 2 views is not necessarily a multiple
21409 of the number of views.
21410 This fixes decoding of MVCNV-4.264.
21412 2014-06-06 17:56:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21414 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21415 decoder: h264: fix inter-view references array growth.
21416 Let the utility layer handle dynamic growth of the inter-view pictures
21417 array. By definition, setting a new size to the array will effectively
21418 grow the array, but would also fill in the newly created elements with
21419 empty entries (NULL), thus also increasing the reported length, which
21422 2014-06-03 17:36:38 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21424 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21425 decoder: h264: reduce ReferenceFrames entries to the essential set.
21426 When decoding Multiview High profile streams with a large number of
21427 views, it is not possible to make the VAPictureParameterBufferH264.
21428 ReferenceFrames[] array hold the complete DPB, with all possibly
21429 active pictures to be used for inter-view prediction in the current
21431 So reduce the scope of the ReferenceFrames[] array to only include
21432 the set of reference pictures that are going to be used for decoding
21433 the current picture. Basically, this is a union of all RefPicListX[]
21434 array, for all slices constituting the decoded picture.
21436 2014-06-04 19:10:44 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21438 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21439 decoder: h264: fix MVC inter-view prediction process.
21440 The inter-view reference components and inter-view only reference
21441 components that are included in the reference picture lists shall
21442 be considered as not being marked as "used for short-term reference"
21443 or "used for long-term reference". This means that reference flags
21444 should all be removed from VAPictureH264.flags.
21445 This fixes decoding of MVCNV-2.264.
21447 2014-06-04 19:03:18 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21449 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21450 decoder: h264: fix detection of profile changes for MVC.
21451 If the VA driver exposes ad-hoc H.264 MVC profiles, then we have to
21452 be careful to detect profiles changes and not reset the underlying
21453 VA context erroneously. In MVC situations, we could indeed get a
21454 profile_idc change for every SPS that gets activated, alternatively
21455 (base-view -> non-base view -> base-view, etc.).
21456 An improved fix would be to characterize the exact profile to use
21457 once and for all when SPS NAL units are parsed. This would also
21458 allow for fallbacks to a base-view decoding only mode.
21460 2014-06-03 14:30:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21462 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21463 encoder: h264: drop extraneous definitions.
21464 Re-use definitions from the codecparser headers instead of duplicating
21465 them here again. That covers NALU definitions and slice types.
21467 2014-04-01 11:26:04 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21469 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21470 encoder: h264: remove unnecessary calcualtion of max_pic_order_cnt.
21471 https://bugzilla.gnome.org/show_bug.cgi?id=727418
21473 2014-04-01 14:23:56 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21475 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21476 encoder: h264: don't allow CABAC with Extended profile.
21477 The H.264 specification does not support CABAC entropy coding for the
21479 https://bugzilla.gnome.org/show_bug.cgi?id=727418
21481 2014-05-07 00:12:39 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21483 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21484 encoder: h264: use packed headers mode for MVC encoding.
21485 Exclusively use VA drivers that support raw packed headers for encoding.
21486 i.e. simply submit packed headers Subset SPS and Prefix NAL units. This
21487 provides for better compatibility accross the various VA drivers and HW
21488 generations since no particular API is needed beyond what readily exists.
21490 2014-05-07 00:09:45 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21492 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21493 encoder: h264: add support for packed slice headers.
21494 https://bugzilla.gnome.org/show_bug.cgi?id=722905
21496 2014-05-07 00:09:19 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21498 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21499 encoder: h264: store subset sps to generate the codec-data
21500 Store the SubsetSPS nal unit which we need for MVC specific
21501 codec_data generation.
21503 2014-05-07 00:08:33 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21505 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21506 encoder: h264: fix MVC pipeline hang while encoding with B-frames.
21507 Since we are encoding each view independently from each other, we
21508 need a higher number of pre-allocated surfaces to be used as the
21509 reconstructed frames. For Stereo High profile encoding, this means
21510 to effectively double the number of frames to be stored in the DPB.
21512 2014-02-17 15:51:43 +0800 Li Xiaowei <xiaowei.a.li@intel.com>
21514 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21515 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
21516 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
21517 * gst/vaapi/gstvaapiencode_h264.c:
21518 encoder: h264: add initial support for H.264 Stereo High profile.
21519 Add initial support for Subset SPS, Prefix NAL and Slice Extension NAL
21520 for non-base-view streams encoding, and the usual SPS, PPS and Slice
21521 NALs for base-view encoding.
21522 The H.264 Stereo High profile encoding mode will be turned on when the
21523 "num-views" parameter is set to 2. The source (raw) YUV frames will be
21524 considered as Left/Right view, alternatively.
21525 Each of the two views has its own frames reordering pool and reference
21526 frames list management system. Inter-view references are not supported
21527 yet, so the views are encoded independently from each other.
21528 Signed-off-by: Li Xiaowei <xiaowei.a.li@intel.com>
21529 [limited to Stereo High profile per the definition of MAX_NUM_VIEWS]
21530 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21532 2014-02-17 11:10:26 +0800 Li Xiaowei <xiaowei.a.li@intel.com>
21534 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21535 encoder: h264: wrap pools for refs and frames reordering.
21536 Create structures to maintain the reference frames list (RefPool) and
21537 frames reordering (ReorderPool) logic.
21538 This is a prerequisite for H.264 MVC support.
21539 Signed-off-by: Li Xiaowei <xiaowei.a.li@intel.com>
21541 2014-02-14 15:33:15 +0800 Li Xiaowei <xiaowei.a.li@intel.com>
21543 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
21544 encoder: h264: add provisional support for subset SPS headers.
21545 Add provisions to write subset SPS headers to the bitstream in view
21546 to supporting the H.264 MVC specification.
21547 This assumes the libva "staging" branch is in use.
21548 Signed-off-by: Li Xiaowei <xiaowei.a.li@intel.com>
21550 2013-12-18 13:47:32 +0800 Li Xiaowei <xiaowei.a.li@intel.com>
21552 * gst-libs/gst/vaapi/gstvaapiprofile.c:
21553 * gst-libs/gst/vaapi/gstvaapiutils.c:
21554 utils: add H.264 MVC profiles.
21555 Add "MultiView High" and "Stereo High" definitions.
21556 Signed-off-by: Li Xiaowei <xiaowei.a.li@intel.com>
21557 [require VA-API >= 0.35.2 for MVC profiles]
21558 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21560 2014-06-02 16:25:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21562 * gst-libs/gst/vaapi/gstvaapiprofile.c:
21563 * gst-libs/gst/vaapi/gstvaapiutils.c:
21564 utils: only enable VP8 profiles for newer VA-API versions.
21565 VP8 decoding API appeared in VA-API >= 0.35.0. So, disable mappings
21566 involving VP8 codec on earlier versions of the API.
21568 2014-05-22 10:04:46 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21570 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21571 decoder: h264: compute view ids only once per slice.
21572 Optimize lookups of view ids / view order indices by caching the result
21573 of the calculatiosn right into the GstVaapiParserInfoH264 struct. This
21574 terribly simplifies is_new_access_unit() and find_first_field() functions.
21576 2014-05-21 17:57:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21578 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21579 decoder: h264: add support for MVC interlaced streams.
21580 Fix support for MVC Stereo High profile streams with interlaced frames.
21581 Also improve the detection logic of the first field.
21583 2014-05-20 18:08:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21585 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21586 decoder: h264: add MVC profiles compatibility logic.
21587 Add safe fallbacks for MVC profiles:
21588 - all MultiView High profile streams with 2 views at most can be decoded
21589 with a Stereo High profile compliant decoder ;
21590 - all Stereo High profile streams with only progressive views can be
21591 decoded with a MultiView High profile compliant decoder ;
21592 - all drivers that support slice-level decoding could normally support
21593 MVC profiles when the DPB holds at most 16 frames.
21595 2014-05-02 14:58:45 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21597 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21598 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
21599 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
21600 decoder: h264: add initial support for MVC.
21601 https://bugzilla.gnome.org/show_bug.cgi?id=721772
21603 2014-05-01 19:16:09 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21605 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21606 decoder: h264: dynamically allocate the DPB.
21607 Dynamically allocate the Decoded Picture Buffer (DPB) and add provisions
21608 for supporting the MVC allocation requirements.
21610 2014-05-01 19:33:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21612 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21613 decoder: h264: fix detection of access unit boundaries.
21614 In order to have a stricter conforming implementation, we need to carefully
21615 detect access unit boundaries. Additional operations could be necessary to
21616 perform at those boundaries.
21618 2013-03-13 11:44:38 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21620 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21621 decoder: h264: detect the first VCL NAL unit of a picture for MVC.
21622 Detect the first VCL NAL unit of a picture for MVC, based on the
21623 view_id as per H.7.4.1.2.4. Note that we only need to detect new
21625 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21627 2013-10-31 19:32:55 +0800 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21629 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21630 decoder: h264: properly handle Prefix NAL units.
21631 Always cache the previous NAL unit so that we could check whether
21632 there is a Prefix NAL unit immediately preceding the current slice
21633 or IDR NAL unit. In that case, the NAL unit metadata is copied into
21634 the current NAL unit. Otherwise, some default values are inferred,
21635 tentatively. e.g. view_id shall be set to 0 and inter_view_flag to 1.
21636 [infer default values for slice if previous NAL was not a Prefix]
21637 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21639 2013-02-28 15:59:55 +0800 Xiaowei Li <xiaowei.a.li@intel.com>
21641 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21642 decoder: h264: add support for MVC base views.
21643 Allow decoding for base views of MVC encoded streams. For now, just skip
21644 the slice extension and prefix NAL units, and skip non-base view frames.
21645 Signed-off-by: Xiaowei Li <xiaowei.a.li@intel.com>
21646 [fixed memory leak, improved check for MVC NAL units]
21647 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21649 2014-05-04 14:49:28 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21651 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21652 decoder: h264: simplify storage of decoded picture into DPB.
21653 Factor out process by which the decoded picture with the lowest POC
21654 is found, and possibly output. Likewise, the storage and marking of
21655 a reference decoded, or non-reference decoded picture, into the DPB
21656 could also be simplified as they mostly share the same operations.
21658 2014-05-02 22:40:16 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21660 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21661 decoder: h264: minor clean-ups.
21662 Make init_picture_ref_lists() more consistent with other functions
21663 related to the reference marking process by supplying the current
21664 picture as argument.
21666 2014-05-20 11:36:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21668 * gst-libs/gst/vaapi/gstvaapidisplay.c:
21669 * gst-libs/gst/vaapi/gstvaapidisplay.h:
21670 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
21671 display: add utility function to query VA driver name.
21672 Add gst_vaapi_display_get_vendor_string() helper function to query
21673 the underlying VA driver name. The display object owns the resulting
21674 string, so it shall not be deallocated.
21675 That function is thread-safe. It could be used for debugging purposes,
21678 2014-03-07 14:50:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21680 * gst-libs/gst/vaapi/gstvaapidisplay.c:
21681 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
21682 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
21683 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
21684 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
21685 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
21686 * gst-libs/gst/vaapi/gstvaapidisplaycache.h:
21687 display: make cache maintenance really MT-safe.
21688 Make sure to initialize one GstVaapiDisplay at a time, even in threaded
21689 environments. This makes sure the display cache is also consistent
21690 during the whole display creation process. In the former implementation,
21691 there were risks that display cache got updated in another thread.
21693 2014-05-03 15:56:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21695 * gst-libs/gst/vaapi/gstvaapicontext.c:
21696 context: allow dynamic growth of VA surfaces pool.
21697 Add support for dynamic growth of the VA surfaces pool. For decoding,
21698 this implies the recreation of the underlying VA context, as per the
21699 requirement from VA-API. Besides, only increases are supported, not
21702 2014-05-03 15:47:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21704 * gst-libs/gst/vaapi/gstvaapicontext.c:
21705 context: reset VA context if VA surfaces set changed.
21706 It is a requirement from VA-API specification that the VA context got
21707 from vaCreateContext(), for decoding purposes, binds the supplied set
21708 of VA surfaces. This means that if the set of VA surfaces is to be
21709 changed for the current decode session, then the VA context needs to
21710 be recreated with the new set of VA surfaces.
21712 2014-05-12 19:23:04 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21714 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21715 decoder: h264: fix assignment of LongTermFrameIdx.
21716 Complement fix committed as e95a42e.
21717 The H.264 AVC standard has to say: if the field is part of a reference
21718 frame or a complementary reference field pair, and the other field of
21719 the same reference frame or complementary reference field pair is also
21720 marked as "used for long-term reference", the reference frame or
21721 complementary reference field pair is also marked as "used for long-term
21722 reference" and assigned LongTermFrameIdx equal to long_term_frame_idx.
21723 This fixes decoding of MR9_BT_B in strict mode.
21724 https://bugs.freedesktop.org/show_bug.cgi?id=64624
21725 https://bugzilla.gnome.org/show_bug.cgi?id=724518
21727 2014-05-10 06:23:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21729 * gst-libs/gst/vaapi/gstvaapicontext.c:
21730 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21731 * gst-libs/gst/vaapi/gstvaapiutils.c:
21732 * gst-libs/gst/vaapi/gstvaapiutils.h:
21733 decoder: h264: properly support grayscale formats.
21734 Request the correct chroma format for decoding grayscale streams.
21735 i.e. make lookups of the VA chroma format more generic, thus possibly
21736 supporting more formats in the future.
21737 This means that, if a VA driver doesn't support grayscale formats,
21738 it is now going to fail. We cannot safely assume that maybe grayscale
21739 was implemented on top of some YUV 4:2:0 with the chroma components
21742 2014-02-06 11:14:09 +0000 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
21744 * gst-libs/gst/vaapi/gstvaapifilter.c:
21745 * gst-libs/gst/vaapi/gstvaapifilter.h:
21746 * tests/test-filter.c:
21747 build: fix source file modes.
21748 A few source files are marked executable in error - fix them
21749 https://bugzilla.gnome.org/show_bug.cgi?id=723748
21750 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
21752 2014-04-29 13:22:47 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
21755 * gst-libs/gst/vaapi/Makefile.am:
21756 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
21757 * gst/vaapi/gstvaapidecode.c:
21758 build: fix conditional compilation of VP8 decoder.
21759 https://bugzilla.gnome.org/show_bug.cgi?id=729170
21760 [added check for VASliceParameterBufferBase fields]
21761 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21763 2014-04-27 08:55:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21765 * gst-libs/gst/codecparsers/Makefile.am:
21766 build: fix make dist for codecparsers.
21768 2014-04-28 09:42:13 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21770 * ext/codecparsers:
21771 codecparsers: update to gst-vaapi-branch commit eaa3f7e.
21772 h264: fix parsing of slice groups for map type = 2
21774 2014-04-26 22:35:49 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21776 * ext/codecparsers:
21777 codecparsers: update to gst-vaapi branch commit f44edfc.
21778 h264: fix derivation of default scaling lists
21780 2013-05-24 19:00:54 +0800 Cong Zhong <congx.zhong@intel.com>
21782 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21783 decoder: h264: fix long-term reference picture marking process.
21784 Fix reference picture marking process with memory_management_control_op
21785 set to 3 and 6, i.e. assign LongTermFrameIdx to a short-term reference
21786 picture, or the current picture.
21787 This fixes decoding of FRExt_MMCO4_Sony_B.
21788 https://bugs.freedesktop.org/show_bug.cgi?id=64624
21789 https://bugzilla.gnome.org/show_bug.cgi?id=724518
21790 [squashed, edited to use GST_VAAPI_PICTURE_IS_COMPLETE() macro]
21791 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21793 2014-04-26 20:21:46 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21795 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21796 decoder: h264: fix initialization of RefPicLists for multiple slices.
21797 The initialization of reference picture lists (8.2.4.2) applies to all
21798 slices. So, the RefPicList0/1 lists need to be constructed prior to
21799 each slice submission to the HW decoder.
21800 This fixes decoding of video sequences where frames are encoded with
21801 multiple slices of different types, e.g. 4 slices in this order I, P,
21802 I, and P. More precisely, CABAST3_Sony_E and CABASTBR3_Sony_B.
21803 https://bugzilla.gnome.org/show_bug.cgi?id=724518
21805 2013-06-04 15:01:46 +0800 Zhong Cong <congx.zhong@intel.com>
21807 * ext/codecparsers:
21808 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21809 decoder: h264: skip SPS extension and auxiliary slice NALs.
21810 When NAL units of type 13 (SPS extension) or type 19 (auxiliary slice)
21811 are present in a video, decoders shall perform the (optional) decoding
21812 process specified for these NAL units or shall ignore them (7.4.1).
21813 Implement option 2 (skip) for now, as alpha composition is not
21814 supported yet during the decoding process.
21815 This fixes decoding of the primary coded video in alphaconformanceG.
21816 https://bugzilla.gnome.org/show_bug.cgi?id=703928
21817 https://bugzilla.gnome.org/show_bug.cgi?id=728869
21818 https://bugzilla.gnome.org/show_bug.cgi?id=724518
21819 [skip NAL units earlier, i.e. at parsing time]
21820 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21822 2013-03-07 11:32:20 +0800 Li Xiaowei <xiaowei.a.li@intel.com>
21824 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21825 decoder: h264: fix slice data bit offset with MVC NAL units.
21826 When MVC slice NAL units (coded slice extension and prefix NAL) are
21827 present, the number of NAL header bytes is 3, not 1 as usual.
21828 Signed-off-by: Li Xiaowei <xiaowei.a.li@intel.com>
21829 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21831 2014-04-25 19:11:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21833 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21834 decoder: h264: fix activation of picture and sequence parameters.
21835 At the time the first VCL NAL unit of a primary coded picture is found,
21836 and if that NAL unit was parsed to be an SPS or PPS, then the entries
21837 in the parser may have been overriden. This means that, when the picture
21838 is to be decoded, slice_hdr->pps could point to an invalid (the next)
21840 So, one way to solve this problem is to not use the parser PPS and
21841 SPS info but rather maintain our own activation chain in the decoder.
21842 https://bugzilla.gnome.org/show_bug.cgi?id=724519
21843 https://bugzilla.gnome.org/show_bug.cgi?id=724518
21845 2014-04-25 16:24:01 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21847 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21848 decoder: h264: retain SEI messages until the end of frame.
21849 Retain the SEI messages that were parsed from the access unit until we
21850 have completely decoded the current frame. This is done so that we can
21851 peek at that data whenever necessary during decoding. e.g. for exposing
21852 3D stereoscopic information at a later stage.
21854 2014-04-25 14:23:11 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21856 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21857 decoder: h264: add support for grayscale encoded clips.
21858 Fix support for grayscale encoded video clips, and possibly others if
21859 the underlying driver supports the non-YUV 4:2:0 formats. i.e. defer
21860 the decision that a surface with the desired chroma format is not
21861 supported to the actual VA driver implementation.
21862 https://bugzilla.gnome.org/show_bug.cgi?id=728144
21864 2014-04-25 14:16:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21866 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
21867 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
21868 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
21869 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
21870 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
21871 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
21872 decoder: default to YUV 4:2:0 VA surfaces.
21873 Cope with context changes to support non-YUV 4:2:0 VA surfaces. Still,
21874 make sure all codecs use YUV 4:2:0 output format for now, by default.
21876 2014-04-25 13:57:02 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21878 * gst-libs/gst/vaapi/gstvaapidecoder.c:
21879 * gst-libs/gst/vaapi/gstvaapidecoder.h:
21880 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
21881 decoder: re-indent GstVaapiDecoder base object.
21883 2014-04-25 13:47:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21885 * gst-libs/gst/vaapi/gstvaapiencoder.c:
21886 encoder: derive chroma type from video format.
21887 Cope with previous VA context change to derive the correct surface chroma
21888 type from the input video format.
21890 2014-04-25 13:45:31 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21892 * gst-libs/gst/vaapi/gstvaapicontext.c:
21893 * gst-libs/gst/vaapi/gstvaapicontext.h:
21894 context: add support for non-YUV 4:2:0 formats.
21895 Don't force allocation of VA surfaces in YUV 4:2:0 format. Rather, allow
21896 for the upper layer to specify the desired chroma type. If the chroma
21897 type field is not set (or yields zero), then YUV 4:2:0 format is used
21900 2014-04-22 19:53:50 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21902 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
21903 vp8: fix per-segment deblocking filter level in relative mode.
21904 Fix possible bug when a per-segment deblocking filter level value
21905 needs to be set in non-absolute mode, i.e. when the loop filter update
21906 value is negative in delta mode.
21907 Also clamp the resulting filter level value to 0..63 range.
21909 2014-04-22 17:25:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21911 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
21912 vp8: fix check for disabling the loop filter (again).
21913 Improve condition to disable the loop filter. The previous heuristic
21914 used to check all filter levels, for all segments. It turns out that
21915 only the base filter_level value defined in the frame header needs
21917 This fixes 00-comprehensive-013.
21919 2014-04-21 18:02:21 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21921 * gst-libs/gst/codecparsers/Makefile.am:
21922 * gst/vaapi/Makefile.am:
21923 build: fix make dist with certain conditionals not met.
21924 Fix generation of source tarballs when certain conditionals are not
21925 met. e.g. always include all buildable codecparsers sources in the
21926 distribution tarball, fix plug-in element sources set to include X11
21929 2014-04-21 17:34:59 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21932 build: add missing files for GStreamer 0.10.
21933 Add missing GstVideoEncoder implementation files to fix build with ancient
21934 GStreamer 0.10 stack.
21935 https://bugzilla.gnome.org/show_bug.cgi?id=723964
21937 2014-04-19 10:17:20 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21940 build: add missing files for VP8 bitstream parser.
21941 Fix make dist for building the VP8 bitstream parser.
21943 2014-04-21 17:49:38 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21946 * ext/libvpx/Makefile.am:
21947 * gst-libs/gst/codecparsers/Makefile.am:
21948 vp8: allow compilation without the built-in libvpx.
21949 The built-in libvpx serves multiple purposes, among which the most
21950 important ones could be: track the most up-to-date, and optimized,
21951 range decoder; allow for future hybrid implementations (non-VLD);
21952 and have a completely independent range decoder implementation.
21954 2014-04-21 17:28:27 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21956 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
21957 vp8: propagate PTS from demux frame.
21958 gst_adapter_prev_pts() is forbidden within libgstvaapi. Besides, the demuxer
21959 or parser would already have determined the PTS from a previous stage.
21961 2014-04-19 07:49:30 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21964 * debian.upstream/libgstvaapi.install.in:
21965 * ext/libvpx/Makefile.am:
21966 * ext/libvpx/sources.frag:
21967 * gst-libs/gst/codecparsers/Makefile.am:
21968 vp8: fix compilation with built-in libvpx.
21969 Apply correct patch from fd.o #722760 to fix several issues: update the
21970 license terms to LGPLv2.1+, fix dependencies to built-in libvpx and fix
21973 2014-02-13 21:17:23 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21979 * ext/libvpx/Makefile.am:
21980 * ext/libvpx/gstlibvpx.c:
21981 * ext/libvpx/gstlibvpx.h:
21982 * ext/libvpx/libgstcodecparsers_vpx.vers:
21983 * ext/libvpx/sources.frag:
21984 * ext/libvpx/upstream:
21985 * gst-libs/gst/codecparsers/Makefile.am:
21986 * gst-libs/gst/codecparsers/gstvaapilibvpx.c:
21987 vp8: use range decoder from libvpx.
21988 Add libvpx submodule that tracks the upstream version 1.3.0. This is
21989 needed to build a libgstcodecparsers_vpx.so library with all symbols
21990 placed into the GSTREAMER namespace.
21992 2014-04-04 19:17:17 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
21994 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
21995 vp8: fix check for disabling the loop filter.
21997 2013-12-27 07:18:24 +0800 Zhao, Halley <halley.zhao@intel.com>
22001 * gst-libs/gst/codecparsers/Makefile.am:
22002 * gst-libs/gst/vaapi/Makefile.am:
22003 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
22004 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
22005 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
22006 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
22007 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.c:
22008 * gst-libs/gst/vaapi/gstvaapidecoder_vp8.h:
22009 * gst-libs/gst/vaapi/gstvaapiprofile.c:
22010 * gst-libs/gst/vaapi/gstvaapiprofile.h:
22011 * gst/vaapi/gstvaapidecode.c:
22012 Add initial VP8 decoder.
22013 https://bugzilla.gnome.org/show_bug.cgi?id=722761
22014 [complete overhaul, fixed support for resolution changes]
22015 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22017 2014-03-21 15:15:37 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22019 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
22020 decoder: h264: cope with new gst_h264_parser_parse_sei() interface.
22021 The gst_h264_parse_parse_sei() function now returns an array of SEI
22022 messages, instead of a single SEI message. Reason: it is allowed to
22023 have several SEI messages packed into a single SEI NAL unit, instead
22024 of multiple NAL units.
22026 2014-04-18 19:36:16 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22028 * ext/codecparsers:
22029 codecparsers: update to gst-vaapi-branch commit a454f86.
22030 b2eb5f6 vp8: rename dboolhuff symbols
22031 b74a881 vp8: add GStreamer native utilities
22032 2940ac6 add VP8 bitstream parser
22034 2014-04-18 19:16:56 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22036 * ext/codecparsers:
22037 codecparsers: update to gst-vaapi-branch commit d459bc5.
22038 d459bc5 h264: set framerate even for interlaced videos
22039 c78b82c h264: add support for Recovery Point SEI message
22040 7693bac h264: add support for Frame Packing Arrangement SEI message
22041 31fafa7 h264: add support for Stereo Video Information SEI message
22042 8b113a6 h264: parse seq_parameter_set_mvc_extension()
22043 040f9b8 h264: parse MVC syntax elements
22044 cc18ef3 h264: add nal_reader_skip_long() helper
22045 7e76a48 h264: fix slice_header() parsing for MVC
22046 caf46d8 h264: add gst_h264_parse_nalu_header() helper
22047 f75074e h264: add gst_h264_parse_sps_data() helper
22048 798c397 h264: clean-up gst_h264_parser_parse_sei_message()
22049 4e36737 h264: fix skipping of unsupported SEI messages
22050 5300766 h264: fix SEI buffering_period() parsing
22052 2014-03-21 15:09:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22054 * ext/codecparsers:
22055 * gst-libs/gst/codecparsers/Makefile.am:
22056 codecparsers: update to gst-vaapi-branch commit 8fadf40.
22057 8fadf40 h264: Fix multiple SEI messages in one SEI RBSP parsing.
22058 644825f h265: remove trailling 0x00 bytes as the spec doesn't allow them
22059 95f9f0f h264: remove trailling 0x00 bytes as the spec doesn't allow them
22060 766007b h265: Initialize pointer correctly that is never assigned but freed in error cases
22061 8ec5816 h265: Fix segfault when parsing HRD parameter
22062 5b1730f h265: Fix segfault when parsing VPS
22063 983b7f7 h265: prevent to overrun chroma_weight_l0_flag
22064 7ba641d h265: Fix debug output
22065 d9f9f9b h264: not all startcodes should have 3-byte 0 prefix
22067 2014-02-04 18:35:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22069 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
22070 decoder: h264: fix robustness patch for bytestream format.
22071 Fix parser and decoder state to sync at the right locations. This is
22072 because we could reset the parser state, while the decoder state was
22073 not copied yet, e.g. when parsing several NAL units from multiple frames
22074 whereas the current frame was not decoded yet.
22075 This is a regression brought in by commit 6fe5496.
22077 2014-02-18 06:56:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22080 * pkgconfig/Makefile.am:
22081 * pkgconfig/gstreamer-vaapi-drm.pc.in:
22082 * pkgconfig/gstreamer-vaapi-glx.pc.in:
22083 * pkgconfig/gstreamer-vaapi-wayland.pc.in:
22084 * pkgconfig/gstreamer-vaapi-x11.pc.in:
22085 build: fix pkgconfig file names (again).
22086 It turns out it is more convenient to have only pkgconfig files named
22087 after the installed GStreamer API version (1.0) instead of using all
22088 possible subsequent names from that (1.0, 1.2, 1.4). i.e. they conflict
22089 altogether anyway, so align pkgconfig file names to that.
22091 2014-02-07 09:43:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22093 * debian.upstream/libgstvaapi-dev.install.in:
22094 * gst-libs/gst/vaapi/Makefile.am:
22095 * pkgconfig/gstreamer-vaapi.pc.in:
22096 build: fix packaging for GStreamer 1.2.
22097 Fix gstreamer-vaapi includedir for GStreamer 1.2 setups. i.e. use
22098 the pkgconfig version (1.0) instead of the intended API version (1.2).
22099 libgstvaapi1.0-dev and libgstvaapi1.2-dev packages will now conflict,
22100 as would core GStreamer 1.0 and GStreamer 1.2 dev packages anyway.
22102 2014-01-24 11:27:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22105 Bump version for development.
22107 2014-01-24 10:55:39 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22109 * debian.upstream/control.in:
22110 debian: fix trailing whitespace in description.
22112 2014-01-23 23:24:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22114 * debian.upstream/control.in:
22115 * debian.upstream/copyright:
22116 debian: fix package description.
22117 Try to improve package description for the compiled plug-in elements
22118 available in there. e.g. only display vaapidownload and vaapiupload
22119 for GStreamer 0.10 builds, display vaapiencode_* elements when VA
22120 encoding is enabled, etc.
22121 Also increase the copyright notice date.
22123 2014-01-23 22:47:19 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22125 * gst-libs/gst/vaapi/gstvaapidecoder.c:
22126 build: fix warnings on 64-bit platforms.
22128 2014-01-23 22:44:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22130 * gst-libs/gst/vaapi/gstvaapicontext.c:
22131 * gst-libs/gst/vaapi/gstvaapiutils.c:
22132 build: fix for older versions of VA-API (< 0.34.0).
22133 Fix build with older versions of VA-API (< 0.34.0), or versions without
22134 good enough headers for encoding support for instance.
22136 2014-01-23 19:36:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22142 2014-01-23 19:32:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22146 VA-API up to 0.34.0 is actually supported. Mention new video encoding
22147 support. Update copyright years, list of supported Intel HD Graphics
22150 2014-01-23 19:18:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22155 2014-01-20 14:16:56 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22157 * tests/test-filter.c:
22158 tests: test-filter: fix "deinterlace" option parse.
22159 Default to GST_VAAPI_DEINTERLACE_METHOD_NONE if no "deinterlace" option
22160 string was provided, i.e. if it remained set to NULL.
22162 2014-01-23 18:41:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22164 * gst-libs/gst/vaapi/Makefile.am:
22165 * gst-libs/gst/vaapi/gstvaapicontext.c:
22166 * gst-libs/gst/vaapi/gstvaapicontext.h:
22167 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22168 * gst-libs/gst/vaapi/gstvaapiutils_core.c:
22169 * gst-libs/gst/vaapi/gstvaapiutils_core.h:
22170 libs: factor out usages of vaGetConfigAttributes().
22171 Add gst_vaapi_get_config_attribute() helper function that takes a
22172 GstVaapiDisplay and the rest of the arguments with VA types. The aim
22173 is to have thread-safe VA helpers by default.
22175 2014-01-23 17:41:02 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22177 * gst-libs/gst/vaapi/gstvaapiutils.c:
22178 * gst-libs/gst/vaapi/gstvaapiutils.h:
22179 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
22180 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
22181 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
22182 * gst-libs/gst/vaapi/gstvaapiutils_h264.h:
22183 * gst-libs/gst/vaapi/gstvaapiutils_h264_priv.h:
22184 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2.h:
22185 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2_priv.h:
22186 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
22187 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
22188 libs: re-indent all source code related to VA utilities.
22190 2014-01-23 17:06:08 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22192 * gst-libs/gst/vaapi/libgstvaapi_priv_check.h:
22193 libs: add missing file (libgstvaapi_priv_check.h).
22195 2014-01-23 15:13:06 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22197 * gst-libs/gst/vaapi/gstvaapicontext.c:
22198 * gst-libs/gst/vaapi/gstvaapicontext.h:
22199 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22200 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22201 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22202 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
22203 encoder: notify the encoder of the submitted packed headers.
22204 Make sure to configure the encoder with the set of packed headers we
22205 intend to generate and submit. i.e. make selection of packed headers
22206 to submit more robust.
22208 2014-01-23 15:10:11 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22210 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22211 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22212 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22213 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
22214 encoder: fix and factor out check for supported rate-control modes.
22215 Cache the first compatible GstVaapiProfile found if the encoder is not
22216 configured yet. Next, factor out the code to check for the supported
22217 rate-control modes by moving out vaGetConfigAttributes() to a separate
22218 function, while also making sure that the attribute type is actually
22219 supported by the encoder.
22220 Also fix the default set of supported rate control modes to not the
22221 "none" variant. It's totally useless to expose it at this point.
22223 2014-01-23 14:01:33 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22225 * gst-libs/gst/vaapi/gstvaapicontext.c:
22226 * gst-libs/gst/vaapi/gstvaapicontext.h:
22227 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22228 context: move rate-control mode to encoder specific config.
22229 Move usage-specific config out of the common GstVaapiContextInfo.
22230 Create a specialized config for encoding and move rate-control mode
22233 2014-01-23 13:30:41 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22235 * gst-libs/gst/vaapi/gstvaapicontext.c:
22236 * gst-libs/gst/vaapi/gstvaapicontext.h:
22237 * gst-libs/gst/vaapi/gstvaapidecoder.c:
22238 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22239 context: introduce concept of usage.
22240 Introduce GstVaapiContextUsage so that to explicitly determine the
22241 usage of a VA context. This is useful in view to simplifying the
22242 creation of VA context for VPP too.
22244 2014-01-23 11:44:12 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22246 * gst-libs/gst/vaapi/gstvaapicontext.c:
22247 context: fix get_attribute() value result.
22248 Unknown attributes, or attributes that are not supported for the given
22249 profile/entrypoint pair have a return value of VA_ATTRIB_NOT_SUPPORTED.
22250 So, return failure in this case.
22252 2014-01-23 10:59:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22254 * gst-libs/gst/vaapi/Makefile.am:
22255 * gst-libs/gst/vaapi/gstvaapicontext.c:
22256 * gst-libs/gst/vaapi/gstvaapicontext.h:
22257 * gst-libs/gst/vaapi/gstvaapicontext_overlay.c:
22258 * gst-libs/gst/vaapi/gstvaapicontext_overlay.h:
22259 * gst-libs/gst/vaapi/gstvaapisurface.c:
22260 context: move overlay composition to separate files.
22261 Move GstVideoOverlayComposition handling to separate source files.
22262 This helps keeing GstVaapiContext core implementation to the bare
22263 minimal, i.e. simpy helpers to create a VA context and handle pool
22264 of associated VA surfaces.
22266 2014-01-23 09:41:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22268 * gst-libs/gst/vaapi/gstvaapicontext.c:
22269 * gst-libs/gst/vaapi/gstvaapicontext.h:
22270 * gst-libs/gst/vaapi/gstvaapidecoder.c:
22271 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22272 context: clean-ups. Strip down APIs.
22273 Improve documentation and debug messages. Clean-up APIs, i.e. strip
22274 them down to the minimal set of interfaces. They are private, so no
22275 need expose getters for instance.
22277 2014-01-23 09:27:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22279 * gst-libs/gst/vaapi/gstvaapicontext.c:
22280 * gst-libs/gst/vaapi/gstvaapicontext.h:
22281 context: re-indent all GstVaapiContext related source code.
22283 2014-01-23 10:20:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22285 * gst-libs/gst/vaapi/Makefile.am:
22286 * gst-libs/gst/vaapi/gstvaapidecoder_unit.h:
22287 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
22288 * gst-libs/gst/vaapi/gstvaapidisplaycache.h:
22289 * gst-libs/gst/vaapi/gstvaapiminiobject.h:
22290 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
22291 * gst-libs/gst/vaapi/gstvaapipixmap_priv.h:
22292 * gst-libs/gst/vaapi/gstvaapiutils.h:
22293 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
22294 * gst-libs/gst/vaapi/gstvaapiutils_h264_priv.h:
22295 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2_priv.h:
22296 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
22297 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
22298 libs: check that private headers remain private.
22299 Make sure that libgstvaapi private headers remain internally used to
22300 build libgstvaapi libraries only. All header dependencies were reviewed
22301 and checks for IN_LIBGSTVAAPI definition were added accordingly.
22302 Also rename GST_VAAPI_CORE definition to IN_LIBGSTVAAPI_CORE to keep
22305 2014-01-22 19:04:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22308 Bump library major version.
22309 Bump the library major version due to API/ABI changes that occurred in
22310 the imaging API. In particular, GstVaapiDisplay interfaces no longer
22311 expose any GstCaps but provide GArray based ones e.g. to determine the
22312 set of supported decode/encode profiles.
22314 2014-01-22 18:54:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22316 * gst-libs/gst/vaapi/glibcompat.h:
22317 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
22318 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
22319 * gst-libs/gst/vaapi/gstvaapicodedbufferpool.c:
22320 * gst-libs/gst/vaapi/gstvaapicodedbufferpool.h:
22321 * gst-libs/gst/vaapi/gstvaapicontext.c:
22322 * gst-libs/gst/vaapi/gstvaapicontext.h:
22323 * gst-libs/gst/vaapi/gstvaapidecoder.c:
22324 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
22325 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
22326 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
22327 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
22328 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22329 * gst-libs/gst/vaapi/gstvaapiencoder.h:
22330 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22331 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
22332 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22333 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
22334 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2_priv.h:
22335 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
22336 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
22337 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
22338 * gst-libs/gst/vaapi/gstvaapifilter.c:
22339 * gst-libs/gst/vaapi/gstvaapipixmap_x11.c:
22340 * gst-libs/gst/vaapi/gstvaapisurface.c:
22341 * gst-libs/gst/vaapi/gstvaapisurface.h:
22342 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
22343 * gst-libs/gst/vaapi/gstvaapitypes.h:
22344 * gst-libs/gst/vaapi/gstvaapiutils.c:
22345 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
22346 * gst-libs/gst/vaapi/gstvaapiutils_h264.h:
22347 * gst-libs/gst/vaapi/gstvaapiutils_h264_priv.h:
22348 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2.c:
22349 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2.h:
22350 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2_priv.h:
22351 * gst-libs/gst/vaapi/gstvaapivalue.c:
22352 * gst-libs/gst/vaapi/gstvaapivalue.h:
22353 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
22354 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
22355 * gst/vaapi/gstvaapidecode.c:
22356 * gst/vaapi/gstvaapiencode.c:
22357 * gst/vaapi/gstvaapiencode.h:
22358 * gst/vaapi/gstvaapiencode_h264.c:
22359 * gst/vaapi/gstvaapiencode_h264.h:
22360 * gst/vaapi/gstvaapiencode_mpeg2.c:
22361 * gst/vaapi/gstvaapiencode_mpeg2.h:
22362 * gst/vaapi/gstvaapipluginbase.c:
22363 * gst/vaapi/gstvaapipluginutil.c:
22364 * gst/vaapi/gstvaapipluginutil.h:
22365 * gst/vaapi/gstvaapipostproc.c:
22366 * gst/vaapi/gstvaapipostproc.h:
22367 * gst/vaapi/gstvaapisink.c:
22368 * gst/vaapi/gstvaapisink.h:
22369 * gst/vaapi/gstvaapivideometa_texture.c:
22370 * tests/simple-decoder.c:
22371 legal: update copyright notice dates.
22373 2014-01-22 18:49:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22375 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22376 * gst-libs/gst/vaapi/gstvaapiencoder.h:
22377 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22378 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
22379 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22380 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
22381 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2_priv.h:
22382 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
22383 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
22384 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
22385 * gst/vaapi/gstvaapiencode.c:
22386 * gst/vaapi/gstvaapiencode.h:
22387 * gst/vaapi/gstvaapiencode_h264.c:
22388 * gst/vaapi/gstvaapiencode_h264.h:
22389 * gst/vaapi/gstvaapiencode_mpeg2.c:
22390 * gst/vaapi/gstvaapiencode_mpeg2.h:
22391 legal: add per-file authorship information.
22393 2014-01-22 18:11:26 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22395 * gst-libs/gst/vaapi/gstvaapidecoder.c:
22396 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
22397 decoder: fix video codec frame number in standalone mode.
22398 Set a valid GstVideoCodecFrame.system_frame_number when decoding a
22399 stream in standalone mode. While we are at it, improve the debugging
22400 messages to also include that frame number.
22402 2014-01-17 16:56:53 +0800 Wind Yuan <feng.yuan@intel.com>
22404 * gst-libs/gst/vaapi/gstvaapidecoder.c:
22405 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
22406 decoder: fix crash on invalid pointer for GST_DEBUG().
22407 When decoding failed, or that the frame was dropped, the associated
22408 surface proxy is not guaranteed to be present. Thus, the GST_DEBUG()
22409 message needs to check whether the proxy is actually present or not.
22410 https://bugzilla.gnome.org/show_bug.cgi?id=722403
22411 [fixed gst_vaapi_surface_proxy_get_surface_id() to return VA_INVALID_ID]
22412 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22414 2014-01-22 17:07:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22416 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22417 encoder: h264: disable NAL HRD parameters for now.
22418 Don't emit NAL HRD parameters for now in the SPS headers because the
22419 SEI buffering_period() and picture_timing() messages are not handled
22420 yet. Some additional changes are necessary to get it right.
22421 https://bugzilla.gnome.org/show_bug.cgi?id=722734
22423 2014-01-21 19:04:41 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22425 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22426 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
22427 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
22428 * gst-libs/gst/vaapi/gstvaapiutils_h264_priv.h:
22429 encoder: h264: fix default CPB buffer size.
22430 Fix default CPB buffer size to something more reasonable (1500 ms)
22431 and that still fits the level limits. This is a non configurable
22432 property for now. The initial CPB removal delay is also fixed to
22434 https://bugzilla.gnome.org/show_bug.cgi?id=722087
22436 2014-01-22 14:43:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22438 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22439 encoder: h264: fix bitrate encoding for HRD conformance.
22440 Round down the calculated, or supplied, bitrate (kbps) into a multiple
22441 of the HRD bitrate scale factor. Use a bitrate scale factor of 64 so
22442 that to have less losses in precision. Likewise, don't round up because
22443 that could be a strict constraint imposed by the user.
22445 2014-01-22 11:25:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22447 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22448 encoder: h264: fix level lookup constraints wrt. bitrate.
22449 Fix the level calculation involving bitrate limits. Since we are
22450 targetting NAL HRD conformance, the check against MaxBR from the
22451 Table A-1 limits shall involve cpbBrNalFactor depending on the
22454 2014-01-21 18:01:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22456 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22457 encoder: h264: submit sequence parameter only once.
22458 Submit sequence parameter buffers only once, or when the bitstream
22459 was reconfigured in a way that requires such. Always submit packed
22460 sequence parameter buffers at I-frame period, if the VA driver needs
22462 https://bugzilla.gnome.org/show_bug.cgi?id=722737
22464 2014-01-21 18:35:17 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22466 * gst-libs/gst/vaapi/gstvaapicontext.c:
22467 * gst-libs/gst/vaapi/gstvaapicontext.h:
22468 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22469 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22470 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
22471 encoder: h264: only submit packed headers when required.
22472 Make sure to submit the packed headers only if the underlying VA driver
22473 requires those. Currently, only handle packed sequence and picture
22475 https://bugzilla.gnome.org/show_bug.cgi?id=722737
22477 2014-01-21 17:35:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22479 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22480 encoder: h264: fix ip_period value in sequence parameter.
22481 The VAEncSequenceParameterBuffer.ip_period value reprents the distance
22482 between the I-frame and the next P-frame. So, this also accounts for
22483 any additional B-frame in the middle of it.
22484 This fixes rate control heuristics for certain VA drivers.
22485 https://bugzilla.gnome.org/show_bug.cgi?id=722735
22487 2014-01-21 17:04:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22489 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22490 encoder: h264: fix level when bitrate is automatically computed.
22491 Fix level characterisation when the bitrate is automatically computed
22492 from the active coding tools. i.e. ensure the bitrate once the profile
22493 is completely characterized but before the level calculation process.
22495 2014-01-21 16:05:22 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22497 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22498 encoder: h264: clean-ups.
22499 Document and rename a few functions here and there. Drop code that
22500 caps num_bframes variable in reset_properties() since they shall
22501 have been checked beforehand, during properties initialization.
22503 2014-01-21 15:28:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22505 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22506 encoder: h264: clean-up bitwriter related utilities.
22507 Clean-up GstBitWriter related utility functions and simplify notations.
22508 While we are at it, also make bitstream writing more robust should an
22509 overflow occur. We could later optimize for writing headers capped to
22510 their maximum possible size by using the _unchecked() helper variants.
22512 2014-01-21 15:23:01 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22514 * gst-libs/gst/vaapi/Makefile.am:
22515 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22516 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
22517 encoder: h264: completely remove private headers.
22518 Drop private header since it was originally used to expose internals
22519 to the plugin element. The proper interface is now the properties API,
22520 thus rendering private headers totally obsolete.
22522 2014-01-15 15:54:32 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22524 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22525 encoder: h264: fix PPS header packing with profile < high.
22526 Fix PPS header packing when profile is below High since 8x8 transform
22527 mode and scaling lists are High Profile features.
22529 2014-01-15 15:46:19 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22531 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22532 encoder: h264: always emit VUI parameters for framerate.
22533 Always emit VUI parameters for timing_info, which includes framerate
22536 2014-01-15 15:10:48 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22538 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22539 encoder: h264: really fix frame cropping rectangle calculation.
22540 Make frame cropping rectangle calculation future proof, i.e. exactly
22541 follow the specification (7-18) to (7-21), and subsampling definitions
22543 https://bugzilla.gnome.org/show_bug.cgi?id=722089
22544 https://bugzilla.gnome.org/show_bug.cgi?id=722238
22546 2014-01-15 12:09:14 +0100 Holger Kaelberer <hk@getslash.de>
22548 * gst/vaapi/gstvaapisink.c:
22549 * gst/vaapi/gstvaapisink.h:
22550 vaapisink: set csc render flags from sinkpad caps.
22551 This maps GstVideoColorimetry information in vaapisink's sinkpad caps
22552 to GST_VAAPI_COLOR_STANDARD_* flags, if per-buffer information was not
22554 https://bugzilla.gnome.org/show_bug.cgi?id=722255
22555 [factored out code, added SMPTE240M, handle per-buffer flags]
22556 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22558 2012-03-28 15:05:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22560 * gst-libs/gst/vaapi/gstvaapisurface.h:
22561 * gst-libs/gst/vaapi/gstvaapiutils.c:
22562 * gst/vaapi/gstvaapipostproc.c:
22563 surface: rework render flags.
22564 Pack render flags per category and provide more flags into the color
22565 standard category. In particular, cover for SMPTE-240M.
22567 2013-12-13 04:14:41 +0800 Zhao, Halley <halley.zhao@intel.com>
22569 * gst-libs/gst/vaapi/gstvaapifilter.c:
22570 * gst/vaapi/gstvaapipostproc.c:
22571 * gst/vaapi/gstvaapipostproc.h:
22572 vaapipostproc: add support for colorbalance filters.
22573 Add support for hue, saturation, brightness and constrat adjustments.
22574 Also fix cap info local copy to match the really expected cap subtype
22576 https://bugzilla.gnome.org/show_bug.cgi?id=720376
22577 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22579 2013-12-12 08:38:12 +0800 Zhao, Halley <halley.zhao@intel.com>
22581 * gst/vaapi/gstvaapipostproc.c:
22582 vaapipostproc: fix support for "sharpen" filter.
22583 Fix copy/paste error when submitting the "sharpen" value to the
22584 GstVaapiFilter instance.
22585 https://bugzilla.gnome.org/show_bug.cgi?id=720375
22586 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22588 2013-12-20 12:05:42 +0000 Lionel Landwerlin <llandwerlin@gmail.com>
22591 * pkgconfig/gstreamer-vaapi-drm.pc.in:
22592 * pkgconfig/gstreamer-vaapi-glx.pc.in:
22593 * pkgconfig/gstreamer-vaapi-wayland.pc.in:
22594 * pkgconfig/gstreamer-vaapi-x11.pc.in:
22595 * pkgconfig/gstreamer-vaapi.pc.in:
22596 pkgconfig: plugin dir should use PKG version not API version.
22597 Fix the pluginsdir and includedir variables in the generated pkgconfig
22598 (.pc) files. The location needs to be built with the PKG version in
22599 mind instead of the API version.
22600 While we are at it, also fix the PKG version for GStreamer >= 1.3.
22601 https://bugzilla.gnome.org/show_bug.cgi?id=720820
22602 [additional fixes for includedir and pkg requirements]
22603 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22605 2014-01-15 10:05:45 +0100 Holger Kaelberer <hk@getslash.de>
22607 * gst/vaapi/gstvaapisink.c:
22608 vaapisink: fix display initialization in GstVideoOverlay implementation.
22609 When gst_vaapisink_video_overlay_set_window_handle() is called early,
22610 before the pipeline has been set to PLAYING, the display has not yet
22611 been initialized and _PLUGIN_BASE_DISPLAY_TYPE() is not yet
22612 up-to-date. For this reason the foreign XID is not attached.
22613 Now _ensure_display() is called earlier.
22614 https://bugzilla.gnome.org/show_bug.cgi?id=722244
22615 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22617 2013-10-09 13:47:54 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
22619 * gst/vaapi/gstvaapisink.c:
22620 vaapisink: expose the raw video formats in static caps template.
22621 Expose all raw video formats in the static caps template since the
22622 vaapisink is supporting raw data. We will get the exact set of formats
22623 supported by the driver dynamically through the _get_caps() routine.
22624 https://bugzilla.gnome.org/show_bug.cgi?id=703271
22625 https://bugzilla.gnome.org/show_bug.cgi?id=720737
22626 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22628 2013-12-11 18:08:26 +0000 Matthieu Bouron <matthieu.bouron@collabora.com>
22630 * gst/vaapi/gstvaapidecode.c:
22631 vaapidecode: query downstream caps features like GLTextureUploadMeta.
22632 Fix vaapidecode to correctly report caps features downstream, when
22633 a custom pipeline is built manually.
22634 https://bugzilla.gnome.org/show_bug.cgi?id=719372
22635 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22637 2013-12-17 15:27:10 +0000 Matthieu Bouron <matthieu.bouron@collabora.com>
22639 * gst/vaapi/gstvaapidecode.c:
22640 vaapidecode: add system memory caps to template caps.
22641 Since vaapidecode provides buffer that can be mapped as regular memory,
22642 those caps should be added to the template caps. That only applies to
22644 https://bugzilla.gnome.org/show_bug.cgi?id=720608
22645 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22647 2013-12-17 10:26:03 +0800 Wind Yuan <feng.yuan@intel.com>
22649 * gst/vaapi/gstvaapidecode.c:
22650 vaapidecode: fix hang on SIGINT.
22651 vaapidecode hangs when pipeline is stopped without any EOS, e.g. when
22652 <Ctrl>+C is pressed, thus causing the srcpad task to keep running and
22653 locked. This fixes a deadlock on state change from PAUSED to READY.
22654 https://bugzilla.gnome.org/show_bug.cgi?id=720584
22655 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22657 2013-12-17 04:23:42 -0500 Wind Yuan <feng.yuan@intel.com>
22659 * gst/vaapi/gstvaapiencode.c:
22660 vaapiencode: fix possible hang on SIGINT.
22661 vaapiencode might hang when the pipeline is stopped without any EOS,
22662 e.g. when <Ctrl>+C is pressed, thus causing the srcpad task to keep
22663 running and locked. This fixes a possible deadlock on state change
22664 from PAUSED to READY.
22665 https://bugzilla.gnome.org/show_bug.cgi?id=720584
22666 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22668 2014-01-14 16:33:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22670 * gst/vaapi/gstvaapiencode.c:
22671 vaapiencode: fix typo in error message.
22672 Fix incomplete error message in gst_vaapiencode_push_frame().
22674 2014-01-14 19:08:36 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22676 * gst/vaapi/gstvaapipluginutil.c:
22677 * gst/vaapi/gstvaapipluginutil.h:
22678 plugins: add helpers to create video caps with features.
22679 Add gst_vaapi_video_format_new_template_caps_with_features() helper
22680 function to add the supplied caps feature string on GStreamer >= 1.2.
22681 Add gst_vaapi_find_preferred_caps_feature() helper function to discover
22682 the "best" caps feature to use for the supplied pad. In practice, we
22683 will always favor memory:VASurface first, then meta:GLTextureUploadMeta,
22684 and finally the system memory caps.
22685 https://bugzilla.gnome.org/show_bug.cgi?id=719372
22687 2014-01-09 11:54:11 +0000 Matthieu Bouron <matthieu.bouron@collabora.com>
22689 * gst/vaapi/gstvaapivideometa_texture.c:
22690 plugins: don't apply overlay composition in GLTextureUpload function.
22691 The GLTextureUpload function is not in charge of doing the overlay
22692 composition if any.
22693 https://bugzilla.gnome.org/show_bug.cgi?id=721859
22694 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22696 2014-01-14 13:47:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22698 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
22699 encoder: re-order submission of VA objects.
22700 Change the submission order of VA objects so that to make that process
22701 more logical. i.e. submit sequence parameter first, if any; next the
22702 packed headers associated to sequece, picture or slices; and finally
22703 the actual picture and associated slices.
22705 2014-01-14 12:01:11 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22707 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22708 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22709 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22710 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
22711 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
22712 encoder: clean-up objects.
22713 Various clean-ups to improve consistency and readability: rename some
22714 variables, drop unused macro definitions, drop initialization of vars
22715 that are zero-initialized from the base class, drop un-necessary casts,
22716 allocate GPtrArrays with a destroy function.
22718 2014-01-13 13:41:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22720 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22721 encoder: h264: fix frame cropping rectangle calculation.
22722 Fix frame cropping rectangle calculation to handle horizontal resolutions
22723 that don't match a multiple of 16 pixels, but also the vertical resolution
22724 that was incorrectly computed for progressive sequences too.
22725 https://bugzilla.gnome.org/show_bug.cgi?id=722089
22727 2014-01-13 11:49:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22729 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22730 encoder: h264: improve automatic bitrate calculation.
22731 For non "Constant-QP" modes, we could provide more reasonable heuristics
22732 for the target bitrate. In general, 48 bits per macroblock with all the
22733 useful coding tools enable looks safe enough. Then, this rate is raised
22734 by +10% to +15% for each coding tool that is disabled.
22735 https://bugzilla.gnome.org/show_bug.cgi?id=719699
22737 2014-01-13 11:11:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22739 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22740 encoder: h264: support "high-compression" tuning option.
22741 Add support for "high-compression" tuning option. First, determine the
22742 largest supported profile by the hardware. Next, check any target limit
22743 set by the user. Then, enable each individual coding tool based on the
22744 resulting profile_idc value to use.
22745 https://bugzilla.gnome.org/show_bug.cgi?id=719696
22747 2014-01-12 22:24:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22749 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22750 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
22751 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
22752 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
22753 * gst-libs/gst/vaapi/gstvaapiutils_h264.h:
22754 * gst/vaapi/gstvaapiencode_h264.c:
22755 encoder: h264: allow target decoder constraints.
22756 Allow user to precise the largest profile to use for encoding due
22757 to target decoder constraints. For instance, if CABAC entropy coding
22758 mode is requested by "constrained-baseline" profile only is desired,
22759 then an error is returned during codec configuration.
22760 Also make sure that the suitable profile we derived actually matches
22761 what the HW can cope with.
22762 https://bugzilla.gnome.org/show_bug.cgi?id=719694
22764 2014-01-12 22:14:11 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22766 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22767 encoder: h264: refine size of coded buffer.
22768 Refine the heuristic to determine the maximum size of a coded buffer
22769 to account for the exact number of slices. set_context_info() is the
22770 last step during codec reconfiguration, no additional change is done
22771 afterwards, so re-using the num_slices field here is fine.
22772 https://bugzilla.gnome.org/show_bug.cgi?id=719953
22774 2013-12-13 17:36:08 +0800 Wind Yuan <feng.yuan@intel.com>
22776 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22777 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
22778 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
22779 encoder: h264: expose more coding tools.
22780 Add new H.264 coding tools to improve compression:
22781 - "cabac": enable CABAC entropy coding (default: FALSE);
22782 - "dct8x8": enable spatial transform 8x8 (default: FALSE).
22783 https://bugzilla.gnome.org/show_bug.cgi?id=719693
22784 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22786 2014-01-10 18:18:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22788 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22789 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
22790 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
22791 encoder: h264: derive profile and level from active coding tools.
22792 Automatically derive the minimum profile and level to be used for
22793 encoding, based on the activated coding tools. The encoder will
22794 be trying to generate a bitstream that has the best chances to be
22795 decoded on most platforms by default.
22796 Also change the default profile to "constrained-baseline" so that
22797 to ensure maximum compatibility when the stream is decoded.
22798 https://bugzilla.gnome.org/show_bug.cgi?id=719691
22800 2014-01-10 17:02:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22802 * gst-libs/gst/vaapi/gstvaapiencoder.h:
22803 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22804 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
22805 encoder: h264: fix hardware profile lookup.
22806 Fix lookup for a suitable HW profile, as to be used by the underlying
22807 hardware, based on heuristics that lead to characterize the SW profile,
22808 i.e. the one used by the SW level encoding logic.
22809 Also fix constraint_set0_flag (A.2.1) and constraint_set1_flag (A.2.2)
22810 as they should respectively match the baseline and main profile.
22811 https://bugzilla.gnome.org/show_bug.cgi?id=719827
22813 2014-01-10 14:46:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22815 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22816 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
22817 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
22818 encoder: h264: support only the byte-stream format.
22819 The libgstvaapi core encoders are meant to support raw bitstreams only.
22820 Henceforth, we are always producing a stream in "byte-stream" format.
22821 However, the "codec-data" buffer which holds SPS and PPS headers is
22822 always available. The "lengthSizeMinusOne" field is always set to 3
22823 so that in-place "byte-stream" format to "avc" format conversion could
22826 2014-01-10 14:05:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22828 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22829 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
22830 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
22831 * gst/vaapi/gstvaapiencode_h264.c:
22832 encoder: h264: clean-ups.
22833 Various clean-ups to improve consistency and readability: rename some
22834 variables, drop unused macro definitions, drop initialization of vars
22835 that are zero-initialized from the base class, drop un-necessary casts.
22837 2014-01-13 17:11:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22839 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22840 encoder: mpeg2: fix hardware profile lookup.
22841 Fix lookup for a suitable HW profile, as to be used by the underlying
22842 hardware, based on heuristics that lead to characterize the SW profile,
22843 i.e. the one used by the SW level encoding logic.
22845 2014-01-13 16:56:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22847 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22848 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2_priv.h:
22849 encoder: mpeg2: derive profile and level from active coding tools.
22850 Automatically derive the minimum profile and level to be used for
22851 encoding, based on the activated coding tools. Improve lookup for
22852 the best suitable level with the new MPEG-2 helper functions.
22853 Also change the default profile to "simple" so that to ensure maximum
22854 compatibility when the stream is decoded.
22855 https://bugzilla.gnome.org/show_bug.cgi?id=719703
22857 2014-01-13 14:41:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22859 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22860 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
22861 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2_priv.h:
22862 encoder: mpeg2: clean-ups.
22863 Various clean-ups to improve consistency and readability: drop unused
22864 macro definitions, drop initialization of vars that are zero-initialized
22865 from the base class, drop un-necessary casts.
22867 2014-01-13 10:48:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22869 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22870 * gst-libs/gst/vaapi/gstvaapiencoder.h:
22871 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22872 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22873 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
22874 encoder: add tuning options API.
22875 Add encoder "tune" option to override the default behaviour that is to
22876 favor maximum decoder compatibility at the expense of lower compression
22878 Expected tuning options to be developed are:
22879 - "high-compression": improve compression, target best-in-class decoders;
22880 - "low-latency": tune for low-latency decoding;
22881 - "low-power": tune for encoding in low power / resources conditions.
22883 2014-01-12 23:17:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22885 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22886 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22887 encoder: fix bitrate units to match kbps.
22888 Bitrate is expressed in kilobits per second (kbps). So, this exactly
22889 means in multiple of 1000 bits, not 1024 bits.
22890 https://bugzilla.gnome.org/show_bug.cgi?id=722086
22892 2014-01-12 21:57:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22894 * gst-libs/gst/vaapi/gstvaapiencoder.h:
22895 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
22896 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
22897 encoder: clean-ups.
22898 Drop obsolete and unused macros. Add a few doc comments. Slightly
22899 improve indentation of a few leftovers.
22901 2014-01-12 18:52:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22903 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22904 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
22905 * gst-libs/gst/vaapi/gstvaapivalue.c:
22906 * gst-libs/gst/vaapi/gstvaapivalue.h:
22907 encoder: filter out the supported set of rate-control properties.
22908 Only expose the exact static set of supported rate-control properties
22909 to the upper layer. For instance, if the GstVaapiEncoderXXX class does
22910 only support CQP rate control, then only add it the the exposed enum
22912 Add helper macros and functions to build a GType for an enum subset.
22914 2014-01-10 13:23:48 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22916 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22917 * gst-libs/gst/vaapi/gstvaapiencoder.h:
22918 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22919 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
22920 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
22921 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22922 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
22923 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2_priv.h:
22924 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
22925 encoder: add keyframe period API.
22926 Add gst_vaapi_encoder_set_keyframe_period() interface to allow the
22927 user control the maximum distance between two keyframes. This new
22928 property can only be set prior to gst_vaapi_encoder_set_codec_state().
22929 A value of zero for "keyframe-period" gets it re-evaluated to the
22930 actual framerate during encoder reconfiguration.
22932 2014-01-10 12:01:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22934 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22935 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22936 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22937 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
22938 encoder: improve codec reconfiguration.
22939 Improve codec reconfiguration to be performed only through a single
22940 function. That is, remove the _set_context_info() hook as subclass
22941 should not alter the parent GstVaapiContextInfo itself. Besides, the
22942 VA context is constructed only at the final stages of reconfigure().
22944 2014-01-10 11:30:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22946 * gst-libs/gst/vaapi/gstvaapicodedbufferpool.c:
22947 * gst-libs/gst/vaapi/gstvaapicodedbufferpool.h:
22948 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22949 encoder: fix possible memory leak of coded buffer pools.
22950 Fix gst_vaapi_encoder_reconfigure_internal() to re-/allocate the coded
22951 buffer pool only if the coded buffer size actually changed.
22953 2014-01-10 10:54:22 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22955 * gst-libs/gst/vaapi/gstvaapiencoder.c:
22956 * gst-libs/gst/vaapi/gstvaapiencoder.h:
22957 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
22958 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
22959 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
22960 * gst/vaapi/gstvaapiencode.c:
22961 * gst/vaapi/gstvaapiencode.h:
22962 encoder: add video codec-state API.
22963 Add interface to communicate the encoder resolution and related info
22964 like framerate, interlaced vs. progressive, etc. This new interface
22965 supersedes gst_vaapi_encoder_set_format() and doesn't use any GstCaps
22966 but rather use GstVideoCodecState.
22967 Note that gst_vaapi_encoder_set_codec_state() is also a synchronization
22968 point for codec config. This means that the encoder is reconfigured
22969 there to match the latest properties.
22971 2014-01-13 17:18:42 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22973 * gst/vaapi/gstvaapiencode.c:
22974 vaapiencode: don't crash on NULL encoder on _finish().
22975 Don't try to destroy an encoder, in GstVideoEncoder::finish() handler,
22976 if it was not created in the first place. Return "not-negotiated" error
22977 since this means we did not even reach GstVideoEncoder::set_format(),
22978 where the encoder could have been created.
22979 This fixes a crash when the vaapiencode_* plug-in elements get deallocated
22980 and that we failed to negotiate either pad.
22981 https://bugzilla.gnome.org/show_bug.cgi?id=719704
22983 2014-01-09 18:20:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22985 * gst/vaapi/gstvaapiencode.c:
22986 * gst/vaapi/gstvaapiencode.h:
22987 * gst/vaapi/gstvaapiencode_h264.c:
22988 vaapiencode: use more GstVaapiPluginBase facilities.
22989 Avoid duplication of pad references or query functions since they are
22990 provided through the GstVaapiPluginBase object.
22992 2014-01-09 18:10:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
22994 * gst/vaapi/gstvaapiencode.c:
22995 * gst/vaapi/gstvaapiencode.h:
22996 * gst/vaapi/gstvaapiencode_h264.c:
22997 * gst/vaapi/gstvaapiencode_h264.h:
22998 * gst/vaapi/gstvaapiencode_mpeg2.c:
22999 vaapiencode: fix negotiation process of output caps.
23000 The specified caps in gst_video_encoder_set_output_state() function
23001 arguments should not contain any resolution, pixel-aspect-ratio,
23002 framerate, codec-data et al. Those rather should be set through the
23003 returned GstVideoCodecState. This means that output caps creation
23004 could be delayed until before gst_video_encoder_finish_frame() is
23006 This greatly simplifies the GstVideoEncoder::set_format() callback
23009 2014-01-08 18:56:23 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23011 * gst/vaapi/gstvaapiencode.c:
23012 vaapiencode: make GstVaapiEncode an abstract type.
23013 Make base GstVaapiEncode class an abstract type so that we cannot
23014 create an instance from it without going through any of the codec
23015 specific derived class.
23017 2014-01-09 10:09:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23019 * gst/vaapi/gstvaapiencode.c:
23020 * gst/vaapi/gstvaapiencode.h:
23021 * gst/vaapi/gstvaapiencode_h264.c:
23022 * gst/vaapi/gstvaapiencode_mpeg2.c:
23023 vaapiencode: rename a few member functions.
23024 Rename a few member functions to make them more consistent:
23025 - alloc_encoder(): now reduced to allocate the encoder object only;
23026 - alloc_buffer(): allocate buffer from srcpad, and copy bitstream.
23028 2014-01-08 18:36:46 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23030 * gst/vaapi/gstvaapiencode.c:
23031 * gst/vaapi/gstvaapiencode.h:
23032 * gst/vaapi/gstvaapiencode_h264.c:
23033 * gst/vaapi/gstvaapiencode_h264.h:
23034 * gst/vaapi/gstvaapiencode_mpeg2.c:
23035 vaapiencode: update for new properties API.
23036 Update MPEG-2 and H.264 encode elements to cope with the new core
23037 libgstvaapi properties API. i.e. all configurable properties are now
23038 directly handled at the GstVaapiEncoder level.
23039 Besides, this also makes sure to not use or modify the GstVaapiEncoder
23040 private definitions directly. Private data need to remain private.
23041 https://bugzilla.gnome.org/show_bug.cgi?id=719529
23043 2014-01-06 17:46:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23045 * gst-libs/gst/vaapi/gstvaapiencoder.c:
23046 * gst-libs/gst/vaapi/gstvaapiencoder.h:
23047 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
23048 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
23049 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23050 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
23051 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
23052 encoder: add properties API.
23053 Add interface to communicate configurable properties to the encoder.
23054 This covers both the common ones (rate-control, bitrate), and the
23055 codec specific properties.
23056 https://bugzilla.gnome.org/show_bug.cgi?id=719529
23058 2014-01-06 18:01:33 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23060 * gst-libs/gst/vaapi/gstvaapiencoder.c:
23061 * gst-libs/gst/vaapi/gstvaapiencoder.h:
23062 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
23063 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
23064 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23065 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2_priv.h:
23066 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
23067 * gst/vaapi/gstvaapiencode.c:
23068 * gst/vaapi/gstvaapiencode_h264.c:
23069 * gst/vaapi/gstvaapiencode_mpeg2.c:
23070 encoder: add bitrate API.
23071 Add gst_vaapi_encoder_set_bitrate() interface to allow the user control
23072 the bitrate for encoding. Currently, changing this parameter is only
23073 valid before the first frame is encoded. Should the value be modified
23074 afterwards, then GST_VAAPI_ENCODER_STATUS_ERROR_OPERATION_FAILED is
23076 https://bugzilla.gnome.org/show_bug.cgi?id=719529
23078 2014-01-06 15:10:36 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23080 * gst-libs/gst/vaapi/gstvaapiencoder.c:
23081 * gst-libs/gst/vaapi/gstvaapiencoder.h:
23082 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
23083 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23084 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2_priv.h:
23085 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
23086 * gst-libs/gst/vaapi/gstvaapitypes.h:
23087 * gst/vaapi/gstvaapiencode.c:
23088 * gst/vaapi/gstvaapiencode.h:
23089 * gst/vaapi/gstvaapiencode_h264.c:
23090 * gst/vaapi/gstvaapiencode_mpeg2.c:
23091 encoder: add rate control API.
23092 Add gst_vaapi_encoder_set_rate_control() interface to request a new
23093 rate control mode for encoding. Changing the rate control mode is
23094 only valid prior to encoding the very first frame. Afterwards, an
23095 error ("operation-failed") is issued.
23096 https://bugzilla.gnome.org/show_bug.cgi?id=719529
23098 2014-01-03 16:57:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23100 * gst/vaapi/gstvaapiencode.c:
23101 * gst/vaapi/gstvaapiencode_h264.c:
23102 * gst/vaapi/gstvaapiencode_mpeg2.c:
23103 vaapiencode: fix indentation.
23105 2014-01-03 16:57:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23107 * gst-libs/gst/vaapi/gstvaapiencoder.c:
23108 * gst-libs/gst/vaapi/gstvaapiencoder.h:
23109 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
23110 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
23111 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
23112 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23113 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
23114 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
23115 * gst/vaapi/gstvaapiencode.h:
23116 encoder: fix indentation.
23118 2014-01-13 16:20:06 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23120 * gst-libs/gst/vaapi/Makefile.am:
23121 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2.c:
23122 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2.h:
23123 * gst-libs/gst/vaapi/gstvaapiutils_mpeg2_priv.h:
23124 utils: add new MPEG-2 helper functions.
23125 Add various helper functions to convert profile, level, chroma formats
23126 from gstreamer-vaapi world and the MPEG-2 specification world.
23128 2014-01-10 19:49:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23130 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
23131 utils: h264: don't use fatal asserts.
23132 Replace g_assert() with a g_debug() so that to not make the program
23133 abort when an unsupported value is supplied.
23135 2014-01-10 19:37:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23137 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
23138 * gst-libs/gst/vaapi/gstvaapiutils_h264.h:
23139 utils: h264: add helpers for profile and level string mappings.
23140 Add profile and level helper functions to convert to/from strings.
23142 2014-01-10 18:27:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23144 * gst-libs/gst/vaapi/Makefile.am:
23145 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
23146 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
23147 * gst-libs/gst/vaapi/gstvaapiutils_h264.h:
23148 * gst-libs/gst/vaapi/gstvaapiutils_h264_priv.h:
23149 utils: h264: expose levels in public header.
23150 Instal <gst/vaapi/gstvaapiutils_h264.h> header but only expose the
23151 H.264 levels in there. The additional helper functions are meant
23152 to be private for now.
23154 2014-01-09 09:27:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23156 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
23157 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
23158 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
23159 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
23160 codec: add helper macros to maintain object refcount.
23161 Add gst_vaapi_mini_object_{ref,unref,replace}() helper macros so that
23162 to avoid explicit casts to GstVaapiMiniObject in all caller sites.
23164 2014-01-09 09:30:49 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23166 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
23167 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
23168 codec: re-indent decoder objects.
23170 2014-01-09 09:10:21 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23172 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
23173 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
23174 codec: re-indent base codec objects.
23176 2014-01-03 12:49:05 +0000 Matthieu Bouron <matthieu.bouron@collabora.com>
23178 * gst/vaapi/gstvaapipluginbase.c:
23179 plugins: do not free debug category in finalize method.
23180 Fixes a crash when multiple vaapidecode elements are finalized since
23181 the debug category is created once in the class init method.
23182 This is a regression from git commit 7e58d60.
23183 https://bugzilla.gnome.org/show_bug.cgi?id=721390
23184 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23186 2014-01-02 11:35:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23188 * gst-libs/gst/vaapi/glibcompat.h:
23189 * tests/simple-decoder.c:
23190 tests: simple-decoder: don't use deprecated g_thread_create().
23191 Use g_thread_try_new() instead of the deprecated g_thread_create()
23192 function. Provide compatibility glue for any GLib version < 2.31.2.
23194 2014-01-02 11:17:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23196 * gst-libs/gst/vaapi/gstvaapidecoder.c:
23197 * gst-libs/gst/vaapi/gstvaapipixmap_x11.c:
23198 * gst-libs/gst/vaapi/gstvaapisurface.c:
23199 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
23200 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
23201 * gst/vaapi/gstvaapiencode.c:
23202 Fix printf()-like formats.
23203 Fix formts for various GST_DEBUG et al. invocations. More precisely,
23204 make size_t arguments use the %zu format specifier accordingly; force
23205 XID formats to be a 32-bit unsigned integer; and fix the format used
23206 for gst_vaapi_create_surface_with_format() error cases since we have
23207 been using strings nowadays.
23209 2013-12-21 07:38:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23211 * gst-libs/gst/vaapi/video-format.c:
23212 * gst-libs/gst/vaapi/video-format.h:
23213 utils: format: drop unused helper functions.
23214 The following helper functions are no longer used, thus are removed:
23215 - gst_vaapi_video_format_from_structure()
23216 - gst_vaapi_video_format_from_caps()
23217 - gst_vaapi_video_format_to_caps()
23219 2013-12-21 07:29:50 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23221 * gst-libs/gst/vaapi/video-format.c:
23222 * gst-libs/gst/vaapi/video-format.h:
23223 utils: re-indent GstVideoFormat related helpers.
23225 2013-12-21 08:27:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23227 * gst/vaapi/gstvaapidownload.c:
23228 download: use GstVideoInfo facilities to build output caps.
23229 Use standard GstVideoInfo related functions to build the output caps,
23230 thus directly preserving additional fields as needed, instead of
23231 manually copying them over through gst_vaapi_append_surface_caps().
23232 Also ensure that the input caps are fixated first.
23234 2013-12-21 10:41:22 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23236 * gst/vaapi/gstvaapidownload.c:
23237 * gst/vaapi/gstvaapipluginutil.c:
23238 * gst/vaapi/gstvaapipluginutil.h:
23239 * gst/vaapi/gstvaapiuploader.c:
23240 plugins: factor out construction of template caps.
23241 Add new helper functions to build video template caps.
23242 - gst_vaapi_video_format_new_template_caps():
23243 create GstCaps with size, frame rate and PAR to full range
23244 - gst_vaapi_video_format_new_template_caps_from_list():
23245 try to create a "simplified" list from the supplied formats
23247 2013-12-21 06:41:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23249 * gst/vaapi/gstvaapipluginutil.c:
23250 * gst/vaapi/gstvaapipluginutil.h:
23251 * gst/vaapi/gstvaapipostproc.c:
23252 plugins: factor out construction of GValue from GstVideoFormat.
23253 Add new helper functions to build GValues from GstVideoFormat:
23254 - gst_vaapi_value_set_format():
23255 build a GValue from the supplied video format
23256 - gst_vaapi_value_set_format_list():
23257 build a GValue list from the supplied array of video formats
23259 2013-12-21 06:22:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23261 * gst/vaapi/gstvaapipluginutil.c:
23262 * gst/vaapi/gstvaapipluginutil.h:
23263 * gst/vaapi/gstvaapivideocontext.c:
23264 * gst/vaapi/gstvaapivideocontext.h:
23265 plugins: re-indent common and video context creation utils.
23267 2013-12-20 15:31:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23269 * gst-libs/gst/vaapi/gstvaapidisplay.c:
23270 * gst-libs/gst/vaapi/gstvaapidisplay.h:
23271 * gst/vaapi/gstvaapidecode.c:
23272 * tests/test-display.c:
23273 display: don't use GstCaps for decode or encode profiles list.
23274 Replace gst_vaapi_display_get_{decode,encode}_caps() APIs with more
23275 more convenient APIs that return an array of GstVaapiProfile instead
23276 of GstCaps: gst_vaapi_display_get_{decode,encode}_profiles().
23278 2013-12-20 15:15:05 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23280 * gst-libs/gst/vaapi/gstvaapidisplay.c:
23281 * gst-libs/gst/vaapi/gstvaapidisplay.h:
23282 * gst/vaapi/gstvaapidownload.c:
23283 * gst/vaapi/gstvaapiuploader.c:
23284 * tests/test-display.c:
23285 display: don't use GstCaps for image or subpicture formats list.
23286 Replace gst_vaapi_display_get_{image,subpicture}_caps() APIs, that
23287 returned GstCaps, with more convenient APIs that return an array of
23288 GstVideoFormat: gst_vaapi_display_get_{image,subpicture}_formats().
23290 2013-12-20 14:01:45 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23292 * gst-libs/gst/vaapi/gstvaapidisplay.c:
23293 * gst-libs/gst/vaapi/gstvaapidisplay.h:
23294 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
23295 display: allocate queried resources on-demand.
23296 Allocate the set of decoders or encoders on-demand, when they are
23297 queried. Likewise for VA display attributes, image and subpicture
23300 2013-12-20 13:27:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23302 * gst-libs/gst/vaapi/gstvaapidisplay.c:
23303 * gst-libs/gst/vaapi/gstvaapidisplay.h:
23304 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
23305 * gst-libs/gst/vaapi/gstvaapidisplay_drm.h:
23306 * gst-libs/gst/vaapi/gstvaapidisplay_drm_priv.h:
23307 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
23308 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
23309 * gst-libs/gst/vaapi/gstvaapidisplay_glx_priv.h:
23310 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
23311 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
23312 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.h:
23313 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
23314 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
23315 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
23316 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
23317 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
23318 * gst-libs/gst/vaapi/gstvaapidisplaycache.h:
23319 display: re-indent all GstVaapiDisplay related source code.
23321 2013-12-20 16:04:19 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23323 * gst-libs/gst/vaapi/gstvaapiprofile.c:
23324 * gst-libs/gst/vaapi/gstvaapiprofile.h:
23325 utils: add helper functions to get codec or profile name.
23327 2013-12-20 17:08:23 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23329 * gst/vaapi/gstvaapipostproc.c:
23330 * gst/vaapi/gstvaapipostproc.h:
23331 * gst/vaapi/gstvaapiuploader.c:
23332 plugins: fix permissions for certain files.
23333 Drop the execute bit for gstvaapiuploader.c and gstvaapipostproc.[ch]
23336 2013-12-12 17:01:29 +0000 Matthieu Bouron <matthieu.bouron@collabora.com>
23338 * gst/vaapi/gstvaapivideometa_texture.c:
23339 plugins: implement GLTextureUploadMeta user data copy.
23340 Makes the copies of a buffer reference their own GLTextureUploadMeta
23341 user data and prevent the original buffer accessing already freed
23342 memory if its copies has been released and freed.
23343 https://bugzilla.gnome.org/show_bug.cgi?id=720336
23344 [Propagate the original meta texture to the copy too]
23345 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23347 2013-12-17 18:52:23 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23349 * gst/vaapi/gstvaapiencode.c:
23350 * gst/vaapi/gstvaapiencode.h:
23351 * gst/vaapi/gstvaapipluginbase.c:
23352 * gst/vaapi/gstvaapipluginbase.h:
23353 * gst/vaapi/gstvaapipostproc.c:
23354 * gst/vaapi/gstvaapipostproc.h:
23355 * gst/vaapi/gstvaapisink.c:
23356 * gst/vaapi/gstvaapisink.h:
23357 plugins: factor out support for raw YUV buffers on sink pads.
23358 Factor out propose_allocation() hooks, creation of video buffer pool
23359 for the sink pad, conversion from raw YUV buffers to VA surface backed
23360 buffers. Update vaapidecode, vaapiencode and vaapipostproc to cope
23361 with the new GstVaapiPluginBase abilities.
23363 2013-12-17 18:46:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23365 * gst-libs/gst/vaapi/gstcompat.h:
23366 * gst/vaapi/gstvaapidecode.c:
23367 * gst/vaapi/gstvaapidecode.h:
23368 * gst/vaapi/gstvaapidownload.c:
23369 * gst/vaapi/gstvaapipluginbase.c:
23370 * gst/vaapi/gstvaapipluginbase.h:
23371 * gst/vaapi/gstvaapipostproc.c:
23372 * gst/vaapi/gstvaapipostproc.h:
23373 * gst/vaapi/gstvaapisink.c:
23374 * gst/vaapi/gstvaapiupload.c:
23375 plugins: factor out pad caps.
23377 2013-12-13 16:03:08 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23379 * gst/vaapi/gstvaapidecode.c:
23380 * gst/vaapi/gstvaapidownload.c:
23381 * gst/vaapi/gstvaapiencode.c:
23382 * gst/vaapi/gstvaapipluginbase.c:
23383 * gst/vaapi/gstvaapipostproc.c:
23384 * gst/vaapi/gstvaapisink.c:
23385 * gst/vaapi/gstvaapiupload.c:
23386 plugins: factor out video context sharing code.
23388 2013-12-13 13:24:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23390 * gst/vaapi/gstvaapidecode.c:
23391 * gst/vaapi/gstvaapidownload.c:
23392 * gst/vaapi/gstvaapiencode.c:
23393 * gst/vaapi/gstvaapipluginbase.c:
23394 * gst/vaapi/gstvaapipluginbase.h:
23395 * gst/vaapi/gstvaapipostproc.c:
23396 * gst/vaapi/gstvaapisink.c:
23397 * gst/vaapi/gstvaapiupload.c:
23398 plugins: factor out GstImplementsInterface.
23400 2013-12-13 12:00:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23402 * gst-libs/gst/vaapi/gstvaapidisplay.c:
23403 * gst-libs/gst/vaapi/gstvaapidisplay.h:
23404 * gst/vaapi/gstvaapipluginbase.c:
23405 * gst/vaapi/gstvaapipluginutil.c:
23406 plugins: check type of display obtained from neighbours.
23407 Fix display creation code to check that any display obtained from a
23408 neighbour actually has the type we expect. Note: if display type is
23409 set to "any", we can then accept any VA display type.
23411 2013-12-13 11:52:47 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23413 * gst/vaapi/gstvaapidecode.c:
23414 * gst/vaapi/gstvaapidownload.c:
23415 * gst/vaapi/gstvaapiencode.c:
23416 * gst/vaapi/gstvaapipluginbase.c:
23417 * gst/vaapi/gstvaapipluginbase.h:
23418 * gst/vaapi/gstvaapipluginutil.c:
23419 * gst/vaapi/gstvaapipluginutil.h:
23420 * gst/vaapi/gstvaapipostproc.c:
23421 * gst/vaapi/gstvaapisink.c:
23422 * gst/vaapi/gstvaapisink.h:
23423 * gst/vaapi/gstvaapiupload.c:
23424 plugins: factor out display creation process.
23425 Move common VA display creation code to GstVaapiPluginBase, with the
23426 default display type remaining "any". Also add a "display-changed"
23427 hook so that subclasses could perform additional tasks when/if the
23428 VA display changed, due to a new display type request for instance.
23429 All plug-ins are updated to cope with the new internal APIs.
23431 2013-12-13 10:24:26 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23433 * gst/vaapi/Makefile.am:
23434 * gst/vaapi/gstvaapidecode.c:
23435 * gst/vaapi/gstvaapidecode.h:
23436 * gst/vaapi/gstvaapidownload.c:
23437 * gst/vaapi/gstvaapidownload.h:
23438 * gst/vaapi/gstvaapiencode.c:
23439 * gst/vaapi/gstvaapiencode.h:
23440 * gst/vaapi/gstvaapipluginbase.c:
23441 * gst/vaapi/gstvaapipluginbase.h:
23442 * gst/vaapi/gstvaapipostproc.c:
23443 * gst/vaapi/gstvaapipostproc.h:
23444 * gst/vaapi/gstvaapisink.c:
23445 * gst/vaapi/gstvaapisink.h:
23446 * gst/vaapi/gstvaapiupload.c:
23447 * gst/vaapi/gstvaapiupload.h:
23448 plugins: add new base object, store display in there.
23449 Introduce a new GstVaapiPluginBase object that will contain all common
23450 data structures and perform all common tasks. First step is to have a
23451 single place to hold VA displays.
23452 While we are at it, also make sure to store and subsequently release
23453 the appropriate debug category for the subclasses.
23455 2013-12-11 14:04:27 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23457 * gst-libs/gst/vaapi/gstvaapiobject.h:
23458 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
23459 * gst/vaapi/gstvaapivideometa_texture.c:
23460 * gst/vaapi/gstvaapivideometa_texture.h:
23461 plugins: fix GLTextureUploadMeta to work with different texture ids.
23462 The GLTextureUploadMeta implementation assumed that for each upload()
23463 sequence, the supplied texture id is always the same as the one that
23464 was previously cached into the underlying GstVaapiTexture. Cope with
23465 any texture id change the expense to recreate the underlying VA/GLX
23467 https://bugzilla.gnome.org/show_bug.cgi?id=719643
23469 2013-12-11 13:25:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23471 * gst/vaapi/gstvaapidecode.c:
23472 * gst/vaapi/gstvaapivideobufferpool.c:
23473 * gst/vaapi/gstvaapivideometa_texture.c:
23474 plugins: allow builds without GLX enabled for GStreamer 1.2.
23475 Don't try to build GLTextureUploadMeta related code if GLX is not
23476 enabled during GStreamer >= 1.2 builds.
23478 2013-11-20 17:20:07 +0000 Matthieu Bouron <matthieu.bouron@collabora.com>
23480 * gst/vaapi/gstvaapidecode.c:
23481 * gst/vaapi/gstvaapivideobufferpool.c:
23482 * gst/vaapi/gstvaapivideobufferpool.h:
23483 * gst/vaapi/gstvaapivideometa_texture.c:
23484 * gst/vaapi/gstvaapivideometa_texture.h:
23485 plugins: request GLTextureUpload meta on buffers in the buffer pool.
23486 Requesting the GLTextureUpload meta on buffers in the bufferpool
23487 prevents such metas from being de-allocated when buffers are released
23489 This is particulary useful in terms of performance when using the
23490 GLTextureUploadMeta API since the GstVaapiTexture associated with
23491 the target texture is stored in the meta.
23492 https://bugzilla.gnome.org/show_bug.cgi?id=712558
23493 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23495 2013-12-11 10:51:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23497 * gst/vaapi/gstvaapivideometa_texture.c:
23498 plugins: robustify GstVideoGLTextureUploadMeta implementation.
23499 Make GstVideoGLTextureUploadMeta::upload() implementation more robust
23500 when the GstVaapiTexture associated with the supplied texture id could
23503 2013-12-10 16:14:27 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23505 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
23506 h264: improve robustness when packets are missing.
23507 Improve robustness when some expected packets where not received yet
23508 or that were not correctly decoded. For example, don't try to decode
23509 a picture if there was no valid frame headers parsed so far.
23510 https://bugs.freedesktop.org/show_bug.cgi?id=57902
23512 2013-12-10 14:20:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23514 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
23515 decoder: h264: fix decoding of BA3_SVA_C.264.
23516 Conformance test Base_Ext_Main_profiles/BA3_SVA_C.264 complys with
23517 extended profile specifications. However, the SPS header has the
23518 constraint_set1_flag syntax element set to 1. This means that, if
23519 a Main profile compliant decoder is available, then it should be
23520 able to decode this stream.
23521 This changes makes it possible to fall-back from Extended profile
23522 to Main profile if constraint_set1_flag is set to 1.
23523 https://bugzilla.gnome.org/show_bug.cgi?id=720190
23525 2013-12-10 11:13:01 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23527 * gst-libs/gst/vaapi/gstvaapiprofile.h:
23528 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
23529 utils: h264: add more profiles.
23530 Add extended profile (A.2.3), high 4:2:2 profile (A.2.6), high 4:2:2
23531 profiles (A.2.7, A.2.10), scalable profiles (G.10.1.1, G.10.1.2) and
23532 multiview profiles (H.10.1.1, H.10.1.2).
23533 Document "Constrained Baseline" and "High 10" profiles.
23535 2013-12-10 15:21:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23537 * ext/codecparsers:
23538 codecparsers: update to gst-vaapi-branch commit e7d0e18.
23539 e7d0e18 h264: complete set of NAL unit types
23541 2013-12-06 15:08:26 +0800 Wind Yuan <feng.yuan@intel.com>
23543 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
23544 * gst-libs/gst/vaapi/gstvaapiprofile.c:
23545 decoder: h264: add support for constrained baseline profile.
23546 Recognize streams marked as conforming to the "Constrained Baseline
23547 Profile". If VA driver supports that as is, fine. Otherwise, fallback
23548 to baseline, main or high profile.
23549 Constrained Baseline Profile conveys coding tools that are common
23550 to baseline profile and main profile.
23551 https://bugzilla.gnome.org/show_bug.cgi?id=719947
23552 [Added fallbacks to main and high profiles]
23553 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23555 2013-12-09 12:46:45 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23557 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
23558 decoder: h264: fix decoding of scaling lists.
23559 The GStreamer codecparser layer now parses the scaling lists in zigzag
23560 scan order, as expected, so that to match the original bitstream layout
23561 and specification. However, further convert the scaling lists into
23562 raster scan order to fit the existing practice in most VA drivers.
23563 https://bugzilla.gnome.org/show_bug.cgi?id=706406
23565 2013-12-09 12:07:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23567 * ext/codecparsers:
23568 codecparsers: update to gst-vaapi-branch commit 177c73b.
23569 a7e3255 add H.265 (HEVC) bitstream parser
23570 177c73b h264: fix picture level scaling lists derivation (rule B)
23571 14733f1 h264: fix parsing of VCL HRD parameters
23572 59a0b47 h264: store quantization matrices in zig-zag order
23573 ffb6e26 h264: add helpers to convert quantization matrices
23574 c78a504 mpeg2: also initialize debug category in parse_sequence_header()
23575 719d1b0 mpeg2: turn internal consistency check into a g_assert()
23576 5241d8e all: remove some unused functions
23577 18eb312 all: fix for GST_DISABLE_GST_DEBUG
23578 963c04a all: make warnings more meaningful
23580 2013-12-06 19:05:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23582 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
23583 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
23584 * gst-libs/gst/vaapi/gstvaapiutils_h264.h:
23585 utils: add helpers for H.264 levels.
23586 - gst_vaapi_utils_h264_get_level():
23587 Returns GstVaapiLevelH264 from H.264 level_idc value
23588 - gst_vaapi_utils_h264_get_level_idc():
23589 Returns H.264 level_idc value from GstVaapiLevelH264
23590 - gst_vaapi_utils_h264_get_level_limits():
23591 Returns level limits as specified in Table A-1 of the H.264 standard
23592 - gst_vaapi_utils_h264_get_level_limits_table():
23593 Returns the Table A-1 specification
23595 2013-12-06 17:34:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23597 * gst-libs/gst/vaapi/gstvaapiprofile.c:
23598 * gst-libs/gst/vaapi/gstvaapiprofile.h:
23599 * gst-libs/gst/vaapi/gstvaapiutils_h264.c:
23600 utils: add new H.264 profiles.
23601 Add "Constrained Baseline Profile" and "High 10 Profile" definitions
23602 and helper functiions.
23604 2013-12-06 17:21:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23606 utils: add new H.264 helper functions.
23608 - gst_vaapi_utils_h264_get_profile():
23609 Returns GstVaapiProfile from H.264 profile_idc value
23610 - gst_vaapi_utils_h264_get_profile_idc():
23611 Returns H.264 profile_idc value from GstVaapiProfile
23613 - gst_vaapi_utils_h264_get_chroma_type():
23614 Returns GstVaapiChromaType from H.264 chroma_format_idc value
23615 - gst_vaapi_utils_h264_get_chroma_format_idc():
23616 Returns H.264 chroma_format_idc value from GstVaapiChromaType
23618 2013-12-03 11:05:17 +0000 Matthieu Bouron <matthieu.bouron@collabora.com>
23620 * gst-libs/gst/base/Makefile.am:
23621 * gst-libs/gst/vaapi/Makefile.am:
23622 Fix missing files in distribution tarball.
23623 https://bugzilla.gnome.org/show_bug.cgi?id=719776
23624 [Additional fixes and clean-ups]
23625 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23627 2013-12-05 18:13:54 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23629 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
23630 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23631 encoder: fix computation of max coded buffer size (again).
23632 The previous fix was only valid to express the maximum size of the
23633 macroblock layer, i.e. without any headers. Now, also account for
23634 the slice headers and top picture header, but also any other header
23635 we might stuff into the VA coded buffer, e.g. sequence headers.
23637 2013-12-04 19:10:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23639 * gst-libs/gst/vaapi/gstvaapiencoder.c:
23640 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
23641 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23642 encoder: fix computation of max coded buffer size.
23643 Fix coded buffer size for each codec. A generic issue was that the
23644 number of macroblocks was incorrectly computed. The second issue was
23645 specific to MPEG-2 were the max number of bits per macroblock, and
23646 as defined by the standard, was incorrectly mapped to the (lower)
23647 H.264 requirement. i.e. 4608 bits vs. 3200 bits limit.
23649 2013-12-04 18:48:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23651 * gst-libs/gst/vaapi/gstvaapiencoder.c:
23652 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
23653 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23654 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
23655 encoder: simplify VA context initialization process.
23656 Change get_context_info() into a set_context_info() function that
23657 initializes common defaults into the base class, thus allowing the
23658 subclasses to specialize the context info further on.
23659 The set_context_info() hook is also the location where additional
23660 context specific data could be initialized. At this point, we are
23661 guaranteed to have valid video resolution size and framerate. i.e.
23662 gst_vaapi_encoder_set_format() was called beforehand.
23664 2013-11-26 14:38:23 +0800 Wind Yuan <feng.yuan@intel.com>
23666 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23667 encoder: fix mpeg2 compilation error.
23668 https://bugzilla.gnome.org/show_bug.cgi?id=719746
23669 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23671 2013-12-04 17:55:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23673 * gst-libs/gst/vaapi/gstvaapiencoder.c:
23674 * gst-libs/gst/vaapi/gstvaapiencoder.h:
23675 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
23676 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23677 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
23678 * gst/vaapi/gstvaapiencode.c:
23679 encoder: clean-ups and document public APIs.
23680 Clean public APIs up so that to better align with the decoder APIs.
23681 Most importantly, gst_vaapi_encoder_get_buffer() is changed to only
23682 return the VA coded buffer proxy. Also provide useful documentation
23683 for the public APIs.
23685 2013-12-04 17:05:17 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23687 * gst-libs/gst/vaapi/gstvaapiencoder.c:
23688 * gst-libs/gst/vaapi/gstvaapiencoder.h:
23689 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
23690 encoder: avoid extra allocations of GstVaapiEncoderSyncPic objects.
23691 Kill GstVaapiEncoderSyncPic objects that are internally and temporarily
23692 allocated. Rather, associate a GstVaapiEncPicture to a coded buffer
23693 through GstVaapiCodedBufferProxy user-data facility.
23694 Besides, use a GAsyncQueue to maintain a thread-safe queue object of
23696 Partial fix for the following report:
23697 https://bugzilla.gnome.org/show_bug.cgi?id=719530
23699 2013-12-03 17:04:43 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23701 * gst-libs/gst/vaapi/gstvaapiencoder.c:
23702 * gst-libs/gst/vaapi/gstvaapiencoder.h:
23703 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
23704 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23705 * gst/vaapi/gstvaapiencode.c:
23706 encoder: refactor status codes.
23707 Drop obsolete or unused status codes. Align some status codes with the
23708 decoder counterparts.
23710 2013-12-04 11:54:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23712 * gst-libs/gst/vaapi/gstvaapiencoder.c:
23713 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
23714 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
23715 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23716 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2_priv.h:
23717 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
23718 encoder: fix subclassing process.
23719 Fix the GstVaapiEncoderClass parent class type. Make sure to validate
23720 subclass hooks as early as possible, i.e. in gst_vaapi_encoder_init(),
23721 thus avoiding useless run-time checks. Also simplify the subclass
23722 initialization process to be less error prone.
23724 2013-12-03 16:11:46 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23726 encoder: rework GstVaapiCodedBuffer and related proxy.
23727 Refactor the GstVaapiCodedBuffer APIs so that to more clearly separate
23728 public and private interfaces. Besides, the map/unmap APIs should not
23729 be exposed as is but appropriate accessors should be provided instead.
23730 * GstVaapiCodedBuffer: VA coded buffer abstraction
23731 - gst_vaapi_coded_buffer_get_size(): get coded buffer size.
23732 - gst_vaapi_coded_buffer_copy_into(): copy coded buffer into GstBuffer
23733 * GstVaapiCodedBufferPool: pool of VA coded buffer objects
23734 - gst_vaapi_coded_buffer_pool_new(): create a pool of coded buffers of
23735 the specified max size, and bound to the supplied encoder
23736 * GstVaapiCodedBufferProxy: pool-allocated VA coded buffer object proxy
23737 - gst_vaapi_coded_buffer_proxy_new_from_pool(): create coded buf from pool
23738 - gst_vaapi_coded_buffer_proxy_get_buffer(): get underlying coded buffer
23739 - gst_vaapi_coded_buffer_proxy_get_buffer_size(): get coded buffer size
23740 Rationale: more optimized transfer functions might be provided in the
23741 future, thus rendering the map/unmap mechanism obsolete or sub-optimal.
23742 https://bugzilla.gnome.org/show_bug.cgi?id=719775
23744 2013-11-29 14:02:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23746 * gst/vaapi/gstvaapidecode.c:
23747 * gst/vaapi/gstvaapiencode.c:
23748 * gst/vaapi/gstvaapipostproc.c:
23749 * gst/vaapi/gstvaapisink.c:
23750 plugins: fix reference leaks of VA display objects.
23751 Fix GstElement::set_context() implementation for all plug-in elements
23752 to avoid leaking an extra reference to the VA display, thus preventing
23753 correct cleanup of VA resources in GStreamer 1.2 builds.
23755 2013-11-29 13:56:12 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23757 * gst/vaapi/gstvaapipluginutil.c:
23758 * gst/vaapi/gstvaapivideocontext.c:
23759 plugins: simplify gst_vaapi_ensure_display().
23760 Return earlier if the creation of a VA display failed. Likewise, simplify
23761 gst_vaapi_video_context_propagate() now that we are guaranteed to have a
23764 2013-11-28 19:08:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23766 * gst/vaapi/gstvaapivideomemory.c:
23767 plugins: fix memory leaks through GstVideoMeta maps.
23768 When GstVideoMeta maps were used, the supporting functions incorrectly
23769 used gst_buffer_get_memory() instead of gst_buffer_peek_memory(), thus
23770 always increasing the associated GstMemory reference count and giving
23771 zero chance to actually release that, and subsequently the VA display.
23773 2013-11-28 14:15:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23775 * gst-libs/gst/vaapi/gstvaapifilter.c:
23776 * gst/vaapi/gstvaapiencode_h264.c:
23777 * gst/vaapi/gstvaapiencode_mpeg2.c:
23778 * gst/vaapi/gstvaapipostproc.c:
23779 * gst/vaapi/gstvaapisink.c:
23780 * gst/vaapi/gstvaapiuploader.c:
23781 plugins: use G_PARAM_STATIC_STRINGS.
23782 This avoids a few string copies during initialization.
23784 2013-11-28 17:28:11 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23786 * gst/vaapi/gstvaapivideometa.c:
23787 plugins: simplify VA video meta to only reference surface proxies.
23788 Simplify GstVaapiVideoMeta to only hold a surface proxy, which is
23789 now allocated from a surface pool. This also means that the local
23790 reference to the VA surface is also gone, as it could be extracted
23791 from the associated surface proxy.
23793 2013-11-28 16:51:37 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23795 * gst/vaapi/gstvaapivideobuffer.c:
23796 * gst/vaapi/gstvaapivideobuffer.h:
23797 * gst/vaapi/gstvaapivideometa.c:
23798 * gst/vaapi/gstvaapivideometa.h:
23799 plugins: drop obsolete functions.
23800 Drop the following functions that are not longer used:
23801 - gst_vaapi_video_buffer_new_with_surface()
23802 - gst_vaapi_video_meta_new_with_surface()
23803 - gst_vaapi_video_meta_set_surface()
23804 - gst_vaapi_video_meta_set_surface_from_pool()
23806 2013-11-28 16:37:31 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23808 * gst/vaapi/gstvaapivideometa.c:
23809 plugins: allow VA video meta to be allocated from surface proxy pools.
23810 Fix gst_vaapi_video_meta_new_from_pool() to allocate VA surface proxies
23811 from surface pools instead of plain VA surfaces. This is to simplify
23812 allocations now that surface proxies are created from a surface pool.
23814 2013-11-28 17:25:05 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23816 * docs/reference/libs/libs-sections.txt:
23817 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
23818 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
23819 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
23820 surfaceproxy: add copy function.
23821 Add gst_vaapi_surface_proxy_copy() function that creates a new surface
23822 proxy with the same information from the parent proxy, except that the
23823 user-defined destroy notify function is not copied over.
23824 The underlying VA surface is pushed back to the video pool only when
23825 the last reference to the parent surface proxy is released.
23827 2013-11-28 15:56:53 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23829 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
23830 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
23831 * gst/vaapi/gstvaapiencode.c:
23832 vaapiencode: optimize _handle_frame() to avoid extra allocation.
23833 Optimize gst_vaapiencode_handle_frame() to avoid extra memory allocation,
23834 and in particular the GstVaapiEncObjUserData object. i.e. directly use
23835 the VA surface proxy from the source buffer. This also makes the user
23836 data attached to the GstVideoCodecFrame more consistent between both
23837 the decoder and encoder plug-in elements.
23839 2013-11-28 15:14:43 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23841 * gst/vaapi/gstvaapiencode.c:
23842 vaapiencode: fix memory leaks in _push_frame() on error.
23843 Simplify gst_vaapiencode_push_frame(), while also removing the call
23844 to gst_video_encoder_negotiate() since this is implicit in _finish()
23845 if caps changed. Also fixed memory leaks that occured on error.
23847 2013-11-28 13:57:54 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23849 * gst/vaapi/gstvaapiencode.c:
23850 * gst/vaapi/gstvaapiencode.h:
23851 * gst/vaapi/gstvaapiencode_h264.c:
23852 * gst/vaapi/gstvaapiencode_h264.h:
23853 * gst/vaapi/gstvaapiencode_mpeg2.c:
23854 vaapiencode: additional clean-ups.
23855 Constify pointers wherever possible. Drop unused variables, and use
23856 consistent variable names. Fix gst_vaapiencode_h264_allocate_buffer()
23857 to correctly report errors, especially when in-place conversion from
23858 bytestream to avcC format failed.
23860 2013-11-28 13:26:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23862 * gst/vaapi/gstvaapiencode.c:
23863 * gst/vaapi/gstvaapiencode.h:
23864 * gst/vaapi/gstvaapiencode_h264.c:
23865 * gst/vaapi/gstvaapiencode_h264.h:
23866 * gst/vaapi/gstvaapiencode_mpeg2.c:
23867 * gst/vaapi/gstvaapiencode_mpeg2.h:
23868 vaapiencode: move common properties to base class.
23869 Move "rate-control" mode and "bitrate" properties to the GstVaapiEncode
23870 base class. The actual range of supported rate control modes is currently
23871 implemented as a plug-in element hook. This ought to be determined from
23872 the GstVaapiEncoder object instead, i.e. from libgstvaapi.
23874 2013-11-28 10:54:36 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23876 * gst/vaapi/gstvaapiencode_h264.c:
23877 * gst/vaapi/gstvaapiencode_mpeg2.c:
23878 vaapiencode: fix plugin description and debug name.
23879 Align the plug-in debug category to its actual name. i.e. enable debug
23880 logs through vaapiencode_<CODEC> where <CODEC> is mpeg2, h264, etc. Fix
23881 the plug-in element description to make it more consistent with other
23884 2013-11-27 16:27:31 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23887 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
23888 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
23889 * gst-libs/gst/video/Makefile.am:
23890 * gst/vaapi/gstvaapiencode.c:
23891 * gst/vaapi/gstvaapiencode.h:
23892 * gst/vaapi/gstvaapiencode_h264.c:
23893 * gst/vaapi/gstvaapiencode_mpeg2.c:
23894 vaapiencode: add initial support for GStreamer 0.10.
23896 2013-11-27 16:25:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23898 * gst-libs/gst/vaapi/gstcompat.h:
23899 libs: add more GstBuffer compat glue for GStreamer 0.10.
23900 Add gst_buffer_new_allocate() and gst_buffer_fill() implementations.
23901 Fix gst_buffer_new_wrapped_full() implementation to handle the destroy
23904 2013-11-27 15:56:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23907 * gst-libs/gst/video/Makefile.am:
23908 libs: always use built-in videoutils for GStreamer 0.10.
23909 GStreamer 0.10.36 is the latest and ultimate version to be released
23910 from the GStreamer 0.10 branch. i.e. no further releases are to be
23911 made. So, we can safely enable the built-in videoutils replacement
23912 now that they are in sync with the 0.10 branch.
23914 2013-11-27 15:47:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23917 videoutils: update to master commit d4a15a5.
23918 d4a15a5 video: fix compiler warning in header with C++11 / clang-3.1
23919 86096cc videodecoder: minor cosmetic changes to align a bit more with master
23920 b4b8b52 videodecoder: allow parse function to not use all data on adapter
23921 2145495 videodecoder: warn if frame list gets long
23922 36c3753 videodecoder: Also use the object lock to protect the output_state
23923 518c93d videodecoder: fix seeking again
23924 185fb63 video: Correct usage of the base class stream lock
23925 170e944 videodecoder: Expose _negotiate function
23927 2013-11-26 12:06:07 +0000 Matthieu Bouron <matthieu.bouron@collabora.com>
23930 * gst-libs/gst/vaapi/Makefile.am:
23931 * tests/Makefile.am:
23932 Fix build with GStreamer >= 1.3.
23933 http://bugzilla.gnome.org/show_bug.cgi?id=715183
23934 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23936 2013-11-26 17:56:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23939 configure: disable encoders with GStreamer 0.10.
23940 Don't try to build video encoders for GStreamer 0.10. Support code is
23941 not there yet, and probably will never for such an ancient version.
23943 2013-11-26 17:26:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23945 * gst/vaapi/gstvaapiencode.c:
23946 vaapiencode: fix error handling while allocating output buffers.
23947 Fix default GstVideoEncoder::allocate_buffer() implementation to properly
23948 unmap the coded buffer prior to returning an error.
23950 2013-11-26 17:11:22 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23952 * gst/vaapi/gstvaapiencode.c:
23953 vaapiencode: fix error handling in _finish() hook.
23954 Fix GstVideoEncoder::finish() implementation to really return possible
23955 errors instead of GST_FLOW_OK. That is, fix check for timeout status.
23957 2013-11-26 16:34:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23959 * gst/vaapi/gstvaapiencode.c:
23960 * gst/vaapi/gstvaapiencode.h:
23961 * gst/vaapi/gstvaapiencode_h264.c:
23962 * gst/vaapi/gstvaapiencode_h264.h:
23963 * gst/vaapi/gstvaapiencode_mpeg2.c:
23964 * gst/vaapi/gstvaapiencode_mpeg2.h:
23965 vaapiencode: minor clean-ups.
23966 Add a GST_VAAPIENCODE_CAST() helper to avoid run-time checks against
23967 the GObject type system. We are guaranteed to only deal with the same
23968 plug-in element object.
23970 2013-11-26 15:31:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23972 * gst/vaapi/gstvaapiencode.c:
23973 * gst/vaapi/gstvaapiencode_h264.c:
23974 * gst/vaapi/gstvaapiencode_mpeg2.c:
23975 vaapiencode: fix support for raw YUV sink buffers.
23976 Allow vaapiencode plug-in elements to encode from raw YUV buffers.
23977 The most efficient way to do so is to let the vaapiencode elements
23978 allocate a buffer pool, and subsequently buffers from it. This means
23979 that upstream elements are expected to honour downstream pools.
23980 If upstream elements insist on providing their own allocated buffers
23981 to the vaapiencode elements, then it possibly would be more efficient
23982 to insert a vaapipostproc element before the vaapiencode element.
23983 This is because vaapipostproc currently has better support than other
23984 elements for "foreign" raw YUV buffers.
23986 2013-11-26 15:12:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23988 * gst/vaapi/gstvaapiencode.c:
23989 vaapiencode: fix support for GStreamer 1.2.
23991 2013-11-07 17:42:21 +0800 Wind Yuan <feng.yuan@intel.com>
23993 * gst/vaapi/gstvaapiencode.c:
23994 * gst/vaapi/gstvaapiencode_h264.c:
23995 * gst/vaapi/gstvaapiencode_mpeg2.c:
23996 vaapiencode: initial port to GStreamer 1.2.
23997 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
23999 2013-11-20 16:21:32 +0800 XuGuangxin <guangxin.xu@intel.com>
24001 * gst/vaapi/Makefile.am:
24002 * gst/vaapi/gstvaapi.c:
24003 * gst/vaapi/gstvaapiencode_mpeg2.c:
24004 * gst/vaapi/gstvaapiencode_mpeg2.h:
24005 plugins: add mpeg2 encoder element.
24006 Add GstVaapiEncodeMPEG2 element object. The actual plug-in element
24007 is called "vaapiencode_mpeg2".
24009 - rate-control: rate control mode (default: cqp - constant QP)
24010 - bitrate: desired bitrate in kbps (default: auto-calculated)
24011 - key-period: maximal distance between two key frames (default: 30)
24012 - max-bframes: number of B-frames between I and P (default: 2)
24013 - quantizer: constant quantizer (default: 8)
24014 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24016 2013-07-29 16:02:56 +0800 Wind Yuan <feng.yuan@intel.com>
24018 * gst/vaapi/Makefile.am:
24019 * gst/vaapi/gstvaapi.c:
24020 * gst/vaapi/gstvaapiencode_h264.c:
24021 * gst/vaapi/gstvaapiencode_h264.h:
24022 plugins: add h264 encoder element.
24023 Add GstVaapiEncodeH264 element object. The actual plug-in element
24024 is called "vaapiencode_h264".
24026 - rate-control: rate control mode (default: none)
24027 - bitrate: desired bitrate in kbps (default: auto-calculated)
24028 - key-period: maximal distance between two key frames (default: 30)
24029 - num-slices: number of slices per frame (default: 1)
24030 - max-bframes: number of B-frames between I and P (default: 0)
24031 - min-qp: minimal quantizer (default: 1)
24032 - init-qp: initial quantizer (default: 26)
24033 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24035 2013-07-29 13:44:48 +0800 Wind Yuan <feng.yuan@intel.com>
24037 * gst/vaapi/Makefile.am:
24038 * gst/vaapi/gstvaapiencode.c:
24039 * gst/vaapi/gstvaapiencode.h:
24040 plugins: add base encoder element.
24041 vaapiencode element is based on GstVideoEncoder APIs.
24042 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24044 2013-11-20 16:20:15 +0800 XuGuangxin <guangxin.xu@intel.com>
24046 * gst-libs/gst/vaapi/Makefile.am:
24047 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.c:
24048 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2.h:
24049 * gst-libs/gst/vaapi/gstvaapiencoder_mpeg2_priv.h:
24050 encoder: add mpeg2 encoder.
24051 Add initial support for MPEG-2 encoding. I/P/B frames are supported.
24052 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24054 2013-07-29 15:46:11 +0800 Wind Yuan <feng.yuan@intel.com>
24056 * gst-libs/gst/vaapi/Makefile.am:
24057 * gst-libs/gst/vaapi/gstvaapiencoder_h264.c:
24058 * gst-libs/gst/vaapi/gstvaapiencoder_h264.h:
24059 * gst-libs/gst/vaapi/gstvaapiencoder_h264_priv.h:
24060 encoder: add h264 encoder.
24061 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24063 2013-07-29 13:34:06 +0800 Wind Yuan <feng.yuan@intel.com>
24066 * gst-libs/gst/vaapi/Makefile.am:
24067 * gst-libs/gst/vaapi/gstvaapiencoder.c:
24068 * gst-libs/gst/vaapi/gstvaapiencoder.h:
24069 * gst-libs/gst/vaapi/gstvaapiencoder_objects.c:
24070 * gst-libs/gst/vaapi/gstvaapiencoder_objects.h:
24071 * gst-libs/gst/vaapi/gstvaapiencoder_priv.h:
24072 Add initial infrastructure for video encoding.
24073 Add initial API for video encoding: only basic interfaces and small
24074 encoder objects are implemented so far.
24075 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24077 2013-07-29 15:41:23 +0800 Wind Yuan <feng.yuan@intel.com>
24080 * gst-libs/gst/Makefile.am:
24081 * gst-libs/gst/base/Makefile.am:
24082 * gst-libs/gst/base/gstbitwriter.c:
24083 * gst-libs/gst/base/gstbitwriter.h:
24084 * gst-libs/gst/vaapi/Makefile.am:
24085 libs: add generic bitstream writer.
24086 GstBitWriter provides a bit writer that can write any number of bits
24087 to a pre-allocated memory buffer. Helper functions are also provided
24088 to write any number of bits from 8, 16, 32 and 64 bit variables.
24089 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24091 2013-07-12 22:07:59 +0800 Wind Yuan <feng.yuan@intel.com>
24093 * gst-libs/gst/vaapi/gstvaapicontext.c:
24094 * gst-libs/gst/vaapi/gstvaapicontext.h:
24095 libs: add support for rate-control to GstVaapiContext.
24096 Extend GstVaapiContextInfo structure to hold the desired rate control
24097 mode for encoding purposes. For decoding purposes, this field is not
24098 used and it is initialized to GST_VAAPI_RATECONTROL_NONE.
24099 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24101 2013-07-12 21:33:32 +0800 Wind Yuan <feng.yuan@intel.com>
24103 * gst-libs/gst/vaapi/gstvaapitypes.h:
24104 * gst-libs/gst/vaapi/gstvaapiutils.c:
24105 * gst-libs/gst/vaapi/gstvaapiutils.h:
24106 * gst-libs/gst/vaapi/gstvaapivalue.c:
24107 * gst-libs/gst/vaapi/gstvaapivalue.h:
24108 libs: add rate-control attributes.
24109 Add GstVaapiRateControl types and GType values in view to supporting
24110 rate controls for encoding. This is meant to be used for instance in
24112 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24114 2013-11-22 11:56:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24117 Bump version for development.
24119 2013-11-22 11:28:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24121 * gst/vaapi/Makefile.am:
24122 build: fix for Wayland headers not in standard include dirs.
24123 Fix build when Wayland headers don't live in plain system include dirs
24124 like /usr/include but rather in /usr/include/wayland for instance.
24125 Original patch written by Dominique Leuenberger <dimstar@opensuse.org>
24126 https://bugzilla.gnome.org/show_bug.cgi?id=712282
24128 2013-11-14 10:58:37 +0000 Ross Burton <ross.burton@intel.com>
24130 * gst-libs/gst/vaapi/Makefile.am:
24131 build: link libgstvaapi-wayland against videoutils.
24132 This library is using symbols that don't exist in GStreamer 0.10 so
24133 it needs to link to built-in implementation (libgstvaapi-videoutils).
24134 https://bugzilla.gnome.org/show_bug.cgi?id=712282
24135 Signed-off-by: Ross Burton <ross.burton@intel.com>
24136 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24138 2013-11-22 11:15:57 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24140 * gst-libs/gst/vaapi/gstvaapifilter.c:
24141 * gst/vaapi/gstvaapipostproc.c:
24142 vaapostproc: fix memory leaks.
24143 Destroy VPP output surface pool on exit. Also avoid a possible crash
24144 in double-free situation caused by insufficiently reference counted
24145 array of formats returned during initialization.
24147 2013-11-22 10:19:06 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24149 * gst/vaapi/gstvaapipostproc.c:
24150 * gst/vaapi/gstvaapipostproc.h:
24151 vaapipostproc: fix and optimize advanced deinterlacing mode.
24152 Fix advanced deinterlacing modes with VPP to track only up to 2 past
24153 reference buffers. This used to be 3 past reference buffers but this
24154 doesn't fit with the existing decode pipeline that only has 4 extra
24156 Also optimize references tracking to be only enabled when needed, i.e.
24157 when advanced deinterlacing mode is used. This means that we don't
24158 need to track past references for basic bob or weave deinterlacing.
24160 2013-11-22 10:04:45 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24162 * gst/vaapi/gstvaapipostproc.c:
24163 vaapipostproc: fix "mixed" mode deinterlacing.
24164 In "mixed" interlaced streams, the buffer contains additional flags that
24165 specify whether the frame contained herein is interlaced or not. This means
24166 that we can alternatively get progressive or interlaced frames. Make sure
24167 to disable deinterlacing at the VPP level when the source buffer is no longer
24170 2013-11-22 09:49:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24172 * gst/vaapi/gstvaapipostproc.c:
24173 vaapipostproc: fix memory leaks with advanced deinterlacing.
24174 Fix memory leaks with advanced deinterlacing, i.e. when we keep track
24175 of past buffers. Completely reset the deinterlace state, thus destroying
24176 any buffer currently held, on _start(), _stop() and _destroy().
24178 2013-11-22 06:59:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24182 - GStreamer 1.2 APIs are supported ;
24183 - Video Processing (VA/VPP) features.
24185 2013-11-22 06:45:22 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24188 README: update for GStreamer >= 1.0.x and VPP features.
24190 2013-11-22 06:37:12 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24192 * gst-libs/gst/vaapi/glibcompat.h:
24193 * gst-libs/gst/vaapi/gstvaapicontext.h:
24194 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
24195 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.h:
24196 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.h:
24197 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.h:
24198 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.h:
24199 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
24200 * gst-libs/gst/vaapi/gstvaapidisplay_drm.h:
24201 * gst-libs/gst/vaapi/gstvaapidisplay_drm_priv.h:
24202 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
24203 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
24204 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
24205 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.h:
24206 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
24207 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
24208 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
24209 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
24210 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
24211 * gst-libs/gst/vaapi/gstvaapidisplaycache.h:
24212 * gst-libs/gst/vaapi/gstvaapiimage.c:
24213 * gst-libs/gst/vaapi/gstvaapiimage.h:
24214 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
24215 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
24216 * gst-libs/gst/vaapi/gstvaapiminiobject.h:
24217 * gst-libs/gst/vaapi/gstvaapiobject.c:
24218 * gst-libs/gst/vaapi/gstvaapiobject.h:
24219 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
24220 * gst-libs/gst/vaapi/gstvaapiprofile.c:
24221 * gst-libs/gst/vaapi/gstvaapisurface.h:
24222 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
24223 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
24224 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
24225 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
24226 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
24227 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
24228 * gst-libs/gst/vaapi/gstvaapitexture.c:
24229 * gst-libs/gst/vaapi/gstvaapitexture.h:
24230 * gst-libs/gst/vaapi/gstvaapitypes.h:
24231 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
24232 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
24233 * gst-libs/gst/vaapi/gstvaapivalue.c:
24234 * gst-libs/gst/vaapi/gstvaapivalue.h:
24235 * gst-libs/gst/vaapi/gstvaapivideopool.c:
24236 * gst-libs/gst/vaapi/gstvaapivideopool.h:
24237 * gst-libs/gst/vaapi/gstvaapivideopool_priv.h:
24238 * gst-libs/gst/vaapi/gstvaapiwindow.c:
24239 * gst-libs/gst/vaapi/gstvaapiwindow.h:
24240 * gst-libs/gst/vaapi/gstvaapiwindow_drm.c:
24241 * gst-libs/gst/vaapi/gstvaapiwindow_drm.h:
24242 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
24243 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
24244 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
24245 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.h:
24246 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
24247 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
24248 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
24249 * gst/vaapi/gstvaapi.c:
24250 * gst/vaapi/gstvaapidecode.h:
24251 * gst/vaapi/gstvaapidownload.h:
24252 * gst/vaapi/gstvaapipluginutil.c:
24253 * gst/vaapi/gstvaapipluginutil.h:
24254 * gst/vaapi/gstvaapipostproc.h:
24255 * gst/vaapi/gstvaapisink.h:
24256 * gst/vaapi/gstvaapivideoconverter_glx.h:
24260 * tests/test-display.c:
24261 * tests/test-jpeg.c:
24262 * tests/test-jpeg.h:
24263 * tests/test-mpeg4.c:
24264 * tests/test-mpeg4.h:
24265 * tests/test-surfaces.c:
24266 * tests/test-windows.c:
24267 legal: update copyright notice dates.
24269 2013-11-22 05:57:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24271 * gst-libs/gst/vaapi/glibcompat.h:
24272 * gst-libs/gst/vaapi/gstcompat.h:
24273 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
24274 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
24275 * gst-libs/gst/vaapi/gstvaapicompat.h:
24276 * gst-libs/gst/vaapi/gstvaapicontext.c:
24277 * gst-libs/gst/vaapi/gstvaapicontext.h:
24278 * gst-libs/gst/vaapi/gstvaapidebug.h:
24279 * gst-libs/gst/vaapi/gstvaapidecoder.c:
24280 * gst-libs/gst/vaapi/gstvaapidecoder.h:
24281 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
24282 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.h:
24283 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
24284 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
24285 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
24286 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.h:
24287 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
24288 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.h:
24289 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
24290 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.h:
24291 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
24292 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
24293 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
24294 * gst-libs/gst/vaapi/gstvaapidecoder_unit.c:
24295 * gst-libs/gst/vaapi/gstvaapidecoder_unit.h:
24296 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
24297 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.h:
24298 * gst-libs/gst/vaapi/gstvaapidisplay.c:
24299 * gst-libs/gst/vaapi/gstvaapidisplay.h:
24300 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
24301 * gst-libs/gst/vaapi/gstvaapidisplay_drm.h:
24302 * gst-libs/gst/vaapi/gstvaapidisplay_drm_priv.h:
24303 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
24304 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
24305 * gst-libs/gst/vaapi/gstvaapidisplay_glx_priv.h:
24306 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
24307 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
24308 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.h:
24309 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
24310 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
24311 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
24312 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
24313 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
24314 * gst-libs/gst/vaapi/gstvaapidisplaycache.h:
24315 * gst-libs/gst/vaapi/gstvaapifilter.c:
24316 * gst-libs/gst/vaapi/gstvaapifilter.h:
24317 * gst-libs/gst/vaapi/gstvaapiimage.c:
24318 * gst-libs/gst/vaapi/gstvaapiimage.h:
24319 * gst-libs/gst/vaapi/gstvaapiimage_priv.h:
24320 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
24321 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
24322 * gst-libs/gst/vaapi/gstvaapiminiobject.c:
24323 * gst-libs/gst/vaapi/gstvaapiminiobject.h:
24324 * gst-libs/gst/vaapi/gstvaapiobject.c:
24325 * gst-libs/gst/vaapi/gstvaapiobject.h:
24326 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
24327 * gst-libs/gst/vaapi/gstvaapiparser_frame.c:
24328 * gst-libs/gst/vaapi/gstvaapiparser_frame.h:
24329 * gst-libs/gst/vaapi/gstvaapipixmap.c:
24330 * gst-libs/gst/vaapi/gstvaapipixmap.h:
24331 * gst-libs/gst/vaapi/gstvaapipixmap_priv.h:
24332 * gst-libs/gst/vaapi/gstvaapipixmap_x11.c:
24333 * gst-libs/gst/vaapi/gstvaapipixmap_x11.h:
24334 * gst-libs/gst/vaapi/gstvaapiprofile.c:
24335 * gst-libs/gst/vaapi/gstvaapiprofile.h:
24336 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
24337 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
24338 * gst-libs/gst/vaapi/gstvaapisurface.c:
24339 * gst-libs/gst/vaapi/gstvaapisurface.h:
24340 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
24341 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
24342 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
24343 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
24344 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
24345 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
24346 * gst-libs/gst/vaapi/gstvaapitexture.c:
24347 * gst-libs/gst/vaapi/gstvaapitexture.h:
24348 * gst-libs/gst/vaapi/gstvaapitypes.h:
24349 * gst-libs/gst/vaapi/gstvaapiutils.c:
24350 * gst-libs/gst/vaapi/gstvaapiutils.h:
24351 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
24352 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
24353 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
24354 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
24355 * gst-libs/gst/vaapi/gstvaapivalue.c:
24356 * gst-libs/gst/vaapi/gstvaapivalue.h:
24357 * gst-libs/gst/vaapi/gstvaapivideopool.c:
24358 * gst-libs/gst/vaapi/gstvaapivideopool.h:
24359 * gst-libs/gst/vaapi/gstvaapivideopool_priv.h:
24360 * gst-libs/gst/vaapi/gstvaapiwindow.c:
24361 * gst-libs/gst/vaapi/gstvaapiwindow.h:
24362 * gst-libs/gst/vaapi/gstvaapiwindow_drm.c:
24363 * gst-libs/gst/vaapi/gstvaapiwindow_drm.h:
24364 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
24365 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
24366 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
24367 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
24368 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.h:
24369 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
24370 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
24371 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
24372 * gst-libs/gst/vaapi/gstvaapiworkarounds.h:
24373 * gst-libs/gst/vaapi/sysdeps.h:
24374 * gst-libs/gst/vaapi/video-format.c:
24375 * gst-libs/gst/vaapi/video-format.h:
24376 * gst/vaapi/gstvaapi.c:
24377 * gst/vaapi/gstvaapidecode.c:
24378 * gst/vaapi/gstvaapidecode.h:
24379 * gst/vaapi/gstvaapidownload.c:
24380 * gst/vaapi/gstvaapidownload.h:
24381 * gst/vaapi/gstvaapipluginutil.c:
24382 * gst/vaapi/gstvaapipluginutil.h:
24383 * gst/vaapi/gstvaapipostproc.c:
24384 * gst/vaapi/gstvaapipostproc.h:
24385 * gst/vaapi/gstvaapisink.c:
24386 * gst/vaapi/gstvaapisink.h:
24387 * gst/vaapi/gstvaapiupload.c:
24388 * gst/vaapi/gstvaapiupload.h:
24389 * gst/vaapi/gstvaapiuploader.c:
24390 * gst/vaapi/gstvaapiuploader.h:
24391 * gst/vaapi/gstvaapivideobuffer.c:
24392 * gst/vaapi/gstvaapivideobuffer.h:
24393 * gst/vaapi/gstvaapivideobufferpool.c:
24394 * gst/vaapi/gstvaapivideobufferpool.h:
24395 * gst/vaapi/gstvaapivideocontext.c:
24396 * gst/vaapi/gstvaapivideocontext.h:
24397 * gst/vaapi/gstvaapivideoconverter_glx.c:
24398 * gst/vaapi/gstvaapivideoconverter_glx.h:
24399 * gst/vaapi/gstvaapivideoconverter_x11.c:
24400 * gst/vaapi/gstvaapivideoconverter_x11.h:
24401 * gst/vaapi/gstvaapivideomemory.c:
24402 * gst/vaapi/gstvaapivideomemory.h:
24403 * gst/vaapi/gstvaapivideometa.c:
24404 * gst/vaapi/gstvaapivideometa.h:
24405 * gst/vaapi/gstvaapivideometa_texture.c:
24406 * gst/vaapi/gstvaapivideometa_texture.h:
24415 * tests/simple-decoder.c:
24416 * tests/test-decode.c:
24417 * tests/test-decode.h:
24418 * tests/test-display.c:
24419 * tests/test-filter.c:
24420 * tests/test-h264.c:
24421 * tests/test-h264.h:
24422 * tests/test-jpeg.c:
24423 * tests/test-jpeg.h:
24424 * tests/test-mpeg2.c:
24425 * tests/test-mpeg2.h:
24426 * tests/test-mpeg4.c:
24427 * tests/test-mpeg4.h:
24428 * tests/test-surfaces.c:
24429 * tests/test-textures.c:
24430 * tests/test-vc1.c:
24431 * tests/test-vc1.h:
24432 * tests/test-windows.c:
24433 legal: add per-file authorship information.
24434 Credit original authors on a per-file basis as we cannot expect people
24435 to know all country-specific rules, or bother browsing through the git
24438 2013-11-21 23:52:43 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24444 2013-11-21 23:51:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24449 2013-11-21 23:17:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24451 * gst-libs/gst/vaapi/gstvaapidecoder.h:
24452 decoder: don't include obsolete headers.
24453 The <gst/vaapi/gstvaapicontext.h> header was removed from the public
24454 set of APIs. So, don't make public headers (gstvaapidecoder.h) depend
24457 2013-11-18 16:20:43 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24459 * gst/vaapi/Makefile.am:
24460 * gst/vaapi/gstvaapi.c:
24461 * gst/vaapi/gstvaapipostproc.c:
24462 vaapipostproc: add initial support for GStreamer 1.2.
24463 Port vaapipostproc element to GStreamer 1.2. Support is quite minimal
24464 right now so that to cope with auto-plugging issues/regressions. e.g.
24465 this happens when the correct set of expected caps are being exposed.
24466 This means that, currently, the proposed caps are not fully accurate.
24468 2013-11-01 10:22:17 +0800 Halley Zhao <halley.zhao@intel.com>
24470 * gst/vaapi/gstvaapipostproc.c:
24471 * gst/vaapi/gstvaapipostproc.h:
24472 vaapipostproc: add support for denoise and sharpen filters.
24473 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24475 2013-11-21 19:52:56 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24477 * gst/vaapi/gstvaapipostproc.c:
24478 * gst/vaapi/gstvaapipostproc.h:
24479 vaapipostproc: add support for advanced deinterlacing.
24480 Add initial support for advanced deinterlacing. The history buffer
24481 size is arbitrarily set to 3 references for now.
24483 2013-11-21 22:32:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24485 * gst/vaapi/gstvaapipostproc.c:
24486 vaapipostproc: fix deinterlacing with VPP.
24487 Fix basic deinterlacing flags provided to gst_vaapi_set_deinterlacing()
24488 for the first field. Render flags were supplied instead of the actual
24489 deinterlacing flags (deint_flags).
24491 2013-11-21 15:08:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24493 * gst/vaapi/gstvaapipostproc.c:
24494 vaapipostproc: fix transform caps.
24495 Fix GstBaseTransform::transform_caps() implementation to always return
24496 the complete set of allowed sink pad caps (unfixated) even if the src
24497 pad caps we are getting are fixated. Rationale: there are just so many
24498 possible combinations, and it was wrong to provide a unique set anyway.
24499 As a side effect, this greatly simplifies the ability to derive src pad
24500 caps from fixated sink pad caps.
24502 2013-11-01 10:31:13 +0800 Halley Zhao <halley.zhao@intel.com>
24504 * docs/reference/libs/libs-sections.txt:
24505 * gst-libs/gst/vaapi/gstvaapifilter.c:
24506 * gst-libs/gst/vaapi/gstvaapifilter.h:
24507 filter: add helper to specify references for deinterlacing.
24508 Add gst_vaapi_fitler_set_deinterlacing_references() API to submit the
24509 list of surfaces used for forward or backward reference in advanced
24510 deinterlacing mode, e.g. Motion-Adaptive, Motion-Compensated.
24511 The list of surfaces used as deinterlacing references shall be live
24512 until the next call to gst_vaapi_filter_process().
24513 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24515 2013-11-21 18:44:46 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24517 * gst-libs/gst/vaapi/gstvaapifilter.c:
24518 * gst-libs/gst/vaapi/gstvaapifilter.h:
24519 * gst-libs/gst/vaapi/gstvaapiutils.c:
24520 * gst/vaapi/gstvaapipostproc.c:
24521 * tests/test-filter.c:
24522 filter: fix semantics of deinterlacing flags.
24523 Fix deinterlacing flags to make more sense. The TFF (top-field-first)
24524 flag is meant to specify the organization of reference frames used in
24525 advanced deinterlacing modes. Introduce the more explicit flag TOPFIELD
24526 to specify that the top-field of the supplied input surface is to be
24527 used for deinterlacing. Conversely, if not set, this means that the
24528 bottom field of the supplied input surface will be used instead.
24530 2013-11-21 17:20:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24532 * docs/reference/libs/libs-sections.txt:
24533 * gst-libs/gst/vaapi/gstvaapifilter.c:
24534 * gst-libs/gst/vaapi/gstvaapifilter.h:
24535 filter: add helpers to check for supported/active operation.
24536 Add a couple of helper functions:
24537 - gst_vaapi_filter_has_operation(): checks whether the VA driver
24538 advertises support for the supplied operation ;
24539 - gst_vaapi_filter_use_operation(): checks whether the supplied
24540 operation was already enabled to its non-default value.
24542 2013-11-20 15:10:17 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24544 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
24545 libs: fix GstVaapiSurfaceProxy destroy notify call site.
24546 The user-defined destroy notify function is meant to be called only when
24547 the surface proxy was fully released, i.e. once it actually released the
24548 VA surface back to the underlying pool.
24550 2013-08-29 13:44:22 +0800 XuGuangxin <guangxin.xu@intel.com>
24552 * gst-libs/gst/vaapi/gstvaapivideopool.c:
24553 * gst-libs/gst/vaapi/gstvaapivideopool_priv.h:
24554 libs: make GstVaapiVideoPool thread-safe.
24555 https://bugzilla.gnome.org/show_bug.cgi?id=707108
24556 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24558 2013-08-29 14:04:06 +0800 XuGuangxin <guangxin.xu@intel.com>
24560 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
24561 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
24562 libs: robustify decoder objects and surface proxy initialization.
24563 Fix GstVaapiPicture, GstVaapiSlice and GstVaapiSurfaceProxy initialization
24564 sequences to have the expected default values set beforehand in case of an
24565 error raising up further during creation. i.e. make it possible to cleanly
24566 destroy those partially initialized objects.
24567 https://bugzilla.gnome.org/show_bug.cgi?id=707108
24568 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24570 2013-11-21 11:01:41 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24572 * gst/vaapi/gstvaapidecode.c:
24573 vaapidecode: fix decoder flush.
24574 There are situations where gst_video_decoder_flush() is called, and
24575 this subsequently produces a gst_video_decoder_reset() that kills the
24576 currently active GstVideoCodecFrame. This means that it no longer
24577 exists by the time we reach GstVideoDecoder::finish() callback, thus
24578 possibly resulting in a crash if we assumed spare data was still
24579 available for decode (current_frame_size > 0).
24580 Try to honour GstVideoDecoder::reset() behaviour from GStreamer 1.0
24581 that means a flush, thus performing the actual operations there like
24582 calling gst_video_decoder_have_frame() if pending data is available.
24584 2013-11-20 19:21:05 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24586 * gst/vaapi/gstvaapidecode.c:
24587 * gst/vaapi/gstvaapidecode.h:
24588 vaapidecode: fix dead-locks with decoder task.
24589 Review all interactions between the main video decoder stream thread
24590 and the decode task to derive a correct sequence of operations for
24591 decoding. Also avoid extra atomic operations that become implicit under
24592 the GstVideoDecoder stream lock.
24594 2013-08-29 14:12:10 +0800 XuGuangxin <guangxin.xu@intel.com>
24596 * gst/vaapi/gstvaapidecode.c:
24597 vaapidecode: fix hard reset for seek cases.
24598 Fix hard reset for seek cases by flushing the GstVaapiDecoder queue
24599 and completely purge any decoded output frame that may come out from
24600 it. At this stage, the GstVaapiDecoder shall be in a complete clean
24601 state to start decoding over new buffers.
24602 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24604 2013-08-29 14:12:10 +0800 XuGuangxin <guangxin.xu@intel.com>
24606 * gst/vaapi/gstvaapidecode.c:
24607 * gst/vaapi/gstvaapidecode.h:
24608 vaapidecode: drop decode timeout, always wait for a free surface.
24609 vaapidecode used to wait up to one second past the expected time of
24610 presentation for the last decoded frame. This is not realistic in
24611 practice when it comes to video pause/resume. Changed behaviour to
24612 unconditionnally wait for a free VA surface prior to continuing the
24613 decoding. The decode task will continue pushing the output frames to
24614 the downstream element while also reporting errors at the same time
24615 to the main thread.
24616 https://bugzilla.gnome.org/show_bug.cgi?id=707108
24617 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24619 2013-11-20 10:56:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24621 * gst/vaapi/gstvaapidecode.c:
24622 vaapidecode: fix srcpad caps for GStreamer 1.2.
24623 The srcpad caps exposed for GStreamer 1.2 were missing any useful info
24624 like framerate, pixel-aspect-ratio, interlace-mode et al. Not to mention
24625 that it relied on possibly un-initialized data. Fix srcpad caps to be
24626 initialized from a sanitized copy of GstVideoDecoder output state caps.
24627 Note: the correct way to expose the srcpad caps triggers an additional
24628 issue in core GStreamer auto-plugging capabilities as the correct caps
24629 to be exposed should be format=ENCODED with memory:VASurface caps feature
24630 at the minimum. In some situations, we could determine the underlying
24631 VA surface format, but this is not always possible. e.g. cases where it
24632 is not allowed to expose the underlying VA surface data, or when the
24633 VA driver implementation cannot actually provide such information.
24635 2013-11-20 10:45:23 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24637 * gst/vaapi/gstvaapidecode.c:
24638 * gst/vaapi/gstvaapisink.c:
24639 plugins: streamline VA formats exposed in caps to a realistic set.
24640 Currently, the decoder only supports YUV 4:2:0 output. So, expose the
24641 output formats for GStreamer 1.2 in caps to a realistic subset. This
24642 means NV12, I420 or YV12 but also ENCODED if we cannot determine the
24643 underlying VA surface format, or if it is actually not allowed to get
24644 access to the surface contents.
24646 2013-11-20 10:37:36 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24648 * gst/vaapi/gstvaapidecode.c:
24649 * gst/vaapi/gstvaapisink.c:
24650 plugins: expose the expected format for GstVideoGLTextureUploadMeta.
24651 Fix vaapidecode srcpad caps to only expose RGBA video format for the
24652 meta:GstVideoGLTextureUploadMeta feature. That's only what is supported
24653 so far. Besides, drop this meta from the vaapisink sinkpad caps since
24654 we really don't support that for rendering.
24655 https://bugzilla.gnome.org/show_bug.cgi?id=711828
24657 2013-11-18 18:25:21 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24660 configure: automatically detect GStreamer API version.
24661 Automatically detect GStreamer API version. The --with-gstreamer-api
24662 configure option now defaults to "autodetect" and configure then tries
24663 to derive the GStreamer API version from the highest version based on
24664 what pkg-config --modversion would report.
24665 https://bugzilla.gnome.org/show_bug.cgi?id=711657
24667 2013-11-01 13:43:11 +0800 Wind Yuan <feng.yuan@intel.com>
24669 * gst/vaapi/gstvaapipostproc.c:
24670 vaapipostproc: fix support for raw YUV data upload on GStreamer 1.0.
24671 Fix raw YUV data uploaded as in the following pipeline:
24672 $ gst-launch-1.0 filesrc video.yuv ! videoparse ! vaapipostproc ! vaapisink
24673 The main reason why it failed was that the videoparse element simply
24674 allocates GstBuffer with raw data chunk'ed off the sink pad without
24675 any prior knowledge of the actual frame info. i.e. it basically just
24676 calls gst_adapter_take_buffer().
24677 We could avoid the extra copy performed in vaapipostproc if the videoparse
24678 element was aware of the downstream pool and bothers copying line by
24679 line, for each plane. This means that, for a single frame per buffer,
24680 the optimizatin will be to allocate the video buffer downstream, map
24681 it, and copy each line that is coming through until we need to fills
24682 in the successive planes.
24683 Still, optimized raw YUV uploads already worked with the following:
24684 $ gst-launch-1.0 videotestsrc ! vaapipostproc ! vaapisink
24685 https://bugzilla.gnome.org/show_bug.cgi?id=711250
24686 [clean-ups, fixed error cases to unmap and unref outbuf]
24687 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24689 2013-11-16 07:02:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24691 * gst/vaapi/gstvaapipostproc.c:
24692 vaapipostproc: try to downgrade deinterlace-method when needed.
24693 If the currently selected deinterlacing method is not supported by the
24694 underlying hardware, then try to downgrade the method to a supported one.
24695 At the minimum, basic bob-deinterlacing shall always be supported.
24697 2013-11-15 19:04:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24699 * gst/vaapi/gstvaapipostproc.c:
24700 vaapipostproc: add initial support for deinterlacing with VPP.
24701 Allow basic bob-deinterlacing to work when VPP is enabled. Currently,
24702 this only covers bob-deinterlacing when the output pixel format is
24705 2013-11-15 17:14:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24707 * gst/vaapi/gstvaapipostproc.c:
24708 vaapipostproc: fix build on 64-bit platforms with GStreamer 0.10.
24709 The size argument for GstBaseTransform::transform_size() hook is a
24710 guint in GStreamer 0.10 APIs but a gsize in GStreamer >= 1.0.X APIs.
24712 2013-10-18 18:08:25 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24714 * gst/vaapi/gstvaapipostproc.c:
24715 * gst/vaapi/gstvaapipostproc.h:
24716 vaapipostproc: add initial support for scaling.
24717 Add initial support for basic scaling with size specified through the
24718 "width" and "height" properties. If either user-provided dimension is
24719 zero and "force-aspect-ratio" is set to true (the default), then the
24720 other dimension is scaled to preserve the aspect ratio.
24722 2013-10-18 18:08:25 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24724 * gst/vaapi/gstvaapipostproc.c:
24725 * gst/vaapi/gstvaapipostproc.h:
24726 vaapipostproc: add initial support for color conversion.
24727 If VPP is available, we always try to implicitly convert the source
24728 buffer to the "native" surface format for the underlying accelerator.
24729 This means that no optimization is performed yet to propagate raw YUV
24730 buffers to the downstream element as is, if VPP is available. i.e. it
24731 will always cause a color conversion.
24733 2013-10-16 11:23:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24735 * gst/vaapi/gstvaapipostproc.c:
24736 vaapipostproc: fix bug when user disabled deinterlacing.
24737 Fix pipeline error / hang when the user disabled deinterlacing through
24738 the deinterlace-mode=disabled property setting.
24740 2013-10-16 11:20:50 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24742 * gst/vaapi/gstvaapipostproc.c:
24743 * gst/vaapi/gstvaapipostproc.h:
24744 vaapipostproc: factor out operations to be applied into flags.
24745 Even if we only support deinterlacing for now, use flags to specify
24746 which filters are to be applied to each frame we receive in transform().
24747 This is preparatory work for integrating new filters.
24749 2013-10-04 15:37:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24751 * gst/vaapi/gstvaapipostproc.c:
24752 * gst/vaapi/gstvaapipostproc.h:
24753 vaapipostproc: add support for raw YUV video source buffers.
24754 Allow video processing from raw YUV buffers coming from the sink pad,
24755 while still producing a VA surface for the downstream elements.
24757 2013-10-04 16:00:56 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24759 * gst/vaapi/gstvaapipluginutil.c:
24760 * gst/vaapi/gstvaapipluginutil.h:
24761 * gst/vaapi/gstvaapipostproc.c:
24762 vaapipostproc: add support for "mixed" interlace mode.
24763 Add support for "mixed" interlace-mode, whereby the video frame buffer
24764 shall be deinterlaced only if its flags mention that's actually an
24765 interlaced frame buffer.
24767 2013-10-03 19:04:07 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24769 * gst-libs/gst/vaapi/gstcompat.h:
24770 * gst/vaapi/gstvaapipostproc.c:
24771 * gst/vaapi/gstvaapipostproc.h:
24772 * gst/vaapi/gstvaapivideobuffer.c:
24773 * gst/vaapi/gstvaapivideobuffer.h:
24774 vaapipostproc: rework plug-in element.
24775 Rewrite the vaapipostproc plug-in element so that it derives from
24776 GstBaseTransform, thus simplifying the caps negotiation process.
24778 2013-10-09 17:25:10 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24780 * gst/vaapi/gstvaapivideobufferpool.c:
24781 * gst/vaapi/gstvaapivideomemory.c:
24782 * gst/vaapi/gstvaapivideomemory.h:
24783 plugins: fix and optimize check for buffer pool allocator params.
24784 Reset the buffer pool allocator only if the config caps changed in a
24785 sensible way: format or resolution change. i.e. don't bother with
24786 other caps like colorimetry et al. as this doesn't affect the way to
24787 allocate VA surfaces or images.
24789 2013-10-09 10:33:55 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24791 * gst/vaapi/gstvaapivideomemory.c:
24792 plugins: enable memory maps for read & write with direct-rendering.
24793 Enable read and write mappings only if direct-rendering is supported.
24794 Otherwise, this means that we may need to download data from the VA
24795 surface first for correctness, even if the VA surface doesn't need to
24796 be read at all. i.e. sometimes, READWRITE mappings are meant for
24797 surfaces that are written to first, and read afterwards for further
24799 https://bugzilla.gnome.org/show_bug.cgi?id=704078
24801 2013-10-09 10:06:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24803 * gst/vaapi/gstvaapivideomemory.c:
24804 plugins: fix check for direct-rendering support.
24805 Fix check for direct-rendering if the creation of VA surfaces with
24806 an explicit pixel format is not support, e.g. VA-API < 0.34.0, and
24807 that we tried to allocate a VA surface based on the corresponding
24808 chroma type. i.e. in that particular case, we have to make sure that
24809 the derived image has actually the expected format.
24811 2013-10-09 09:47:18 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24813 * gst/vaapi/gstvaapivideobufferpool.c:
24814 * gst/vaapi/gstvaapivideomemory.c:
24815 * gst/vaapi/gstvaapivideomemory.h:
24816 plugins: fix buffer pool reset_buffer() to reset memory resources.
24817 Fix GstVaapiVideoBufferPool::reset_buffer() to reset the underlying
24818 memory resources, and more particularly the VA surface proxy. Most
24819 importantly, the GstVaapiVideoMeta is retained. Cached surface in
24820 memory are released, thus triggering a new allocation the next time
24821 we need to map the buffer.
24823 2013-10-09 09:33:56 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24825 * gst/vaapi/gstvaapivideomemory.c:
24826 * gst/vaapi/gstvaapivideomemory.h:
24827 plugins: fix GstVaapiVideoMemory to allocate VA surface proxies.
24828 Make sure GstVaapiVideoMemory allocates VA surface proxies from a
24829 pool stored in the parent VA memory allocator.
24830 This fixes the following scenario:
24831 - VA video buffer 1 is allocated from a buffer pool
24832 - Another video buffer is created, and inherits info from buffer 1
24833 - Buffer 1 is released, thus pushing it back to the buffer pool
24834 - New buffer alloc request comes it, this yields buffer 1 back
24835 - At this stage, buffers 1 and 2 still share the same underlying VA
24836 surface, but buffer 2 was already submitted downstream for further
24837 processing, thus conflicting with additional processing we were
24838 about to perform on buffer 1.
24839 Maybe the core GstBufferPool implementation should have been fixed
24840 instead to actually make sure that the returned GstBuffer memory we
24841 found from the pool is writable?
24843 2013-10-04 19:34:32 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24845 * gst/vaapi/gstvaapiuploader.c:
24846 plugins: create a proxy for GstVaapiUploader allocated buffers.
24847 Always make sure to allocate a VA surface proxy for GstVaapiUploader
24848 allocated buffers, i.e. make gst_vaapi_uploader_get_buffer() allocate
24850 This fixes cases where we want to retain the underlying surface longer,
24851 instead of releasing it back to the surface pool right away.
24853 2013-10-04 19:30:36 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24855 * gst/vaapi/gstvaapidecode.c:
24856 * gst/vaapi/gstvaapipluginutil.c:
24857 * gst/vaapi/gstvaapipluginutil.h:
24858 * gst/vaapi/gstvaapipostproc.c:
24859 plugins: add helper function to disable deinterlacing in caps.
24860 Add gst_caps_set_interlaced() helper function that would reset the
24861 interlace-mode field to "progressive" for GStreamer >= 1.0, or the
24862 interlaced field to "false" for GStreamer 0.10.
24864 2013-10-01 18:26:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24866 * gst-libs/gst/vaapi/gstvaapifilter.c:
24867 filter: fix memory leak of VPP operations.
24868 Fix ensure_operations() to release the VPP operations array if non
24869 NULL, prior to returning to the caller. The former function was also
24870 renamed to a more meaningful get_operations() since the caller owns
24871 the returned array that needs to be released.
24873 2013-09-04 13:53:25 +0800 Zhao Halley <halley.zhao@intel.com>
24875 * gst-libs/gst/vaapi/gstvaapifilter.c:
24876 filter: fix first-time operation lookup.
24877 Fix first-time operation lookup through find_operation() if the set
24878 of supported operations was not initially determined through the
24879 gst_vaapi_filter_get_operations() helper function.
24880 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24882 2013-09-04 13:53:25 +0800 Zhao Halley <halley.zhao@intel.com>
24884 * gst-libs/gst/vaapi/gstvaapifilter.c:
24885 filter: fix colorbalance related subtypes.
24886 Fix intiialization of GstVaapiFilterOpData for colorbalance related
24887 operations. In particular, fill in the va_subtype field accordingly.
24888 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24890 2013-09-30 17:08:12 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
24892 * gst-libs/gst/vaapi/gstvaapifilter.c:
24893 filter: fix VA-API 0.34.0 symbol guards.
24894 VASurfaceAttrib and VAProcFilterParameterBufferType are symbols
24895 that need to be guarded for libva 0.34 and 0.33, respectively.
24896 https://bugzilla.gnome.org/show_bug.cgi?id=709102
24897 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24899 2013-10-01 17:57:11 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
24901 * gst/vaapi/gstvaapidecode.c:
24902 * gst/vaapi/gstvaapipluginutil.c:
24903 * gst/vaapi/gstvaapisink.c:
24904 plugins: hanle the context query in any pad.
24905 Also this patch simplifies the code, since now the query is common for the
24906 decoder and the sink.
24907 https://bugzilla.gnome.org/show_bug.cgi?id=709200
24909 2013-10-01 12:09:44 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
24911 * gst/vaapi/gstvaapivideocontext.c:
24912 plugins: query upstream element for a GstContext.
24913 Fix gst_vaapi_video_context_prepare() to also query upstream elements
24914 for a valid GstContext. Improve comments regarding the steps used to
24915 lookup or build that context, thus conforming to the GstContext API
24917 https://bugzilla.gnome.org/show_bug.cgi?id=709112
24918 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24920 2013-09-26 15:21:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24923 * debian.upstream/control.in:
24924 Fix detection and packaging of GStreamer 1.2.x builds.
24925 The GStreamer 1.2.x packages sticked to the naming convention for 1.0.x
24926 packages, i.e. -1.0 suffix. However, for gstreamer-vaapi packaging
24927 purposes, update the versioning to -1.2 suffix instead.
24929 2013-07-15 13:41:00 +0200 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
24931 * gst/vaapi/Makefile.am:
24932 * gst/vaapi/gstvaapidecode.c:
24933 * gst/vaapi/gstvaapidecode.h:
24934 * gst/vaapi/gstvaapisink.c:
24935 * gst/vaapi/gstvaapivideometa_texture.c:
24936 * gst/vaapi/gstvaapivideometa_texture.h:
24937 plugins: add support for GstVideoGLTextureUploadMeta.
24938 If the allocation meta GST_VIDEO_GL_TEXTURE_UPLOAD_META_API_TYPE is
24939 requested, and more specifically under a GLX configuration, then add
24940 the GstVideoGLTextureUploadMeta to the output buffer.
24941 https://bugzilla.gnome.org/show_bug.cgi?id=703236
24942 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24944 2013-07-04 11:03:52 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
24946 * gst/vaapi/gstvaapidecode.c:
24947 * gst/vaapi/gstvaapisink.c:
24948 * gst/vaapi/gstvaapivideomemory.h:
24949 plugins: add support for GstCaps features.
24950 Move VA video buffer memory from "video/x-surface,type=vaapi" format,
24951 as expressed in caps, to the more standard use of caps features. i.e.
24952 add "memory:VASurface" feature attribute to the associated caps.
24953 https://bugzilla.gnome.org/show_bug.cgi?id=703271
24954 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24956 2013-07-12 12:58:57 -0400 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
24958 * gst/vaapi/gstvaapidecode.c:
24959 * gst/vaapi/gstvaapisink.c:
24960 plugins: improve ::query() debugging messages.
24961 Fix gst_vaapidecode_query() to correctly display the query type name,
24962 instead of randomly displaying that we shared the underlying display.
24963 Also add debug info for the GstVaapiSink::query() handler, i.e. the
24964 supplied query type name actually.
24965 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24967 2013-07-12 12:58:57 -0400 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
24969 * gst/vaapi/gstvaapidecode.c:
24970 * gst/vaapi/gstvaapisink.c:
24971 plugins: add support for GstContext API.
24972 Add support for the new GstContext API from GStreamer 1.2.x.
24973 - implement the GstElement::set_context() hook ;
24974 - reply to the `context' query from downstream elements.
24975 https://bugzilla.gnome.org/show_bug.cgi?id=703235
24976 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24978 2013-05-22 12:07:52 -0400 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
24980 * gst/vaapi/Makefile.am:
24981 * gst/vaapi/gstvaapipluginutil.c:
24982 * gst/vaapi/gstvaapivideocontext.c:
24983 * gst/vaapi/gstvaapivideocontext.h:
24984 plugins: add compat layer for GstVideoContext.
24985 Add thin compatibility layer for the deprecated GstVideoContext API.
24986 For GStreamer API >= 1.2, this involves the following two functions:
24987 - gst_vaapi_video_context_prepare(): queries if a context is already
24988 set in the pipeline ;
24989 - gst_vaapi_video_context_propagate(): propagates the newly-created
24990 context to the rest of the pipeline.
24991 https://bugzilla.gnome.org/show_bug.cgi?id=703235
24992 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
24994 2013-05-21 12:42:39 -0400 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
24997 * gst/vaapi/Makefile.am:
24998 * gst/vaapi/gstvaapi.c:
24999 * gst/vaapi/gstvaapidecode.c:
25000 * gst/vaapi/gstvaapipluginutil.c:
25001 * gst/vaapi/gstvaapisink.c:
25002 * gst/vaapi/gstvaapivideobuffer.c:
25003 plugins: initial port to GStreamer 1.2.
25004 Port vaapidecode and vaapisink plugins to GStreamer API >= 1.2. This
25005 is rather minimalistic so that to test the basic functionality.
25006 Disable vaapipostproc plugin for now as further polishing is needed.
25007 Also disable GstVideoContext interface support since this API is now
25008 gone in 1.2.x. This is preparatory work for GstContext support.
25009 https://bugzilla.gnome.org/show_bug.cgi?id=703235
25010 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25012 2013-09-24 16:21:11 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25014 * tests/simple-decoder.c:
25015 tests: simple-decoder: fix for non-X11 backends.
25016 Don't try to create pixmaps if we have not requested that feature. This
25017 fixes execution for non-X11 backends, and most specifically DRM video
25020 2013-09-24 16:22:59 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25022 * ext/codecparsers:
25023 codecparsers: update to gst-vaapi-branch commit b33bd32.
25024 b33bd32 jpeg: fix and optimize scan for next marker code
25026 2013-09-23 19:14:56 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25028 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
25029 jpeg: fix calculation of MCU count.
25030 Fix calculation of MCU count for image sizes that are not a multiple
25031 of 8 pixels in either dimension, but also for non-common sampling
25032 factors like 4:2:2 in non-interleaved mode.
25034 2013-09-23 16:49:41 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25036 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
25037 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
25038 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
25039 jpeg: add support for multiscan images.
25040 Add support for images with multiple scans per frame. The Huffman table
25041 can be updated before SOS, and thus possibly requiring multiple uploads
25042 of Huffman tables to the VA driver. So, the latter must be able to cope
25043 with multiple VA buffers of type 'huffman-table' and with the correct
25046 2013-09-23 11:41:52 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25048 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
25049 jpeg: improve robustness when packets are missing.
25050 Improve robustness when some expected packets where not received yet
25051 or that were not correctly decoded. For example, don't try to decode
25052 a picture if there was no valid frame headers.
25054 2013-09-20 16:46:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25056 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
25057 jpeg: minor clean-ups.
25058 Improve debugging and error messages. Rename a few variables to fit the
25059 existing naming conventions. Change some fatal asserts to non-fatal
25062 2013-09-20 10:12:08 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25064 * gst-libs/gst/vaapi/gstvaapidecoder.c:
25065 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
25066 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
25067 jpeg: rework and optimize parser.
25068 Split the input buffer data into decoder units that represent a JPEG
25069 segment. Handle scan decoder unit specifically so that it can include
25070 both the scan header (SOS) but also any other ECS or RSTi segment.
25071 That way, we parse the input buffer stream only once at the gst-vaapi
25072 level instead of (i) in gst_vaapi_decoder_jpeg_parse() to split the
25073 stream into frames SOI .. EOI and (ii) in decode_buffer() to further
25074 determine segment boundaries and decode them.
25075 In practice, this is a +15 to +25% performance improvement.
25077 2013-09-17 14:29:54 +0800 Junfeng Xu <jun.feng.xu@intel.com>
25079 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
25080 jpeg: handle comment segments.
25081 Fix decode_buffer() function to gracefully skip comment (COM) segments.
25082 This fixes decoding of streams generated by certain cameras, e.g. like
25083 the Logitech Pro C920.
25084 https://bugzilla.gnome.org/show_bug.cgi?id=708208
25085 Signed-off-by: Junfeng Xu <jun.feng.xu@intel.com>
25087 2013-09-18 17:59:44 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25089 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
25090 jpeg: fix determination of image bounds.
25091 Look for the exact image bounds characterised by the <SOI> and <EOI>
25092 markers. Use the gst_jpeg_parse() codec parser utility function to
25093 optimize the lookup for the next marker segment.
25094 https://bugzilla.gnome.org/show_bug.cgi?id=707447
25096 2013-09-10 15:46:09 +0800 Junfeng Xu <jun.feng.xu@intel.com>
25098 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
25099 jpeg: fix calculation of offset to next marker segment.
25100 Fix calculation of the offset to the next marker segment since the
25101 correction of the codecparser part to match the API specification.
25102 i.e. the GstJpegMarkerSegment.size field represents the size in bytes
25103 of the segment minus any marker prefix.
25104 https://bugzilla.gnome.org/show_bug.cgi?id=707447
25105 Signed-off-by: Junfeng Xu <jun.feng.xu@intel.com>
25107 2013-09-20 18:30:18 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25109 * ext/codecparsers:
25110 codecparsers: update to gst-vaapi-branch commit 23c7dde.
25111 23c7dde jpeg: fix calculation of segment size
25113 2013-08-31 16:00:05 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25116 Bump version for development.
25118 2013-08-31 15:47:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25124 2013-08-31 15:46:25 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25129 2013-08-15 17:59:37 +0800 Wind Yuan <feng.yuan@intel.com>
25132 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
25133 configure: fix detection of VA/JPEG decoding API.
25134 Fix detection of VA/JPEG decoding API with non-standard libva packages.
25135 More precisely, some packages were shipping with a <va/va.h> header that
25136 did not include <va/va_dec_jpeg.h>.
25137 https://bugzilla.gnome.org/show_bug.cgi?id=706055
25138 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25140 2013-04-18 19:49:42 +0800 Zhao Halley <halley.zhao@intel.com>
25142 * gst/vaapi/gstvaapisink.c:
25143 vaapisink: ensure the uploader is setup for upstream allocated buffers.
25144 In GStreamer 0.10 builds, make sure that the GstVaapiUploader helper
25145 is setup in case upstream elements allocate buffers themselves without
25146 honouring our GstVaapiSink::bufer_alloc() hook.
25147 In particular, this fixes support for OGG video streams with WebKit.
25148 https://bugzilla.gnome.org/show_bug.cgi?id=703934
25149 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25151 2013-08-29 19:07:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25153 * gst/vaapi/gstvaapisink.c:
25154 vaapisink: simplify get_render_buffer() for GStreamer 0.10 builds.
25155 Implement and use gst_vaapisink_get_render_buffer() for GStreamer 0.10
25158 2013-08-29 18:34:57 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25160 * gst/vaapi/gstvaapisink.c:
25161 * gst/vaapi/gstvaapisink.h:
25162 vaapisink: handle raw buffers not created from VA video buffer pool.
25163 Handle raw video buffers that were not created from a VA video buffer
25164 pool. Use the generic GstVideo API to copy buffers in GStreamer 1.0.x
25165 builds instead of the GstVaapiUploader.
25166 https://bugs.freedesktop.org/show_bug.cgi?id=55818
25168 2013-08-29 19:33:02 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25170 * gst/vaapi/gstvaapidecode.c:
25171 vaapidecode: remove extraneous size information from allowed caps.
25172 Fix _getcaps() implementation to not report codecs with size information
25173 filled in the returned caps. That's totally useless nowadays. Ideally,
25174 this is a hint to insert a video parser element, thus allowing future
25175 optimizations, but this is not a strict requirement for gstreamer-vaapi,
25176 which is able to parse the elementary bitstreams itself.
25177 https://bugzilla.gnome.org/show_bug.cgi?id=704734
25179 2013-07-30 14:05:39 +0800 Guangxin.Xu <Guangxin.Xu@intel.com>
25181 * gst/vaapi/gstvaapidecode.c:
25182 * gst/vaapi/gstvaapidecode.h:
25183 vaapidecode: submit the last frame from output adapter to decoder.
25184 If there is no frame delimiter at the end of the stream, e.g. no
25185 end-of-stream or end-of-sequence marker, and that the current frame
25186 was fully parsed correctly, then assume that last frame is complete
25187 and submit it to the decoder.
25188 https://bugzilla.gnome.org/show_bug.cgi?id=705123
25189 Signed-off-by: Guangxin.Xu <Guangxin.Xu@intel.com>
25190 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25192 2013-08-29 11:55:05 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25194 * gst/vaapi/gstvaapidecode.c:
25195 * gst/vaapi/gstvaapidecode.h:
25196 vaapidecode: push all decoded frames from within the task.
25197 Make sure to push all decoded frames from the task so that the unlying
25198 VA surfaces could all be rendered from the same thread.
25200 2013-08-27 18:24:12 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25202 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
25203 wayland: render the raw surface if VPP failed.
25204 As a last resort, if video processing capabilities (VPP) are not available,
25205 or they did not produce anything conclusive enough, then try to fallback to
25206 the original rendering code path whereby the whole VA surface is rendered
25207 as is, no matter of video cropping or deinterlacing requests.
25208 Note: under those conditions, the visual outcome won't be correct but at
25209 least, something gets displayed instead of bailing out.
25211 2013-08-27 18:20:08 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25213 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
25214 wayland: add supporting for video processing.
25215 Try to use VA/VPP processing capabilities to handle video cropping and
25216 additional rendering flags that may not be directly supported by the
25217 underlying hardware when exposing a suitable Wayland buffer for the
25218 supplied VA surface. e.g. deinterlacing, different color primaries than
25221 2013-08-27 16:26:22 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25223 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
25224 wayland: add new frame redraw infrastructure.
25225 Update the frame redraw infrastructure with a new FrameState stucture
25226 holds all the necessary information used to display the next pending
25228 While we are at it, delay the sync operation down to when it is actually
25229 needed. That way, we keep performing additional tasks meanwhile.
25231 2013-08-27 18:06:10 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25233 * docs/reference/libs/libs-sections.txt:
25234 * gst-libs/gst/vaapi/gstvaapifilter.c:
25235 * gst-libs/gst/vaapi/gstvaapifilter.h:
25236 filter: allow specification of render target regions.
25237 Add support for rendering the source surface to a particular region within
25238 the supplied target surface. The default background color is black.
25240 2013-08-26 17:14:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25242 * gst/vaapi/gstvaapivideobuffer.c:
25243 decode: fix creation of GLX video buffers for GStreamer 0.10.
25244 Fix creation of GstVaapiVideoBuffer objects (i) to have that type for real;
25245 and (ii) to correctly extract the GstSurfaceConverter from the video buffer
25247 This fixes support for cluttersink with GStreamer 0.10 builds.
25249 2013-08-26 16:15:49 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25251 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
25252 mpeg2: disable video cropping as picture_display_extension() is missing.
25253 Disable video cropping in MPEG-2 codec because it is partially implemented
25254 and actually because nobody implements it that way, and the standard spec
25255 does not specify the display process either anyway.
25256 Most notably, there are two possible use cases for sequence_display_extension()
25257 horizontal_display_size & vertical_display_size: (i) guesstimating the
25258 pixel-aspect-ratio, or (ii) implement some kind of span & scan process
25259 in conjunction with picture_display_extension() information.
25260 https://bugzilla.gnome.org/show_bug.cgi?id=704848
25262 2013-08-16 16:58:58 +0100 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
25264 * gst/vaapi/gstvaapisink.c:
25265 * gst/vaapi/gstvaapisink.h:
25266 vaapisink: allow scaling to ignore aspect ratio.
25267 Other GStreamer sinks, like xvimagesink, have a force-aspect-ratio property,
25268 which allows you to say that you don't want the sink to respect aspect
25269 ratio. Add the same property to vaapisink.
25270 http://lists.freedesktop.org/archives/libva/2012-September/001298.html
25271 Signed-off-by: Simon Farnsworth <simon.farnsworth at onelan.co.uk>
25273 2013-05-14 15:19:04 +0800 Wind Yuan <feng.yuan@intel.com>
25275 * gst/vaapi/gstvaapisink.c:
25276 vaapisink: fix memory leak of GstVaapiUploader instance.
25277 Make sure gst_vaapisink_ensure_uploader() checks for the existence
25278 of a former GstVaapiUploader instance prior to forcibly creating a
25280 https://bugzilla.gnome.org/show_bug.cgi?id=703980
25282 2013-07-31 16:49:20 +0800 Guangxin.Xu <Guangxin.Xu@intel.com>
25284 * gst/vaapi/gstvaapisink.c:
25285 vaapisink: fix get_caps() implementation for GStreamer 1.0.
25286 Fix GstBaseSink::get_caps() implementation for GStreamer 1.0.X builds
25287 by honouring the filter caps argument. More precisely, this fixes the
25288 following pipeline: gst-launch-1.0 videotestsrc ! vaapisink
25289 https://bugzilla.gnome.org/show_bug.cgi?id=705192
25290 Signed-off-by: Guangxin.Xu <Guangxin.Xu@intel.com>
25291 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25293 2013-08-26 11:31:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25295 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
25296 mpeg4: fix double definition of GstVaapiDecoderMpeg4Class.
25297 This fixes the following issue:
25298 CC libgstvaapi_0.10_la-gstvaapidecoder_mpeg4.lo
25299 gstvaapidecoder_mpeg4.c:113: error: redefinition of typedef
25300 'GstVaapiDecoderMpeg4Class'
25301 gstvaapidecoder_mpeg4.c:44: note: previous declaration of
25302 'GstVaapiDecoderMpeg4Class' was here
25303 make[5]: *** [libgstvaapi_0.10_la-gstvaapidecoder_mpeg4.lo] Error 1
25304 make[5]: Leaving directory
25305 `/builddir/build/BUILD/gstreamer-vaapi-0.5.5.1/gst-libs/gst/vaapi'
25306 https://bugzilla.gnome.org/show_bug.cgi?id=705148
25308 2013-07-30 15:59:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25310 * tests/test-filter.c:
25311 tests: filter: add support for deinterlacing.
25312 Add --deinterlace option to enable deinterlacing through explicit VA/VPP
25313 deinterlacing filter. However, if --deinterlace option is not set but the
25314 --deinterlace-flags option is set with "top-field-first", then the very
25315 basic bob deinterlacing filter is set through VA/VPP proc pipeline flags.
25317 2013-07-17 17:29:41 +0800 Zhao Halley <halley.zhao@intel.com>
25319 * tests/test-filter.c:
25320 tests: filter: add support for denoising and sharpening.
25321 Add --denoise option to enable noise reduction with the level specified
25322 as the option value (float). Likewise, add --sharpen option to enable
25324 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25326 2013-07-24 14:31:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25328 * tests/test-filter.c:
25329 tests: filter: add support for frame cropping.
25330 Add support for frame cropping through the --crop-rect|-c argument.
25331 The format used is either <WIDTH> 'x' <HEIGHT>, with origin at (0,0) ;
25332 or full specification with '('? <X> ',' <Y> ')'? <WIDTH> 'x' <HEIGHT>.
25334 2013-07-23 18:00:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25336 * tests/test-filter.c:
25337 tests: filter: dump supported operations and formats.
25339 2013-07-08 16:54:55 +0800 Zhao Halley <halley.zhao@intel.com>
25341 * tests/Makefile.am:
25342 * tests/test-filter.c:
25343 tests: add initial test for video processing.
25344 Add minimal test case for video processing: scaling and color format
25346 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25348 2013-07-29 09:23:50 +0800 Zhao Halley <halley.zhao@intel.com>
25350 * docs/reference/libs/libs-sections.txt:
25351 * gst-libs/gst/vaapi/gstvaapifilter.c:
25352 * gst-libs/gst/vaapi/gstvaapifilter.h:
25353 * gst-libs/gst/vaapi/gstvaapiutils.c:
25354 * gst-libs/gst/vaapi/gstvaapiutils.h:
25355 * gst/vaapi/gstvaapipostproc.c:
25356 * gst/vaapi/gstvaapipostproc.h:
25357 filter: add initial support for deinterlacing.
25358 Add basic deinterlacing support, i.e. bob-deinterlacing whereby only
25359 the selected field from the input surface is kept for the target surface.
25360 Setting gst_vaapi_filter_set_deinterlacing() method argument to
25361 GST_VAAPI_DEINTERLACE_METHOD_NONE means to disable deinterlacing.
25362 Also move GstVaapiDeinterlaceMethod definition from vaapipostproc plug-in
25363 to libgstvaapi core library.
25364 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25366 2013-07-17 17:40:41 +0800 Zhao Halley <halley.zhao@intel.com>
25368 * docs/reference/libs/libs-sections.txt:
25369 * gst-libs/gst/vaapi/gstvaapifilter.c:
25370 * gst-libs/gst/vaapi/gstvaapifilter.h:
25371 filter: add support for color balance adjustment.
25372 Add ProcAmp (color balance) adjustments for hue, saturation, brightness
25373 and contrast. The respective range for each filter shall be the same as
25374 for the VA display attributes.
25375 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25377 2013-07-17 17:37:16 +0800 Zhao Halley <halley.zhao@intel.com>
25379 * docs/reference/libs/libs-sections.txt:
25380 * gst-libs/gst/vaapi/gstvaapifilter.c:
25381 * gst-libs/gst/vaapi/gstvaapifilter.h:
25382 filter: add support for sharpening.
25383 Sharpening is configured with a float value. The supported range is
25384 -1.0 .. 1.0 with 0.0 being the default, and that means no sharpening
25386 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25388 2013-07-17 17:29:41 +0800 Zhao Halley <halley.zhao@intel.com>
25390 * docs/reference/libs/libs-sections.txt:
25391 * gst-libs/gst/vaapi/gstvaapifilter.c:
25392 * gst-libs/gst/vaapi/gstvaapifilter.h:
25393 filter: add support for denoising.
25394 Noise reduction is configured with a float value. The supported range
25395 is 0.0 .. 1.0 with 0.0 being the default, and that means no denoise
25397 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25399 2013-07-24 14:22:28 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25401 * docs/reference/libs/libs-sections.txt:
25402 * gst-libs/gst/vaapi/gstvaapifilter.c:
25403 * gst-libs/gst/vaapi/gstvaapifilter.h:
25404 filter: add support for frame cropping.
25405 Frame cropping is defined with a GstVaapiRectangle value. The default
25406 behaviour is to treat the source surface as a whole
25408 2013-07-25 13:55:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25410 * gst-libs/gst/vaapi/gstvaapifilter.c:
25411 filter: add helper functions.
25412 Add helper functions to ensure an operation VA buffer is allocated to
25413 the right size; that filter caps get parsed and assigned to the right
25414 operation too; and that float parameters are correctly scaled to fit
25415 the reported range from the VA driver.
25417 2013-07-23 15:52:45 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25420 * docs/reference/libs/libs-docs.xml.in:
25421 * docs/reference/libs/libs-sections.txt:
25422 * gst-libs/gst/vaapi/Makefile.am:
25423 * gst-libs/gst/vaapi/gstvaapidisplay.c:
25424 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
25425 * gst-libs/gst/vaapi/gstvaapifilter.c:
25426 * gst-libs/gst/vaapi/gstvaapifilter.h:
25427 Add initial infrastructure for video processing.
25428 Add initial API for video processing: only scaling and color format
25429 conversion operations are supported.
25431 2013-07-24 11:53:38 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25433 * gst-libs/gst/vaapi/video-format.c:
25434 * gst-libs/gst/vaapi/video-format.h:
25435 libs: add gst_vaapi_video_format_from_string() helper.
25436 Add gst_vaapi_video_format_from_string() helper function to convert from
25437 a video format string representation to a suitable GstVideoFormat. This
25438 is just an alias to gst_video_format_from_string() for GStreamer 1.0.x
25439 builds, and a proper iteration over all GstVideoFormat string representations
25440 otherwise for earlier GStreamer 0.10.x builds.
25442 2013-07-24 11:37:23 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25444 * gst-libs/gst/vaapi/video-format.c:
25445 * gst-libs/gst/vaapi/video-format.h:
25446 libs: add gst_vaapi_video_format_from_va_fourcc() helper.
25447 Add gst_vaapi_video_format_from_va_fourcc() helper that converts from a
25448 VA fourcc value to a suitable GstVideoFormat.
25450 2013-07-24 11:41:05 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25452 * gst-libs/gst/vaapi/gstvaapivalue.c:
25453 * gst-libs/gst/vaapi/gstvaapivalue.h:
25454 libs: add type definitions for GstVaapiPoint and GstVaapiRectangle.
25455 Add helper functions to describe GstVaapiPoint and GstVaapiRectangle
25456 structures as a standard GType. This could be useful to have them
25457 described as a GValue later on.
25459 2013-07-26 13:57:35 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25461 * docs/reference/libs/libs-docs.xml.in:
25462 * docs/reference/libs/libs-sections.txt:
25463 * gst-libs/gst/vaapi/Makefile.am:
25464 * gst-libs/gst/vaapi/gstvaapicontext.h:
25465 * gst-libs/gst/vaapi/gstvaapidisplaycache.h:
25466 libs: drop some public APIs.
25467 Don't expose GstVaapiContext APIs and make them totally private to
25468 libgstvaapi core library. That API would also tend to disappear in
25469 a future revision. Likewise, don't expose GstVaapiDisplayCache API
25470 but keep symbols visible so that the various render backends could
25471 share a common display cache implementation in libgstvaapi.
25472 Try to clean-up the documentation from any stale entry too.
25474 2013-08-23 18:35:42 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25478 tests: image: allow creation of images with interleaved patterns.
25479 Add image_generate_full() function to create interleaved color rectangles.
25480 If flags is zero, the whole frame is generated with a unique pattern. If
25481 flags is non-zero, then each field is handled individually.
25483 2013-08-23 16:25:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25486 tests: image: fix conversion from RGB to YUV.
25487 Fix RGB to YUV conversion to preserve full data range.
25489 2013-07-26 13:12:28 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25492 tests: image: try to upload images through vaDeriveImage() too.
25493 On some platforms, vaPutImage() would fail even if it does not involve
25494 color format conversion or scaling, whereas copying raw pixels through
25495 vaDeriveImage() could work instead.
25497 2013-07-26 10:05:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25500 tests: image: add support for packed YUV formats.
25501 Add support for packed YUV 4:2:2 formats, i.e. YUY2 and UYVY.
25503 2013-07-25 18:10:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25506 tests: image: fix generation of I420/YV12 images.
25507 U/V planes were reversed, thus producing invalid images.
25509 2013-07-24 13:55:04 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25512 tests: image: fix string representation for GstVideoFormat.
25514 2013-07-26 12:57:19 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25516 * docs/reference/libs/libs-sections.txt:
25517 * gst-libs/gst/vaapi/Makefile.am:
25518 * gst-libs/gst/vaapi/gstvaapiimage.c:
25519 * gst-libs/gst/vaapi/gstvaapiimage.h:
25520 * gst-libs/gst/vaapi/gstvaapiimage_priv.h:
25521 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
25522 * gst-libs/gst/vaapi/gstvaapisurface.c:
25523 image: clean image API up.
25524 Don't expose functions that reference a GstVaapiImageRaw, those are
25525 meant to be internal only for implementing subpictures sync. Also add
25526 a few private definitions to avoid functions calls for retrieving
25527 image size and format information.
25529 2013-07-26 11:43:49 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25531 * docs/reference/libs/libs-sections.txt:
25532 * gst-libs/gst/vaapi/gstvaapiimage.c:
25533 * gst-libs/gst/vaapi/gstvaapiimage.h:
25534 image: add gst_vaapi_image_copy() helper.
25535 Add gst_vaapi_image_copy() helper function to copy images of same format
25538 2013-07-22 14:53:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25540 * gst/vaapi/gstvaapivideoconverter_x11.c:
25541 plugins: handle video cropping in X11 pixmap converter.
25542 Use GstVideoCropMeta in GStreamer 1.0 or any other render rectangle
25543 we could decode from the stream.
25545 2013-07-22 11:58:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25547 * gst/vaapi/Makefile.am:
25548 * gst/vaapi/gstvaapivideobuffer.c:
25549 * gst/vaapi/gstvaapivideoconverter_glx.c:
25550 * gst/vaapi/gstvaapivideoconverter_x11.c:
25551 * gst/vaapi/gstvaapivideoconverter_x11.h:
25552 plugins: add support for "x11-pixmap" video converter type.
25553 Install a new video converter that supports X11 pixmap targets for X11
25554 backends only, or make the GLX converter creation function chain up to
25555 the X11 converter whenever requested.
25557 2013-07-22 09:36:08 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25559 * tests/simple-decoder.c:
25560 tests: simple-decoder: add support for pixmap API.
25561 Add support for the new render-to-pixmap API. Avoid flickering on
25562 platforms supporting video overlay by keeping up to 2 intermediate
25565 2013-07-22 09:12:21 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25567 * tests/simple-decoder.c:
25568 tests: simple-decoder: add support for video cropping.
25569 Handle video cropping information attached to a VA surface proxy.
25571 2013-07-22 09:03:30 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25575 * tests/test-decode.c:
25576 tests: add support for render-to-pixmap.
25577 Add --pixmap option to test-decode so that to allow copies of VA
25578 surface to an intermediate pixmap and rendering from that pixmap.
25579 Only X11 backends are supported for now.
25581 2013-07-22 09:00:38 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25584 * gst-libs/gst/vaapi/Makefile.am:
25585 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
25586 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
25587 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
25588 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
25589 x11: implement pixmap rendering with RENDER extension.
25590 Use hardware accelerated XRenderComposite() function, from the RENDER
25591 extension, to blit a pixmap to screen. Besides, this can also support
25592 cropping and scaling.
25594 2013-07-19 15:05:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25596 * docs/reference/libs/libs-docs.xml.in:
25597 * docs/reference/libs/libs-sections.txt:
25598 * gst-libs/gst/vaapi/Makefile.am:
25599 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
25600 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
25601 * gst-libs/gst/vaapi/gstvaapipixmap_x11.c:
25602 * gst-libs/gst/vaapi/gstvaapipixmap_x11.h:
25603 x11: implement pixmap API.
25604 Implement the new render-to-pixmap API. The only supported pixmap format
25605 that will work is xRGB, with native byte ordering. Others might work but
25606 they were not tested.
25608 2013-07-22 10:10:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25610 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
25611 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
25612 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
25613 x11: update x11_get_geometry() helper function with depth output.
25614 Allow x11_get_geometry() utility function to also return the depth
25615 assigned to the X drawable.
25617 2013-07-22 10:00:21 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25619 * docs/reference/libs/libs-docs.xml.in:
25620 * docs/reference/libs/libs-sections.txt:
25621 * gst-libs/gst/vaapi/Makefile.am:
25622 * gst-libs/gst/vaapi/gstvaapipixmap.c:
25623 * gst-libs/gst/vaapi/gstvaapipixmap.h:
25624 * gst-libs/gst/vaapi/gstvaapipixmap_priv.h:
25625 * gst-libs/gst/vaapi/gstvaapiwindow.c:
25626 * gst-libs/gst/vaapi/gstvaapiwindow.h:
25627 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
25628 Add initial Pixmap API.
25629 Add API to transfer VA urfaces to native pixmaps. Also add an API to
25630 render a native pixmap, for completeness. In general, rendering to
25631 pixmap would only be useful to certain VA drivers and use cases on
25632 X11 display servers. e.g. GLX_EXT_texture_from_pixmap (TFP) handled
25635 2013-07-22 15:15:48 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25637 * docs/reference/libs/libs-sections.txt:
25638 * gst-libs/gst/vaapi/gstvaapiimage.c:
25639 * gst-libs/gst/vaapi/gstvaapisurface.c:
25640 * gst-libs/gst/vaapi/video-format.c:
25641 * gst-libs/gst/vaapi/video-format.h:
25642 libs: add and expose gst_vaapi_video_format_to_string() helper.
25643 This is just a wrapper over gst_video_format_to_string() for older
25644 GStreamer 0.10 builds.
25646 2013-07-18 02:54:54 -0300 Emilio López <emilio@elopez.com.ar>
25648 * gst/vaapi/gstvaapipluginutil.c:
25649 plugins: fix display type comparison in gst_vaapi_create_display().
25650 After the code got moved to create the gst_vaapi_create_display() helper,
25651 this comparison was not updated to dereference the newly-created
25652 pointer, so the code was comparing the pointer itself to the type, and
25653 therefore failing to retrieve the VA display.
25654 This fixes the following error (and gets gst-vaapi decoding again):
25655 ERROR vaapidecode gstvaapidecode.c:807:gst_vaapidecode_ensure_allowed_caps: failed to retrieve VA display
25656 https://bugzilla.gnome.org/show_bug.cgi?id=704410
25657 Signed-off-by: Emilio López <emilio@elopez.com.ar>
25659 2013-07-17 11:07:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25662 Bump version for development.
25664 2013-07-15 17:49:31 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25666 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
25667 mpeg2: don't output dummy pictures.
25668 Mark dummy pictures as output already so that we don't try to submit
25669 them to the upper layer since this is purely internal / temporary
25670 picture for helping the decoder.
25672 2013-07-15 17:43:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25674 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
25675 decoder: dispose GstVideoCodecFrame earlier.
25676 Once the picture was output, it is no longer necessary to keep an extra
25677 reference to the underlying GstVideoCodecFrame. So, we can release it
25678 earlier, and maybe subsequently release the associate surface proxy
25681 2013-07-15 14:47:01 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25687 2013-07-15 14:42:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25689 * docs/reference/libs/libs-sections.txt:
25690 * gst-libs/gst/vaapi/Makefile.am:
25691 * gst-libs/gst/vaapi/gstvaapidisplay.c:
25692 * gst-libs/gst/vaapi/gstvaapiimage.c:
25693 * gst-libs/gst/vaapi/gstvaapisurface.c:
25694 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
25695 * gst-libs/gst/vaapi/video-format.c:
25696 * gst-libs/gst/vaapi/video-format.h:
25697 * gst/vaapi/gstvaapidownload.c:
25698 * gst/vaapi/gstvaapiuploader.c:
25700 * tests/test-display.c:
25701 Fix new video format API.
25702 Fix new internal video format API, based on GstVideoFormat, to not
25703 clobber with system symbols. So replace the gst_video_format_* prefix
25704 with gst_vaapi_video_format_ prefix, even if the format type remains
25707 2013-07-15 14:05:45 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25710 Bump library major version.
25711 Bump the library major version due to API/ABI changes that occurred in
25712 the imaging API. In particular, GstVaapiImageFormat type was replaced
25713 with the standard GstVideoFormat type. All dependent APIs were updated
25714 to match this change.
25716 2013-07-15 13:44:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25721 2013-06-11 15:11:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25723 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
25724 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
25725 decoder: fix memory leak when processing interlaced pictures.
25726 Fix memory leak when processing interlaced pictures and that occurs
25727 because the first field, represented as a GstVideoCodecFrame, never
25728 gets released. i.e. when the picture is completed, this is generally
25729 the case when the second field is successfully decoded, we need to
25730 propagate the GstVideoCodecFrame of the first field to the original
25731 GstVideoDecoder so that it could reclaim memory.
25732 Otherwise, we keep accumulating the first fields into GstVideoDecoder
25733 private frames list until the end-of-stream is reached. The frames
25734 are eventually released there, but too late, i.e. too much memory
25735 may have been consumed.
25736 https://bugzilla.gnome.org/show_bug.cgi?id=701257
25738 2013-07-15 11:58:31 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25740 * gst/vaapi/gstvaapipluginutil.c:
25741 plugins: simlpify gst_vaapi_create_display() helper.
25742 Simplify gst_vaapi_create_display() helper as gst_vaapi_display_XXX_new()
25743 performs the necessary validation checks for the underlying VA display
25744 prior to returning to the caller. So, if an error occurred, then NULL is
25745 really returned in that case.
25747 2013-05-24 05:04:01 -0400 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
25749 * gst/vaapi/gstvaapipluginutil.c:
25750 plugins: add gst_vaapi_create_display() helper.
25751 https://bugzilla.gnome.org/show_bug.cgi?id=703235
25752 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25754 2013-07-12 17:47:07 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25756 * gst/vaapi/gstvaapivideobufferpool.c:
25757 plugins: don't reallocate pool allocator for the same caps.
25758 If the video buffer pool config doesn't have new caps, then it's not
25759 necessary to reinstantiate the allocator. That could be a costly
25760 operation as we could do some extra heavy checking in there.
25762 2013-07-12 17:14:49 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25764 * gst/vaapi/gstvaapivideomemory.c:
25765 plugins: fix ref counting of GstVaapiVideoMemory allocator.
25766 Fix reference counting issue whereby gst_memory_init() does not hold
25767 an extra reference to the GstAllocator. So, there could be situations
25768 where the last instance of GstVaapiVideoAllocator gets released before
25769 a dangling GstVaapiVideoMemory object, thus possibly leading to a crash.
25771 2013-07-12 15:15:07 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25773 * gst/vaapi/gstvaapiuploader.c:
25774 vaapiupload: use implicit color conversion to NV12.
25775 Always perform conversion of sources buffers to NV12 since this is
25776 the way we tested for this capability in ensure_allowed_caps(). This
25777 also saves memory bandwidth for further rendering. However, this may
25778 not preserve quality since the YUV buffers are down-sampled to 4:2:0.
25780 2013-07-12 15:01:01 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25782 * gst-libs/gst/vaapi/gstvaapivideopool.c:
25783 pool: fix deallocation of video pools.
25784 The queue of free objects to used was deallocated with g_queue_free_full().
25785 However, this convenience function shall only be used if the original queue
25786 was allocated with g_queue_new(). This caused memory corruption, eventually
25787 leading to a crash.
25788 The correct solution is to pair the g_queue_init() with the corresponding
25789 g_queue_clear(), while iterating over all free objects to deallocate them.
25791 2013-03-13 17:44:52 +0800 Wind Yuan <feng.yuan@intel.com>
25793 * gst/vaapi/gstvaapidownload.c:
25794 vaapidownload: fix src caps format error.
25795 This fixes direct linking of vaapidownload element to xvimagesink with
25796 VA drivers supporting vaGetImage() from the native VA surface format to
25797 a different VA image format. i.e. color conversion during download.
25798 http://bugzilla.gnome.org/show_bug.cgi?id=703937
25799 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25801 2013-07-11 18:26:37 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25803 * gst/vaapi/gstvaapidownload.c:
25804 vaapidownload: fix debug string for image formats.
25805 The image is now expressed as a standard GstVideoFormat, which is not
25806 a FOURCC but rather a regular enum value.
25807 This is a regression introduced in commit 09397fa.
25809 2013-04-24 10:39:03 +0800 Wind Yuan <feng.yuan@intel.com>
25811 * gst-libs/gst/vaapi/gstvaapiimage.c:
25812 image: add support for raw YUY2/UYVY image copies.
25813 Implement raw image copies for YUY2 format. Add support for UYVY format
25814 too, with the same copy function as for YUY2. Even though components
25815 ordering differs, copying line strides is essentially the same.
25816 https://bugzilla.gnome.org/show_bug.cgi?id=703939
25817 https://bugzilla.gnome.org/show_bug.cgi?id=703940
25818 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25820 2013-07-10 15:15:11 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25822 * gst/vaapi/gstvaapiuploader.c:
25823 plugins: clean-up video uploader helper.
25824 Fix gst_vaapi_uploader_get_buffer() to not assign caps since they
25825 were already negotiated beforehand, and they are not used from the
25826 buffer in upstream elements.
25827 Clean-up gst_vaapi_uploader_ensure_caps() to use the new image caps
25828 represented as a GstVideoInfo.
25830 2013-07-10 15:03:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25832 * gst/vaapi/gstvaapiuploader.c:
25833 plugins: use GstVideoInfo in video uploader helper.
25835 2013-07-10 10:34:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25837 * gst/vaapi/gstvaapivideomemory.c:
25838 plugins: allow creation of VA surfaces with explicit pixel format.
25839 Adapt GstVaapiVideoMemory allocator to support creation of VA surfaces
25840 with an explicit pixel format. This allows for direct rendering to
25841 VA surface memory from a software decoder.
25843 2013-07-10 14:20:30 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25845 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
25846 surface: fix surface pool creation with an explicit pixel format.
25847 Fix creation of surface pool objects to honour explicit pixel format
25848 specification. If this operation is not supported, then fallback to
25849 the older interface with chroma format.
25851 2013-07-10 13:58:55 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25853 * gst-libs/gst/vaapi/gstvaapisurface.c:
25854 surface: try to determine the underlying VA surface format.
25855 If a VA surface was allocated with the chroma-format interface, try to
25856 determine the underlying pixel format on gst_vaapi_surface_get_format(),
25857 or return GST_VIDEO_FORMAT_ENCODED if this is not a supported operation.
25859 2013-07-09 19:08:37 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25861 * docs/reference/libs/libs-sections.txt:
25862 * gst-libs/gst/vaapi/gstvaapisurface.c:
25863 * gst-libs/gst/vaapi/gstvaapisurface.h:
25864 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
25865 surface: allow creation with explicit pixel format.
25866 Make it possible to create VA surfaces with a specific pixel format.
25867 This is a new capability brought in by VA-API >= 0.34.0. If that
25868 capability is not built-in (e.g. using VA-API < 0.34.0), then
25869 gst_vaapi_surface_new_with_format() will return NULL.
25871 2013-07-10 09:48:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25873 * docs/reference/libs/libs-sections.txt:
25874 * gst-libs/gst/vaapi/video-format.c:
25875 * gst-libs/gst/vaapi/video-format.h:
25876 surface: add helper function to get chroma type from GstVideoFormat.
25877 Add gst_video_format_get_chroma_type() helper function to determine
25878 the GstVaapiChromaType from a standard GStreamer video format. It is
25879 possible to reconstruct that from GstVideoFormatInfo but it is much
25880 simpler (and faster?) to use the local GstVideoFormatMap table.
25882 2013-07-09 19:13:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25884 * gst-libs/gst/vaapi/gstvaapisurface.c:
25885 * gst-libs/gst/vaapi/gstvaapisurface.h:
25886 * gst-libs/gst/vaapi/gstvaapiutils.c:
25887 * gst-libs/gst/vaapi/gstvaapiutils.h:
25888 surface: add new chroma formats.
25889 Add new chroma formats available with VA-API >= 0.34.0. In particular,
25890 this includes "RGB" chroma formats, and more YUV subsampled formats.
25891 Also add a new from_GstVaapiChromaType() helper function to convert
25892 libgstvaapi chroma type to VA chroma format.
25894 2013-07-10 13:32:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25896 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
25897 pool: fix image pool to check for the video format to use.
25898 Make gst_vaapi_image_pool_new() succeed, and thus returning a valid
25899 image pool object, only if the underlying VA display does support the
25900 requested VA image format.
25902 2013-07-10 13:07:37 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25904 * gst-libs/gst/vaapi/gstvaapicontext.c:
25905 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
25906 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
25907 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
25908 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
25909 * gst-libs/gst/vaapi/gstvaapivideopool.h:
25910 * gst/vaapi/gstvaapidownload.c:
25911 * gst/vaapi/gstvaapiuploader.c:
25912 * tests/Makefile.am:
25913 * tests/test-surfaces.c:
25914 Use GstVideoInfo for video pools.
25915 Get rid of GstCaps to create surface/image pool, and use GstVideoInfo
25916 structures instead. Those are smaller, and allows for streamlining
25919 2013-07-09 18:03:36 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25921 * gst-libs/gst/vaapi/gstvaapiimage.c:
25922 * gst-libs/gst/vaapi/video-format.c:
25923 Add more video formats.
25924 Add new video format mappings to VA image formats:
25925 - YUV: packed YUV (YUY2, UYVY), grayscale (Y800) ;
25926 - RGB: 32-bit RGB without alpha channel (XRGB, XBGR, RGBX, BGRX).
25928 2013-07-10 15:52:20 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25930 * gst-libs/gst/vaapi/gstvaapiimage.c:
25931 image: fix debug message with video format.
25932 Fix debug message string with image format expressed with GstVideoFormat
25933 instead of the obsolete format that turned out to be a fourcc.
25934 This is a regression from git commit e61c5fc.
25936 2013-07-09 15:28:31 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25940 * tests/test-display.c:
25941 * tests/test-textures.c:
25942 * tests/test-windows.c:
25943 tests: port to new video format API.
25945 2013-07-09 15:44:35 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25947 * gst/vaapi/gstvaapidownload.c:
25948 * gst/vaapi/gstvaapiuploader.c:
25949 * gst/vaapi/gstvaapivideomemory.c:
25950 plugins: port to new video format API.
25952 2013-07-09 16:26:11 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25954 * gst-libs/gst/vaapi/gstvaapiimage.c:
25955 libs: use GstVideoInfo wherever possible.
25956 In particular, use gst_video_info_from_caps() helper function in VA image
25957 for implementating gst_vaapi_image_get_buffer() [vaapidownload] and
25958 gst_vaapi_image_update_from_buffer() [subpictures] in GStreamer 0.10 builds.
25960 2013-07-09 16:38:05 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25962 * docs/reference/libs/libs-docs.xml.in:
25963 * docs/reference/libs/libs-sections.txt:
25964 * gst-libs/gst/vaapi/Makefile.am:
25965 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
25966 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
25967 libs: drop GstVaapiImageFormat helpers.
25968 Drop GstVaapiImageFormat helpers since everything was moved to the new
25969 GstVideoFormat based API. Don't bother with backwards compatibility and
25970 just bump the library major version afterwards.
25972 2013-07-09 14:03:01 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25974 * gst-libs/gst/vaapi/gstvaapidisplay.c:
25975 * gst-libs/gst/vaapi/gstvaapidisplay.h:
25976 * gst-libs/gst/vaapi/gstvaapiimage.c:
25977 * gst-libs/gst/vaapi/gstvaapiimage.h:
25978 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
25979 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
25980 libs: port to new video format API.
25982 2013-07-09 15:29:59 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25984 * docs/reference/libs/libs-docs.xml.in:
25985 * docs/reference/libs/libs-sections.txt:
25986 * gst-libs/gst/vaapi/Makefile.am:
25987 * gst-libs/gst/vaapi/video-format.c:
25988 * gst-libs/gst/vaapi/video-format.h:
25989 Add new video format API.
25990 Leverage GstVideoFormat utilities from core GStreamer to provide an
25991 adaptation layer to VA image formats.
25993 2013-07-09 11:13:59 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
25998 2013-07-08 18:32:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26000 * gst/vaapi/gstvaapisink.c:
26001 vaapisink: fix creation of GLX texture.
26002 Fix creation of GLX texture, to not depend on the GstCaps video size that
26003 could be wrong, especially in presence of frame cropping. So, use the size
26004 from the source VA surfaces.
26005 An optimization could be to reduce the texture size to the actual visible
26006 size on screen. i.e. scale down the texture size to match the screen dimensions,
26007 while preserving the VA surface aspect ratio. However, some VA drivers don't
26010 2013-02-18 16:28:27 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
26012 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
26013 mpeg2: add support for video cropping.
26014 If the stream has a sequence_display_extenion, then attach the
26015 display_horizontal/display_vertical dimension as the cropping
26016 rectangle width/height to the GstVaapiPicture.
26017 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26019 2013-02-18 15:05:37 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
26021 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
26022 vc1: add support for video cropping.
26023 If the Advanced profile has display_extension fields, then set the display
26024 width/height dimension as cropping rectangle to the GstVaapiPicture.
26025 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26027 2013-02-15 18:50:26 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
26029 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
26030 h264: add support for video cropping.
26031 If the encoded stream has the frame_cropping_flag set, then associate
26032 the cropping rectangle to GstVaapiPicture.
26033 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26035 2013-07-08 17:01:21 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26039 * tests/test-decode.c:
26040 * tests/test-subpicture.c:
26041 tests: add basic support for video cropping.
26042 Change generic decoder of sample I-frame to return a GstVaapiSurfaceProxy
26043 instead of a plain GstVaapiSurface. This means that we can now retrieve
26044 the frame cropping rectangle from the surface proxy, along with additional
26045 information if ever needed.
26047 2013-07-08 14:50:42 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26049 * gst/vaapi/gstvaapidecode.c:
26050 * gst/vaapi/gstvaapisink.c:
26051 * gst/vaapi/gstvaapivideometa.c:
26052 plugins: add support for video cropping.
26053 Add support for GstVideoCropMeta in GStreamer >= 1.0.x builds and gst-vaapi
26054 specific meta information to hold video cropping details. Make the sink
26055 support video cropping in X11 and GLX modes.
26057 2013-02-15 18:24:24 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
26059 * gst/vaapi/gstvaapivideometa.c:
26060 * gst/vaapi/gstvaapivideometa.h:
26061 plugins: add helper functions to set the render rectangle.
26062 Some video clips may have a clipping region that needs to propogate to
26063 the renderer. These helper functions make it possible to attach that
26064 clipping region, as a GstVaapiRectangle, the the video meta associated
26066 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
26067 signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26069 2013-07-08 14:47:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26071 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
26072 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
26073 surfaceproxy: allow for NULL cropping rectangle.
26074 Make it possible associate an empty cropping rectangle to the surface
26075 proxy, thus resetting any cropping rectangle that was previously set.
26076 This allows for returning plain NULL when no cropping rectangle was
26077 initially set up to the surface proxy, or if it was reset to defaults.
26079 2013-07-08 11:41:59 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26081 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
26082 surfaceproxy: clean-up helper macros.
26083 Always use the GST_VAAPI_SURFACE_PROXY() helper macro to cast from a
26084 proxy macro argument to a GstVaapiSurfaceProxy pointer.
26086 2013-07-08 11:43:27 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26088 * gst-libs/gst/vaapi/gstvaapisurface.c:
26089 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
26090 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
26091 * gst-libs/gst/vaapi/gstvaapiwindow.c:
26092 surface: add simple surface info accessors as helper macros.
26093 Add helper macros to retrieve the VA surface information like size
26094 (width, height) or chroma type. This is a micro-optimization to avoid
26095 useless function calls and NULL pointer re-checks in internal routines.
26097 2013-02-15 18:42:12 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
26099 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
26100 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
26101 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
26102 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
26103 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
26104 decoder: add support for video cropping.
26105 Add gst_vaapi_picture_set_crop_rect() helper function to copy the video
26106 cropping information from raw bitstreams to each picture being decoded.
26107 Also add helper function to surface proxy to propagate that information
26108 outside of libgstvaapi. e.g. plug-in elements or standalone applications.
26109 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
26110 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26112 2013-07-08 17:30:30 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26114 * ext/codecparsers:
26115 codecparsers: update to gst-vaapi-branch commit f90de0a.
26116 f90de0a h264: fix calculation of the frame cropping rectangle
26117 535515c h264: parse the cropping rectangle separately
26119 2013-07-05 19:03:41 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26121 * ext/codecparsers:
26122 codecparsers: update to gst-vaapi-branch commit 0f68a71.
26123 0f68a71 mpeg2: fix video packet header size checks
26125 2013-06-07 20:08:43 +0800 Zhong Cong <congx.zhong@intel.com>
26127 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
26128 mpeg2: reset quantization matrices on new sequence headers.
26129 The MPEG-2 standard specifies (6.3.7) that all quantisation matrices
26130 shall be reset to their default values when a Sequence_Header() is
26132 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26134 2013-07-05 15:49:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26137 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
26138 mpeg2: cope with latest codecparser changes.
26139 Fix build with newer MPEG-2 codecparser where GstMpegVideoPacket are
26140 used in individual header parsers. Also use the new slice parsing API.
26142 2013-07-05 17:51:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26144 * ext/codecparsers:
26145 codecparsers: update to gst-vaapi-branch commit dddd182.
26146 dddd182 mpeg2: add slice header parsing API
26147 94e6228 mpeg2: add sequence scalable extension parsing API
26148 531134f mpeg2: add new API that takes GstMpegVideoPacket arguments
26149 4b135d3 h264: fix the return value type for the SEI palyload parsing methods
26151 2013-06-27 12:25:44 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
26153 * gst/vaapi/gstvaapisink.c:
26154 vaapisink: expose the raw video formats in static caps template.
26155 Expose all raw video formats in the static caps template since the
26156 vaapisink is supporting raw data. We will get the exact set of formats
26157 supported by the driver dynamically through the _get_caps() routine.
26158 This also fixes an inconsistency wrt. GStreamer 0.10 builds.
26159 https://bugzilla.gnome.org/show_bug.cgi?id=702178
26160 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26162 2013-06-27 13:53:46 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26164 * gst/vaapi/gstvaapisink.c:
26165 * gst/vaapi/gstvaapisink.h:
26166 vaapisink: add "use-glx" property for OpenGL rendering.
26167 Now that VA/GLX capable buffers are generated by default on X11, thus
26168 depending on a VA/GLX display, we stil want to use vaPutSurface() for
26169 rendering since it is faster.
26170 Anyway, OpenGL rendering in vaapisink was only meant for testing and
26171 enabling "fancy" effects to play with. This has no real value. So,
26172 disable OpenGL rendering by default.
26174 2013-06-06 05:36:03 -0400 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
26176 * gst/vaapi/gstvaapipluginutil.c:
26177 plugins: try to allocate a GLX display first over an X11 one.
26178 If the gstreamer-vaapi plug-in elements are built with GLX support, then
26179 try to allocate a GstVaapiDisplayGLX first before resorting to a VA/X11
26181 https://bugzilla.gnome.org/show_bug.cgi?id=701742
26183 2013-04-25 17:07:13 +0100 Lionel Landwerlin <llandwerlin@gmail.com>
26186 configure: use GST_PLUGIN_PATH_1_0 instead of GST_PLUGIN_PATH for Gst 1.0.
26187 jhbuild sets $GST_PLUGIN_PATH_1_0 which overrides $GST_PLUGIN_PATH.
26188 https://bugzilla.gnome.org/show_bug.cgi?id=698858
26189 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26191 2013-04-27 15:15:49 +0800 Wind Yuan <feng.yuan@intel.com>
26193 * gst-libs/gst/vaapi/gstvaapiimage.c:
26194 image: fix wrong check for rect bounds in copy_image().
26196 2013-06-14 13:41:14 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26199 Bump version for development.
26201 2013-06-14 11:47:50 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26207 2013-06-14 11:43:46 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26212 2013-06-14 11:39:54 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26215 * gst-libs/gst/codecparsers/Makefile.am:
26216 configure: always build the MPEG-4 parser.
26217 Always build the MPEG-4 parser for now as there are also core fixes
26218 included in the parser that cannot be tested for with API checks.
26220 2013-06-14 11:32:36 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26223 configure: add --enable-builtin-codecparsers [default="yes"] option.
26224 Add flag to have all codecparsers built-in, thus ensuring that the
26225 resulting binaries have all the necessary bug fixes and this is what
26226 the QA has been testing anyway.
26227 Of course, for a completely up-to-date Linux distribution, you could
26228 also opt for --disable-builtin-codecparsers and use the system ones.
26229 Though, some core fixes could be missing, and those cannot be tested
26230 for with API checks.
26232 2013-06-14 11:14:23 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26234 * ext/codecparsers:
26235 codecparsers: update to gst-vaapi-branch commit 843ce3e.
26236 843ce3e jpeg: fix default Huffman tables generation.
26237 8655187 mpeg2: fix the pixel-aspect-ratio calculation
26238 21099dc mpeg2: actually store video bitrate values
26239 dd02087 mpeg2: fix picture packet extension size check
26240 25948e9 mpeg2: increase min size for picture coding ext
26241 f1f5a40 ensure the debug category is properly initialized
26243 2013-06-12 14:16:17 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26245 * debian.upstream/Makefile.am:
26246 debian: fix list of generated files for .deb packaging.
26248 2013-06-12 13:48:26 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26251 * debian.upstream/Makefile.am:
26252 * debian.upstream/control.in:
26253 debian: fix libgstvaapi -dev package name.
26254 Fix libgstvaapi -dev package name so that to allow installation of both
26255 GStreamer 0.10 and 1.0.x based packages.
26257 2013-06-05 17:42:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26262 2013-05-31 11:09:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26264 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
26265 wayland: fix memory leak of display resources.
26267 2013-06-04 07:14:22 +0800 Zhao Halley <halley.zhao@intel.com>
26269 * gst/vaapi/gstvaapisink.c:
26270 vaapisink: fix build without VA/GLX support.
26272 2013-06-05 11:01:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26274 * gst/vaapi/gstvaapivideomemory.c:
26275 * gst/vaapi/gstvaapivideomemory.h:
26276 plugins: allow buffer mappings to GstVaapiSurfaceProxy.
26277 Allow plain gst_buffer_map() interface to work with gstreamer-vaapi
26278 video buffers, i.e. expose the underlying GstVaapiSurfaceProxy to the
26279 caller. This is the only sensible enough thing to do in this mode as
26280 the underlying surface pixels need to be extracted through an explicit
26281 call to the gst_video_frame_map() function instead.
26282 A possible use-case of this is to implement a "handoff" signal handler
26283 to fakesink or identity element for further processing.
26285 2013-06-03 10:22:44 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26287 * gst/vaapi/gstvaapivideomemory.c:
26288 plugins: silence check for direct-rendering mode in video memory.
26289 Fix gst_vaapi_video_allocator_new() to silently check for direct-rendering
26290 mode support, and not trigger fatal-criticals if either test surface or
26291 image could not be created. Typical case: pixel format mismatch, e.g. NV12
26292 supported by most hardware vs. I420 supported by most software decoders.
26294 2013-06-03 10:06:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26296 * gst/vaapi/gstvaapivideomemory.c:
26297 plugins: improve video memory flags safety checks.
26298 On map, ensure we have GST_MAP_WRITE flags since this is only what we
26299 support for now. Likewise, on unmap, make sure that the VA image is
26300 unmapped for either read or write, while still committing it to the
26301 VA surface if write was requested.
26303 2013-05-30 18:17:07 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26305 * gst-libs/gst/vaapi/gstvaapisurface.c:
26306 surface: fix memory leak through unreleased parent context.
26307 Break the circular references between GstVaapiContext and its children
26308 GstVaapiSurfaces. Since the VA surfaces held an extra reference to the
26309 context, which holds a reference to its VA surfaces, then none of those
26311 How does this impact support for subpictures?
26312 The only situation when the parent context needs to disappear is when
26313 it is replaced with another one because of a resolution change in the
26314 video stream for instance, or a normal destroy. In this case, it does
26315 not really matter to apply subpictures to the peer surfaces since they
26316 are either gone, or those that are left in the pipe can probably bear
26317 a reinstantiation of the subpictures for it.
26318 So, parent_context is set to NULL when the parent context is destroyed,
26319 other VA surfaces can still get subpictures attached to them, individually
26320 not as a whole. i.e. subpictures for surface S1 will be created from
26321 active composition buffers and associated to S1, subpictures for S2 will
26322 be created from the next active composition buffers, etc. We don't try
26323 to cache the subpictures in those cases (pending surfaces until EOS
26324 is reached, or pending surfaces until new surfaces matching new VA context
26325 get to be used instead).
26327 2013-05-27 14:01:48 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26329 * gst/vaapi/gstvaapisink.c:
26330 vaapisink: fix one-time initialization when display property is set.
26331 Fix gst_vaapisink_ensure_display() to perform one-time initialization
26332 tasks even if the `display' property was explicitly set.
26334 2013-05-27 15:59:08 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26336 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
26337 window: fix GLX window initialization.
26338 Make sure to create the GLX context once the window object has completed
26339 its creation. Since gl_resize() relies on the newly created window size,
26340 then we cannot simply overload the GstVaapiWindowClass::create() hook.
26341 So, we just call into gst_vaapi_window_glx_ensure_context() once the
26342 window object is created in the gst_vaapi_window_glx_new*() functions.
26344 2013-05-27 17:18:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26346 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
26347 * gst-libs/gst/vaapi/gstvaapidisplay_drm_priv.h:
26348 * gst-libs/gst/vaapi/gstvaapidisplay_glx_priv.h:
26349 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
26350 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
26351 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
26352 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
26353 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
26354 * gst-libs/gst/vaapi/gstvaapitexture.c:
26355 * gst-libs/gst/vaapi/gstvaapiwindow_drm.c:
26356 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
26357 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
26358 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
26359 display: validate display types.
26361 2013-05-27 16:13:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26363 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
26364 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
26365 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
26366 display: drop internal NAME_PREFIX, store the real display name.
26367 Always store a valid display name/device path, instead of adding a
26368 particular prefix. i.e. make it simply a strdup(), or "" if it was
26371 2013-05-27 13:17:31 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26373 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
26374 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
26375 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
26376 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
26377 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
26378 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
26379 * gst-libs/gst/vaapi/gstvaapidisplaycache.h:
26380 display: make it possible to lookup the display cache by type.
26381 Make it possible to add extra an extra filter to most of display cache
26382 lookup functions so that the GstVaapiDisplay instance can really match
26383 a compatible and existing display by type, instead of relying on extra
26384 string tags (e.g. "X11:" prefix, etc.).
26386 2013-05-24 16:19:23 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26388 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
26389 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
26390 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
26391 display: cope with new display cache API.
26393 2013-05-24 16:12:01 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26395 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
26396 * gst-libs/gst/vaapi/gstvaapidisplaycache.h:
26397 display: rework display cache API.
26398 Simplify display cache API, while making it more flexible. We can now create
26399 custom lookup functions with gst_vaapi_display_cache_lookup_custom().
26401 2013-05-24 15:05:45 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26403 * tests/test-display.c:
26404 tests: improve check for display cache.
26405 Improve check for display cache infrastructure. In particular, for X11 and
26406 GLX backends, we need to make sure that we can create a GstVaapiDisplayX11
26407 from another GstVaapiDisplayGLX, i.e. underlying X11 and VA displays can be
26408 shared. Besides, allocating a GstVaapiDisplayGLX while a GstVaapiDisplayX11
26409 already exists will have to generate different VA displays.
26411 2013-05-15 10:33:16 +0800 Zhao Halley <halley.zhao@intel.com>
26413 * gst/vaapi/gstvaapiuploader.c:
26414 uploader: fix memory leak in GStreamer 0.10 builds.
26415 In GStreamer 0.10 builds, gst_vaapi_uploader_get_buffer() was used
26416 but it exhibited a memory leak because the surface generated for the
26417 GstVaapiVideoMeta totally lost its parent video pool. So, it was not
26418 possible to release that surface back to the parent pool when the meta
26419 gets released, and the memory consumption kept growing.
26420 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26422 2013-05-23 18:56:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26424 * gst/vaapi/gstvaapivideometa.c:
26425 plugins: fix gst_vaapi_video_meta_new_from_pool().
26426 Since GST_VAAPI_IS_xxx_VIDEO_POOL() was only testing for NULL and not
26427 the underlying object type, the gst_vaapi_video_meta_new_from_pool()
26428 was hereby totally broken. Fixed this regression by using the newly
26429 provided gst_vaapi_video_pool_get_object_type() function.
26431 2013-05-23 18:22:50 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26433 * gst/vaapi/gstvaapiuploader.c:
26434 * gst/vaapi/gstvaapivideomemory.c:
26435 * gst/vaapi/gstvaapivideometa.c:
26436 plugins: cope with GST_VAAPI_IS_xxx() macros removal.
26438 2013-05-23 18:19:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26441 tests: cope with GST_VAAPI_IS_xxx() macros removal.
26443 2013-05-23 18:45:23 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26445 * docs/reference/libs/libs-sections.txt:
26446 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
26447 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
26448 * gst-libs/gst/vaapi/gstvaapivideopool.c:
26449 * gst-libs/gst/vaapi/gstvaapivideopool.h:
26450 * gst-libs/gst/vaapi/gstvaapivideopool_priv.h:
26451 libs: add query for GstVaapiVideoPool object types.
26452 Add API to identify the underlying GstVaapiVideoPool object type.
26454 2013-05-23 18:15:48 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26456 * docs/reference/libs/libs-sections.txt:
26457 * gst-libs/gst/vaapi/gstvaapicontext.c:
26458 * gst-libs/gst/vaapi/gstvaapicontext.h:
26459 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26460 * gst-libs/gst/vaapi/gstvaapidecoder.h:
26461 * gst-libs/gst/vaapi/gstvaapidisplay.c:
26462 * gst-libs/gst/vaapi/gstvaapidisplay.h:
26463 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
26464 * gst-libs/gst/vaapi/gstvaapidisplay_drm.h:
26465 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
26466 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
26467 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.h:
26468 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
26469 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
26470 * gst-libs/gst/vaapi/gstvaapiimage.c:
26471 * gst-libs/gst/vaapi/gstvaapiimage.h:
26472 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
26473 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
26474 * gst-libs/gst/vaapi/gstvaapiobject.c:
26475 * gst-libs/gst/vaapi/gstvaapiobject.h:
26476 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
26477 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
26478 * gst-libs/gst/vaapi/gstvaapisurface.c:
26479 * gst-libs/gst/vaapi/gstvaapisurface.h:
26480 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
26481 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
26482 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
26483 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
26484 * gst-libs/gst/vaapi/gstvaapitexture.c:
26485 * gst-libs/gst/vaapi/gstvaapivideopool.c:
26486 * gst-libs/gst/vaapi/gstvaapivideopool.h:
26487 * gst-libs/gst/vaapi/gstvaapiwindow.c:
26488 * gst-libs/gst/vaapi/gstvaapiwindow.h:
26489 * gst-libs/gst/vaapi/gstvaapiwindow_drm.h:
26490 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
26491 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
26492 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
26493 libs: drop GST_VAAPI_IS_xxx() helper macros.
26494 Drop obsolete GST_VAAPI_IS_xxx() helper macros since we are no longer
26495 deriving from GObject and so those were only checking for whether the
26496 argument was NULL or not. This is now irrelevant, and even confusing
26497 to some extent, because we no longer have type checking.
26498 Note: this incurs more type checking (review) but the libgstvaapi is
26499 rather small, so this is manageable.
26501 2013-05-07 18:52:28 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26504 Bump library major version.
26505 The whole libgstvaapi libraries got a major refresh to get rid of GObject.
26506 This is a fundamental change that requires a new SONAME. More changes are
26507 underway to streamline the core libraries.
26508 So far, the net result is a reduction of .text size (code) by 32KB, i.e. -10%.
26509 On one particular test (sintel HD trailer), the total number of executed
26510 instruction was reduced by 8%.
26512 2013-05-07 18:37:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26514 * docs/reference/libs/Makefile.am:
26515 * docs/reference/libs/libs-docs.xml.in:
26516 * docs/reference/libs/libs-sections.txt:
26517 * docs/reference/libs/libs.core.types:
26518 * docs/reference/libs/libs.glx.types:
26519 * docs/reference/libs/libs.x11.types:
26520 docs: cope with removed APIs.
26521 Some APIs are dead because they are no longer based on GObject.
26523 2013-05-06 14:43:38 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26525 * gst/vaapi/gstvaapidecode.c:
26526 * gst/vaapi/gstvaapidownload.c:
26527 * gst/vaapi/gstvaapipluginutil.c:
26528 * gst/vaapi/gstvaapipostproc.c:
26529 * gst/vaapi/gstvaapisink.c:
26530 * gst/vaapi/gstvaapiupload.c:
26531 * gst/vaapi/gstvaapiuploader.c:
26532 * gst/vaapi/gstvaapivideobufferpool.c:
26533 * gst/vaapi/gstvaapivideoconverter_glx.c:
26534 * gst/vaapi/gstvaapivideomemory.c:
26535 * gst/vaapi/gstvaapivideometa.c:
26536 plugins: cope with new GstVaapiMiniObject objects.
26538 2013-05-07 11:45:10 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26543 * tests/simple-decoder.c:
26544 * tests/test-decode.c:
26545 * tests/test-display.c:
26546 * tests/test-subpicture.c:
26547 * tests/test-surfaces.c:
26548 * tests/test-textures.c:
26549 * tests/test-windows.c:
26550 tests: cope with new GstVaapiMiniObject objects.
26552 2013-05-07 15:38:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26554 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
26555 display: fix set_synchronous() to lock display.
26557 2013-05-03 19:02:23 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26559 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
26560 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
26561 videopool: simplify creation of video objects pool.
26563 2013-05-07 18:17:10 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26565 * gst-libs/gst/vaapi/gstvaapiobject.c:
26566 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
26567 * gst-libs/gst/vaapi/gstvaapitypes.h:
26568 libs: simplify GstVaapiID definitions.
26569 Make GstVaapiID a gsize instead of guessing an underlying integer large
26570 enough to hold all bits of a pointer. Also drop GST_VAAPI_ID_NONE since
26571 this is plain zero and that it is no longer passed as varargs.
26573 2013-05-02 16:11:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26575 * gst-libs/gst/vaapi/Makefile.am:
26576 * gst-libs/gst/vaapi/gstvaapi_priv.h:
26577 * gst-libs/gst/vaapi/gstvaapiparamspecs.c:
26578 * gst-libs/gst/vaapi/gstvaapiparamspecs.h:
26579 * gst-libs/gst/vaapi/gstvaapivalue.c:
26580 * gst-libs/gst/vaapi/gstvaapivalue.h:
26581 libs: drop obsolete function helpers and objects.
26582 Drop obsolete GstVaapiID related function helpers for passing them as
26585 2013-05-07 11:39:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26587 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26588 * gst-libs/gst/vaapi/gstvaapidisplay.c:
26589 * gst-libs/gst/vaapi/gstvaapidisplay.h:
26590 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
26591 * gst-libs/gst/vaapi/gstvaapidisplay_drm.h:
26592 * gst-libs/gst/vaapi/gstvaapidisplay_drm_priv.h:
26593 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
26594 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
26595 * gst-libs/gst/vaapi/gstvaapidisplay_glx_priv.h:
26596 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
26597 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
26598 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.h:
26599 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
26600 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
26601 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
26602 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
26603 * gst-libs/gst/vaapi/gstvaapiobject.c:
26604 * gst-libs/gst/vaapi/gstvaapivideopool.c:
26605 libs: use GstVaapiMiniObject for display objects.
26607 2013-05-06 14:07:17 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26609 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
26610 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26611 * gst-libs/gst/vaapi/gstvaapidecoder.h:
26612 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
26613 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
26614 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
26615 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.h:
26616 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
26617 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.h:
26618 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
26619 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.h:
26620 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
26621 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
26622 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
26623 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.h:
26624 libs: use GstVaapiMiniObject for video decoders.
26625 Port GstVaapiDecoder and GstVaapiDecoder{MPEG2,MPEG4,JPEG,H264,VC1} to
26626 GstVaapiMiniObject. Add gst_vaapi_decoder_set_codec_state_changed_func()
26627 helper function to let the user add a callback to a function triggered
26628 whenever the codec state (e.g. caps) changes.
26630 2013-05-03 11:01:12 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26632 * docs/reference/libs/libs-sections.txt:
26633 * gst-libs/gst/vaapi/Makefile.am:
26634 * gst-libs/gst/vaapi/gstvaapicontext.c:
26635 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
26636 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
26637 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
26638 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
26639 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
26640 * gst-libs/gst/vaapi/gstvaapivideopool.c:
26641 * gst-libs/gst/vaapi/gstvaapivideopool.h:
26642 * gst-libs/gst/vaapi/gstvaapivideopool_priv.h:
26643 libs: use GstVaapiMiniObject for video object pools.
26644 Port GstVaapiVideoPool, GstVaapiSurfacePool and GstVaapiImagePool to
26645 GstVaapiMiniObject. Drop gst_vaapi_video_pool_get_caps() since it was
26646 no longer used for a long time. Make object allocators static, i.e.
26647 local to the shared library.
26649 2013-04-30 17:22:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26651 * gst-libs/gst/vaapi/gstvaapitexture.c:
26652 * gst-libs/gst/vaapi/gstvaapitexture.h:
26653 libs: use GstVaapiObject for texture objects.
26655 2013-04-30 17:20:14 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26657 * gst-libs/gst/vaapi/Makefile.am:
26658 * gst-libs/gst/vaapi/gstvaapiwindow.c:
26659 * gst-libs/gst/vaapi/gstvaapiwindow.h:
26660 * gst-libs/gst/vaapi/gstvaapiwindow_drm.c:
26661 * gst-libs/gst/vaapi/gstvaapiwindow_drm.h:
26662 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
26663 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
26664 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
26665 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
26666 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.h:
26667 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
26668 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
26669 * gst-libs/gst/vaapi/gstvaapiwindow_x11_priv.h:
26670 libs: use GstVaapiObject for window objects.
26672 2013-04-30 17:22:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26674 * gst-libs/gst/vaapi/gstvaapicontext.c:
26675 * gst-libs/gst/vaapi/gstvaapicontext.h:
26676 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26677 * gst-libs/gst/vaapi/gstvaapiimage.c:
26678 * gst-libs/gst/vaapi/gstvaapiimage.h:
26679 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
26680 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
26681 * gst-libs/gst/vaapi/gstvaapisurface.c:
26682 * gst-libs/gst/vaapi/gstvaapisurface.h:
26683 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
26684 * gst-libs/gst/vaapi/gstvaapivideopool.c:
26685 libs: use GstVaapiObject for VA objects.
26687 2013-04-30 17:20:46 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26689 * gst-libs/gst/vaapi/Makefile.am:
26690 * gst-libs/gst/vaapi/gstvaapiobject.c:
26691 * gst-libs/gst/vaapi/gstvaapiobject.h:
26692 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
26693 Port GstVaapiObject to GstVaapiMiniObject.
26695 2013-04-30 10:28:30 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26697 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
26698 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
26699 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
26700 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
26701 * gst-libs/gst/vaapi/gstvaapiminiobject.c:
26702 * gst-libs/gst/vaapi/gstvaapiminiobject.h:
26703 libs: refine GstVaapiMiniObject.
26704 Drop support for user-defined data since this capability was not used
26705 so far and GstVaapiMiniObject represents the smallest reference counted
26706 object type. Add missing GST_VAAPI_MINI_OBJECT_CLASS() helper macro.
26707 Besides, since GstVaapiMiniObject is a libgstvaapi internal object, it
26708 is also possible to further simplify the layout of the object. i.e. merge
26709 GstVaapiMiniObjectBase into GstVaapiMiniObject.
26711 2013-05-07 16:43:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26713 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26714 decoder: update picture size from the bitstream.
26715 Propagate the picture size from the bitstream to the GstVaapiDecoder,
26716 and subsequent user who installed a signal on notify::caps. This fixes
26717 decoding of TS streams when the demuxer failed to extract the required
26720 2013-04-25 14:16:01 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26722 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26723 decoder: fix raw decoding mode.
26724 Fix gst_vaapi_decoder_get_surface() to actually transfer ownership of the
26725 surface proxy to the caller.
26727 2013-04-25 13:56:18 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26729 * docs/reference/libs/libs-sections.txt:
26730 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26731 * gst-libs/gst/vaapi/gstvaapidecoder.h:
26732 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
26733 * gst/vaapi/gstvaapidecode.c:
26734 decoder: add gst_vaapi_decoder_get_frame_with_timeout().
26735 Add gst_vaapi_decoder_get_frame_with_timeout() helper function that will
26736 wait for a frame to be decoded, until the specified timeout in microseconds,
26737 prior to returning to the caller.
26738 This is a fix to performance regression from 851cc0, whereby the vaapidecode
26739 loop executed on the srcpad task was called to often, thus starving all CPU
26742 2013-04-19 14:38:59 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26745 Bump version for development.
26747 2013-04-18 19:09:45 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26753 2013-04-18 19:08:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26758 2013-04-18 15:55:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26760 * gst/vaapi/gstvaapidecode.c:
26761 vaapidecode: rework heuristics to detect decode timeout.
26762 Rework heuristics to detect when downstream element ran into errors,
26763 and thus failing to release any VA surface in due time for the current
26764 frame to get decoded. In particular, recalibrate the render time base
26765 when the first frame gets submitted downstream, or when there is no
26766 timestamp that could be inferred.
26768 2013-04-18 15:50:02 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26770 * gst-libs/gst/vaapi/gstcompat.h:
26771 * gst/vaapi/gstvaapidecode.c:
26772 vaapidecode: rework GstVideoDecoder::handle_frame() with a task.
26773 Rework GstVideoDecoder::handle_frame() to decode the current frame,
26774 while possibly waiting for a free surface, and separately submit all
26775 decoded frames from a task. This makes it possible to pop and render
26776 decoded frames as soon as possible.
26778 2013-04-18 10:06:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26780 * gst/vaapi/gstvaapidecode.c:
26781 * gst/vaapi/gstvaapidownload.c:
26782 * gst/vaapi/gstvaapisink.c:
26783 * gst/vaapi/gstvaapiupload.c:
26784 plugins: use gst_object_unref() wherever applicable.
26785 Use gst_object_unref() wherever applicable, e.g. objects derived from
26786 GstElement, GstVideoPool, etc.
26788 2013-04-17 14:21:16 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26790 * docs/reference/plugins/plugins-docs.xml.in:
26791 * docs/reference/plugins/plugins-sections.txt:
26792 * docs/reference/plugins/plugins.types:
26793 docs: drop obsolete plug-ins.
26794 Drop documentation for obsolete plug-ins, even for GStreamer 0.10.
26795 i.e. vaapiupload and vaapidownload are no longer the recommended
26798 2013-04-17 13:17:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26800 * debian.upstream/rules:
26801 debian: fix build of GStreamer 0.10 packages.
26802 Fix build of Debian packages to scan the actual GStreamer API version
26803 from the generated changelog file.
26805 2013-04-17 10:58:04 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26807 * gst/vaapi/gstvaapipostproc.c:
26808 vaapipostproc: minor clean-ups.
26809 Use g_clear_object() wherever appropriate and remove dead-code.
26811 2013-04-17 10:53:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26813 * gst/vaapi/gstvaapipostproc.c:
26814 vaapipostproc: fix reference counting buf for passthrough mode.
26815 Fix reference counting bug for passthrough mode, whereby the input buffer
26816 was propagated as is downstream through gst_pad_push() without increasing
26817 its reference count before. The was a problem when gst_pad_push() returns
26818 an error and we further decrease the reference count of the input buffer.
26820 2013-04-17 10:18:45 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26822 * gst-libs/gst/vaapi/gstcompat.h:
26823 * gst/vaapi/Makefile.am:
26824 * gst/vaapi/gstvaapi.c:
26825 * gst/vaapi/gstvaapidecode.c:
26826 * gst/vaapi/gstvaapipluginutil.h:
26827 * gst/vaapi/gstvaapipostproc.c:
26828 vaapipostproc: port to GStreamer 1.0.
26829 Add support for interlaced streams with GStreamer 1.0 too. Basically,
26830 this enables vaapipostproc, though it is not auto-plugged yet. We also
26831 make sure to reply to CAPS queries, and happily handle CAPS events.
26833 2013-04-17 10:14:55 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26835 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26836 decoder: fix GstVideoCodecFrame flags for interlaced contents.
26837 Fix support for interlaced contents with GStreamer 0.10. In particular,
26838 propagate GstVaapiSurfaceProxy frame flags to GstVideoCodecFrame flags
26840 This is a regression from commit 87e5717.
26842 2013-04-16 13:23:41 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26844 * gst-libs/gst/vaapi/Makefile.am:
26845 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26846 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
26847 * gst-libs/gst/vaapi/gstvaapiparser_frame.c:
26848 * gst-libs/gst/vaapi/gstvaapiparser_frame.h:
26849 decoder: rename GstVaapiDecoderFrame to GstVaapiParserFrame.
26850 Rename GstVaapiDecoderFrame to GstVaapiParserFrame because this data
26851 structure was only useful to parsing and a proper GstvaapiDecoderFrame
26852 instance will be created instead.
26854 2013-04-16 19:09:30 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26856 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26857 decoder: export presentation timestamp for raw decoding mode.
26858 Fix regression from 0.4-branch whereby GstVaapiSurfaceProxy no longer
26859 held any information about the expected presentation timestamp, frame
26860 duration or additional flags like interlaced or top-field-first.
26862 2013-04-16 18:56:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26864 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26865 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
26866 decoder: use new GstVaapiSurfaceProxy utility functions.
26867 Use new GstVaapiSurfaceProxy internal helper functions to propagate the
26868 necessary GstVideoCodecFrame flags to vaapidecode (GStreamer 0.10).
26869 Also make GstVaapiDecoder push_frame() operate similarly to drop_frame().
26870 i.e. increase the GstVideoCodecFrame reference count in push_frame rather
26871 than gst_vaapi_picture_output().
26873 2013-04-16 18:35:48 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26875 * docs/reference/libs/libs-sections.txt:
26876 * gst-libs/gst/vaapi/Makefile.am:
26877 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
26878 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
26879 * gst-libs/gst/vaapi/gstvaapisurfaceproxy_priv.h:
26880 surfaceproxy: add more attributes for raw decoding modes.
26881 Add more attributes for raw decoding modes, i.e. directly through the
26882 libgstvaapi helper library. In particular, add presentation timestamp,
26883 duration and a couple of flags (interlaced, TFF, RFF, one-field).
26885 2013-04-16 13:48:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26887 * docs/reference/libs/libs-sections.txt:
26888 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
26889 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
26890 * gst/vaapi/gstvaapidecode.c:
26891 * tests/simple-decoder.c:
26892 surfaceproxy: drop user-data support from GstVaapiSurfaceProxy.
26893 Drop user-data support from GstVaapiSurfaceProxy. Rather make it explicit
26894 to call some user-provided function when the surface proxy is released.
26896 2013-04-15 12:52:51 +0400 Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
26899 * gst-libs/gst/vaapi/Makefile.am:
26900 build: link libgstvaapi-glx-1.0.so against libdl.
26901 Ensure libgstvaapi-glx*.so builds against libdl since dlsym() is used
26902 to resolve glXGetProcAddress() from GLX libraries. This fix builds on
26904 https://bugzilla.gnome.org/show_bug.cgi?id=698046
26905 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26907 2013-04-15 14:22:57 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26909 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26910 decoder: fix gst_vaapi_decoder_get_codec_state().
26911 Fix previous commit whereby gst_vaapi_decoder_get_codec_state() was
26912 supposed to make GstVaapiDecoder own the return GstVideoCodecState
26913 object. Only comment was updated, not the actual code.
26915 2013-04-15 13:58:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26917 * gst-libs/gst/vaapi/gstvaapidecoder.c:
26918 * gst/vaapi/gstvaapidecode.c:
26919 decoder: make gst_vaapi_decoder_get_codec_state() return the original state.
26920 Make gst_vaapi_decoder_get_codec_state() return the original codec state,
26921 i.e. make the GstVaapiDecoder object own the return state so that callers
26922 that want an extra reference to it would just gst_video_codec_state_ref()
26923 it before usage. This aligns the behaviour with what we had before with
26924 gst_vaapi_decoder_get_caps().
26925 This is an ABI incompatible change, library major version was bumped from
26926 previous release (0.5.2).
26928 2013-04-15 13:52:19 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26930 * gst/vaapi/gstvaapivideobufferpool.h:
26931 * gst/vaapi/gstvaapivideoconverter_glx.h:
26932 plugins: mark a few more functions as internal.
26933 Mark the following functions are internal, i.e. private to the vaapi plug-in:
26934 - gst_vaapi_video_buffer_pool_get_type()
26935 - gst_vaapi_video_converter_glx_get_type()
26936 - gst_vaapi_video_converter_glx_new()
26938 2013-04-15 13:48:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26940 * gst/vaapi/gstvaapivideobuffer.c:
26941 plugins: implement GstSurfaceMeta API.
26942 Implement GstSurfaceMeta API for GStreamer 1.0.x. Even though this is
26943 an unstable/deprecated API, this makes it possible to support Clutter
26944 sink with minimal changes. Tested against clutter-gst 1.9.92.
26946 2013-04-12 17:12:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26948 * gst/vaapi/gstvaapisink.c:
26949 vaapisink: optimize GstVideoOverlayInterface::expose().
26950 When render-mode is "overlay", then it is not really useful to peek into
26951 the GstBaseSink::last_buffer, since we have our own video_buffer already
26952 recorded and maintained into GstVaapiSink.
26954 2013-04-12 17:05:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26956 * gst/vaapi/gstvaapisink.c:
26957 vaapisink: fix memory leak of GstSample objects.
26958 Fix memory leak of GstSample objects in GstVideoOverlayInterface::expose().
26959 This also fixes extra unreferencing of the underlying GstBuffer in the common
26960 path afterwards (for both 0.10 or 1.0).
26962 2013-04-12 13:44:52 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26964 * gst-libs/gst/vaapi/gstcompat.h:
26965 * gst/vaapi/gstvaapi.c:
26966 plugins: fix description for gst-inspect.
26967 Fix the name of the plug-in element reported to gst-inspect-1.0. i.e. we
26968 need an explicit definition for GStreamer >= 1.0 because the GST_PLUGIN_DEFINE
26969 incorrectly uses #name for creating the plug-in name, instead of using macro
26970 expansion (and let further expansion of macros) through e.g. G_STRINGIFY().
26972 2013-04-11 09:24:44 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26976 Update build requirements for GStreamer 1.0.x support. Add section for
26977 ways to report bugs.
26979 2013-04-10 16:54:01 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26984 2013-04-10 15:31:41 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26986 * gst-libs/gst/vaapi/Makefile.am:
26987 * gst/vaapi/Makefile.am:
26988 Fix make dist to include all source files, in any case.
26989 Fix make dist to allow build for either GStreamer 0.10 or 1.0. i.e. make
26990 sure to include all source files in either case while generating source
26993 2013-04-10 15:21:57 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
26996 Bump library major version.
26997 Bump library major version, while preserving a major version of 0 for
26998 GStreamer 1.0 based libraries, and a major version of 2 for GStreamer
26999 0.10 based librarieS.
27001 2013-04-10 14:37:42 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27003 * gst/vaapi/gstvaapivideomemory.c:
27004 * gst/vaapi/gstvaapivideomemory.h:
27005 plugins: implement direct-rendering mode for raw YUV buffer uploads.
27006 Allow direct-rendering (writes) into target VA surfaces.
27008 2013-04-09 16:02:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27010 * gst/vaapi/gstvaapivideobufferpool.c:
27011 * gst/vaapi/gstvaapivideomemory.c:
27012 * gst/vaapi/gstvaapivideomemory.h:
27013 plugins: implement uploads from raw YUV buffers for GStreamer 1.0.
27014 Implement GstVideoMeta::{,un}map() to support raw YUV buffer upload when
27015 the last component is unmapped. Downloads are not supported yet. The aim
27016 was to first support SW decoding + HW accelerated rendering (vaapisink).
27019 2013-04-03 11:10:41 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27021 * gst/vaapi/gstvaapidecode.c:
27022 vaapidecode: submit all decoded frames before decoding a new one.
27023 Make sure to purge all pending frames that were already decoded prior
27024 to decoding a new one. This helps release VA surfaces as early as
27027 2013-04-02 16:12:16 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27029 * gst/vaapi/gstvaapidecode.c:
27030 vaapidecode: reply to CAPS queries.
27031 Handle GST_QUERY_CAPS, which is the GStreamer 1.0 mechanism to retrieve
27032 the set of allowed caps, i.e. it works similar to GstPad::get_caps().
27033 This fixes fallback to SW decoding if no HW decoder is available.
27035 2013-03-20 11:26:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27037 * gst-libs/gst/vaapi/gstvaapidecoder.c:
27038 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
27039 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
27040 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
27041 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
27042 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27043 decoder: fix unpaired GstBuffer map/unmaps.
27044 This possibly fixes a few memory leaks along the way.
27046 2013-03-20 14:40:57 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27049 * gst-libs/gst/vaapi/gstcompat.h:
27050 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
27051 * gst-libs/gst/vaapi/gstvaapiprofile.c:
27052 * gst/vaapi/Makefile.am:
27053 * gst/vaapi/gstvaapi.c:
27054 * gst/vaapi/gstvaapidecode.c:
27055 * gst/vaapi/gstvaapisink.c:
27056 * gst/vaapi/gstvaapisink.h:
27057 * gst/vaapi/gstvaapiuploader.c:
27059 Allow build against either GStreamer API (0.10 or 1.0).
27060 Introduce a new configure option --with-gstreamer-api that determines
27061 the desired GStreamer API to use. By default, GStreamer 1.0 is selected.
27062 Also integrate more compatibility glue into gstcompat.h and plugins.
27064 2012-11-08 16:41:22 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
27066 * gst/vaapi/gstvaapidecode.c:
27067 * gst/vaapi/gstvaapisink.c:
27068 * gst/vaapi/gstvaapisink.h:
27069 plugins: use new video buffer pools.
27070 Use new GstVaapiVideoBufferPool to maintain video buffers. Implement
27071 GstBaseSink::propose_allocation() to expose that pool to upstream
27072 elements; and also implement GstVideoDecoder::decide_allocation() to
27073 actually use that pool (from downstream), if any, or create one.
27074 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27076 2012-11-08 16:41:22 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
27078 * gst/vaapi/Makefile.am:
27079 * gst/vaapi/gstvaapivideobufferpool.c:
27080 * gst/vaapi/gstvaapivideobufferpool.h:
27081 * gst/vaapi/gstvaapivideomemory.c:
27082 * gst/vaapi/gstvaapivideomemory.h:
27083 plugins: add GstVaapiVideoMemory and GstVaapiVideoBufferPool objects.
27084 Add initial support for GstVaapiVideoMemory backed buffer pool. The memory
27085 object currently holds a reference to GstVaapiVideoMeta.
27086 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27088 2013-04-04 17:36:45 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27090 * gst/vaapi/gstvaapivideometa.c:
27091 * gst/vaapi/gstvaapivideometa.h:
27092 plugins: allow copies of GstVaapiVideoMeta objects.
27093 Make it possible to copy GstVaapiVideoMeta objects, unless they contain VA
27094 objects created from GstVaapiVideoPool. This is mostly useful to clone a
27095 GstVaapiVideoMeta object containing a VA surface proxy so that to alter its
27098 2013-04-04 16:16:31 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27100 * gst/vaapi/gstvaapivideometa.c:
27101 plugins: make it possible to clear VA objects from GstVaapiVideoMeta.
27102 Fix GstVaapiVideoMeta to allow VA objects to be destroyed when they are
27103 reset to NULL. i.e. make gst_vaapi_video_meta_set_{image,surface}() and
27104 gst_vaapi_video_meta_set_surface_proxy() actually clear VA objects when
27107 2012-09-03 14:00:25 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
27109 * gst/vaapi/Makefile.am:
27110 * gst/vaapi/gstvaapi.c:
27111 * gst/vaapi/gstvaapidecode.c:
27112 * gst/vaapi/gstvaapidownload.c:
27113 * gst/vaapi/gstvaapipluginutil.c:
27114 * gst/vaapi/gstvaapipostproc.c:
27115 * gst/vaapi/gstvaapisink.c:
27116 * gst/vaapi/gstvaapiupload.c:
27117 * gst/vaapi/gstvaapiuploader.c:
27118 * gst/vaapi/gstvaapivideobuffer.c:
27119 * gst/vaapi/gstvaapivideoconverter_glx.c:
27120 * gst/vaapi/gstvaapivideometa.c:
27121 * gst/vaapi/gstvaapivideometa.h:
27122 plugins: initial port to GStreamer 1.0.
27123 Port vaapidecode and vaapisink plugins to GStreamer API >= 1.0. This
27124 is rather minimalistic so that to test the basic functionality.
27125 Disable vaapiupload, vaapidownload and vaapipostproc plugins. The latter
27126 needs polishing wrt. to GStreamer 1.x functionality and the former are
27127 totally phased out in favor of GstVaapiVideoMemory map/unmap facilities,
27128 which are yet to be implemented.
27129 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27131 2013-03-21 10:12:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27133 * gst-libs/gst/vaapi/gstcompat.h:
27136 * tests/simple-decoder.c:
27137 * tests/test-subpicture.c:
27138 tests: add support for GStreamer 1.0.
27140 2012-09-04 15:12:18 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
27143 * gst-libs/gst/vaapi/gstcompat.h:
27144 * gst-libs/gst/vaapi/gstvaapicontext.c:
27145 * gst-libs/gst/vaapi/gstvaapidecoder.c:
27146 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
27147 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
27148 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
27149 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
27150 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
27151 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27152 * gst-libs/gst/vaapi/gstvaapidisplay.c:
27153 * gst-libs/gst/vaapi/gstvaapiimage.c:
27154 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
27155 * gst-libs/gst/vaapi/gstvaapiprofile.c:
27156 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
27157 * gst-libs/gst/vaapi/gstvaapisurface.h:
27158 Add initial support for GStreamer 1.0.
27159 This integrates support for GStreamer API >= 1.0 only in the libgstvaapi
27160 core decoding library. The changes are kept rather minimal here so that
27161 the library retains as little dependency as possible on core GStreamer
27163 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27165 2013-04-03 15:58:57 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27167 * gst/vaapi/gstvaapisink.c:
27168 vaapisink: improve check for raw YUV format mode.
27169 Improve check for raw YUV format modes by avoiding checks against strings
27170 ("video/x-raw-yuv") for each new GstBuffer allocation. In the usual case,
27171 GstBaseSink::set_caps() is called first and if VA surface format mode is
27172 used, then GstBaseSink::buffer_alloc() is not called. If the latter is
27173 called before set_caps(), then we just make a full check. This one is
27174 pretty rare though, e.g. it usually happens once for custom pipelines.
27176 2013-04-03 15:06:46 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27178 * gst/vaapi/gstvaapipluginutil.c:
27179 plugins: don't fail if there is no overlay composition to apply.
27180 Fix gst_vaapi_apply_composition() to not fail if no overlay composition
27181 was found. i.e. return success (TRUE). This was harmless though extra
27182 debug messages are not nice.
27183 This is a regression introduced by commit 95b8659.
27185 2013-04-03 14:59:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27187 * gst/vaapi/gstvaapidecode.c:
27188 * gst/vaapi/gstvaapidecode.h:
27189 vaapidecode: expose the exact set of supported HW decoders.
27190 Don't return static caps that don't mean anything for the underlying codecs
27191 that are actually supported for decoding. i.e. always allocate a VA display
27192 and retrieve the exact set of HW decoders available. That VA display may be
27193 re-used later on during negotiation through GstVideoContext "prepare-context".
27194 This fixes fallback to SW decoding if no HW decoder is available.
27196 2013-04-03 13:08:55 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27198 * docs/reference/libs/libs-sections.txt:
27199 * gst-libs/gst/vaapi/gstvaapicontext.c:
27200 * gst-libs/gst/vaapi/gstvaapicontext.h:
27201 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
27202 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
27203 decoder: drop obsolete functions.
27204 Drop the following functions that are now obsolete:
27205 - gst_vaapi_context_get_surface()
27206 - gst_vaapi_context_put_surface()
27207 - gst_vaapi_context_find_surface_by_id()
27208 - gst_vaapi_surface_proxy_new()
27209 - gst_vaapi_surface_proxy_get_context()
27210 - gst_vaapi_surface_proxy_set_context()
27211 - gst_vaapi_surface_proxy_set_surface()
27212 This is an API change.
27214 2013-04-03 13:14:59 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27216 * gst-libs/gst/vaapi/gstvaapidecoder.c:
27217 decoder: delegate surface size check to VA context reset.
27218 Now that the surface pool is reference counted in the surface proxy wrapper,
27219 we can safely ignore surface size checks in gst_vaapi_decoder_ensure_context().
27220 Besides, this check is already performed in gst_vaapi_context_reset_full().
27222 2013-04-03 11:37:44 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27224 * docs/reference/libs/libs-sections.txt:
27225 * gst-libs/gst/vaapi/gstvaapicontext.c:
27226 * gst-libs/gst/vaapi/gstvaapicontext.h:
27227 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
27228 * gst-libs/gst/vaapi/gstvaapisurface.h:
27229 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
27230 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
27231 decoder: simplify acquisition/release of spare surface.
27232 Introduce gst_vaapi_surface_proxy_new_from_pool() to allocate a new surface
27233 proxy from the context surface pool. This change also makes sure to retain
27234 the parent surface pool in the proxy.
27235 Besides, it was also totally useless to attach/detach parent context to
27236 VA surface each time we acquire/release it. Since the whole context owns
27237 all associated VA surfaces, we can mark this as such only once and for all.
27239 2013-03-29 10:39:37 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27242 Bump version for development.
27244 2013-03-28 10:18:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27250 2013-03-28 10:15:53 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27255 2013-03-26 18:57:00 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27257 * gst/vaapi/gstvaapipluginutil.c:
27258 * gst/vaapi/gstvaapisink.c:
27259 plugins: fix usage of gst_vaapi_reply_to_query().
27260 Make gst_vaapi_reply_to_query() first check whether the query argument
27261 is actually a video-context query, i.e. with type GST_QUERY_TYPE_CUSTOM.
27262 Then, make sure vaapisink propagates the query to the parent class if
27263 it is not a video-context query.
27265 2013-03-26 18:45:53 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27267 * gst/vaapi/gstvaapivideobuffer.c:
27268 * gst/vaapi/gstvaapivideobuffer.h:
27269 plugins: streamline video buffers.
27270 Add new gst_vaapi_video_buffer_new() helper function that allocates a video
27271 buffer from a GstVaapiVideoMeta. Also remove obsolete and useless function
27272 gst_vaapi_video_buffer_get_meta().
27274 2013-03-26 10:31:10 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27276 * docs/reference/libs/libs-sections.txt:
27277 * gst-libs/gst/vaapi/Makefile.am:
27278 * gst/vaapi/Makefile.am:
27279 * gst/vaapi/gstvaapidecode.c:
27280 * gst/vaapi/gstvaapidownload.c:
27281 * gst/vaapi/gstvaapipostproc.c:
27282 * gst/vaapi/gstvaapisink.c:
27283 * gst/vaapi/gstvaapiupload.c:
27284 * gst/vaapi/gstvaapiuploader.c:
27285 * gst/vaapi/gstvaapivideobuffer.h:
27286 * gst/vaapi/gstvaapivideoconverter_glx.c:
27287 * gst/vaapi/gstvaapivideometa.c:
27288 * gst/vaapi/gstvaapivideometa.h:
27289 plugins: integrate GstVaapiVideoMeta from libgstvaapi.
27290 Move GstVaapiVideoMeta from core libgstvaapi decoding library to the
27291 actual plugin elements. That's only useful there. Also inline reference
27292 counting code from GstVaapiMiniObject.
27294 2013-03-21 17:17:53 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27296 * gst/vaapi/Makefile.am:
27297 * gst/vaapi/gstvaapidecode.c:
27298 * gst/vaapi/gstvaapidownload.c:
27299 * gst/vaapi/gstvaapipluginbuffer.c:
27300 * gst/vaapi/gstvaapipluginbuffer.h:
27301 * gst/vaapi/gstvaapipostproc.c:
27302 * gst/vaapi/gstvaapiupload.c:
27303 * gst/vaapi/gstvaapiuploader.c:
27304 * gst/vaapi/gstvaapivideobuffer.c:
27305 * gst/vaapi/gstvaapivideobuffer.h:
27306 plugins: drop gstvaapipluginbuffer.[ch] helper files.
27307 Move all gst_vaapi_video_buffer_new*() helpers from gstvaapipluginbuffer.[ch]
27308 to gstvaapivideobuffer.[ch], and drop the obsolete files.
27310 2013-03-21 17:06:43 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27312 * docs/reference/libs/libs-sections.txt:
27313 * docs/reference/libs/libs.core.types:
27314 * gst-libs/gst/vaapi/Makefile.am:
27315 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
27316 * gst/vaapi/Makefile.am:
27317 * gst/vaapi/gstvaapidownload.h:
27318 * gst/vaapi/gstvaapipluginbuffer.c:
27319 * gst/vaapi/gstvaapipostproc.h:
27320 * gst/vaapi/gstvaapivideobuffer.c:
27321 * gst/vaapi/gstvaapivideobuffer.h:
27322 * gst/vaapi/gstvaapivideoconverter_glx.h:
27323 plugins: integrate GstVaapiVideoBuffer from libgstvaapi.
27324 Move GstVaapiVideoBuffer from core libgstvaapi decoding library to the
27325 actual plugin elements. That's only useful there.
27327 2013-03-21 16:32:43 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27329 * gst/vaapi/gstvaapipluginutil.c:
27330 * gst/vaapi/gstvaapipluginutil.h:
27331 * gst/vaapi/gstvaapisink.c:
27332 * gst/vaapi/gstvaapivideoconverter_glx.c:
27333 plugins: use common helper function to apply compositions.
27334 Use common gst_vaapi_apply_composition() helper function to apply compositions
27335 attached to a buffer in vaapisink or GstVaapiVideoConverterGLX.
27337 2013-03-21 16:09:42 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27339 * gst-libs/gst/vaapi/Makefile.am:
27340 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
27341 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.h:
27342 * gst/vaapi/Makefile.am:
27343 * gst/vaapi/gstvaapipluginbuffer.c:
27344 * gst/vaapi/gstvaapivideoconverter_glx.c:
27345 * gst/vaapi/gstvaapivideoconverter_glx.h:
27346 plugins: integrate GstVaapiVideoConverterGLX from libgstvaapi.
27347 Make sure libgstvaapi core decoding library doesn't include un-needed
27348 dependencies. So, move out GstVaapiVideoConverterGLX to plugins instead.
27349 Besides, even if the vaapisink element is not used, we are bound to have
27350 a correctly populated GstSurfaceBuffer from vaapidecode.
27351 Also clean-up the file along the way.
27353 2013-03-21 13:32:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27355 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27356 vc1: fix use of possibly uninitialized variable.
27357 In decode_codec_data(), force initialization of format to zero so that
27358 we can catch up cases where codec-data has neither "format" nor "wmvversion"
27359 fields, thus making it possible to gracefully fail in this case.
27361 2013-03-21 13:43:46 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27363 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
27364 jpeg: propagate buffer data as a const guchar * pointer (cosmetics).
27366 2013-03-21 14:36:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27368 * gst-libs/gst/vaapi/gstvaapidecoder.c:
27369 * gst-libs/gst/vaapi/gstvaapidecoder.h:
27370 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
27371 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
27372 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
27373 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27374 decoder: sanitize codec-data decoding.
27375 Add a new GstVaapiDecoder::decode_codec_data() hook to actually decode
27376 codec-data in the decoder sub-class. Provide a common shared helper
27377 function to do the actual work and delegating further to the sub-class.
27379 2013-03-21 13:41:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27381 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
27382 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
27383 * gst-libs/gst/vaapi/gstvaapidecoder_unit.c:
27384 * gst-libs/gst/vaapi/gstvaapidecoder_unit.h:
27385 decoder: get rid of GstVaapiDecoderUnit::buffer field.
27386 Drop GstVaapiDecoderUnit buffer field (GstBuffer) since it's totally
27387 useless nowadays as creating sub-buffers doesn't bring any value. It
27388 actually means more memory allocations. We can't do without that in
27389 JPEG and MPEG-4:2 decoders.
27391 2013-03-21 13:28:05 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27393 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
27394 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
27395 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
27396 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
27397 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27398 decoder: sanitize uses of codec frame input buffer (cosmetics).
27399 Alias GST_VAAPI_DECODER_CODEC_FRAME(decoder)->input_buffer to a simple
27402 2013-03-20 17:34:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27404 * gst/vaapi/gstvaapisink.c:
27405 vaapisink: add helper function to apply a composition buffer.
27406 Simplify application of a composition buffer to a GstVaapiSurface, and
27407 all its peers, until that function is eventually promoted to libgstvaapi.
27409 2013-03-20 13:42:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27411 * gst/vaapi/gstvaapisink.c:
27412 vaapisink: fix support for raw YUV buffers.
27413 If the raw YUV buffer was created from vaapisink, through the buffer_alloc()
27414 hook, then it will have a valid GstVaapiVideoMeta object attached to it.
27415 However, we previously assumed in that case that it was a "native" VA buffer,
27416 thus not calling into GstVaapiUploader::process().
27418 2013-03-20 18:41:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27420 * gst-libs/gst/vaapi/gstcompat.h:
27421 * gst/vaapi/gstvaapidecode.c:
27422 * gst/vaapi/gstvaapidownload.c:
27423 * gst/vaapi/gstvaapipostproc.c:
27424 * gst/vaapi/gstvaapisink.c:
27425 * gst/vaapi/gstvaapiupload.c:
27426 plugins: use modern GstElement metadata information.
27427 Use gst_element_class_set_static_metadata() from GStreamer 1.0, which
27428 basically is the same as gst_element_class_set_details_simple() in
27429 GStreamer 0.10 context.
27431 2013-03-20 18:04:39 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27433 * gst/vaapi/gstvaapidecode.c:
27434 * gst/vaapi/gstvaapidownload.c:
27435 * gst/vaapi/gstvaapipostproc.c:
27436 * gst/vaapi/gstvaapisink.c:
27437 * gst/vaapi/gstvaapiupload.c:
27438 plugins: move up interfaces (cosmetics).
27439 Move GstImplementsInterface and GstVideoContext support functions up
27440 so that to keep a clear separation between the plugin element and its
27443 2013-03-20 12:57:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27445 * gst/vaapi/gstvaapisink.c:
27446 * gst/vaapi/gstvaapiuploader.c:
27447 plugins: upgrade to newer APIs (GstVideoInfo based helpers).
27448 Use GstVideoInfo and gst_video_info_from_caps() helper wherever possible.
27449 Also use the newly added gst_vaapi_image_format_from_structure() helper
27450 in GstVaapiUploader::ensure_allowed_caps().
27452 2013-03-20 14:02:48 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27454 * gst/vaapi/gstvaapipluginbuffer.c:
27455 plugins: fix creation of video buffer from another source buffer.
27456 gst_vaapi_video_buffer_new_from_buffer() needs to reference the source
27457 buffer video meta since it would be unreference'd from the get_buffer()
27458 helper function. For other cases, we still use (steal) the newly created
27461 2013-03-20 11:57:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27463 * gst/vaapi/gstvaapipluginbuffer.c:
27464 * gst/vaapi/gstvaapipluginutil.c:
27465 plugins: include "sysdeps.h" header instead of "config.h".
27467 2013-03-20 18:33:23 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27469 * gst-libs/gst/vaapi/gstcompat.h:
27471 tests: modernize GstTypeFind functions.
27472 Use the GstTypeFind hooks from GStreamer 1.0. They look safer and
27473 exactly correspond to the expected behaviour.
27475 2013-03-20 11:57:57 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27479 * tests/test-decode.c:
27480 * tests/test-display.c:
27481 * tests/test-h264.c:
27482 * tests/test-h264.h:
27483 * tests/test-jpeg.c:
27484 * tests/test-jpeg.h:
27485 * tests/test-mpeg2.c:
27486 * tests/test-mpeg2.h:
27487 * tests/test-mpeg4.c:
27488 * tests/test-mpeg4.h:
27489 * tests/test-textures.c:
27490 * tests/test-vc1.c:
27491 * tests/test-vc1.h:
27492 * tests/test-windows.c:
27493 tests: fix license templates.
27495 2013-03-20 11:53:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27497 * tests/test-display.c:
27498 tests: use gst_vaapi_image_format_from_structure() in test-display.
27499 Use gst_vaapi_image_format_from_structure() helper in test-display and
27500 then extract a VAImageFormat from it instead of relying on GstCaps for
27501 YUV and RGB formats.
27503 2013-03-20 11:50:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27508 * tests/test-decode.c:
27509 * tests/test-display.c:
27510 * tests/test-subpicture.c:
27511 * tests/test-textures.c:
27512 * tests/test-windows.c:
27513 tests: include "sysdeps.h" header instead of "config.h".
27515 2013-03-20 18:25:05 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27517 * gst-libs/gst/vaapi/gstcompat.h:
27518 * gst-libs/gst/vaapi/gstvaapicontext.c:
27519 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
27520 subpicture: use gst_video_overlay_rectangle_get_pixels_unscaled_raw().
27521 Use newer gst_video_overlay_rectangle_get_pixels_unscaled_raw() helper
27522 function with GStreamer 0.10 compatible semantics, or that tries to
27523 approach the current meaning. Basically, this is also just about moving
27524 the helper to gstcompat.h.
27526 2013-03-20 11:10:31 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27528 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
27529 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
27530 image: add gst_vaapi_image_format_from_structure() helper.
27531 Add helper function to convert video formats from a GstStructure to a
27532 plain GstVaapiImageFormat.
27534 2013-03-20 18:12:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27536 * gst-libs/gst/vaapi/Makefile.am:
27537 * gst-libs/gst/vaapi/gstcompat.h:
27538 * gst-libs/gst/vaapi/sysdeps.h:
27539 sysdeps: split out GStreamer API compatibility glue to "gstcompat.h".
27541 2013-03-20 11:56:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27543 * gst-libs/gst/vaapi/sysdeps.h:
27544 sysdeps: add more standard includes by default.
27546 2013-03-20 14:43:46 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27549 configure: improve GStreamer API version checks.
27551 2013-03-20 11:44:10 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27554 * debian.upstream/Makefile.am:
27555 * debian.upstream/changelog.in:
27556 * debian.upstream/control.in:
27557 * debian.upstream/gstreamer-vaapi-doc.install.in:
27558 * debian.upstream/libgstvaapi-dev.install.in:
27559 * debian.upstream/libgstvaapi-drm.install.in:
27560 * debian.upstream/libgstvaapi-glx.install.in:
27561 * debian.upstream/libgstvaapi-wayland.install.in:
27562 * debian.upstream/libgstvaapi-x11.install.in:
27563 * debian.upstream/libgstvaapi.install.in:
27564 * docs/reference/libs/Makefile.am:
27565 * docs/reference/libs/libs-docs.xml.in:
27566 * docs/reference/plugins/Makefile.am:
27567 * docs/reference/plugins/plugins-docs.xml.in:
27568 * gst-libs/gst/vaapi/Makefile.am:
27569 * gst/vaapi/Makefile.am:
27570 * pkgconfig/Makefile.am:
27571 * pkgconfig/gstreamer-vaapi-drm.pc.in:
27572 * pkgconfig/gstreamer-vaapi-glx.pc.in:
27573 * pkgconfig/gstreamer-vaapi-wayland.pc.in:
27574 * pkgconfig/gstreamer-vaapi-x11.pc.in:
27575 * pkgconfig/gstreamer-vaapi.pc.in:
27576 * tests/Makefile.am:
27577 configure: rename GST_MAJORMINOR to GST_API_VERSION.
27579 2013-03-20 11:28:06 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27582 configure: improve check for H.264 codecparser.
27584 2013-02-26 00:38:24 +0100 Holger Kaelberer <hk@getslash.de>
27586 * gst/vaapi/gstvaapiuploader.c:
27587 vaapiupload: fix illegal write in ensure_image().
27588 Fix ensure_image() to only zero-initialize the first line of each plane.
27589 Properly initializing each plane to their full vertical resolution would
27590 require to actually compute it based on the image format.
27591 In particular, for NV12 images, the UV plane has half vertical resolution
27592 vs. the Y plane. So using the full image height to initialize the UV plane
27593 will obviously lead to a buffer overflow. Likewise for other YUV format.
27594 Since ensure_image() is only a helper function to initialize something,
27595 and not necessarily the whole thing, it is fine to initializ the first
27596 line only. Besides, the target surface is not rendered either.
27597 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27599 2013-02-17 16:28:47 +0800 Xiang, Haihao <haihao.xiang@intel.com>
27601 * gst-libs/gst/codecparsers/Makefile.am:
27602 * gst-libs/gst/video/Makefile.am:
27603 build: fix compiling of local GstVideoDecoder and codecparsers.
27604 Generated source files were missing a dependency on the complete set of
27605 generated header files. e.g. gstvideodecoder.c requires gstvideoutils.h
27606 to build and almost every codec parser source depends on parserutils.h.
27607 https://bugs.freedesktop.org/show_bug.cgi?id=59575
27608 Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
27609 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27611 2013-02-08 11:56:54 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27613 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
27614 h264: set {luma,chroma}_log2_weight_denom to 0 if no pred_weight_table().
27615 Force luma_log2_weight_denom and chroma_log2_weight_denom to zero if
27616 there is no pred_weight_table() that was parsed.
27617 This is a workaround for the VA intel-driver on Ivy Bridge.
27619 2013-02-07 15:42:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27621 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
27622 h264: use new profile definitions from codecparsers.
27624 2013-02-07 15:29:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27626 * ext/codecparsers:
27627 codecparsers: update to gst-vaapi-branch commit 500bc02.
27628 500bc02 h264: add profile enums
27630 2013-02-06 15:27:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27635 2013-02-06 15:21:27 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27637 * ext/codecparsers:
27638 codecparsers: update to gst-vaapi-branch commit 31b1c57.
27639 8957fb7 mpeg2: add helpers to convert quantization matrices
27640 07c4034 mpeg2: store quantization matrices in zigzag scan order
27642 2013-01-31 11:32:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27644 * tests/simple-decoder.c:
27645 tests: simple-decoder: fix build on older platforms.
27646 Make simple-decoder build and execute correctly on older platforms,
27647 and more precisely older versions of glib.
27649 2013-01-31 11:30:12 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27651 * gst-libs/gst/vaapi/glibcompat.h:
27652 glibcompat: add replacement for g_async_queue_timeout_pop().
27653 g_async_queue_timeout_pop() appeared in glib 2.31.18. Implement it as
27654 g_async_queue_timed_pop() with a GTimeVal as the final time to wait for
27655 new data to arrive in the queue.
27657 2013-01-31 11:25:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27659 * gst-libs/gst/vaapi/glibcompat.h:
27660 glibcompat: add replacement for g_cond_wait().
27662 2013-01-30 18:38:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27664 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
27665 mpeg2: fix decoding of 4K videos.
27666 Account for slice_vertical_position_extension when vertical_size > 2800.
27668 2013-01-30 18:54:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27670 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
27671 mpeg2: fix decoding of sequence_end().
27672 There shall be only one place to call decode_current_picture(), and this
27673 is in the end_frame() hook. The EOS unit is processed after end_frame()
27674 so this means we cannot have a valid picture to decode/output at this
27677 2013-01-30 15:10:06 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27679 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
27680 mpeg2: improve robustness when packets are missing.
27681 Improve robustness when some expected packets where not received yet
27682 or that were not correctly decoded. For example, don't try to decode
27683 a picture if there was no valid sequence or picture headers.
27685 2013-01-30 18:58:01 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27687 * gst-libs/gst/vaapi/gstvaapidecoder.c:
27688 decoder: handle decode-only frames in raw API mode.
27689 Fix gst_vaapi_decoder_get_surface() to only return frames with a valid
27690 surface proxy, i.e. with a valid VA surface. This means that any frame
27691 marked as decode-only is simply skipped.
27693 2013-01-30 16:33:48 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27695 * gst-libs/gst/vaapi/gstvaapidecoder.c:
27696 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
27697 decoder: allow frames to be dropped.
27698 If the decoder was not able to decode a frame because insufficient
27699 information was available, e.g. missing sequence or picture header,
27700 then allow the frame to be gracefully dropped without generating
27702 It is also possible that a frame is not meant to be displayed but
27703 only used as a reference, so dropping that frame is also a valid
27704 operation since GstVideoDecoder base class has extra references to
27705 that GstVideoCodecFrame that needs to be released.
27707 2013-01-30 16:26:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27709 * gst/vaapi/gstvaapidecode.c:
27710 vaapidecode: handle decode-only frames.
27711 Decode-only frames may not have a valid surface proxy. So, simply discard
27712 them gracefully, i.e. don't create meta data information. GstVideoDecoder
27713 base class will properly handle this case and won't try to push any buffer
27714 to downstream elements.
27716 2013-01-24 00:49:17 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
27718 * gst/vaapi/gstvaapidecode.c:
27719 vaapidecode: add support for post-seek semantics reset.
27720 Implement GstVideoDecoder::reset() as a destruction of the VA decoder
27721 and the creation of a new VA decoder.
27722 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27724 2013-01-30 09:38:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27727 Bump version for development.
27729 2013-01-30 09:37:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27735 2013-01-24 00:48:26 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
27737 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
27738 mpeg2: implement GstVaapiDecoder::flush() as a DPB flush.
27740 2013-01-24 17:34:43 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
27742 * gst-libs/gst/vaapi/gstvaapidecoder_frame.h:
27743 decoder: fix documentation for GstVaapiDecoderFrame.
27744 Drop superfluous reference to prev_slice member.
27746 2013-01-29 16:18:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27748 * gst-libs/gst/vaapi/gstvaapidecoder.c:
27749 decoder: assume current frame is complete at end-of-stream.
27750 Assume we got a complete frame when the end-of-stream is reached and that
27751 the current codec frame contains at least one slice data unit.
27753 2013-01-29 14:14:45 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27757 * debian.upstream/copyright:
27758 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
27759 * gst-libs/gst/vaapi/gstvaapicontext.c:
27760 * gst-libs/gst/vaapi/gstvaapidecoder.c:
27761 * gst-libs/gst/vaapi/gstvaapidecoder.h:
27762 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
27763 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.h:
27764 * gst-libs/gst/vaapi/gstvaapidecoder_frame.c:
27765 * gst-libs/gst/vaapi/gstvaapidecoder_frame.h:
27766 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
27767 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
27768 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
27769 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
27770 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
27771 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
27772 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
27773 * gst-libs/gst/vaapi/gstvaapidecoder_unit.c:
27774 * gst-libs/gst/vaapi/gstvaapidecoder_unit.h:
27775 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27776 * gst-libs/gst/vaapi/gstvaapidisplay.c:
27777 * gst-libs/gst/vaapi/gstvaapidisplay.h:
27778 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
27779 * gst-libs/gst/vaapi/gstvaapiminiobject.c:
27780 * gst-libs/gst/vaapi/gstvaapiprofile.h:
27781 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
27782 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
27783 * gst-libs/gst/vaapi/gstvaapisurface.c:
27784 * gst-libs/gst/vaapi/gstvaapiutils.c:
27785 * gst-libs/gst/vaapi/gstvaapiutils.h:
27786 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
27787 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
27788 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
27789 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
27790 * gst-libs/gst/vaapi/sysdeps.h:
27791 * gst/vaapi/gstvaapidecode.c:
27792 * gst/vaapi/gstvaapidownload.c:
27793 * gst/vaapi/gstvaapipluginbuffer.c:
27794 * gst/vaapi/gstvaapipluginbuffer.h:
27795 * gst/vaapi/gstvaapipostproc.c:
27796 * gst/vaapi/gstvaapisink.c:
27797 * gst/vaapi/gstvaapiupload.c:
27798 * gst/vaapi/gstvaapiuploader.c:
27800 * tests/test-decode.c:
27801 * tests/test-subpicture.c:
27802 legal: fix year for some copyright notices (2013).
27804 2013-01-29 14:03:27 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27806 * gst-libs/gst/vaapi/gstvaapicontext.h:
27807 * gst-libs/gst/vaapi/gstvaapidecoder.h:
27808 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
27809 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.h:
27810 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
27811 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.h:
27812 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27813 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.h:
27814 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
27815 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
27816 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
27817 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
27818 * gst-libs/gst/vaapi/gstvaapisurface.c:
27819 * gst-libs/gst/vaapi/gstvaapisurface.h:
27820 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
27821 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
27822 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
27823 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
27824 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
27825 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.h:
27826 * gst/vaapi/gstvaapidecode.h:
27827 * gst/vaapi/gstvaapisink.h:
27828 * tests/test-subpicture.c:
27829 legal: fix year for some copyright notices (2012).
27831 2013-01-29 14:00:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27833 * gst-libs/gst/vaapi/gstvaapicompat.h:
27834 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
27835 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
27836 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
27837 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
27838 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
27839 * gst-libs/gst/vaapi/gstvaapiobject.c:
27840 * gst-libs/gst/vaapi/gstvaapiobject.h:
27841 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
27842 * gst-libs/gst/vaapi/gstvaapiparamspecs.c:
27843 * gst-libs/gst/vaapi/gstvaapiparamspecs.h:
27844 * gst-libs/gst/vaapi/gstvaapiprofile.c:
27845 * gst-libs/gst/vaapi/gstvaapiprofile.h:
27846 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
27847 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
27848 * gst-libs/gst/vaapi/gstvaapitexture.c:
27849 * gst-libs/gst/vaapi/gstvaapitexture.h:
27850 * gst-libs/gst/vaapi/gstvaapitypes.h:
27851 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
27852 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
27853 * gst-libs/gst/vaapi/gstvaapivalue.c:
27854 * gst-libs/gst/vaapi/gstvaapivalue.h:
27855 * gst-libs/gst/vaapi/gstvaapivideopool.c:
27856 * gst-libs/gst/vaapi/gstvaapivideopool.h:
27857 * gst-libs/gst/vaapi/gstvaapiwindow.h:
27858 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
27859 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
27860 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
27861 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
27862 * tests/test-display.c:
27863 * tests/test-surfaces.c:
27864 * tests/test-windows.c:
27865 legal: add Intel copyright on modified files.
27867 2013-01-29 13:37:41 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27872 2013-01-28 18:09:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27874 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
27875 wayland: use a local event queue to avoid lock contention.
27876 This improves performance when rendering several surfaces from within
27877 the same process. e.g. a tee of vaapidecode'd buffers to vaapisink.
27879 2013-01-28 17:28:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27881 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
27882 wayland: fix thread-safe issues.
27883 The Wayland API is not fully thread-safe and client applications shall
27884 perform locking themselves on key functions. Besides, make sure to
27885 release the lock if the _render() function fails.
27887 2013-01-28 16:37:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27889 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
27890 wayland: really wait until the pending redraw completed.
27891 Introduce gst_vaapi_window_wayland_sync() helper function to wait for
27892 the completion of the redraw request. Use it in _render() function to
27893 actually block until the previous draw request is completed.
27895 2013-01-23 10:10:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27897 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
27898 wayland: fix frame_redraw callback.
27899 The redraw callback needs to be attached to the surface prior to the
27900 commit. Otherwise, the callback notifies the next surface repaint,
27901 which is not the desired behaviour. i.e. we want to be notified for
27902 the surface we have just filled.
27903 Another isse was the redraw_pending was reset before the actual completion
27904 of the frame redraw callback function, thus causing concurrency issues.
27905 e.g. the callback could have been called again, but with a NULL buffer.
27907 2013-01-28 14:45:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27909 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
27910 wayland: fix display sharing.
27911 When the Wayland display is shared, we still have to create our own local
27912 shell and compositor objects, since they are not propagated from the cache.
27913 Likewise, we also need to determine the display size or vaapisink would
27914 fail to account for the display aspect ratio, and will try to create a 0x0
27917 2013-01-24 17:38:53 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27919 * ext/codecparsers:
27920 codecparsers: update to gst-vaapi-branch commit 21a098e.
27921 21a098e vc1: fix bitplanes decoding (DIFF6 or NORM6) [residual]
27922 f8c836a vc1: fix bitplanes decoding (DIFF6 or NORM6)
27924 2013-01-23 16:38:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27926 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27927 vc1: handle frames with multiple slices.
27929 2013-01-23 17:01:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27931 * ext/codecparsers:
27932 codecparsers: update to gst-vaapi-branch commit 3fba492.
27933 3fba492 vc1: add API to parse slice headers
27935 2013-01-23 11:11:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27937 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27938 vc1: handle CLOSED_ENTRY.
27939 When CLOSED_ENTRY == 0, and if the B pictures that follow an entry-point
27940 lack a reference anchor picture, these B pictures shall be discarded.
27941 https://bugs.freedesktop.org/show_bug.cgi?id=59505
27943 2013-01-23 10:25:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27945 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27946 vc1: cope with latest codecparser changes.
27947 Fix build with newer VC-1 codecparser where dqsbedge was renamed to
27948 dqbedge, and now represents either DQSBEDGE or DQDBEDGE depending on
27949 the actual value of DQPROFILE.
27951 2013-01-23 10:24:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27953 * ext/codecparsers:
27954 codecparsers: update to gst-vaapi-branch commit 3d2c67c.
27955 3d2c67c vc1: simplify GstVC1VopDquant structure
27957 2013-01-22 10:51:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27959 * ext/codecparsers:
27960 codecparsers: update to gst-vaapi-branch commit 5d33da8.
27961 5d33da8 vc1: fix bitplanes decoding
27962 562bdc4 vc1: fix VOPDQUANT parser for DQUANT == 2
27963 0b13d2b vc1: fix calculation of ALTPQUANT
27964 ba88e63 vc1: fix parser for DQPROFILE in VOPDQUANT
27966 2013-01-22 15:47:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27968 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27969 vc1: fix size of encapsulated BDU.
27970 Fix size of encapsulated BDUs since GstVC1BDU.size actually represents
27971 the size of the BDU data, starting from offset, i.e. after any start
27973 This fixes a buffer overflow during the unescaping process.
27975 2013-01-11 17:08:00 +0800 Wind Yuan <feng.yuan@intel.com>
27977 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27978 vc1: fix decoding of WMV3 videos in AVI format.
27979 The AVI demuxer (avidemux) does not set a proper "format" attribute
27980 to the generated caps. So, try to recover the video codec format from
27981 the "wmvversion" property instead.
27982 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27984 2013-01-22 13:28:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27986 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27987 vc1: review and report errors accordingly.
27988 Use GST_ERROR() to report real errors instead of hiding them into
27991 2013-01-22 13:50:39 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
27993 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
27994 vc1: don't create GstBuffers for all decoder units.
27995 Don't create temporary GstBuffers for all decoder units, even if they
27996 are lightweight "sub-buffers", since it is not really necessary to keep
27997 the buffer data around.
27999 2013-01-22 16:03:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28001 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
28002 vc1: implement flush() hook.
28003 Make it a simple DPB flush.
28005 2013-01-22 13:44:32 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28007 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
28008 vc1: implement {start,end}_frame() hooks.
28009 Implement GstVaapiDecoder.start_frame() and end_frame() semantics so
28010 that to create new VA context earlier and submit VA pictures to the
28011 HW for decoding as soon as possible. i.e. don't wait for the next
28012 frame to start decoding the previous one.
28014 2013-01-22 09:30:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28016 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
28017 vc1: fix next POC for new sequence layers.
28018 Fix next POC when a new sequence layer is reached. At this point, we
28019 need to reset any previous reference picture, i.e. non B-frame.
28021 2012-08-02 17:15:26 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
28023 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
28024 vc1: port to common GstVaapiDpb interface.
28025 Use GstVaapiDpb interface instead of maintaining our own prev and next
28026 picture pointers. While doing so, try to derive a sensible POC value.
28027 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28029 2013-01-15 17:10:56 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28031 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
28032 vc1: fix decode_sequence_end() to return success, not EOS.
28034 2013-01-18 17:00:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28036 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28037 decoder: simplify gst_vaapi_decoder_get_surface().
28038 Avoid extraenous branches, i.e. immediately return with success once we
28039 have a decoded frame available.
28041 2013-01-18 16:56:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28043 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28044 decoder: optimize and clean decode_step() up.
28045 Avoid usage of goto. Simplify decode_step() process to first accumulate all
28046 pending buffers into the GstAdapter, and then parse and decode units from
28047 that input adapter. Stop the process once a frame is fully decoded or an
28050 2013-01-18 14:46:23 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28052 * gst-libs/gst/vaapi/gstvaapidisplay.c:
28053 display: move "vaapi" debug init to libgstvaapi_init_once().
28055 2013-01-18 14:17:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28058 * gst-libs/gst/vaapi/Makefile.am:
28059 * gst-libs/gst/vaapi/gstvaapidisplay.c:
28060 * gst-libs/gst/vaapi/gstvaapiversion.h.in:
28061 display: dump gstreamer-vaapi version for debugging purposes.
28063 2013-01-18 14:30:48 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28065 * tests/Makefile.am:
28066 tests: simple-decoder: fix build with built-in videoutils.
28067 Fix build with built-in videoutils, i.e. when system GStreamer installation
28068 does not know about GstVideoDecoder API.
28070 2013-01-18 10:35:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28072 * tests/simple-decoder.c:
28073 tests: simple-decoder: flush decoded frames at EOS.
28074 Flush the remaining decoded frames when an end-of-stream is reached.
28076 2013-01-18 10:25:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28078 * tests/simple-decoder.c:
28079 tests: simple-decoder: drop use of GstVaapiVideoMeta.
28080 Don't use GstVaapiVideoMeta since that object is not guaranteed to live
28081 in libgstvaapi forever. Rather, that'd move to plugin elements at some
28084 2013-01-16 13:53:43 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28086 * tests/simple-decoder.c:
28087 tests: simple-decoder: add benchmark mode.
28088 Add --benchmark option to enable benchmark mode where rendering is not
28089 synchronized with presentation timestamps of the decoded surfaces.
28091 2013-01-16 13:29:06 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28093 * tests/Makefile.am:
28094 * tests/simple-decoder.c:
28095 tests: simple-decoder: honour framerate from the bitstream.
28096 Try to honour the framerate from the bitstream, or cap the playback to
28099 2013-01-15 18:49:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28101 * tests/simple-decoder.c:
28102 tests: simple-decoder: set window size to the surface dimensions.
28103 Set the window size to the decoded surface dimensions, if the user has
28104 not requested the application to run in full-screen mode. Besides, no
28105 effort is made to preserve aspect ratio or to center the video within
28108 2013-01-15 17:33:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28110 * tests/Makefile.am:
28111 * tests/simple-decoder.c:
28112 tests: add simple decoder application.
28113 Add simple decoder application to show off decoding capabilities from
28114 raw bitstreams, for debugging or performance evaluation purposes.
28116 2013-01-15 17:30:57 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28118 * tests/Makefile.am:
28121 tests: add codec helper utils.
28122 Add helper functions to determine the codec type from a specific file
28123 or utility functions to convert from codec type to GstCaps or from
28124 codec name to codec type.
28126 2013-01-15 17:47:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28129 tests: allow fullscreen mode.
28130 Add new --fullscreen|-f option to create new windows in fullscreen mode.
28132 2013-01-17 18:35:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28134 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28135 h264: implement GstVaapiDecoder::flush() as a DPB flush.
28137 2013-01-17 18:07:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28139 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28140 h264: handle end-of-stream NALU.
28141 Handle <end-of-stream> NAL unit to actually flush any pending picture
28144 2013-01-17 18:22:49 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28146 * gst/vaapi/gstvaapidecode.c:
28147 vaapidecode: handle EOS events.
28148 Flush all decoded frames to downstream when EOS is received. This is
28149 performed by implementing GstVideoDecoder::finish() hook.
28151 2013-01-17 18:19:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28153 * gst/vaapi/gstvaapidecode.c:
28154 vaapidecode: split gvd_handle_frame() into decode/push frames.
28155 Split GstVideoDecoder::handle_frame() implementation into two functions:
28156 (i) one for decoding the provided GstVideoCodecFrame and (ii) another one
28157 for purging all decoded frames and submit them downstream.
28159 2013-01-17 18:33:32 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28161 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28162 * gst-libs/gst/vaapi/gstvaapidecoder.h:
28163 decoder: add GstVaapiDecoder::flush() hook.
28165 2013-01-15 17:21:50 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28167 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28168 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
28169 decoder: fix check for end-of-stream in raw API mode.
28170 Make sure to immediately return GST_VAAPI_DECODER_STATUS_END_OF_STREAM
28171 if the end-of-stream was already reached at the previous iteration.
28173 2013-01-15 16:55:29 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28175 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28176 decoder: make decode_step() return once the frame is decoded.
28177 Make sure we always have a free surface left to use for decoding the
28178 current frame. This means that decode_step() has to return once a frame
28179 gets decoded. If the current adapter contains more buffers with valid
28180 frames, they will get parsed and decoded on subsequent iterations.
28182 2013-01-17 15:47:17 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28184 * ext/codecparsers:
28185 codecparsers: update to gst-vaapi-branch commit b47983a.
28186 8840c2d h264: zero-initialize SPS VUI parameters
28188 2013-01-15 09:21:36 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28191 Bump version for development.
28193 2013-01-15 09:21:08 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28199 2013-01-14 11:48:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28201 * docs/reference/libs/libs-docs.xml.in:
28202 * docs/reference/libs/libs-sections.txt:
28203 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
28204 * gst-libs/gst/vaapi/gstvaapidisplay.c:
28205 * gst-libs/gst/vaapi/gstvaapiprofile.h:
28206 docs: expose new interfaces.
28208 2013-01-14 12:58:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28213 2013-01-14 10:58:49 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28215 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
28216 dpb: cosmetics (clean-ups).
28218 2013-01-14 10:46:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28220 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
28221 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.h:
28222 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28223 dpb: port to GstVaapiMiniObject.
28225 2013-01-14 10:21:53 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28227 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
28228 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.h:
28229 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28230 dpb: drop GstVaapiDpb2 interface, keep only one class.
28231 Keep only one DPB interface and rename gst_vaapi_dpb2_get_references()
28232 to gst_vaapi_dpb_get_neighbours() so that to retrieve pictures in DPB
28233 around the specified picture POC.
28235 2012-08-02 15:56:54 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
28237 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
28238 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.h:
28239 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28240 dpb: rename GstVaapiDpbMpeg2 to GstVaapiDpb2.
28241 Move GstVaapiDpbMpeg2 API to a more generic version that could also be
28242 useful to other decoders that require 2 reference pictures, e.g. VC-1.
28243 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28245 2013-01-11 16:04:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28248 Bump version for pre-release.
28250 2013-01-11 15:57:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28255 2012-07-20 12:36:33 +0200 Holger Kaelberer <holger.k@elberer.de>
28258 * gst-libs/gst/vaapi/gstvaapicontext.c:
28259 * gst-libs/gst/vaapi/gstvaapiutils.c:
28260 * gst-libs/gst/vaapi/sysdeps.h:
28261 * tests/test-subpicture.c:
28262 overlay: fix build without advanced GstVideoOverlayFormatFlags.
28263 Check for global-alpha support in GstVideoOverlayComposition API.
28264 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28266 2013-01-04 10:19:56 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28268 * tests/test-subpicture.c:
28269 tests: add support for global-alpha subpictures.
28270 Add --global-alpha option to test-subpicture.
28272 2013-01-10 13:09:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28274 * tests/Makefile.am:
28275 * tests/test-subpicture.c:
28276 tests: use GstVideoOverlayComposition API for subpicture test.
28278 2013-01-10 11:26:17 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28280 * tests/Makefile.am:
28281 * tests/test-subpicture.c:
28282 tests: use common decoder helpers for subpicture test.
28283 Use common decoder helpers for subpicture test, thus allowing to decode
28284 sample images in an alternate format.
28286 2013-01-10 11:22:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28288 * tests/Makefile.am:
28291 * tests/test-decode.c:
28292 tests: add decoder helpers.
28294 2013-01-11 15:19:45 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28296 * gst-libs/gst/vaapi/gstvaapicontext.c:
28297 overlay: fix ordering of composition layers.
28298 Make sure to maintain the association order of composition layers when
28299 GstVideoOverlayRectangle objects are kept around (cached).
28301 2012-05-15 10:24:08 +0200 Holger Kaelberer <holger.k@elberer.de>
28303 * gst-libs/gst/vaapi/gstvaapicontext.c:
28304 overlay: fix support for global-alpha.
28305 Fix support for global-alpha subpictures. The previous changes brought
28306 the ability to check for GstVideoOverlayRectangle changes by comparing
28307 the underlying pixel buffer pointers. If sequence number and pixel data
28308 did not change, then this is an indication that only the global-alpha
28309 value changed. Now, try to update the underlying VA subpicture global-alpha
28311 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28313 2013-01-11 11:53:05 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28315 * gst-libs/gst/vaapi/gstvaapicontext.c:
28316 overlay: detect render-rect changes.
28317 Don't re-upload VA subpicture if only the render rectangle changed.
28318 Rather deassociate the subpicture and re-associate it with the new
28321 2013-01-11 11:12:26 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28323 * gst-libs/gst/vaapi/gstvaapicontext.c:
28324 overlay: fix check for pixels buffer change.
28325 A GstVideoOverlayRectangle is created whenever the underlying pixels data
28326 change. However, when global-alpha is supported, it is possible to re-use
28327 the same GstVideoOverlayRectangle but with a change to the global-alpha
28328 value. This process causes a change of sequence number, so we can no longer
28330 Still, if sequence numbers did not change, then there was no change in
28331 global-alpha either. So, we need a way to compare the underlying GstBuffer
28332 pointers. There is no API to retrieve the original pixels buffer from
28333 a GstVideoOverlayRectangle. So, we use the following heuristics:
28334 1. Use gst_video_overlay_rectangle_get_pixels_unscaled_argb() with the same
28335 format flags from which the GstVideoOverlayRectangle was created. This
28336 will work if there was no prior consumer of the GstVideoOverlayRectangle
28337 with alternate (non-"native") format flags.
28338 2. In overlay_rectangle_has_changed_pixels(), we have to use the same
28339 gst_video_overlay_rectangle_get_pixels_unscaled_argb() function but
28340 with flags that match the subpicture. This is needed to cope with
28341 platforms that don't support global-alpha in HW, so the gst-video
28342 layer takes care of that and fixes this up with a possibly new
28343 GstBuffer, and hence pixels data (or) in-place by caching the current
28344 global-alpha value applied. So we have to determine the rectangle
28345 was previously used, based on what previous flags were used to
28346 retrieve the ARGB pixels buffer.
28348 2013-01-10 18:42:37 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28350 * gst-libs/gst/vaapi/gstvaapicontext.c:
28351 overlay: optimize cache at the GstVideoOverlayRectangle level.
28352 We previously assumed that an overlay composition changed if the number
28353 of overlay rectangles in there actually changed, or that the rectangle
28354 was updated, and thus its seqnum was also updated.
28355 Now, we can cope with cases where the GstVideoOverlayComposition grew
28356 by one or a few more overlay rectangles, and the initial overlay rectangles
28359 2013-01-10 13:41:39 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28361 * gst-libs/gst/vaapi/gstvaapicontext.c:
28362 overlay: simplify caching of GstVideoOverlayComposition objects.
28363 Create the GPtrArray once in the _init() function and destroy it only
28364 in the _finalize() function. Then use overlay_clear() to remove all
28365 subpicture associations for intermediate updates, don't recreate the
28367 Make GstVaapiOverlayRectangle a reference counted object. Also make
28368 sure that overlay_rectangle_new() actually creates and associates the
28371 2012-05-15 10:24:08 +0200 Holger Kaelberer <holger.k@elberer.de>
28373 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
28374 * gst-libs/gst/vaapi/gstvaapiutils.c:
28375 * gst-libs/gst/vaapi/gstvaapiutils.h:
28376 overlay: add support for global-alpha.
28377 Handle global-alpha from GstVideoOverlayComposition API. Likewise,
28378 the same code path could also work for premultiplied-alpha but this
28380 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28382 2012-05-15 10:24:08 +0200 Holger Kaelberer <holger.k@elberer.de>
28384 * gst-libs/gst/vaapi/gstvaapidisplay.c:
28385 * gst-libs/gst/vaapi/gstvaapidisplay.h:
28386 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
28387 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
28388 * gst-libs/gst/vaapi/gstvaapisurface.c:
28390 * tests/test-subpicture.c:
28391 subpicture: add support for global-alpha.
28392 Add the necessary helpers in GstVaapiDisplay to determine whether subpictures
28393 with global alpha are supported or not. Also add accessors in GstVaapiSubpicture
28394 to address this feature.
28395 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28397 2013-01-04 09:41:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28399 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
28400 * gst-libs/gst/vaapi/gstvaapiutils.c:
28401 * gst-libs/gst/vaapi/gstvaapiutils.h:
28402 subpicture: add premultiplied-alpha and global-alpha feature flags.
28403 Add premultiplied-alpha and global-alpha feature flags, along with converters
28404 between VA-API and gstreamer-vaapi definitions. Another round of helpers is
28405 also necessary for GstVideoOverlayComposition API.
28407 2013-01-03 18:02:49 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28409 * gst-libs/gst/vaapi/gstvaapidisplay.c:
28410 display: allow image/subpicture formats with additional flags.
28411 Introduce new GstVaapiFormatInfo to store the actual GstVaapiImageFormat
28412 and any additional flags needed. Currently, all flags are set to zero.
28414 2013-01-11 13:34:45 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28416 * gst-libs/gst/codecparsers/Makefile.am:
28417 * gst-libs/gst/video/Makefile.am:
28418 * tests/Makefile.am:
28419 libs: fix build of submodule wrappers.
28420 Make sure to build codecparsers/ and videoutils/ sources against the
28421 newly generated headers when out-of-source builds are used.
28423 2013-01-11 14:11:39 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28426 configure: fix checks for packages installed in non-standard roots.
28428 2013-01-10 10:12:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28430 * gst-libs/gst/vaapi/gstvaapiminiobject.c:
28431 decoder: fix mini object implementation on 64-bit systems.
28432 Use GPOINTER_TO_SIZE() instead of GPOINTER_TO_UINT() while manipulating
28433 pointers. The latter is meant to be 32-bit only, not uintptr_t like size.
28434 Only a gsize can hold all bits of a pointer.
28435 Thanks to Ouping Zhang for spotting this error.
28437 2013-01-09 16:05:39 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28439 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28440 mpeg2: optimize scan for the end of the frame.
28441 Heuristic: if the second start-code is available, check whether that
28442 one marks the start of a new frame because e.g. this is a sequence
28443 or picture header. This doesn't save much, since we already cache the
28446 2013-01-09 13:44:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28448 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28449 mpeg2: optimize scan for start codes.
28450 Accelerate scan for start codes by skipping up to 3 bytes per iteration.
28451 A start code prefix is defined by the following bytes: 00 00 01. Thus,
28452 for any group of 3 bytes (xx yy zz), we have the following possible cases:
28453 1. If zz != 1, this cannot be a start code, then skip 3 bytes;
28454 2. If yy != 0, this cannot be a start code, then skip 2 bytes;
28455 3. If xx != 0 or zz != 1, this cannot be a start code, then skip 1 byte;
28456 4. xx == 00, yy == 00, zz == 1, we have match!
28457 This algorithm requires to peek bytes from the adapter. This increases the
28458 amount of bytes copied to a temporary buffer, but this process is much faster
28459 than scanning for all the bytes and using shift/masks. So, overall, this is
28462 2013-01-08 16:41:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28464 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28465 mpeg2: drop useless gst_adapter_peek().
28466 Drop useless gst_adapter_peek() since the returned buffer was not used
28467 and this could incur superfluous memcpy().
28469 2013-01-07 16:07:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28471 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28472 mpeg2: cosmetics: move parse_slice() down.
28474 2013-01-07 15:24:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28476 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28477 mpeg2: avoid too many allocations of parser info objects.
28478 Move parsing back to decoding step, but keep functions separate for now.
28479 This is needed for future optimizations that may introduce some meta data
28480 for parsed info attached to codec frames.
28482 2013-01-07 14:04:22 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28484 * gst-libs/gst/vaapi/gstvaapidecoder_unit.c:
28485 * gst-libs/gst/vaapi/gstvaapidecoder_unit.h:
28486 decoder: decoder units are no longer dynamically allocated objects.
28488 2013-01-07 13:59:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28490 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28491 * gst-libs/gst/vaapi/gstvaapidecoder_frame.c:
28492 * gst-libs/gst/vaapi/gstvaapidecoder_frame.h:
28493 decoder: optimize pre-allocation of decoder units.
28494 Optimize pre-allocation of decoder units, thus avoiding un-necessary
28495 memory reallocations. The heuristic used is that we could have around
28496 one slice unit per macroblock line.
28498 2013-01-07 13:41:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28500 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28501 * gst-libs/gst/vaapi/gstvaapidecoder_frame.c:
28502 * gst-libs/gst/vaapi/gstvaapidecoder_frame.h:
28503 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
28504 * gst-libs/gst/vaapi/gstvaapidecoder_unit.c:
28505 decoder: use an array of units instead of a single-linked list.
28506 Use a GArray to hold decoder units in a frame, instead of a single-linked
28507 list. This makes 'append' calls faster, but not that much. At least, this
28508 makes things clearer.
28510 2013-01-07 11:13:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28512 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28513 * gst-libs/gst/vaapi/gstvaapidecoder.h:
28514 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28515 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
28516 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28517 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
28518 * gst-libs/gst/vaapi/gstvaapidecoder_unit.c:
28519 * gst-libs/gst/vaapi/gstvaapidecoder_unit.h:
28520 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
28521 decoder: refactor decoder unit API.
28522 Allocate decoder unit earlier in the main parse() function and don't
28523 delegate this task to derived classes. The ultimate purpose is to get
28524 rid of dynamic allocation of decoder units.
28526 2013-01-07 10:48:27 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28528 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28529 mpeg2: introduce parser info instead of MPEG-2 specific decoder unit.
28530 Use a new GstVaapiParserInfoMpeg2 data structure instead of deriving
28531 from GstVaapiDecoderUnit for MPEG-2 specific parser information.
28533 2013-01-07 10:22:54 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28535 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28536 h264: introduce parser info instead of H.264 specific decoder unit.
28537 Use a new GstVaapiParserInfoH264 data structure instead of deriving
28538 from GstVaapiDecoderUnit for H.264 specific parser information.
28540 2013-01-05 12:33:06 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
28542 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28543 h264: set default values for some header fields.
28544 The SPS, PPS and slice headers are not fully zero-initialized in the
28545 codecparsers/ library. Rather, the standard upstream behaviour is to
28546 initialize only certain syntax elements with some inferred values if
28547 they are not present in the bitstream.
28548 At the gstreamer-vaapi decoder level, we need to further initialize
28549 certain syntax elements with some sensible default values so that to
28550 not complicate VA drivers that just pass those verbatim to the HW,
28551 and also avoid an memset() of the whole decoder unit.
28552 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
28553 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28555 2013-01-06 19:05:49 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28557 * ext/codecparsers:
28558 codecparsers: update to gst-vaapi-rebased commit b47983a.
28559 b47983a h264: add inferred value for slice_beta_offset_div2
28561 2013-01-05 17:55:47 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28563 * gst/vaapi/gstvaapidecode.c:
28564 * gst/vaapi/gstvaapidownload.c:
28565 * gst/vaapi/gstvaapipluginbuffer.c:
28566 * gst/vaapi/gstvaapipluginbuffer.h:
28567 * gst/vaapi/gstvaapipostproc.c:
28568 * gst/vaapi/gstvaapisink.c:
28569 * gst/vaapi/gstvaapiupload.c:
28570 * gst/vaapi/gstvaapiuploader.c:
28571 plugins: cope with new GstVaapiVideoMeta API.
28572 Update plugin elements with the new GstVaapiVideoMeta API.
28573 This also fixes support for subpictures/overlay because GstVideoDecoder
28574 generates a sub-buffer from the GstVaapiVideoBuffer. So, that sub-buffer
28575 is marked as read-only. However, when comes in the textoverlay element
28576 for example, it checks whether the input buffer is writable. Since that
28577 buffer read-only, then a new GstBuffer is created. Since gst_buffer_copy()
28578 does not preserve the parent field, the generated buffer in textoverlay
28579 is not exploitable because we lost all VA specific information.
28580 Now, with GstVaapiVideoMeta information attached to a standard GstBuffer,
28581 all information are preserved through gst_buffer_copy() since the latter
28582 does copy metadata (qdata in this case).
28584 2013-01-05 17:37:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28586 * gst-libs/gst/vaapi/Makefile.am:
28587 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
28588 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
28589 * gst-libs/gst/vaapi/gstvaapivideobuffer_glx.c:
28590 * gst-libs/gst/vaapi/gstvaapivideobuffer_glx.h:
28591 * gst-libs/gst/vaapi/gstvaapivideobuffer_priv.h:
28592 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
28593 videobuffer: wrap video meta into a surface buffer.
28594 Make GstVaapiVideoBuffer a simple wrapper for video meta. This buffer is
28595 no longer necessary but for compatibility with GStreamer 0.10 APIs or users
28596 expecting a GstSurfaceBuffer like Clutter.
28598 2013-01-05 08:31:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28600 * gst-libs/gst/vaapi/Makefile.am:
28601 * gst-libs/gst/vaapi/gstvaapivideometa.c:
28602 * gst-libs/gst/vaapi/gstvaapivideometa.h:
28603 videobuffer: add video meta information.
28604 Add new GstVaapiVideoMeta object that holds all information needed to
28605 convey gst-vaapi specific data as a GstBuffer.
28607 2013-01-03 13:10:33 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28609 * gst/vaapi/gstvaapidecode.c:
28610 vaapidecode: fix calculation of the time-out value.
28611 Fix calculation of the time-out value for cases where no VA surface is
28612 available for decoding. In this case, we need to wait until downstream
28613 sink consumed at least one surface. The time-out was miscalculated as
28614 it was always set to <current-time> + one second, which is not suitable
28615 for streams with larger gaps.
28617 2013-01-03 13:05:47 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28619 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
28620 decoder: always use the calculated presentation timestamp.
28621 Use PTS value computed by the decoder, which could also be derived from
28622 the GstVideoCodecFrame PTS. This makes it possible to fix up the PTS if
28623 the original one was miscomputed or only represented a DTS instead.
28625 2013-01-02 17:33:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28627 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28628 h264: don't create sub-buffer for slice data.
28630 2013-01-03 11:16:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28632 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28633 decoder: create new context when encoded resolution changes.
28634 Create a new VA context if the encoded surface size changes because we
28635 need to keep the underlying surface pool until the last one was released.
28636 Otherwise, either of the following cases could have happened: (i) release
28637 a VA surface to an inexistent pool, or (ii) release VA surface to an
28638 existing surface pool, but with different size.
28640 2013-01-02 17:23:53 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28642 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28643 mpeg2: don't create sub-buffer for slice data.
28644 Avoid creating a GstBuffer for slice data. Rather, directly use the codec
28645 frame input buffer data. This is possible because the codec frame is valid
28646 until end_frame() where we submit the VA buffers for decoding. Anyway, the
28647 slice data buffer is copied into the VA buffer when it is created.
28649 2013-01-02 14:45:50 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28651 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28652 mpeg2: minor clean-ups.
28653 Drop explicit initialization of most fields that are implicitly set to
28654 zero. Remove some useless checks for NULL pointers.
28656 2013-01-02 14:18:31 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28658 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28659 mpeg2: optimize scan for the second start code.
28660 Optimize scan for the second start code, on the next parse() call so that
28661 to avoid scanning again earlier bytes where we didn't find any start code.
28663 2013-01-02 14:10:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28665 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28666 mpeg2: use sequence_display_extension() to compute PAR.
28667 Also compute pixel-aspect-ratio from sequence_display_extension(),
28668 should it exist in the bitstream.
28670 2013-01-02 14:02:29 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28672 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28673 mpeg2: handle sequence_display_extension().
28675 2012-12-27 15:18:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28677 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28678 mpeg2: implement {start,end}_frame() hooks.
28679 Implement GstVaapiDecoder.start_frame() and end_frame() semantics so
28680 that to create new VA context earlier and submit VA pictures to the
28681 HW for decoding as soon as possible. i.e. don't wait for the next
28682 frame to start decoding the previous one.
28684 2012-12-27 14:54:29 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28686 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28687 mpeg2: parse slice() header earlier.
28688 Parse slice() header and first macroblock position earlier in _parse()
28689 function instead of waiting for the _decode() stage. This doesn't change
28690 anything but readability.
28692 2012-12-27 14:41:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28694 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28695 mpeg2: add codec specific decoder unit.
28696 Introduce new GstVaapiDecoderUnitMpeg2 object, which holds the standard
28697 GstMpegVideoPacket and additional parsed header info. Besides, we now
28698 parse as early as in the _parse() function so that to avoid un-necessary
28699 creation of sub-buffers in _decode() for video packets that are not slices.
28701 2012-12-27 18:52:43 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28703 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28704 * gst-libs/gst/vaapi/gstvaapidecoder_frame.c:
28705 * gst-libs/gst/vaapi/gstvaapidecoder_frame.h:
28706 * gst-libs/gst/vaapi/gstvaapidecoder_unit.h:
28707 decoder: introduce lists of units to decode before/after frame.
28708 Theory of operations: all units marked as "slice" are moved to the "units"
28709 list. Since this list only contains slice data units, the prev_slice pointer
28710 was removed. Besides, we now maintain two extra lists of units to be decoded
28711 before or after slice data units.
28712 In particular, all units in the "pre_units" list will be decoded before
28713 GstVaapiDecoder::start_frame() is called and units in the "post_units"
28714 list will be decoded after GstVaapiDecoder::end_frame() is called.
28716 2013-01-02 16:06:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28718 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
28719 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
28720 decoder: drop useless checks for codec objects.
28721 Codec objects are used internally only and they are bound to be created
28722 with a valid GstVaapiDecoder object.
28724 2012-12-27 10:35:45 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28726 * gst/vaapi/gstvaapidecode.c:
28727 vaapidecode: use GST_ERROR to print error messages.
28729 2012-12-27 09:55:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28731 * gst/vaapi/gstvaapidecode.c:
28732 vaapidecode: avoid double release of frame on error.
28733 Don't call gst_video_decoder_drop_frame() if gst_video_decoder_finish_frame()
28734 was already called before and it returned an error. In that case, we were
28735 releasing the frame again, thus leading to a "double-free" condition.
28737 2012-12-21 14:29:01 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28744 * gst-libs/gst/Makefile.am:
28745 * gst-libs/gst/vaapi/Makefile.am:
28746 * gst-libs/gst/video/Makefile.am:
28747 * gst/vaapi/Makefile.am:
28748 Add videoutils submodule for GstVideoDecoder APIs.
28750 2012-12-18 16:36:01 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28753 configure: check for GstVideoDecoder API.
28754 GstVideoDecoder API is part of an unreleased GStreamer 0.10 stack. In particular,
28755 this is only available in git 0.10 branch or GStreamer >= 1.0 stack. Interested
28756 parties may either use upstream git 0.10 branch or backport the necessary support
28757 for GstVideoDecoder API, thus including helper tools like GstVideoCodecFrame et al.
28759 2012-12-18 16:21:31 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28761 * docs/reference/libs/libs.core.types:
28762 docs: remove obsolete gst_vaapi_surface_proxy_get_type().
28763 GstVaapiSurfaceProxy is no longer based on the GType system.
28765 2012-12-18 16:17:22 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28767 * docs/reference/libs/libs-sections.txt:
28768 docs: fix entries for GstVaapiSurfaceProxy.
28770 2012-12-18 15:29:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28775 2012-12-18 15:15:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28778 * gst-libs/gst/vaapi/Makefile.am:
28779 Bump library major version.
28780 Increase library major so that to cope with API/ABI incompatible changes
28781 since 0.4.x series and avoid user issues.
28783 2012-12-13 16:02:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28785 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
28786 surfaceproxy: minor clean-ups.
28788 2012-12-13 15:51:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28790 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
28791 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
28792 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
28793 surfaceproxy: drop accessors to obsolete attributes.
28794 Make GstVaapiSurfaceProxy only a thin wrapper around a VA context and a
28795 VA surface. i.e. drop any other attribute like timestamp, duration,
28796 interlaced or top-field-first.
28798 2012-12-13 15:34:10 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28800 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28801 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
28802 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
28803 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
28804 * gst/vaapi/gstvaapidecode.c:
28805 decoder: maintain decoded frames as GstVideoCodecFrame objects.
28806 Maintain decoded surfaces as GstVideoCodecFrame objects instead of
28807 GstVaapiSurfaceProxy objects. The latter will tend to be reduced to
28808 the strict minimum: a context and a surface.
28810 2012-12-13 14:30:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28812 * gst/vaapi/gstvaapidecode.c:
28813 vaapidecode: output all decoded frames as soon as possible.
28814 Make sure to push all decoded frames downstream as soon as possible.
28815 This makes sure we don't need to wait for a new frame to be ready to
28816 be decoded before receiving new decoded frames.
28817 This also separates the decode process and the output process. The latter
28818 could be moved to a specific GstTask later on.
28820 2012-12-13 14:27:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28822 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28823 * gst-libs/gst/vaapi/gstvaapidecoder.h:
28824 decoder: add gst_vaapi_decoder_get_frame() API.
28825 Add new gst_vaapi_decoder_get_frame() function meant to be used with
28826 gst_vaapi_decoder_decode(). The purpose is to return the next decoded
28827 frame as a GstVideoCodecFrame and the associated GstVaapiSurfaceProxy
28828 as the user-data object.
28830 2012-12-13 15:47:27 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28832 * gst/vaapi/gstvaapipostproc.c:
28833 vaapipostproc: use GstBuffer flags for TFF.
28834 Determine whether the buffer represents the top-field only by checking for
28835 the GST_VIDEO_BUFFER_TFF flag instead of relying on the GstVaapiSurfaceProxy
28836 flag. Also trust "interlaced" caps to determine whether the input frame
28837 is interleaved or not.
28839 2012-12-13 13:27:33 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28841 * gst/vaapi/gstvaapipostproc.c:
28842 vaapipostproc: handle video sub-buffers.
28843 Intermediate elements may produce a sub-buffer from a valid GstVaapiVideoBuffer
28844 for non raw YUV cases. Make sure vaapipostproc now understands those buffers.
28846 2012-12-18 14:57:36 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28848 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28849 h264: optimize initialization process of decoder units.
28850 Decoder units were zero-initialized, including the SPS/PPS/slice headers.
28851 The latter don't require zero-initialization since the codecparsers/ lib
28852 will do so for key variables already. This is not a great value per se but
28853 at least it makes it possible to check whether the default initialization
28854 decisions made in the codecparsers/ lib were right or not.
28855 This can be reverted if this exposes too many issues.
28857 2012-12-13 11:48:06 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28859 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28860 h264: minor clean-ups.
28861 Drop explicit initialization of most fields that are implicitly set to
28862 zero. Drop helper macros for casting to GstVaapiPictureH264 or
28863 GstVaapiFrameStore. Also remove some useless checks for NULL pointers.
28865 2012-12-07 17:45:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28867 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28868 h264: drop GstVaapiSliceH264 object.
28869 Use standard GstVaapiSlice object from now on since we already have
28870 parsed and recorded the slice headers (GstH264SliceHdr decode units).
28872 2012-12-13 10:47:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28874 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28875 h264: detect new pictures from decode-units.
28876 Update is_new_picture() to cope with GstVaapiDecoderUnitH264, instead
28877 of assuming frame boundaries when first_mb_in_slice is zero.
28879 2012-12-13 10:21:46 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28881 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28882 h264: implement {start,end}_frame() hooks.
28883 Implement GstVaapiDecoder.start_frame() and end_frame() semantics so
28884 that to create new VA context earlier and submit VA pictures to the
28885 HW for decoding as soon as possible. i.e. don't wait for the next
28886 frame to start decoding the previous one.
28888 2012-12-12 18:33:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28890 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28891 h264: optimize scan for the second start code.
28892 Optimize scan for the second start code, on the next parse() call so that
28893 to avoid scanning again earlier bytes where we didn't find any start code.
28895 2012-12-06 17:25:01 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28897 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28898 h264: add codec specific decoder unit.
28899 Introduce new GstVaapiDecoderUnitH264 object, which holds the standard
28900 NAL unit header (GstH264NalUnit) and additional parsed header info.
28901 Besides, we now parse headers as early as in the _parse() function so
28902 that to avoid un-necessary creation of sub-buffers in _decode() for
28903 NAL units that are not slices.
28904 This is a performance win by ~+1.1% only.
28906 2012-12-04 11:01:42 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28908 * gst/vaapi/gstvaapisink.c:
28909 vaapisink: handle sub video-buffers.
28910 Intermediate elements may produce a sub-buffer from a valid GstVaapiVideoBuffer
28911 for non raw YUV cases. Make sure vaapisink now understands those buffers.
28913 2012-12-12 15:22:32 +0100 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
28915 * gst/vaapi/gstvaapidecode.c:
28916 vaapidecode: use gst_vaapi_decoder_get_codec_state().
28917 Directly use the GstVideoCodecState associated with the VA decoder
28918 instead of parsing caps again.
28919 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
28920 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28922 2012-12-04 14:53:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28924 * gst/vaapi/gstvaapidecode.c:
28925 vaapidecode: use more standard helpers.
28926 Use g_clear_object() [glib >= 2.28] and gst_caps_replace() helper functions
28929 2012-12-04 14:45:29 +0100 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
28931 * gst/vaapi/gstvaapidecode.c:
28932 * gst/vaapi/gstvaapidecode.h:
28933 vaapidecode: move to GstVideoDecoder base class.
28934 Make vaapidecode derive from the standard GstVideoDecoder base element
28935 class. This simplifies the code to the strict minimum for the decoder
28936 element and makes it easier to port to GStreamer 1.x API.
28937 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
28938 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28940 2012-12-06 14:02:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28942 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
28943 jpeg: initial port to new GstVaapiDecoder API
28945 2012-12-06 14:02:21 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28947 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
28948 vc1: initial port to new GstVaapiDecoder API
28950 2012-12-06 14:02:17 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28952 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
28953 h264: initial port to new GstVaapiDecoder API
28955 2012-12-17 09:47:20 -0800 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28957 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
28958 mpeg4: initial port to new GstVaapiDecoder API
28960 2012-12-06 14:01:46 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28962 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
28963 mpeg2: initial port to new GstVaapiDecoder API.
28965 2012-12-12 15:09:21 +0100 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
28967 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28968 * gst-libs/gst/vaapi/gstvaapidecoder.h:
28969 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
28970 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
28971 * gst-libs/gst/vaapi/gstvaapiprofile.c:
28972 * gst-libs/gst/vaapi/gstvaapiprofile.h:
28973 decoder: use GstVideoCodecState.
28974 Use standard GstVideoCodecState throughout GstVaapiDecoder and expose
28975 it with a new gst_vaapi_decoder_get_codec_state() function. This makes
28976 it possible to drop picture size (width, height) information, framerate
28977 (fps_n, fps_d) information, pixel aspect ratio (par_n, par_d) information,
28978 and interlace mode (is_interlaced field).
28979 This is a new API with backwards compatibility maintained. In particular,
28980 gst_vaapi_decoder_get_caps() is still available.
28981 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
28982 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28984 2012-12-12 13:44:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28986 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28987 * gst-libs/gst/vaapi/gstvaapidecoder.h:
28988 * tests/test-decode.c:
28989 * tests/test-subpicture.c:
28990 decoder: update gst_vaapi_decoder_get_surface() semantics.
28991 Align gst_vaapi_decoder_get_surface() semantics with the rest of the
28992 API. That is, return a GstVaapiDecoderStatus and the decoded surface
28993 as a handle to GstVaapiSurfaceProxy in parameter.
28994 This is an API/ABI change.
28996 2012-12-07 16:40:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
28998 * gst-libs/gst/vaapi/gstvaapidecoder.c:
28999 decoder: use standard helper functions.
29000 Use g_clear_object(), gst_buffer_replace() and gst_caps_replace()
29001 whenever necessary.
29003 2012-11-29 15:06:00 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29005 * gst-libs/gst/vaapi/gstvaapidecoder.c:
29006 * gst-libs/gst/vaapi/gstvaapidecoder.h:
29007 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
29008 decoder: expose new parse/decode API.
29009 Introduce new decoding process whereby a GstVideoCodecFrame is created
29010 first. Next, input stream buffers are accumulated into a GstAdapter,
29011 that is then passed to the _parse() function. The GstVaapiDecoder object
29012 accumulates all parsed units and when a complete frame or field is
29013 detected, that GstVideoCodecFrame is passed to the _decode() function.
29014 Ultimately, the caller receives a GstVaapiSurfaceProxy if decoding
29015 process was successful.
29017 2012-12-13 10:20:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29019 * gst-libs/gst/vaapi/gstvaapidecoder.c:
29020 * gst-libs/gst/vaapi/gstvaapidecoder.h:
29021 decoder: add {start,end}_frame() hooks.
29022 The start_frame() hook is called prior to traversing all decode-units
29023 for decoding. The unit argument represents the first slice in the frame.
29024 Some codecs (e.g. H.264) need to wait for the first slice in order to
29025 determine the actual VA context parameters.
29027 2012-12-06 13:57:42 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29029 * gst-libs/gst/vaapi/gstvaapidecoder.c:
29030 * gst-libs/gst/vaapi/gstvaapidecoder.h:
29031 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
29032 decoder: add new GstVaapiDecoder API.
29033 Split decoding process into two steps: (i) parse incoming bitstreams
29034 into simple decoder-units until the frame or field is complete; and
29035 (ii) decode the whole frame or field at once.
29036 This is an ABI change.
29038 2012-12-05 10:51:41 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29040 * gst-libs/gst/vaapi/Makefile.am:
29041 * gst-libs/gst/vaapi/gstvaapidecoder_frame.c:
29042 * gst-libs/gst/vaapi/gstvaapidecoder_frame.h:
29043 decoder: add new "decoder-frame" object.
29044 Introduce a new GstVaapiDecoderFrame that is just a list of decoder units
29045 (GstVaapiDecoderUnit objects) that constitute a frame. This object is just
29046 an extension to GstVideoCodecFrame for VA decoder purposes. It is available
29047 as the user-data member element.
29048 This is a libgstvaapi internal object.
29050 2012-12-06 09:44:01 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29052 * gst-libs/gst/vaapi/Makefile.am:
29053 * gst-libs/gst/vaapi/gstvaapidecoder_unit.c:
29054 * gst-libs/gst/vaapi/gstvaapidecoder_unit.h:
29055 decoder: add new "decoder-unit" object.
29056 Introduce GstVaapiDecoderUnit which represents a fragment of the source
29057 stream to be decoded. For instance, a decode-unit will be a NAL unit for
29058 H.264 streams, an EBDU for VC-1 streams, and a video packet for MPEG-2
29060 This is a libgstvaapi internal object.
29062 2012-12-03 14:09:01 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29064 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29065 Port GstVaapiFrameStore to GstVaapiMiniObject.
29067 2012-12-03 11:19:08 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29069 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
29070 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
29071 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29072 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
29073 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
29074 Port codec objects to GstVaapiMiniObject.
29076 2012-12-03 13:46:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29078 * gst-libs/gst/vaapi/gstvaapidecoder.c:
29079 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
29080 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
29081 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
29082 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
29083 * gst/vaapi/gstvaapidecode.c:
29084 * gst/vaapi/gstvaapipluginbuffer.c:
29085 * tests/test-decode.c:
29086 * tests/test-subpicture.c:
29087 surfaceproxy: port to GstVaapiMiniObject.
29088 GstVaapiSurfaceProxy does not use any particular functionality from
29089 GObject. Actually, it only needs a basic object type with reference
29091 This is an API and ABI change.
29093 2012-11-30 17:25:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29095 * gst-libs/gst/vaapi/Makefile.am:
29096 * gst-libs/gst/vaapi/gstvaapiminiobject.c:
29097 * gst-libs/gst/vaapi/gstvaapiminiobject.h:
29098 Add GstVaapiMiniObject.
29099 Introduce a new reference counted object that is very lightweight and
29100 also provides flags and user-data functionalities. Initialization and
29101 finalization times are reduced by up to a factor 5x vs GstMiniObject
29102 from GStreamer 0.10 stack.
29103 This is a libgstvaapi internal object.
29105 2012-12-17 02:51:17 -0800 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29107 * tests/Makefile.am:
29108 * tests/test-decode.c:
29109 * tests/test-mpeg4.c:
29110 * tests/test-mpeg4.h:
29111 tests: add test for MPEG-4:2 decoding.
29113 2012-12-17 04:42:29 -0800 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29115 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29116 h264: initialize VA context before allocating the first slice.
29117 Fix decode_slice() to ensure a VA context exists prior to creating a
29118 new GstVaapiSliceH264, which invokes vaCreateBuffer() with some VA
29119 context ID. i.e. the latter was not initialized, thus causing failures
29120 on Cedar Trail for example.
29122 2012-12-05 09:15:32 +0800 Zhao Halley <halley.zhao@intel.com>
29125 configure: install plugin elements in GST_PLUGIN_PATH, if set.
29126 If GST_PLUGIN_PATH environment variable exists and points to a valid
29127 directory, then use it as the system installation path for gst-vaapi
29129 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29131 2012-12-17 14:27:56 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29134 configure: downgrade glib required version to 2.28.
29136 2012-12-17 09:41:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29138 * gst-libs/gst/vaapi/glibcompat.h:
29139 * gst/vaapi/gstvaapi.c:
29140 libs: fix compatibility with glib 2.28.
29141 Always prefer non deprecated APIs by default and provide compatibility
29142 glue for older glib versions when necessary.
29144 2012-12-17 10:10:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29146 * gst-libs/gst/vaapi/gstvaapidisplay.c:
29147 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
29148 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
29149 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
29150 libs: use glib >= 2.32 semantics for mutexes.
29151 Use glib >= 2.32 semantics for GMutex and GRecMutex wrt. initialization
29152 and termination. Basically, the new mutex objects can be used as static
29153 mutex objects from the deprecated APIs, e.g. GStaticMutex and GStaticRecMutex.
29155 2012-12-17 04:15:53 -0800 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29157 * gst-libs/gst/vaapi/Makefile.am:
29158 libs: only export gst_vaapi_*() symbols.
29159 This fixes symbol clashes between the gst-vaapi built-in codecparsers/
29160 library and the system-provided one, mainly used by videoparses/. Now,
29161 only symbols with the gst_vaapi_* prefix will be exported, if they are
29162 not marked as "hidden" to libgstvaapi.
29164 2012-11-20 18:21:41 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29166 * gst/vaapi/gstvaapiuploader.c:
29167 vaapiupload: reset direct-rendering to zero when changing caps.
29168 Make sure to reset direct-rendering flag to zero when caps are changed,
29169 and only derive it to one when the next checks succeed.
29171 2012-11-20 14:42:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29173 * gst/vaapi/gstvaapiupload.c:
29174 vaapiupload: fix sink caps to report the supported set of YUV caps.
29175 Try to allocate the GstVaapiUploader helper object prior to listing the
29176 supported image formats. Otherwise, only a single generic caps is output
29177 with no particular pixel format referenced in there.
29179 2012-11-20 14:32:40 +0100 Zhao Halley <halley.zhao@intel.com>
29181 * gst/vaapi/gstvaapiupload.c:
29182 * gst/vaapi/gstvaapiupload.h:
29183 vaapiupload: use new GstVaapiUploader helper.
29184 Use GstVaapiUploader helper that automatically handles direct rendering
29185 mode, thus making the "direct-rendering" property obsolete and hence it
29187 The "direct-rendering" level 2, i.e. exposing VA surface buffers, was never
29188 really well supported and it could actually trigger degraded performance.
29189 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29191 2012-11-20 15:50:56 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29193 * gst/vaapi/gstvaapisink.c:
29194 * gst/vaapi/gstvaapiuploader.c:
29195 * gst/vaapi/gstvaapiuploader.h:
29196 vaapisink: compute and expose the supported set of YUV caps.
29197 Make vaapisink expose only the set of supported caps for raw YUV buffers.
29198 Add gst_vaapi_uploader_get_caps() helper function to determine the set
29199 of supported YUV caps as source (for images). This function actually
29200 tries to zero and upload each image to a 64x64 test surface. Of course,
29201 this relies on VA drivers to not claim success if vaPutImage() is not
29202 correctly supported.
29204 2012-11-20 14:28:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29207 * gst/vaapi/Makefile.am:
29208 * gst/vaapi/gstvaapisink.c:
29209 * gst/vaapi/gstvaapisink.h:
29210 * gst/vaapi/gstvaapiuploader.c:
29211 * gst/vaapi/gstvaapiuploader.h:
29212 vaapisink: add support for raw YUV buffers.
29213 Add new GstVaapiUploader helper to upload raw YUV buffers to VA surfaces.
29214 It is up to the caller to negotiate source caps (for images) and output
29215 caps (for surfaces). gst_vaapi_uploader_has_direct_rendering() is available
29216 to help decide between the creation of a GstVaapiVideoBuffer or a regular
29217 GstBuffer on sink pads.
29218 Signed-off-by: Zhao Halley <halley.zhao@intel.com>
29219 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29221 2012-11-20 14:36:29 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29223 * gst-libs/gst/vaapi/gstvaapiimage.c:
29224 image: fix GstVaapiImage map and unmap.
29225 Fix gst_vaapi_image_map() to return TRUE and the GstVaapiImageRaw
29226 structure correctly filled in if the image was already mapped.
29227 Likewise, make gst_vaapi_image_unmap() return TRUE if the image
29228 was already unmapped.
29230 2012-10-30 13:15:45 +0800 Wind Yuan <feng.yuan@intel.com>
29233 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
29234 videobuffer: fix memory leak for surface and image.
29235 Fix reference leak of surface and image in GstVaapiVideoBuffer wrapper,
29236 thus resulting on actual memory leak of GstVaapiImage when using them
29237 for downloads/uploads from VA surfaces and more specifically surfaces
29238 when the pipeline is shutdown. i.e. vaTerminate() was never called
29239 because the resources were not unreferenced, and thus not deallocated
29241 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29243 2012-11-19 10:04:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29248 2012-11-16 18:00:10 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29250 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29251 h264: fix picture size in macroblocks.
29252 The picture size signalled by sps->{width,height} is the actual size with
29253 cropping applied, not the original size derived from pic_width_in_mbs_minus1
29254 and pic_height_in_map_units_minus1. VA driver expects that original size,
29256 There is another issue pending: frame cropping information needs to be
29259 2012-11-16 16:18:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29261 * gst-libs/gst/codecparsers/Makefile.am:
29262 codecparsers: always build parserutils first.
29263 Fix commit 18245b4 so that to link and build parserutils.[ch] first.
29264 This is needed since that's the common dependency for actual codec
29265 parsers (gstvc1parser.c for instance).
29267 2012-11-15 17:50:45 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29269 * gst-libs/gst/codecparsers/Makefile.am:
29270 codecparsers: always build the VC-1 parser library.
29271 ... this is useful to make sure pixel-aspect-ratio and framerate
29272 information are correctly parsed since we have no means to detect
29273 that at configure time.
29275 2012-11-08 11:40:47 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
29278 * gst-libs/gst/codecparsers/Makefile.am:
29279 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
29280 mpeg2: fix PAR calculation from commit bd11bae.
29281 Invoke gst_mpeg_video_finalise_mpeg2_sequence_header() to get the
29282 correct PAR values. While doing so, require a newer version of the
29283 bitstream parser library.
29284 Note: it may be necessary to also parse the Sequence_Display_Extension()
29286 Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>
29287 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29289 2012-11-15 15:00:43 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29291 * gst-libs/gst/vaapi/Makefile.am:
29292 Fix build with the GNU gold linker.
29293 In particular, fix libgstvaapi-glx DSO dependencies to include libgstbase
29294 and libgstvideo libs, e.g. for gst_video_buffer_get_overlay_composition().
29296 2012-11-02 18:18:37 +0000 Rob Bradford <rob@linux.intel.com>
29299 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
29300 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
29301 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
29302 wayland: port to 1.0 version of the protocol.
29303 This patch updates to relect the 1.0 version of the protocol. The main
29304 changes are the switch to wl_registry for global object notifications
29305 and the way that the event queue and file descriptor is processed.
29306 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29308 2012-11-14 19:22:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29310 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29311 h264: fix incorrect integration of previous commit (4d31e1e).
29312 git am got confused somehow, though the end result doesn't change at
29313 all since we require both SPS and PPS to be parsed prior to decoding
29316 2012-11-14 18:40:47 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29318 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29319 h264: start decoding slices after first SPS/PPS activation.
29320 Only start decoding slices when at least one SPS and PPS got activated.
29321 This fixes cases when a source represents a substream of another stream
29322 and no SPS and PPS was inserted before the first slice of the generated
29325 2012-11-14 14:25:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29327 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29328 h264: fix VAPictureParameterBufferH264.ReferenceFrames[] construction.
29329 ... for interlaced streams. The short_ref[] and long_ref[] arrays may
29330 contain up to 32 fields but VA ReferenceFrames[] array expects up to
29331 16 reference frames, thus including both fields.
29333 2012-11-14 10:27:12 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29335 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29336 h264: fix interlaced stream decoding with MMCO.
29337 Fix decoding of interlaced streams when adaptive_ref_pic_marking_mode_flag
29338 is equal to 1, i.e. when memory management control operations are used. In
29339 particular, when field_pic_flag is set to 0, the new reference flags shall
29340 be applied to both fields.
29342 2012-11-13 17:14:39 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29344 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29345 h264: add initial support for interlaced streams.
29346 Decoded frames are only output when they are complete, i.e. when both
29347 fields are decoded. This also means that the "interlaced" caps is not
29348 propagated to vaapipostproc or vaapisink elements. Another limitation
29349 is that interlaced bitstreams with MMCO are unlikely to work.
29351 2012-11-13 16:35:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29353 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29354 h264: split remove_reference_at() into finer units.
29355 Split remove_reference_at() into a function that actually removes the
29356 specified entry from the short-term or long-term reference picture array,
29357 and a function that sets reference flags to the desired value, possibly
29358 zero. The latters marks the picture as "unused for reference".
29360 2012-10-23 14:04:22 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29362 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
29363 decoder: fix gst_vaapi_picture_new_field() object type.
29364 Fix gst_vaapi_picture_new_field() to preserve the original picture type.
29365 e.g. gst_vaapi_picture_new_field() with a GstVaapiPictureH264 argument
29366 shall generate a GstVaapiPictureH264 object.
29368 2012-11-13 14:04:31 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29370 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29371 h264: add picture structure for reference picture marking process.
29372 Introduce new `structure' field to the H.264 specific picture structure
29373 so that to simplify the reference picture marking process. That local
29374 picture structure is derived from the original picture structure, as
29375 defined by the syntax elements field_pic_flag and bottom_field_flag.
29377 2012-11-02 15:14:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29379 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29380 h264: introduce new frame store structure.
29381 The frame store represents a Decoded Picture Buffer entry, which can
29382 hold up to two fields. So far, the frame store is only used to hold
29385 2012-11-13 10:10:31 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29387 * ext/codecparsers:
29388 codecparsers: update to gst-vaapi-rebased commit 73d6aab.
29389 73d6aab h264: fix rbsp_more_data() implementation
29390 25d04cf h264: fix error code for invalid size parsed in SPS
29391 84798e5 fix FSF address
29393 2012-10-31 16:37:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29395 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29396 h264: minor clean-ups.
29397 Move DPB flush up if the current picture to decode is an IDR. Besides,
29398 don't bother to check for IDR pictures in dpb_add() function since an
29399 explicit DPB flush was already performed in this case.
29401 2012-10-31 14:24:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29403 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29404 h264: simplify reference picture marking process.
29405 ... to build the short_ref[] and long_ref[] lists from the DPB, instead
29406 of maintaining them separately. This avoids refs/unrefs while making it
29407 possible to generate the list based on the actual picture structure.
29408 This also ensures that the list of generated ReferenceFrames[] actually
29409 matches what reference frames are available in the DPB. i.e. short_ref[]
29410 and long_ref[] entries are implied from the DPB, so there is no risk of
29411 having "dangling" references.
29413 2012-10-31 11:52:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29415 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29416 h264: introduce per-field POC in GstVaapiPictureH264.
29417 Use the POC member available in the GstVaapiPicture base class and
29418 get rid of the dependency on the local VAPictureH264 TopFieldOrderCnt
29419 and BottomFieldOrderCnt. Rather, use a simple field_poc[] array
29420 initialized to INT_MAX, so that to simplify picture POC calculation
29421 for non frame pictures.
29423 2012-10-31 11:45:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29425 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29426 h264: introduce GST_VAAPI_PICTURE_{SHORT,LONG}_TERM_REFERENCE flags.
29427 Further get rid of GstVaapiPictureH264-local VAPictureH264.flags for
29428 reference bits, thus simplifying the reference picture marking process
29429 to only track a single set of reference flags. Also introduce a new
29430 long_term_frame_idx member.
29432 2012-10-31 11:33:40 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29434 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29435 h264: introduce GST_VAAPI_PICTURE_FLAG_IDR flag.
29437 2012-10-31 10:56:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29439 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29440 h264: fill in GstVaapiPicture structure.
29441 ... and get rid of local VAPictureH264.flags fields in GstVaapiPictureH264.
29443 2012-10-31 11:07:48 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29445 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29446 h264: add vaapi_fill_picture() helper.
29447 Add vaapi_fill_picture() helper function to convert GstVaapiPictureH264
29448 to VAPictureH264 structure. This is preparatory work to get rid of the
29449 local VAPictureH264 member in GstVaapiPictureH264.
29451 2012-10-26 16:12:05 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29453 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29454 h264: fix activation order of picture and sequence parameters.
29455 Delay ensure_context() until we actually need a VA context for allocating
29456 new VA surfaces, and then GstVaapiPictures, but also when a real activation
29457 of a new picture parameter set occurs, thus also implying an activation
29458 of the related sequence parameter set.
29459 The most important thing was to drop the global pps and sps pointers since
29460 they may not have matched the currently activated picture parameter or
29461 sequence parameter sets at the specified decode point.
29462 Anoter positive side-effect is that this cleans up all occurrences of
29463 decode_current_picture() to only keep those useful in decode_picture(),
29464 before a new picture is allocated, or in decode_sequence_end() when
29465 an end-of-stream or end-of-sequence condition occurred.
29467 2012-10-26 13:17:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29469 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29470 h264: fix scaling list generation.
29471 ... aka fix regression from efaab79. In particular, ScalingList8x8[]
29472 array was partially copied to the VAIQMatrixBufferH264. While we are
29473 at it, also improve bounds checking and avoid copying 8x8 scaling
29474 lists if transform_8x8_mode_flag is set to 0.
29476 2012-10-24 18:23:09 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29478 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29479 h264: fix detection of picture boundaries.
29480 Strictly follow the standard (7.4.1.2.4) to detect the first VCL NAL
29481 unit of a primary coded picture.
29483 2012-10-23 14:50:14 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29485 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29486 h264: optimize handling of scaling lists.
29487 Don't copy scaling lists twice to an intermediate state. Rather, directly
29488 use the scaling lists from GstH264PPS since they would match those provided
29489 by SPS header, if necessary. i.e. if PPS-specific scaling lists are not
29490 available in the bitstream.
29492 2012-10-23 10:33:50 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29494 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29495 h264: simplify code when MMCO is 5.
29496 Remove exit_picture() and exit_picture_poc() since PicOrderCnt(CurrPic)
29497 is now updated accordingly to the standard. Besides, MMCO = 5 specific
29498 operations are moved up to exec_ref_pic_marking_adaptive_mmco_5().
29500 2012-10-22 11:52:13 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29502 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29503 h264: fix MMCO-based reference picture marking process.
29504 Fix adaptive memory control decoded reference picture marking process
29505 implementation for operations 2 to 6, thus also fixing support for
29506 long-term reference pictures.
29508 2012-10-22 10:50:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29510 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29511 h264: move MMCO handlers out of the loop (cosmetics).
29512 This change only splits each individual MMCO handler into several functions
29513 dedicated for each operation. This is needed to perform further work later
29516 2012-10-17 15:49:23 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29519 debian: fix make dist for packaging.
29520 bzip2 tarballs are now used, so update the deb.upstream dependencies
29521 to include dist-bzip2 instead of plain old dist, and use the correct
29522 tar extract options to handle that format.
29524 2012-10-17 15:42:17 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29527 configure: generate bzip2 tarballs in ustar format by default.
29529 2012-10-17 15:38:14 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29532 configure: bump glib required version to 2.31.2.
29533 Use new Thread API. In particular, g_mutex_init() and g_cond_init()
29534 rather than g_mutex_new() and g_cond_new() respectively.
29536 2012-10-04 17:39:53 +0100 Rob Bradford <rob@linux.intel.com>
29538 * gst/vaapi/gstvaapidecode.c:
29539 * gst/vaapi/gstvaapidecode.h:
29540 vaapidecode: adopt non-deprecrated glib locking primitive pattern.
29541 The use of heap allocated GMutex/GCond is deprecated. Instead place them
29542 inside the structure they are locking.
29543 These changes switch to use g_mutex_init/g_cond_init rather than the heap
29544 allocation functions.
29545 Because we cannot test for a NULL pointer for the GMutex/GCond we must
29546 initialise inside the GObject _init function and clear inside the _finalize
29547 which is guaranteed to only be called once and after the object is no longer
29550 2012-10-17 14:52:35 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29552 * gst/vaapi/gstvaapidecode.c:
29553 vaapidecode: fix compiler warnings.
29554 Don't care of the return value for gst_vaapi_decoder_put_buffer()
29555 during destruction of the element. Don't print out (uninitialised)
29556 error code when allocation of video buffer failed.
29558 2012-10-16 16:52:04 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29560 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29561 h264: add flag to compile with strict DPB ordering mode.
29562 Allow build with strict DPB ordering mode whereby evicted entries
29563 are replaced by the next entries, in order instead of optimizing
29564 it away with the last entry in the DPB.
29565 This is only useful for debugging purpose, against a reference SW
29566 decoder for example.
29568 2012-10-16 16:46:17 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29570 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29571 h264: drop extra code covered by built-in codecparsers.
29572 GstH264SliceHdr.n_emulation_prevention_bytes is bound to exist now that
29573 a newer version of codecparsers/ are used if the system provided one is
29574 now recent enough to have those required extensions.
29576 2012-10-16 16:43:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29578 * gst-libs/gst/codecparsers/Makefile.am:
29579 codecparsers: fix generation of symlinks.
29580 Try to improve dependencies while generating symlinks to externally
29581 maintained copy of codecparsers (derived from upstream git master
29584 2012-10-11 15:04:12 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29586 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
29587 vc1: use framerate information from bitstream parser.
29589 2012-09-27 18:05:46 +0100 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
29591 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
29592 vc1: use pixel-aspect-ratio from bitstream parser.
29593 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
29594 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29596 2012-09-27 18:05:46 +0100 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
29598 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
29599 mpeg2: use pixel-aspec-ratio information from bitstream parser.
29600 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
29601 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29603 2012-10-11 13:49:14 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29606 * gst-libs/gst/codecparsers/Makefile.am:
29607 codecparsers: h264: use submodule sources.
29608 Use newer sources from the codecparsers/ submodule for
29609 - GstH264SliceHdr.n_emulation_prevention_bytes: EPBs;
29610 - GstH264VUIParams.{par_n,par_d}: pixel-aspect-ratio.
29612 2012-10-11 13:23:02 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29616 * gst-libs/gst/codecparsers/Makefile.am:
29617 * gst-libs/gst/codecparsers/gstjpegparser.c:
29618 * gst-libs/gst/codecparsers/gstjpegparser.h:
29619 * gst-libs/gst/vaapi/Makefile.am:
29620 codecparsers: jpeg: use submodule sources.
29622 2012-10-11 10:03:14 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29629 * ext/codecparsers:
29630 Add codecparsers submodule.
29632 2012-10-11 14:17:12 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29635 .gitignore: updates.
29637 2012-10-11 13:40:37 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29640 autogen: fix check for gtkdocize and autoreconf.
29641 If gtkdocize or autoreconf programs were not found, then the autogen.sh
29642 script would fail to report that correctly because test -z was not passed
29643 any argument (empty string "" in this case).
29645 2012-09-27 18:05:46 +0100 Simon Farnsworth <simon.farnsworth@onelan.co.uk>
29647 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29648 h264: use pixel-aspect-ratio from SPS header.
29649 Propagate pixel-aspect-ratio determined by the GStreamer codecparser
29650 from the sequence headers.
29651 Signed-off-by: Simon Farnsworth <simon.farnsworth@onelan.co.uk>
29652 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29654 2012-10-10 10:35:20 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29656 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29657 h264: add decode_nalu() helper function.
29658 Split decode_buffer() into the core infrastructure that determines
29659 the NAL units contained in the adapter and the actual function that
29660 decodes the NAL unit.
29662 2012-10-10 10:31:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29664 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29665 h264: fix end-of-stream conditions (flush).
29666 Decode pending data in the adapter prior to processing the actual
29667 code for end-of-stream.
29669 2012-10-10 09:45:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29671 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
29672 mpeg2: add decode_packet() helper function.
29673 Split decode_buffer() into the core infrastructure that determines
29674 the packets contained in the adapter and the actual function that
29675 decodes the packet data.
29677 2012-10-09 15:34:18 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29679 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
29680 mpeg2: fix end-of-stream conditions (flush).
29681 Decode pending data in the adapter prior to processing the actual
29682 code for end-of-stream. Initial code from Feng Yuan.
29684 2012-10-09 15:40:49 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29686 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
29687 mpeg2: fix memory leak of empty packets.
29688 Fix memory leakage of empty packets, i.e. packets that only contain
29689 the start code prefix. In particular, free empty user-data packets.
29690 Besides, the codec parser will already fail gracefully if the packet
29691 to parse does not have the minimum required size. So, we can also
29692 completely drop the block of code that used to handle packets of size 4
29693 (including the start code).
29695 2012-10-09 15:01:38 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29697 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
29698 mpeg2: fix return value for "no-data" conditions.
29699 Fix return value when the second scan for start code fails. This means
29700 there is not enough data to determine the full extents of the current
29701 packet and the function shall return GST_VAAPI_DECODER_STATUS_ERROR_NO_DATA
29702 in this case, instead of GST_VAAPI_DECODER_STATUS_SUCCESS.
29704 2012-10-09 14:48:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29706 * gst-libs/gst/vaapi/gstvaapidecoder.c:
29707 decoder: refine semantics of gst_vaapi_decoder_put_buffer().
29708 Improve the semantics for gst_vaapi_decoder_put_buffer() when an empty
29709 buffer is passed on. An empty buffer is a buffer with a NULL data pointer
29710 or with a size equals to zero. In this case, that buffer is simply
29711 skipped and the function returns TRUE. A NULL buffer argument still
29712 marks the end-of-stream.
29714 2012-10-09 14:40:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29716 * gst-libs/gst/vaapi/gstvaapidecoder.c:
29717 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
29718 decoder: drop unused functions.
29720 2012-08-26 22:29:04 -0400 Wind Yuan <feng.yuan@intel.com>
29722 * gst/vaapi/gstvaapidecode.c:
29723 vaapidecode: flush buffers when receiving EOS.
29724 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29726 2012-10-05 13:36:27 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29728 * debian.upstream/Makefile.am:
29729 debian: fix make dist for packaging.
29731 2012-10-05 12:06:27 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29733 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
29734 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
29735 wayland: cosmetics (remove tabs).
29737 2012-10-04 17:39:52 +0100 Rob Bradford <rob@linux.intel.com>
29739 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
29740 wayland: add support for windowed mode.
29741 Rather than always making the surface fullscreen instead implement the
29742 set_fullscreen vfunc on GstVaapiWindow and then set the shell surface
29743 fullscreen on not depending on that.
29744 Reviewed-by: Joe Konno <joe.konno@intel.com>
29745 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29747 2012-10-01 09:21:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29750 Bump version for development.
29752 2012-09-28 17:54:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29756 Fix and document build dependencies better.
29758 2012-09-28 17:41:42 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29760 * debian.upstream/control.in:
29761 debian: fix GStreamer build dependencies.
29763 2012-09-28 17:39:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29766 * debian.upstream/control.in:
29767 debian: fix Wayland build dependencies.
29769 2012-09-28 17:38:17 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29771 * debian.upstream/control.in:
29772 debian: fix conditional build of packages.
29773 Make it still possible to build package even if one of the build dependencies
29774 for a specific video backend is not available.
29776 2012-09-27 11:08:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29778 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
29779 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
29780 utils: drop unused GLX helpers.
29781 Remove helpers for GL_ARB_fragment_program and GL_ARB_multitexture
29782 extensions since they are not used throughout gstreamer-vaapi.
29784 2012-09-27 11:04:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29786 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
29787 utils: fix build with <GL/glext.h> version >= 85.
29788 Mesa recently updated the <GL/glext.h> header version to Khronos version 85.
29789 This caused the PFNGLMULTITEXCOORD2FPROC definition to be moved out of the
29790 GL_VERSION_1_3_DEPRECATED block. However, since <GL/gl.h> also defines
29791 GL_VERSION_1_3 to 1, the definitions in <GL/glext.h> are then not enabled,
29792 thus leaving PFNGLMULTITEXCOORD2FPROC undefined as well.
29793 Provide a PFNGLMULTITEXCOORD2FPROC replacement as an interim solution for
29794 newer versions of the <GL/glext.h> header.
29796 2012-09-26 16:33:16 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29799 configure: update VA-API version requirements.
29800 VA/DRM and VA/Wayland API are now promoted to VA-API 0.33.0 (libva 1.1.0).
29802 2012-09-21 16:43:38 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29804 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29805 h264: review and report errors accordingly.
29806 Use GST_ERROR() to report real errors instead of hiding them into
29809 2012-09-20 17:58:21 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29811 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29812 h264: exclusively use GstAdapter, drop sub-buffer hack.
29813 Maintaining the sub-buffer is rather suboptimal especially since we
29814 were also maintaining a GstAdapter. Now, we only use the GstAdapter
29815 thus requiring minor extra parsing when receiving avcC buffers.
29817 2012-09-20 16:18:27 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29822 2012-09-20 16:02:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29827 2012-09-20 14:38:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29829 * debian.upstream/gstreamer-vaapi.install.in:
29830 debian: fix packaging on recent Ubuntu platforms.
29831 Use explicit GStreamer plugins path.
29833 2012-09-17 17:55:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29835 * docs/reference/libs/libs-docs.xml.in:
29836 * docs/reference/libs/libs.core.types:
29837 docs: fix build for make dist.
29839 2012-09-14 10:30:35 -0400 Kristian Høgsberg <krh@bitplanet.net>
29841 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
29842 wayland: set opaque region for YUV surface.
29843 This allows the compositor to optimize redraws and cull away changes
29844 obscured by the video surface.
29845 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29847 2012-09-14 17:30:19 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29850 configure: fix check for libva-glx and libva-drm.
29852 2012-09-12 13:42:49 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29854 * gst-libs/gst/vaapi/glibcompat.h:
29855 glibcompat: add replacement for g_cond_wait_until().
29857 2012-09-12 13:41:47 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29859 * gst/vaapi/gstvaapidecode.c:
29860 * gst/vaapi/gstvaapidownload.c:
29861 * gst/vaapi/gstvaapipostproc.c:
29862 * gst/vaapi/gstvaapisink.c:
29863 * gst/vaapi/gstvaapiupload.c:
29864 plugins: include "sysdeps.h" instead of "config.h".
29866 2012-09-12 10:40:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29868 * gst-libs/gst/codecparsers/gstjpegparser.c:
29869 codecparsers: jpeg: add missing includes.
29871 2012-09-11 17:03:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29873 * gst-libs/gst/vaapi/gstvaapidecoder.c:
29874 * gst-libs/gst/vaapi/gstvaapidecoder.h:
29875 * gst/vaapi/gstvaapidecode.c:
29876 vaapidecode: don't reset decoder if codec type is the same.
29877 Reset, i.e. destroy then create, the decoder in _setcaps() handler only
29878 if the underlying codec type actually changed. This makes it possible
29879 to be more tolerant with certain MPEG-2 streams that get parsed to
29880 form caps that are compatible with the previous state but minor changes
29883 2012-09-11 16:41:32 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29885 * gst/vaapi/gstvaapidecode.c:
29886 vaapidecode: simplify codec lookup from caps.
29887 Add new gst_vaapi_codec_from_caps() helper to determine codec type from
29888 the specified caps. Don't globally expose this function since this is
29889 really trivial and only used in the vaapidecode element.
29891 2012-09-11 15:54:20 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29893 * gst/vaapi/gstvaapidecode.c:
29894 * gst/vaapi/gstvaapidecode.h:
29895 vaapidecode: improve "no free surface" conditions.
29896 Previously, vaapidecode would wait up to one second until a free surface
29897 is available, or it aborts decoding. Now, vaapidecode waits until the
29898 last decoded surface was to be presented, plus one second. Besides, end
29899 times are now expressed relative to the monotonic clock.
29901 2012-09-11 10:59:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29903 * gst-libs/gst/vaapi/gstvaapidecoder.c:
29904 * gst/vaapi/gstvaapidecode.c:
29905 decoder: propagate buffer duration downstream.
29907 2012-09-11 10:59:10 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29909 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
29910 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
29911 surfaceproxy: add "duration" property.
29913 2012-09-10 18:26:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29915 * gst-libs/gst/vaapi/gstvaapidecoder.c:
29916 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29917 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
29918 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
29919 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
29920 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
29921 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
29922 decoder: cope with new GstVaapiContextInfo based API.
29923 Update decoders to report the maximum number of reference frames to use.
29925 2012-09-10 18:17:10 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29927 * gst-libs/gst/vaapi/gstvaapicontext.c:
29928 context: JPEG codec does not need any reference frame.
29930 2012-09-10 18:15:02 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29932 * gst-libs/gst/vaapi/gstvaapicontext.c:
29933 * gst-libs/gst/vaapi/gstvaapicontext.h:
29934 context: allow number of reference frames to be set.
29935 Make it possible to specify the maximum number of references to use within
29936 a single VA context. This helps reducing GPU memory allocations to the useful
29937 number of references to be used.
29939 2012-09-07 16:41:16 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29941 * gst/vaapi/gstvaapipostproc.c:
29942 vaapipostproc: fix deinterlace-{mode,method} types definition.
29944 2012-09-07 16:15:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29946 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
29947 mpeg4: fix debug info for unsupported profile.
29949 2012-09-07 16:14:11 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29951 * gst-libs/gst/vaapi/gstvaapicontext.c:
29952 * gst-libs/gst/vaapi/gstvaapidecoder.c:
29953 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
29954 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
29955 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
29956 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
29957 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
29958 * gst-libs/gst/vaapi/gstvaapidisplay.c:
29959 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
29960 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
29961 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
29962 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
29963 * gst-libs/gst/vaapi/gstvaapiimage.c:
29964 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
29965 * gst-libs/gst/vaapi/gstvaapiobject.c:
29966 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
29967 * gst-libs/gst/vaapi/gstvaapisurface.c:
29968 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
29969 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
29970 * gst-libs/gst/vaapi/gstvaapitexture.c:
29971 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
29972 * gst-libs/gst/vaapi/gstvaapivalue.c:
29973 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
29974 * gst-libs/gst/vaapi/gstvaapivideobuffer_glx.c:
29975 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
29976 * gst-libs/gst/vaapi/gstvaapivideopool.c:
29977 * gst-libs/gst/vaapi/gstvaapiwindow.c:
29978 * gst-libs/gst/vaapi/gstvaapiwindow_drm.c:
29979 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
29980 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
29981 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
29982 libs: fix build in strict ISO C mode.
29984 2012-09-07 16:11:12 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29986 * gst/vaapi/gstvaapi.c:
29987 * gst/vaapi/gstvaapidecode.c:
29988 * gst/vaapi/gstvaapidownload.c:
29989 * gst/vaapi/gstvaapipostproc.c:
29990 * gst/vaapi/gstvaapisink.c:
29991 * gst/vaapi/gstvaapiupload.c:
29992 plugins: fix build in strict ISO C mode.
29994 2012-09-07 15:31:09 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
29996 * pkgconfig/gstreamer-vaapi-glx.pc.in:
29997 * pkgconfig/gstreamer-vaapi-x11.pc.in:
29998 * pkgconfig/gstreamer-vaapi.pc.in:
29999 pkgconfig: fix dependencies and slightly improve description.
30000 Drop @LIBVA_EXTRA_{CFLAGS,LIBS}@ substitutions and slightly improve
30001 descriptions with clearer renderer names.
30003 2012-09-04 13:54:19 +0200 Philip Lorenz <philip@bithub.de>
30005 * gst/vaapi/gstvaapidecode.c:
30006 vaapidecode: acquire lock only if the mutex exists.
30007 When playback stops the GstVaapiDecode object is reset into a clean
30008 state. However, surfaces may still be referenced by library users and
30009 unreferencing them after the reset triggers an access to an unset mutex.
30010 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30012 2012-09-07 11:58:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30014 * gst-libs/gst/vaapi/gstvaapidecoder.c:
30015 decoder: drop extraneous return for void function.
30017 2012-09-07 11:57:59 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30019 * gst-libs/gst/vaapi/gstvaapiimage.c:
30020 image: don't use (void *) pointer arithmetic.
30022 2012-09-04 13:40:04 +0200 Philip Lorenz <philip@bithub.de>
30024 * gst-libs/gst/vaapi/gstvaapidecoder.h:
30025 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
30026 * gst-libs/gst/vaapi/gstvaapidisplay.h:
30027 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
30028 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
30029 * gst-libs/gst/vaapi/gstvaapiprofile.h:
30030 * gst-libs/gst/vaapi/gstvaapisurface.h:
30031 * gst-libs/gst/vaapi/gstvaapitypes.h:
30032 * gst/vaapi/gstvaapipostproc.h:
30033 Do not forward declare enums.
30034 Forward declaring enums is not allowed by the C standard and aborts
30035 compilation if the header file is included in a C++ project.
30036 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30038 2012-09-07 11:44:44 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30041 configure: fix check for VA/DRM API.
30043 2012-09-04 11:53:18 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30045 * gst/vaapi/gstvaapisink.c:
30046 vaapisink: fix calculation of window size.
30047 If either dimension is out-of-bounds, then scale window to fit the
30048 display size, even if the output is to be rotated. Use the standard
30049 gst_video_sink_center_rect() function to center and scale the window
30050 wrt. the outer (display) bounds.
30052 2012-08-28 02:45:22 -0400 Wind Yuan <feng.yuan@intel.com>
30054 * gst/vaapi/gstvaapisink.c:
30055 * gst/vaapi/gstvaapisink.h:
30056 vaapisink: add video rotation support.
30057 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30059 2012-09-06 11:47:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30061 * gst/vaapi/gstvaapipluginutil.h:
30062 pluginutils: add G_PRIMITIVE_SWAP() helper macro.
30063 This macro helps swapping variables while maintaining the correct underlying
30064 and primitive type.
30066 2012-09-06 11:51:41 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30068 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30069 display: fix display aspect ratio when display is rotated.
30071 2012-09-06 11:50:21 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30074 * gst-libs/gst/vaapi/Makefile.am:
30075 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
30076 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
30077 display: fix physical display size when display is rotated.
30079 2012-08-30 16:27:56 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30081 * gst-libs/gst/vaapi/gstvaapivalue.c:
30082 display: fix GstVaapiRotation enumeration of values.
30084 2012-08-29 13:18:05 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30086 * docs/reference/libs/libs-docs.xml.in:
30087 * docs/reference/libs/libs-sections.txt:
30088 * gst-libs/gst/vaapi/Makefile.am:
30089 * gst-libs/gst/vaapi/gstvaapivideosink.c:
30090 * gst-libs/gst/vaapi/gstvaapivideosink.h:
30091 * gst/vaapi/gstvaapidecode.c:
30092 * gst/vaapi/gstvaapidownload.c:
30093 * gst/vaapi/gstvaapipostproc.c:
30094 * gst/vaapi/gstvaapisink.c:
30095 * gst/vaapi/gstvaapiupload.c:
30096 vaapisink: drop obsolete GstVaapiVideoSink interface.
30097 This interface was deprecated since 0.3.x series when the GstVideoContext
30098 interface was added to the main GStreamer APIs.
30100 2012-08-27 18:34:27 +0300 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30102 * gst/vaapi/gstvaapisink.c:
30103 * gst/vaapi/gstvaapisink.h:
30104 vaapisink: automatically detect overlay rendering mode.
30105 Retain the VA surface until another surface is to be displayed only
30106 if VA display rendering mode is determined to be "overlay" mode.
30108 2012-08-24 16:30:33 +0300 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30110 * gst/vaapi/gstvaapisink.c:
30111 * gst/vaapi/gstvaapisink.h:
30112 vaapisink: retain VA surface until another one is displayed.
30113 Keep VA surface proxy associated with the surface that is currently
30114 being displayed. This makes sure that surface is not released back
30115 to the pool of surfaces free to use for decoding. This is necessary
30116 with VA driver implementations that support rendering to an overlay
30117 pipe. Otherwise, there could be cases where we are decoding into a
30118 surface that is being displayed, hence some flickering.
30120 2012-08-24 14:54:16 +0300 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30122 * gst/vaapi/gstvaapisink.h:
30123 vaapisink: fix build with older toolchains.
30124 Don't re-declare GstVaapiTexture if USE_GLX mode is set.
30126 2012-08-29 10:13:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30128 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30129 display: partially revert 8ebe4d6.
30130 Don't try to fix up the initial values, this could make things worse.
30131 Simply assume the driver does not support the capability in this case.
30133 2012-08-28 16:08:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30135 * tests/test-display.c:
30136 tests: dump VA display properties.
30138 2012-08-28 18:11:32 +0300 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30140 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30141 display: fix validation process of properties during discovery.
30142 Some VA drivers (e.g. EMGD) can have completely random values for initial
30143 display attributes. So, try to improve the discovery process to check the
30144 initial display attribute values actually fall within valid bounds. If not,
30145 try to reset those to some sensible values like the default value reported
30146 through vaQueryDisplayAttributes().
30148 2012-08-28 13:59:50 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30150 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30151 * gst-libs/gst/vaapi/gstvaapidisplay.h:
30152 display: add color balance properties.
30153 Add support for hue, saturation, brightness and contrast attributes.
30155 2012-08-28 14:05:16 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30157 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30158 display: initialize default attribute values.
30159 Ensure the display attribute is actually supported by trying to retrieve
30160 its current value during GstVaapiDisplay creation.
30162 2012-08-28 11:09:56 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30164 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30165 display: raise "notify" for property changes.
30167 2012-08-28 10:55:59 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30169 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30170 * gst-libs/gst/vaapi/gstvaapidisplay.h:
30171 display: expose display attributes as GObject properties.
30172 Expose VA display "render-mode" and "rotation" attributes as standard
30173 GObject properties.
30175 2012-08-28 16:24:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30177 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30178 display: install properties in batch.
30179 Use g_object_class_install_properties() to install GstVaapiDisplay properties.
30180 It is useful to maintain properties as GParamSpec so that to be able to raise
30181 "notify" signals by id instead of by name in the future.
30183 2012-08-27 19:00:37 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30185 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30186 display: fix gst_vaapi_display_has_property().
30187 Append the "render-mode" and "rotation" properties, should they be supported
30188 by the underlying VA driver.
30190 2012-08-22 02:18:11 -0400 Wind Yuan <feng.yuan@intel.com>
30192 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30193 * gst-libs/gst/vaapi/gstvaapidisplay.h:
30194 * gst-libs/gst/vaapi/gstvaapitypes.h:
30195 * gst-libs/gst/vaapi/gstvaapiutils.c:
30196 * gst-libs/gst/vaapi/gstvaapiutils.h:
30197 * gst-libs/gst/vaapi/gstvaapivalue.c:
30198 * gst-libs/gst/vaapi/gstvaapivalue.h:
30199 display: add support for rotation modes.
30200 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30202 2012-08-27 18:11:37 +0300 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30204 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30205 * gst-libs/gst/vaapi/gstvaapidisplay.h:
30206 * gst-libs/gst/vaapi/gstvaapitypes.h:
30207 * gst-libs/gst/vaapi/gstvaapivalue.c:
30208 * gst-libs/gst/vaapi/gstvaapivalue.h:
30209 display: add support for rendering modes.
30210 A rendering mode can be "overlay" or "texture"'ed blit.
30211 The former mode implies that a VA surface used for rendering can't be
30212 re-used right away for decoding, so the sink shall make provisions to
30213 retain the associated surface proxy until the next surface is to be
30215 The latter mode implies that the VA surface is implicitly copied to an
30216 intermediate backing store, or back buffer of a frame buffer, so the
30217 associated surface proxy can be disposed right away.
30219 2012-08-27 17:02:49 +0300 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30221 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30222 * gst-libs/gst/vaapi/gstvaapidisplay.h:
30223 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
30224 display: add initial support for display attributes.
30225 The VA display attributes are mapped to properties so that to maintain the
30226 GStreamer terminology. Properties are to be identified by name, but internal
30227 functions are available to lookup the property by the actual VA display
30230 2012-08-24 11:36:16 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
30232 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
30233 jpeg: fix end-of-image (EOI) handler.
30234 decode_current_picture() was converted to return a gboolean instead
30235 of a GstVaapiDecoderStatus, so we were not getting out of the decode
30236 loop as expected, or could cause an error instead.
30237 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30239 2012-08-24 18:41:47 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30241 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
30242 mpeg2: include start code into VA slice data buffer.
30243 Integrate the start code prefix in the slice data buffer that is submitted
30244 to the hardware. VA-API specifies that slice_data_offset is the offset to
30245 the first byte of slice data. And, for MPEG-2, slice() data begins with
30246 the slice_start_code. Some VA driver implementations (EMGD) expect this.
30248 2012-06-28 01:08:03 +0900 Javier Jardón <jjardon@gnome.org>
30251 autogen: fix configure script generation when srcdir != builddir.
30252 This patch allows for regenerating the configure script from a build
30253 directory that is not the actual source directory.
30254 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30256 2012-06-28 00:22:03 +0900 Javier Jardón <jjardon@gnome.org>
30259 configure: use new libtool syntax.
30260 This now requires libtool >= 2.2 to regenerate the configure script.
30261 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30263 2012-08-08 12:50:41 +0900 Javier Jardón <jjardon@gnome.org>
30265 * gst-libs/gst/vaapi/gstvaapidecoder.c:
30266 decoder: use g_object_notify_by_pspec().
30267 Use g_object_notify_by_pspec() instead of g_object_notify() so that to
30268 avoid a property name lookup. i.e. this makes notifications faster to
30269 the `vaapidecode' element.
30270 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30272 2012-08-06 19:21:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30274 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
30275 wayland: handle de-interlacing flags.
30276 VA/Wayland API was updated to allow flags for bob deinterlacing.
30277 More elaborated filters will require a complete VA/VPP pipeline.
30279 2012-08-02 18:27:48 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30281 * gst-libs/gst/codecparsers/gstjpegparser.c:
30282 jpeg: fix default quantization tables.
30283 Two elements in the luminance quantization table were wrong. So,
30284 gst_jpeg_get_default_quantization_tables() now reconstructs tables
30285 in zig-zag order from the standard ones (Tables K.1 and K.2).
30287 2012-08-02 15:17:57 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30289 * gst-libs/gst/codecparsers/gstjpegparser.c:
30290 jpeg: compute default Huffman tables.
30291 ... instead of having them pre-calculated. This saves around 1.5 KB
30292 of data in the DSO but requires gst_jpeg_get_default_huffman_tables()
30293 to do more work. Though, the client application may have to call that
30294 function at most once, only.
30296 2012-08-01 18:30:27 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30298 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30299 display: drop VAProfileNone entries from debug messages.
30301 2012-07-31 18:24:14 +0800 Yan Yin <yan.yin@intel.com>
30303 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30304 display: query for supported display attributes.
30305 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30307 2012-07-31 18:22:48 +0800 Yan Yin <yan.yin@intel.com>
30309 * gst-libs/gst/vaapi/gstvaapiutils.c:
30310 * gst-libs/gst/vaapi/gstvaapiutils.h:
30311 utils: add string_of_VADisplayAttributeType() helper.
30312 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30314 2012-08-01 15:46:35 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30316 * gst/vaapi/gstvaapisink.c:
30317 vaapisink: handle VA/DRM API.
30318 This is not useful in practice but for raw performance evaluation when
30319 the sink is invoked with display=drm sync=false. fakesink could also be
30322 2012-08-01 15:46:19 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30324 * gst/vaapi/Makefile.am:
30325 * gst/vaapi/gstvaapipluginutil.c:
30326 plugins: add support for headless pipelines.
30328 2012-08-01 15:44:49 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30330 * tests/Makefile.am:
30332 * tests/test-display.c:
30333 * tests/test-windows.c:
30334 tests: add support for headless decoding.
30336 2012-08-01 15:44:02 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30340 * debian.upstream/Makefile.am:
30341 * debian.upstream/control.in:
30342 * debian.upstream/libgstvaapi-drm.install.in:
30343 * gst-libs/gst/vaapi/Makefile.am:
30344 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30345 * gst-libs/gst/vaapi/gstvaapidisplay.h:
30346 * gst-libs/gst/vaapi/gstvaapidisplay_drm.c:
30347 * gst-libs/gst/vaapi/gstvaapidisplay_drm.h:
30348 * gst-libs/gst/vaapi/gstvaapidisplay_drm_priv.h:
30349 * gst-libs/gst/vaapi/gstvaapiwindow_drm.c:
30350 * gst-libs/gst/vaapi/gstvaapiwindow_drm.h:
30351 * pkgconfig/Makefile.am:
30352 * pkgconfig/gstreamer-vaapi-drm.pc.in:
30353 Add initial support for VA/DRM.
30355 2012-07-31 17:58:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30358 * gst-libs/gst/vaapi/Makefile.am:
30359 Fix build without X11 (again).
30360 Don't try to build libgstvaapi-x11.so.* if X11 was disabled. Also shuffle
30361 files list wrt. x11, glx and wayland backends.
30363 2012-07-31 11:51:57 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30366 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
30367 * gst-libs/gst/vaapi/gstvaapicompat.h:
30368 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
30369 jpeg: update to the latest VA-API changes (0.32.1+).
30371 2012-07-27 14:27:05 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30373 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
30374 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
30375 * gst/vaapi/gstvaapisink.c:
30376 wayland: implement display ::get_size*() hooks.
30378 2012-07-27 10:45:41 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30380 * gst/vaapi/gstvaapipluginutil.c:
30381 plugins: prefer X11 rendering over GLX.
30382 Prefer X11 display over GLX so that "vaapisink" uses X11, i.e. vaPutSurface(),
30383 for rendering instead of texturing.
30385 2012-07-26 09:28:51 -0400 Kristian Høgsberg <krh@bitplanet.net>
30387 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
30388 wayland: use scale fullscreen method.
30389 This makes the compositor scale the surface to fit and preserves aspect
30391 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30393 2012-07-26 09:27:47 -0400 Kristian Høgsberg <krh@bitplanet.net>
30395 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
30396 wayland: respond to ping/pong protocol so we're not deemed unresponsive.
30397 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30399 2012-07-25 10:39:04 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30401 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
30402 wayland: fix double disconnect of display.
30404 2012-07-24 19:58:55 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30406 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
30407 wayland: mangle display name for cache lookups.
30409 2012-07-24 15:43:44 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30412 * debian.upstream/Makefile.am:
30413 * debian.upstream/control.in:
30414 * debian.upstream/libgstvaapi-wayland.install.in:
30415 * pkgconfig/Makefile.am:
30416 * pkgconfig/gstreamer-vaapi-wayland.pc.in:
30417 wayland: add packaging files.
30419 2012-07-24 15:07:48 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30421 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30422 * gst/vaapi/gstvaapipluginutil.c:
30423 * gst/vaapi/gstvaapisink.c:
30424 * pkgconfig/Makefile.am:
30425 Fix build without X11.
30427 2012-07-24 09:45:25 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30429 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30430 * gst-libs/gst/vaapi/gstvaapidisplay.h:
30431 * gst/vaapi/Makefile.am:
30432 * gst/vaapi/gstvaapipluginutil.c:
30433 * gst/vaapi/gstvaapisink.c:
30434 plugins: add support for Wayland.
30436 2012-07-23 12:56:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30438 * tests/Makefile.am:
30440 * tests/test-display.c:
30441 * tests/test-windows.c:
30442 tests: add support for Wayland.
30444 2012-07-19 10:27:23 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
30448 * gst-libs/gst/vaapi/Makefile.am:
30449 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.c:
30450 * gst-libs/gst/vaapi/gstvaapidisplay_wayland.h:
30451 * gst-libs/gst/vaapi/gstvaapidisplay_wayland_priv.h:
30452 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
30453 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.c:
30454 * gst-libs/gst/vaapi/gstvaapiwindow_wayland.h:
30455 Add initial support for VA/Wayland.
30456 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30458 2012-07-25 15:11:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30460 * gst/vaapi/gstvaapipluginbuffer.c:
30461 plugins: fix creation of video buffer from surface proxy.
30462 Fix a regression introduced with commit 8ef490a.
30464 2012-07-25 14:51:28 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30466 * gst/vaapi/gstvaapipluginbuffer.c:
30467 * gst/vaapi/gstvaapipluginutil.c:
30468 plugins: use new display types more.
30469 In particular, simplify gst_vaapi_reply_to_query() with display types.
30470 Likewise for creating new video buffers.
30472 2012-07-25 10:02:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30474 * gst/vaapi/gstvaapidecode.c:
30475 * gst/vaapi/gstvaapidownload.c:
30476 * gst/vaapi/gstvaapipluginutil.c:
30477 * gst/vaapi/gstvaapipluginutil.h:
30478 * gst/vaapi/gstvaapipostproc.c:
30479 * gst/vaapi/gstvaapisink.c:
30480 * gst/vaapi/gstvaapiupload.c:
30481 plugins: fix display type selection and propagation.
30482 If vaapisink is in the GStreamer pipeline, then we shall allocate a
30483 unique GstVaapiDisplay and propagate it upstream. i.e. subsequent
30484 queries from vaapidecode shall get a valid answer from vaapisink.
30486 2012-07-25 11:37:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30488 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30489 display: fix destruction of mutex.
30491 2012-07-25 09:16:02 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30493 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30494 * gst-libs/gst/vaapi/gstvaapidisplay.h:
30495 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
30496 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
30497 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
30498 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
30499 * gst/vaapi/gstvaapipluginutil.c:
30500 * gst/vaapi/gstvaapipluginutil.h:
30501 * gst/vaapi/gstvaapisink.c:
30502 display: add display types.
30503 Move display types from gstvaapipluginutil.* to gstvaapidisplay.* so that
30504 we could simplify characterization of a GstVaapiDisplay. Also rename "auto"
30505 type to "any", and add a "display-type" attribute.
30507 2012-07-24 19:43:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30509 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
30510 display: use prefixed display names for cache lookups.
30511 This improves display name comparisons by always allocating a valid display
30512 name. This also helps to disambiguate lookups by name in the global display
30513 cache, should a new backend be implemented.
30515 2012-07-24 16:14:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30517 * gst/vaapi/gstvaapipluginutil.h:
30518 plugins: declare helper functions as internal.
30520 2012-07-24 14:31:25 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30522 * docs/reference/libs/libs-sections.txt:
30523 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
30524 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
30525 * gst-libs/gst/vaapi/gstvaapivideobuffer_glx.c:
30526 * gst-libs/gst/vaapi/gstvaapivideobuffer_glx.h:
30527 * gst/vaapi/Makefile.am:
30528 * gst/vaapi/gstvaapidecode.c:
30529 * gst/vaapi/gstvaapidownload.c:
30530 * gst/vaapi/gstvaapipluginbuffer.c:
30531 * gst/vaapi/gstvaapipluginbuffer.h:
30532 * gst/vaapi/gstvaapipostproc.c:
30533 * gst/vaapi/gstvaapiupload.c:
30534 videobuffer: drop deprecated functions.
30535 Move video buffer creation routines to plugin elements. That exclusively
30536 uses *_typed_new*() variants.
30538 2012-07-24 14:09:09 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30540 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
30541 * gst-libs/gst/vaapi/gstvaapivideobuffer_glx.h:
30542 videobuffer: mark video buffer creation routines as deprecated.
30543 The vdeo buffer creation routines shall actually be internal to gstreamer-vaapi
30544 plugin elements. So deprecate any explicit creation routines that are not the
30545 new *_typed_new*() variants.
30547 2012-07-24 13:52:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30549 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
30550 * gst-libs/gst/vaapi/gstvaapivideobuffer_glx.c:
30551 * gst-libs/gst/vaapi/gstvaapivideobuffer_priv.h:
30552 videobuffer: factor out base and GLX implementations.
30553 Introduce new typed constructors internal to gstreamer-vaapi plugin elements.
30554 This avoids duplication of code, and makes it possible to further implement
30555 generic video buffer creation routines that automatically map to base or GLX
30558 2012-07-24 10:58:32 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30560 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
30561 utils: fix gl_create_context() with parent context set.
30562 If GLX window was created from a foreign Display, then that same Display shall
30563 be used for subsequent glXMakeCurrent(). This means that gl_create_context()
30564 will now use the same Display that the parent, if available.
30565 This fixes cluttersink with the Intel GenX VA driver.
30567 2012-07-23 18:37:38 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30569 * gst/vaapi/gstvaapidecode.c:
30570 * gst/vaapi/gstvaapidownload.c:
30571 * gst/vaapi/gstvaapipluginutil.c:
30572 * gst/vaapi/gstvaapipluginutil.h:
30573 * gst/vaapi/gstvaapipostproc.c:
30574 * gst/vaapi/gstvaapisink.c:
30575 * gst/vaapi/gstvaapiupload.c:
30576 pluginutils: improve automatic display type selection.
30578 2012-07-23 18:01:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30580 * gst/vaapi/gstvaapipluginutil.c:
30581 * gst/vaapi/gstvaapipluginutil.h:
30582 pluginutils: cosmetics (indentation fixes).
30584 2012-07-23 17:54:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30587 configure: simplify video outputs summary.
30589 2012-07-23 17:49:08 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30592 * gst/vaapi/gstvaapipluginutil.c:
30593 * gst/vaapi/gstvaapipluginutil.h:
30594 * gst/vaapi/gstvaapisink.c:
30595 * gst/vaapi/gstvaapisink.h:
30596 configure: drop check for --enable-vaapisink-glx.
30597 vaapisink is now built with support for multiple display types, whenever
30598 they are enabled. The new "display" attribute is used to select a particular
30601 2012-07-23 16:15:38 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30604 * gst-libs/gst/vaapi/gstvaapicompat.h:
30605 * gst/vaapi/Makefile.am:
30606 * gst/vaapi/gstvaapidecode.c:
30607 * gst/vaapi/gstvaapidownload.c:
30608 * gst/vaapi/gstvaapipluginutil.c:
30609 * gst/vaapi/gstvaapipostproc.c:
30610 * gst/vaapi/gstvaapiupload.c:
30611 * tests/test-display.c:
30612 configure: drop check for --enable-vaapi-glx.
30613 This flag is obsolete. It was meant to explicitly enable/disable VA/GLX API
30614 support, or fallback to TFP+FBO if this API is not found. Now, we check for
30615 the VA/GLX API by default if --enable-glx is set. If this API is not found,
30616 we now default to use TFP+FBO.
30617 Note: TFP+FBO, i.e. using vaPutSurface() is now also a deprecated usage and
30618 will be removed in the future. If GLX rendering is requested, then the VA/GLX
30619 API shall be used as it covers most usages. e.g. AMD driver can't render to
30622 2012-07-23 15:20:23 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30625 tests: allow GLX output, if available and selected.
30627 2012-07-23 15:17:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30629 * tests/Makefile.am:
30632 * tests/test-decode.c:
30633 * tests/test-subpicture.c:
30634 * tests/test-surfaces.c:
30635 tests: use common display and window creation routines.
30636 Add new --output option to select the renderer. Use --list-outputs to
30637 print a list of supported renderers.
30639 2012-07-23 14:15:42 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30641 * tests/Makefile.am:
30642 tests: move encoded bitstreams to libutils.la.
30644 2012-07-23 14:11:16 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30646 * tests/Makefile.am:
30647 tests: build convenience library for common utilities.
30649 2012-07-20 16:37:01 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30651 * tests/Makefile.am:
30652 * tests/test-display.c:
30653 tests: simplify build with various display options.
30655 2012-07-23 13:28:42 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30658 configure: improve checks for X11.
30660 2012-07-20 15:57:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30662 * tests/Makefile.am:
30663 configure: fix previous commit for GLX deps.
30665 2012-07-20 14:44:27 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30668 * gst-libs/gst/vaapi/Makefile.am:
30669 configure: improve checks for GLX.
30671 2012-07-20 11:45:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30674 * gst-libs/gst/vaapi/gstvaapicompat.h:
30675 * gst-libs/gst/vaapi/gstvaapidisplay.h:
30676 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
30677 * gst-libs/gst/vaapi/gstvaapiutils.h:
30678 Drop support for obsolete VA-API versions < 0.30.4.
30680 2012-07-20 11:16:27 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30683 * gst-libs/gst/Makefile.am:
30684 * gst-libs/gst/gstutils_version.h.in:
30685 * gst/vaapi/gstvaapisink.c:
30686 vaapisink: drop checks for new APIs used by default.
30687 GStreamer -base plugins >= 0.10.31 are now required, so the checks for
30688 new APIs like GstXOverlay::set_window_handle() and ::set_render_rectangle()
30689 are no longer necessary.
30691 2012-07-20 14:05:23 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30694 configure: cosmetics and some minor changes.
30695 - Better grouping of feature checks
30696 - Sort list of config files to generate
30698 2012-07-19 17:55:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30701 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
30702 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.h:
30703 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
30704 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
30705 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
30706 * gst-libs/gst/vaapi/gstvaapiutils.h:
30707 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
30708 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
30709 Use standard G_GNUC_INTERNAL keyword instead of attribute_hidden.
30711 2012-07-19 17:41:25 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30713 * gst-libs/gst/vaapi/Makefile.am:
30714 * gst-libs/gst/vaapi/gstvaapiutils_tsb.c:
30715 * gst-libs/gst/vaapi/gstvaapiutils_tsb.h:
30716 Drop obsolete GstVaapiTSB.
30717 It has been replaced with a GstAdapter and gst_adapter_prev_pts().
30719 2012-07-19 17:27:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30721 * docs/reference/libs/libs-sections.txt:
30722 docs: add missing entries for the JPEG decoder.
30724 2012-07-19 17:16:28 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30727 * gst-libs/gst/vaapi/Makefile.am:
30728 * gst/vaapi/gstvaapidecode.c:
30729 * tests/test-decode.c:
30730 * tests/test-subpicture.c:
30731 Drop all references to USE_CODEC_PARSERS.
30733 2012-07-19 17:00:36 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30737 * docs/reference/libs/libs-sections.txt:
30738 * docs/reference/libs/libs.core.types:
30739 * gst-libs/gst/vaapi/Makefile.am:
30740 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
30741 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.h:
30742 * gst/vaapi/gstvaapidecode.c:
30743 * gst/vaapi/gstvaapidecode.h:
30744 * tests/test-decode.c:
30745 * tests/test-subpicture.c:
30746 Drop FFmpeg-based decoders.
30747 GStreamer codecparsers-based decoders are the only supported decoders now.
30748 Though, FFmpeg decoders are still available in gstreamer-vaapi 0.3.x series.
30750 2012-07-01 05:55:05 +0900 Javier Jardón <jjardon@gnome.org>
30753 * debian.upstream/control.in:
30754 configure: bump glib required version to 2.28.
30755 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30757 2012-06-29 08:45:47 +0900 Javier Jardón <jjardon@gnome.org>
30759 * gst/vaapi/gstvaapidecode.h:
30760 * gst/vaapi/gstvaapidownload.h:
30761 * gst/vaapi/gstvaapipostproc.h:
30762 * gst/vaapi/gstvaapisink.h:
30763 * gst/vaapi/gstvaapiupload.h:
30764 plugins: declare _get_type() functions as const.
30765 Declaring a function as const enables better optimization of calls to
30767 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30769 2012-07-01 05:50:17 +0900 Javier Jardón <jjardon@gnome.org>
30771 * gst/vaapi/gstvaapidecode.c:
30772 * gst/vaapi/gstvaapidownload.c:
30773 * gst/vaapi/gstvaapipostproc.c:
30774 * gst/vaapi/gstvaapisink.c:
30775 * gst/vaapi/gstvaapiupload.c:
30776 plugins: use g_clear_object() wherever applicable.
30777 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30779 2012-06-29 15:19:51 +0900 Javier Jardón <jjardon@gnome.org>
30781 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
30782 * gst-libs/gst/vaapi/gstvaapicontext.h:
30783 * gst-libs/gst/vaapi/gstvaapidecoder.h:
30784 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.h:
30785 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.h:
30786 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
30787 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.h:
30788 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.h:
30789 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.h:
30790 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
30791 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.h:
30792 * gst-libs/gst/vaapi/gstvaapidisplay.h:
30793 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
30794 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
30795 * gst-libs/gst/vaapi/gstvaapiimage.h:
30796 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
30797 * gst-libs/gst/vaapi/gstvaapiobject.h:
30798 * gst-libs/gst/vaapi/gstvaapiparamspecs.h:
30799 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
30800 * gst-libs/gst/vaapi/gstvaapisurface.h:
30801 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
30802 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
30803 * gst-libs/gst/vaapi/gstvaapitexture.h:
30804 * gst-libs/gst/vaapi/gstvaapivalue.h:
30805 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
30806 * gst-libs/gst/vaapi/gstvaapivideobuffer_glx.h:
30807 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.h:
30808 * gst-libs/gst/vaapi/gstvaapivideopool.h:
30809 * gst-libs/gst/vaapi/gstvaapivideosink.h:
30810 * gst-libs/gst/vaapi/gstvaapiwindow.h:
30811 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
30812 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
30813 libs: declare _get_type() functions as const.
30814 Declaring a function as const enables better optimization of calls
30816 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30818 2012-07-01 05:34:15 +0900 Javier Jardón <jjardon@gnome.org>
30820 * gst-libs/gst/vaapi/glibcompat.h:
30821 * gst-libs/gst/vaapi/gstvaapicontext.c:
30822 * gst-libs/gst/vaapi/gstvaapidisplay.c:
30823 * gst-libs/gst/vaapi/gstvaapiobject.c:
30824 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
30825 * gst-libs/gst/vaapi/gstvaapisurface.c:
30826 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
30827 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
30828 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
30829 * gst-libs/gst/vaapi/gstvaapivideopool.c:
30830 libs: use g_clear_object() wherever applicable.
30831 This is a preferred thread-safe version. Also add an inline version of
30832 g_clear_object() if compiling with glib < 2.28.
30833 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30835 2012-07-01 06:02:22 +0900 Javier Jardón <jjardon@gnome.org>
30839 * docs/reference/libs/Makefile.am:
30840 * gst-libs/gst/vaapi/Makefile.am:
30841 * gst-libs/gst/vaapi/gstvaapimarshal.list:
30842 * gst-libs/gst/vaapi/gstvaapiobject.c:
30843 libs: use generic g_cclosure_marshal_VOID__VOID().
30844 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30846 2012-07-19 14:29:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30849 * gst-libs/gst/vaapi/glibcompat.h:
30850 glibcompat: drop explicit check for g_list_free_full().
30852 2012-07-19 13:58:31 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30855 Bump version for development.
30857 2012-07-19 13:57:05 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30859 * gst-libs/gst/codecparsers/Makefile.am:
30860 jpeg: fix make dist.
30862 2012-06-28 00:39:10 +0900 Javier Jardón <jjardon@gnome.org>
30866 * docs/Makefile.am:
30867 configure: fix build without gtk-doc support.
30868 Also do not generate tamplate files as all the documentation is inline.
30869 Drop un-needed code in autogen.sh as well.
30870 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30872 2012-06-28 00:27:31 +0900 Javier Jardón <jjardon@gnome.org>
30877 configure: put m4 macros and autogenerated files into m4/ directory.
30878 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30880 2012-06-28 00:20:12 +0900 Javier Jardón <jjardon@gnome.org>
30883 configure: drop deprecated autoconf macros.
30884 Bump autoconf required version to 2.58, needed for AS_HELP_STRING macro.
30885 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30887 2012-06-28 00:04:19 +0900 Javier Jardón <jjardon@gnome.org>
30890 configure: don't use AC_SUBST for some variables.
30891 PKG_CHECK_MODULES already does this for us.
30892 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30894 2012-07-19 11:43:03 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30896 * gst/vaapi/gstvaapidecode.c:
30897 * gst/vaapi/gstvaapidownload.c:
30898 * gst/vaapi/gstvaapipostproc.c:
30899 * gst/vaapi/gstvaapiupload.c:
30900 plugins: add support for GstImplementsInterface.
30902 2012-07-01 02:58:36 +0900 Javier Jardón <jjardon@gnome.org>
30905 * gst/vaapi/gstvaapidecode.c:
30906 * gst/vaapi/gstvaapidownload.c:
30907 * gst/vaapi/gstvaapipostproc.c:
30908 * gst/vaapi/gstvaapisink.c:
30909 * gst/vaapi/gstvaapiupload.c:
30910 plugins: use G_DEFINE_TYPE_* instead of deprecated GST_BOILERPLATE_*.
30911 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30913 2012-07-01 03:57:13 +0900 Javier Jardón <jjardon@gnome.org>
30916 * gst/vaapi/gstvaapisink.c:
30917 plugins: do not use deprecated GStreamer -base symbols.
30918 Bump GStreamer plugins -base required version to 0.10.31, needed for
30919 gst_x_overlay_got_window_handle().
30920 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30922 2012-07-01 03:57:13 +0900 Javier Jardón <jjardon@gnome.org>
30925 * gst/vaapi/gstvaapidecode.c:
30926 * gst/vaapi/gstvaapidownload.c:
30927 * gst/vaapi/gstvaapipostproc.c:
30928 * gst/vaapi/gstvaapisink.c:
30929 * gst/vaapi/gstvaapiupload.c:
30930 plugins: do not use deprecated core GStreamer symbols.
30931 Bump GStreamer required version to 0.10.14, needed for
30932 gst_element_class_set_details_simple().
30933 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30935 2012-07-19 10:54:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30937 * tests/test-decode.c:
30938 tests: fix build without JPEG decoder support.
30940 2012-07-17 13:44:45 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30942 * tests/Makefile.am:
30943 * tests/test-decode.c:
30944 * tests/test-jpeg.c:
30945 * tests/test-jpeg.h:
30946 tests: add test for JPEG decoding.
30948 2012-07-17 13:43:32 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30950 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
30951 jpeg: update to match latest parser API.
30953 2012-07-16 17:35:19 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30955 * gst-libs/gst/codecparsers/gstjpegparser.c:
30956 * gst-libs/gst/codecparsers/gstjpegparser.h:
30957 codecparsers: jpeg: tweak parser API.
30958 ... to allow for more consistent parsing API among various codec parsers.
30959 In particular, drop use of GList.
30961 2012-07-16 16:24:04 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30964 jpeg: fix configure check for VA/JPEG decoding API.
30966 2012-06-26 15:18:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30968 * gst-libs/gst/vaapi/gstvaapiprofile.c:
30969 jpeg: fix build with VA-API < 0.32.0.
30971 2012-06-26 15:04:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30976 2012-06-26 15:02:44 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30979 Bump version for development.
30981 2012-06-26 14:46:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30987 2012-06-26 13:34:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30992 2012-06-25 16:07:55 +0800 Yan Yin <yan.yin@intel.com>
30994 * gst/vaapi/gstvaapipluginutil.c:
30995 vaapiplugin: fix build when compiling without GLX.
30996 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
30998 2012-06-26 11:03:25 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31001 configure: disable FFmpeg-based decoders.
31002 FFmpeg decoders are still available through the --enable-ffmpeg option
31003 but are no longer maintained.
31005 2012-06-25 17:25:44 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31007 * gst-libs/gst/vaapi/Makefile.am:
31008 Fix build with recent GStreamer stack.
31010 2012-06-25 17:10:49 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31013 * gst-libs/gst/vaapi/gstvaapicompat.h:
31014 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
31015 jpeg: update to current VA/JPEG decoding API.
31017 2012-06-21 16:06:47 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31019 * gst-libs/gst/codecparsers/gstjpegparser.c:
31020 * gst-libs/gst/codecparsers/gstjpegparser.h:
31021 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
31022 codecparsers: jpeg: track valid quantization and Huffman tables.
31023 Add valid flag to GstJpegQuantTable and GstJpegHuffmanTable so that
31024 to determine whether a table actually changed since the last user
31025 synchronization point. That way, this makes it possible for some
31026 hardware accelerated decoding solution to upload only those tables
31029 2012-06-05 10:10:22 +0800 Wind Yuan <feng.yuan@intel.com>
31031 * gst-libs/gst/codecparsers/gstjpegparser.c:
31032 codecparsers: jpeg: use U_READ_UINT*() wherever possible.
31033 Use GstByteReader *_unchecked() variants as much as possible.
31034 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31036 2012-06-04 16:20:13 +0800 Wind Yuan <feng.yuan@intel.com>
31038 * gst-libs/gst/codecparsers/gstjpegparser.c:
31039 * gst-libs/gst/codecparsers/gstjpegparser.h:
31040 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
31041 jpeg: add new GstJpegHuffmanTables structure.
31042 Add new GstJpegHuffmanTables helper structure to hold all possible
31043 AC/DC Huffman tables available to all components.
31044 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31046 2012-06-04 15:52:19 +0800 Wind Yuan <feng.yuan@intel.com>
31048 * gst-libs/gst/codecparsers/gstjpegparser.c:
31049 * gst-libs/gst/codecparsers/gstjpegparser.h:
31050 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
31051 jpeg: make gst_jpeg_parse() support multiple scans.
31052 gst_jpeg_parse() now gathers all scans available in the supplied
31053 buffer. A scan comprises of the scan header and any entropy-coded
31054 segments or restart marker following it. The size and offset to
31055 the associated data (ECS + RST segments) are append to a new
31056 GstJpegScanOffsetSize structure.
31057 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31059 2012-04-19 23:50:14 +0800 Wind Yuan <feng.yuan@intel.com>
31061 * gst-libs/gst/codecparsers/gstjpegparser.c:
31062 * gst-libs/gst/codecparsers/gstjpegparser.h:
31063 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
31064 jpeg: update to match latest parser API.
31065 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31067 2012-04-13 01:58:39 -0400 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31069 * gst-libs/gst/codecparsers/gstjpegparser.h:
31070 jpeg: simplify and optimize parser API.
31072 2012-04-18 22:30:45 -0400 Wind Yuan <feng.yuan@intel.com>
31074 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31075 mpeg: fix picture used to determine backward_reference_vop_coding_type.
31076 Complete fix brought by bf9f77b1afb0829b97e2d502057aec973c5fd7f5
31077 but Gwenole did not apply all the bits.
31078 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31080 2012-04-27 04:13:00 -0400 Wind Yuan <feng.yuan@intel.com>
31082 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31083 mpeg4: map Simple_Scalable profile to Advanced_Simple profile.
31084 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31086 2012-04-27 04:10:17 -0400 Wind Yuan <feng.yuan@intel.com>
31088 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31089 mpeg4: handle skipped frames (vop_hdr->coded = 0).
31090 Gracefully skip non VOP coded frames.
31091 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31093 2012-04-26 04:00:41 -0400 Wind Yuan <feng.yuan@intel.com>
31095 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31096 mpeg4: fix timestamp issues on too fast playback.
31097 Improve generation of presentation timestamps to be less sensitive
31098 to input stream errors. In practise, GOP is also a synchronization
31099 point for PTS calculation.
31100 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31102 2012-04-16 10:02:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31104 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
31105 Fix build without JPEG decoder.
31107 2012-04-12 11:48:24 +0200 Wind Yuan <feng.yuan@intel.com>
31109 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31110 mpeg4: fix VOP coding type of backward reference pictures.
31111 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31113 2012-04-11 23:02:45 -0400 Wind Yuan <feng.yuan@intel.com>
31115 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31116 mpeg4: fix handling of temporal reference distances.
31117 TRD and TRB fields are not large enough to hold the difference of PTS
31118 expressed with nanosecond resolution. So, compute them from the original
31120 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31122 2012-04-12 11:00:22 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31124 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
31125 dpb: mpeg2: cosmetics.
31126 Define MAX_MPEG2_REFERENCES to 2 and avoid magic numbers all around.
31128 2012-02-10 00:21:04 +0800 Wind Yuan <feng.yuan@intel.com>
31131 * gst-libs/gst/vaapi/Makefile.am:
31132 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
31133 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
31134 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.c:
31135 * gst-libs/gst/vaapi/gstvaapidecoder_jpeg.h:
31136 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31137 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
31138 * gst-libs/gst/vaapi/gstvaapiprofile.c:
31139 * gst-libs/gst/vaapi/gstvaapiprofile.h:
31140 * gst/vaapi/gstvaapidecode.c:
31141 Add initial JPEG decoder.
31142 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31144 2012-02-10 00:21:04 +0800 Wind Yuan <feng.yuan@intel.com>
31147 * gst-libs/gst/Makefile.am:
31148 * gst-libs/gst/codecparsers/Makefile.am:
31149 * gst-libs/gst/codecparsers/gstjpegparser.c:
31150 * gst-libs/gst/codecparsers/gstjpegparser.h:
31151 * gst-libs/gst/vaapi/Makefile.am:
31152 codecparsers: add JPEG parser.
31153 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31155 2012-04-10 13:29:10 +0200 Wind Yuan <feng.yuan@intel.com>
31157 * gst/vaapi/gstvaapidecode.c:
31158 vaapidecode: fix VA display type.
31159 Fix typo whereby plain VADisplay type was used instead of the GstVaapiDisplay
31161 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31163 2012-04-10 14:28:31 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31165 * gst/vaapi/gstvaapidecode.c:
31166 vaapidecode: fix includes when compiling for a single API.
31168 2012-04-02 18:42:12 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31171 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31172 mpeg2: fix calculation of macroblock_offset.
31173 Fix decoding of streams with extra slice() information before the first
31174 macroblock(). e.g. this fixes sony-ct3.bs from conformance test.
31176 2012-04-02 18:09:21 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31178 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31179 mpeg2: fix interpolation of GOP TSN from new PTS.
31180 New GOP TSN base could be mis-calculated. In particular, this fixes
31181 decoding of uruseiyatsura.vob from <http://samples.mplayerhq.hu/>.
31183 2012-04-02 16:07:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31186 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31187 * gst-libs/gst/vaapi/gstvaapiprofile.h:
31188 mpeg2: fix decoding of high profile streams.
31189 Allow MPEG-2 High profile streams only if the HW supports that profile
31190 or no High profile specific bits are used, and thus Main profile could
31191 be used instead. i.e. chroma_format is 4:2:0, intra_dc_precision is not
31192 set to 11 and no sequence_scalable_extension() was parsed.
31194 2012-04-02 14:51:06 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31197 * gst/vaapi/gstvaapidecode.c:
31198 vaapidecode: report unsupported codec profiles.
31199 Try to gracefully abort when the HW does not support the requested
31200 profile. There is no fallback unless profiles are correctly parsed
31201 and matched through caps beforehand.
31203 2012-02-07 15:23:22 +0100 Holger Kaelberer <holger.k@elberer.de>
31206 * gst/vaapi/gstvaapisink.c:
31207 vaapisink: don't resize a 'foreign' X-window.
31208 Don't forcibly resize foreign X windows. The user is responsible for
31209 their size and vaapisink shall not change this.
31210 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31212 2012-02-07 15:21:05 +0100 Holger Kaelberer <holger.k@elberer.de>
31215 * gst/vaapi/gstvaapisink.c:
31216 * gst/vaapi/gstvaapisink.h:
31217 vaapisink: recalculate render rect only if caps are negotiated.
31218 Fix gst_vaapisink_xoverlay_set_window_handle() when it is called before
31219 caps got negotiated. Besides, when a foreign window is provided by the
31220 user, so should the render rect.
31221 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31223 2012-04-02 13:07:34 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31226 Bump version for development.
31228 2012-04-02 10:07:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31233 2012-04-02 12:52:54 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31235 * docs/reference/libs/libs-docs.xml.in:
31236 * docs/reference/libs/libs-sections.txt:
31237 * docs/reference/plugins/plugins-docs.xml.in:
31238 * docs/reference/plugins/plugins-sections.txt:
31239 * docs/reference/plugins/plugins.types:
31240 * gst-libs/gst/vaapi/gstvaapiimage.c:
31241 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
31242 * gst/vaapi/gstvaapipostproc.c:
31243 Fix a few documentation issues.
31245 2012-04-02 10:05:57 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31249 Update introduction and changelog.
31251 2012-04-02 11:29:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31253 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
31254 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.h:
31255 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31256 mpeg2: allocate dummy picture for first field based I-frame.
31257 In P-pictures, prediction shall be made from the two most recently
31258 decoded reference fields. However, when the first I-frame is a field,
31259 the next field of the current picture could be a P-picture but only a
31260 single field was decoded so far. In this case, create a dummy picture
31261 with POC = -1 that will be used as reference.
31262 Some VA drivers would error out if P-pictures don't have a forward
31263 reference picture. This is true in general but not in this very specific
31266 2012-04-02 10:43:30 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31268 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31269 mpeg2: fix simple to main profile fallback.
31270 Allow fallback from simple to main profile when the HW decoder does
31271 not support the former profile and that no sequence_header_extension()
31272 is available to point out this.
31274 2012-03-30 03:04:40 -0400 Wind Yuan <feng.yuan@intel.com>
31276 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31277 mpeg4: improve error checking while decoding packets.
31278 decode_picture() could return an error when an MPEG-4 profile is not
31279 supported for example. In this case, the underlying VA context is not
31280 allocated and no other proper action can be taken. Likewise on exit
31281 from decode_slice().
31282 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31284 2012-03-30 17:03:28 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31286 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
31287 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31288 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31289 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
31290 decoder: use POC to maintain the DPB.
31291 Introduce a POC field in GstVaapiPicture so that to store simpler sequential
31292 numbers. A signed 32-bit integer should be enough for 1 year of continuous
31293 video streaming at 60 Hz.
31294 Use this new POC value to maintain the DPB, instead of 64-bit timestamps.
31295 This also aligns with H.264 that will be migrated to GstVaapiDpb infrastructure.
31297 2012-03-30 16:23:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31299 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31300 mpeg2: rework generation of presentation timestamps.
31301 Always prefer PTS from the demuxer layer for GOP times. If this is invalid,
31302 i.e. demuxer could not determine the PTS or the generated PTS is lower than
31303 max PTS from past pictures, then try to fix it up based on the duration of
31305 For picture PTS, simply use the GOP PTS formerly computed then use TSN to
31306 reconstruct a current time. Also now handle wrapped TSN correctly.
31308 2012-03-30 17:07:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31310 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31311 mpeg2: ignore empty user-data packets.
31312 Fix tcela-8.bits conformance test.
31314 2012-03-29 11:13:20 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31316 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31317 mpeg2: review and report errors accordingly.
31318 Use GST_ERROR() to report real errors instead of hiding them into
31321 2012-03-28 19:15:47 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31323 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31324 mpeg2: fix invalid interlaced frame in progressive sequence.
31325 Some streams, badly constructed, could have signaled an interlaced
31326 frame while the sequence was meant to be progressive. Warn and force
31327 frame to be progressive in this case.
31329 2012-03-28 16:08:29 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31331 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31332 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
31333 mpeg2: add support for interlaced streams.
31334 Pictures are submitted to the HW for rendering only when both fields
31335 are decoded or current picture is a full frame.
31337 2012-03-28 14:36:30 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31339 * gst-libs/gst/vaapi/Makefile.am:
31340 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.c:
31341 * gst-libs/gst/vaapi/gstvaapidecoder_dpb.h:
31342 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31343 mpeg2: add new decoded picture buffer infrastructure.
31344 Decoded pictures are now maintained into DPB, similarly to H.264.
31345 The same mechanism could be re-used for VC-1 and MPEG-4:2 codecs.
31347 2012-03-28 17:50:28 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31349 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31350 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31351 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
31352 decoder: add first-field flag to picture.
31353 Add first-field (FF) flag to GstVaapiPicture, thus not requiring is_first_field
31354 member in each decoder. Rather, when a GstVaapiPicture is created, it is considered
31355 as the first field. Any subsequent allocated field will become the second field.
31357 2012-03-28 16:05:58 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31359 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
31360 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
31361 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31362 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
31363 decoder: allow pictures to be cloned for field decoding.
31364 Add gst_vaapi_picture_new_field() function that clones a picture, while
31365 preserving the parent picture surface. i.e. the surface proxy reference
31366 count is increased and other fields copied as is. Besides, the picture
31367 is reset into a "non-output" mode.
31369 2012-03-28 16:07:44 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31371 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31372 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31373 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
31374 decoder: rework picture field flags.
31375 Add top-field-first (TFF) and interlaced flags to GstVaapiPicture so they
31376 could be propagated to the surface proxy when it is pushed for rendering.
31377 Besides, top and bottom fields are now expressed with picture structure flags
31378 from GstVaapiSurfaceRenderFlags.
31380 2012-03-28 14:28:26 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31382 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31383 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
31384 decoder: add OUTPUT flag to pictures.
31385 Allow pictures to be marked as output gst_vaapi_picture_output().
31387 2012-03-28 14:24:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31389 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31390 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
31391 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
31392 decoder: fix semantics of SKIPPED pictures.
31393 If GstVaapiPicture has flag SKIPPED set, this means gst_vaapi_picture_output()
31394 will not push the underlying surface for rendering. Besides, VC-1 skipped P-frame
31395 has nothing to do with rendering. This only means that the currently decoded
31396 picture is just a copy of its reference picture.
31398 2012-03-28 15:16:17 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31400 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
31401 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
31402 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
31403 * gst/vaapi/gstvaapipostproc.c:
31404 vaapipostproc: get "interlaced" attribute from surface proxy.
31405 Add new "interlaced" attribute to GstVaapiSurfaceProxy. Use this in
31406 vaapipostproc so that to handles cases where bitstream is interlaced
31407 but almost only frame pictures are generated. In this case, we should
31408 not be alternating between top/bottom fields.
31410 2012-03-26 14:37:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31413 * debian.upstream/control.in:
31414 * gst/vaapi/Makefile.am:
31415 * gst/vaapi/gstvaapi.c:
31416 * gst/vaapi/gstvaapipostproc.c:
31417 * gst/vaapi/gstvaapipostproc.h:
31418 vaapipostproc: add new element for video postprocessing.
31419 Add vaapipostproc element for video postprocessing. So far, only basic
31420 bob deinterlacing is implemented. Interlaced mode is automatically
31421 detected based on sink caps ("interlaced" field).
31423 2012-03-26 12:01:36 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31425 * docs/reference/libs/libs-sections.txt:
31426 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
31427 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
31428 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
31429 * gst/vaapi/gstvaapisink.c:
31430 videobuffer: add surface render flags.
31431 Allow rendering flags, as a combination of GstVaapiSurfaceRenderFlags,
31432 to be set to the video buffer. In particular, this is mostly useful for
31433 basic deinterlacing.
31435 2012-03-23 17:13:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31437 * gst-libs/gst/vaapi/gstvaapicompat.h:
31438 compat: add compatibility glue with VA-API 0.34+ (WIP).
31440 2012-03-23 17:11:18 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31442 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31443 h264: skip all Filler Data NALs.
31445 2012-03-22 03:28:22 -0400 Wind Yuan <feng.yuan@intel.com>
31447 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31448 mpeg4: check for decoder status prior to decoding packet.
31449 Make sure there is a VA surface free prior to decoding the current frame.
31450 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31452 2012-03-15 04:58:04 -0400 Wind Yuan <feng.yuan@intel.com>
31454 * gst/vaapi/gstvaapidecode.c:
31455 * gst/vaapi/gstvaapidecode.h:
31456 decode: delay NEWSEGMENT event if vaapidecode element was not linked.
31457 Rationale: playbin2 links all elements at run-time. Once vaapidecode
31458 is created and a NEWSEGMENT event arrives, downstream element may not
31459 be ready yet. So, delay this event until next element is chained in,
31460 otherwise basesink could output "Received buffer without a new-segment.
31461 Assuming timestamps start from 0".
31462 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31464 2012-03-13 20:33:41 -0400 Wind Yuan <feng.yuan@intel.com>
31466 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31467 mpeg2: fix GOP timestamps when incorrect data is received.
31468 Some streams have incorrect GOP timestamps, or nothing set at all.
31469 i.e. GOP time is 00:00:00 for all GOPs. Try to recover in this case
31470 from demuxer timestamps, which are monotonic.
31471 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31473 2012-03-13 02:03:31 -0400 Wind Yuan <feng.yuan@intel.com>
31475 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31476 mpeg2: don't decode anything before the first sequence_header().
31477 Skip all pictures prior to the first sequence_header(). Besides,
31478 skip all picture_data() if there was no prior picture_header().
31479 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31481 2012-02-07 15:57:14 +0100 Holger Kaelberer <holger.k@elberer.de>
31483 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
31484 ffmpeg: add support for interlaced streams.
31485 Evaluate interlaced stream properties.
31486 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31488 2012-02-07 15:54:15 +0100 Holger Kaelberer <holger.k@elberer.de>
31490 * gst/vaapi/gstvaapidecode.c:
31491 vaapidecode: propagate interlaced and TFF properties downstream.
31492 Propagate "interlaced" caps downstream and set "tff" buffer flag
31493 appropriately to output buffers for interlaced pictures.
31494 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31496 2012-02-07 15:54:15 +0100 Holger Kaelberer <holger.k@elberer.de>
31498 * gst-libs/gst/vaapi/gstvaapidecoder.c:
31499 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
31500 decoder: maintain caps for interlaced streams.
31501 Extend GstVaapiDecoder base object to maintain caps with "interlaced"
31503 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31505 2012-02-07 15:54:15 +0100 Holger Kaelberer <holger.k@elberer.de>
31507 * docs/reference/libs/libs-sections.txt:
31508 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
31509 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
31510 surfaceproxy: add TFF property.
31511 Add TFF (top-field-first) property to GstVaapiSurfaceProxy.
31512 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31514 2012-03-16 14:21:36 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31517 AUTHORS: update to match current authors.
31519 2012-02-28 11:58:21 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31521 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31522 mpeg2: fix is_first_field calculation.
31523 Reset is_first_field for frame pictures. Factor out locations where
31524 the flag is updated.
31526 2012-02-24 12:56:48 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31528 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31529 mpeg2: catch incorrect picture_structure from bitstreams.
31530 Assume "frame" picture structure if the syntax element was zero or if
31531 progressive_frame is set.
31533 2012-02-24 12:53:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31535 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31536 mpeg2: fix slice_vertical_position calculation (again).
31537 VA-API expects slice_vertical_position as the initial position from the
31538 bitstream. i.e. the direct slice() information. VA drivers will be fixed
31541 2012-03-02 15:03:57 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31543 * gst/vaapi/gstvaapidecode.c:
31544 Revert "vaapidecode: fix another pad template ref leak" (Holger Kaelberer)
31545 This reverts commit 2f127d6af473afd647a2c88f75faafd1cd718437.
31546 For gst_element_class_get_pad_template(), no unreferencing is necessary
31547 according to the GStreamer documentation.
31549 2012-03-02 13:41:16 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31552 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31553 h264: fix slice_data_bit_offset calculation.
31554 Unlike what VA-API documentation defines, the slice_data_bit_offset
31555 represents the offset to the first macroblock in the slice data, minus
31556 any emulation prevention bytes in the slice_header().
31557 This fix copes with binary-only VA drivers that won't be fixed any
31558 time soon. Besides, this aligns with the current FFmpeg behaviour
31559 that was based on those proprietary drivers implementing the API
31562 2012-02-21 02:11:20 -0500 Wind Yuan <feng.yuan@intel.com>
31564 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31565 h264: skip all Access Unit (AU) NALs.
31566 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31568 2012-02-29 03:08:46 -0500 Wind Yuan <feng.yuan@intel.com>
31570 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31571 h264: fix modification process of reference picture lists.
31572 Construction of RefPicList0/1 could be off by one element.
31573 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31575 2012-02-12 11:21:52 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
31577 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31578 mpeg2: fix size calculation from sequence_extension().
31579 Original values from sequence_header() are 12-bit and the remaining
31580 2 most significant bits are coming from sequence_extension().
31581 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31583 2012-02-23 16:39:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31585 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31586 mpeg2: fix slice_vertical_position calculation.
31587 Make sure to adjust slice_vertical_position if picture structure
31588 is a top or bottom field.
31590 2012-02-23 16:23:27 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31592 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
31593 decoder: add picture structure flags.
31595 2012-02-23 14:42:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31597 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31598 mpeg2: fix decoding at end-of-sequence.
31600 2012-02-23 14:17:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31602 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31603 mpeg2: fix slice_horizontal_position calculation.
31605 2012-02-23 16:14:02 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31607 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31608 mpeg2: drop useless mb_y and mb_height members.
31610 2012-02-23 11:19:48 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31612 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31613 mpeg2: fix decoding of multiple slices with same slice_vertical_position.
31614 6.3.15 says that "some slices may have the same slice_vertical_position,
31615 since slices may start and finish anywhere". So, we can't submit the current
31616 picture to the HW right away since subsequent slices would be missing.
31618 2012-02-15 14:08:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31620 * gst-libs/gst/vaapi/gstvaapiimage.c:
31621 image: fix source stride in picture copy.
31623 2012-02-13 10:10:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31625 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31626 decoder: fix double buffer free with some VA drivers.
31627 vaRenderPicture() implicitly disposes VA buffers. Some VA drivers would
31628 push the VA buffer object into a list of free buffers to be re-used. However,
31629 reference pictures (and data) that was kept would explicitly release the VA
31630 buffer object later on, thus possibly destroying a valid (re-used) object.
31631 Besides, some other VA drivers don't support correctly the vaRenderPicture()
31632 semantics for VA buffers disposal and would leak memory if there is no explicit
31633 vaDestroyBuffer(). The temporary workaround is to explcitily destroy VA buffers
31634 right after vaRenderPicture(). All VA drivers need to be aligned.
31636 2012-02-08 18:08:49 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31638 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31639 h264: complete any current picture decoder before SPS / PPS change.
31640 This ensures the VA context is clear when the encoded resolution
31641 changes. i.e. make sure older picture is decoded with the older
31642 VA context before it changes.
31644 2012-02-08 18:07:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31646 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31647 h264: create VA context earlier when SPS is parsed.
31649 2012-02-08 17:57:29 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31651 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31652 h264: don't allocate too big data structures on stack.
31654 2012-02-07 11:07:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31656 * gst-libs/gst/vaapi/glibcompat.h:
31657 * gst-libs/gst/vaapi/gstvaapidisplay.c:
31658 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
31659 glib: map deprecated API to glib >= 2.32 equivalents.
31660 GStaticMutex and GStaticRecMutex are now replaced with GMutex and
31661 GRecMutex, which no longer require any prior call to g_thread_init().
31663 2012-02-07 10:01:01 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31665 * gst-libs/gst/vaapi/gstvaapiutils.h:
31666 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
31667 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
31668 glib: fix includes.
31670 2012-02-07 10:05:53 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31672 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
31673 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31674 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31675 * gst-libs/gst/vaapi/gstvaapidisplay.c:
31676 cosmetics: fix warnings (drop unused variables).
31678 2012-02-06 16:11:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31680 * gst/vaapi/gstvaapidecode.c:
31681 vaapidecode: fix another pad template ref leak.
31683 2012-02-06 15:54:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31685 * gst-libs/gst/vaapi/gstvaapidecoder.c:
31686 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31687 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
31688 mpeg2: fix crash when there is no free surface to decode into.
31690 2012-01-31 16:38:58 +0800 Zhao Halley <halley.zhao@intel.com>
31692 * gst-libs/gst/vaapi/gstvaapidisplay.c:
31693 display: skip profiles which have no entrypoints.
31694 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31696 2012-02-05 18:28:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31698 * gst/vaapi/gstvaapiupload.c:
31699 vaapiupload: use g_object_unref() for GstVaapiImage.
31701 2012-02-05 18:24:08 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31703 * gst/vaapi/gstvaapidecode.c:
31704 * gst/vaapi/gstvaapidownload.c:
31705 * gst/vaapi/gstvaapisink.c:
31706 * gst/vaapi/gstvaapiupload.c:
31707 plugins: fix pad template ref leaks.
31709 2012-02-02 09:23:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31712 Bump version for development.
31714 2012-02-01 23:34:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31720 2012-02-01 23:32:47 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31724 Mention codecparsers-based decoders, FFmpeg is now optional. Update
31725 list of support HW.
31727 2012-02-01 23:28:23 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31732 2012-01-31 11:34:17 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31734 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31735 decoder: allocate proxy surface earlier.
31736 This simplifies gst_vaapi_picture_output() to only update the presentation
31737 timestamp and submit the proxy to the decoder for output.
31739 2012-01-31 11:26:37 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31741 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31742 decoder: fix memory leak of VA objects on exit.
31743 On sequence end, if the last decoded picture is not output for rendering,
31744 then the proxy surface is not created. In this case, the original surface
31745 must be released explicitly to the context.
31747 2012-01-31 10:47:36 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31749 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31750 mpeg2: use GstAdapter to track input sequence.
31751 This fixes possible memory leaks and improves performance by removing
31754 2012-01-30 18:25:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31757 * gst-libs/gst/vaapi/Makefile.am:
31758 * gst-libs/gst/vaapi/glibcompat.h:
31759 * gst-libs/gst/vaapi/sysdeps.h:
31760 Add glib compatibility glue for older versions.
31762 2012-01-30 18:12:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31764 * gst-libs/gst/vaapi/Makefile.am:
31765 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
31766 * gst-libs/gst/vaapi/gstvaapicontext.c:
31767 * gst-libs/gst/vaapi/gstvaapidecoder.c:
31768 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
31769 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31770 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31771 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31772 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31773 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
31774 * gst-libs/gst/vaapi/gstvaapidisplay.c:
31775 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
31776 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
31777 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
31778 * gst-libs/gst/vaapi/gstvaapiimage.c:
31779 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
31780 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
31781 * gst-libs/gst/vaapi/gstvaapiobject.c:
31782 * gst-libs/gst/vaapi/gstvaapiparamspecs.c:
31783 * gst-libs/gst/vaapi/gstvaapiprofile.c:
31784 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
31785 * gst-libs/gst/vaapi/gstvaapisurface.c:
31786 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
31787 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
31788 * gst-libs/gst/vaapi/gstvaapitexture.c:
31789 * gst-libs/gst/vaapi/gstvaapiutils.c:
31790 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
31791 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
31792 * gst-libs/gst/vaapi/gstvaapivalue.c:
31793 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
31794 * gst-libs/gst/vaapi/gstvaapivideobuffer_glx.c:
31795 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
31796 * gst-libs/gst/vaapi/gstvaapivideopool.c:
31797 * gst-libs/gst/vaapi/gstvaapivideosink.c:
31798 * gst-libs/gst/vaapi/gstvaapiwindow.c:
31799 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
31800 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
31801 * gst-libs/gst/vaapi/sysdeps.h:
31802 Add header for system-dependent definitions.
31804 2012-01-30 10:15:32 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31806 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
31807 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31808 * gst-libs/gst/vaapi/gstvaapiutils.c:
31809 * gst-libs/gst/vaapi/gstvaapiutils.h:
31810 decoder: optimize slice data buffers initialization.
31811 VA drivers may have a faster means to transfer user buffers to GPU
31812 buffers than using memcpy(). In particular, on Intel Gen graphics, we
31813 can use pwrite(). This provides for faster upload of bitstream and can
31814 help higher bitrates.
31815 vaapi_create_buffer() helper function was also updated to allow for
31816 un-mapped buffers and pre-initialized data for buffers.
31818 2012-01-27 17:28:50 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31820 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31821 h264: simplify RefPicList reconstruction.
31823 2012-01-27 16:08:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31825 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31826 h264: flush DPB when the end of the sequence is reached.
31828 2012-01-24 15:38:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31830 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31831 h264: handle Decoded Picture Buffer (DPB).
31833 2012-01-24 09:20:25 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31835 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31836 h264: fix pred_weight_table() reconstruction.
31837 Only the explicit pred_weight_table(), possibly with the inferred default
31838 values, shall be required. e.g. don't fill in the table if weighted_pred_flag
31839 is not set for P/SP slices.
31841 2012-01-23 15:03:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31843 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31844 h264: execute reference picture marking process (MMCO).
31846 2012-01-23 15:20:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31848 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31849 h264: fix presentation timestamps.
31851 2012-01-18 13:38:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31853 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31854 h264: execute reference picture marking process (sliding window).
31856 2012-01-17 10:42:49 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31858 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31859 h264: handle avcC format for decoding buffers.
31861 2011-11-25 14:37:00 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
31863 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31864 h264: handle codec-data.
31865 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31867 2011-08-12 17:43:55 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31869 * gst-libs/gst/vaapi/Makefile.am:
31870 * gst-libs/gst/vaapi/gstvaapidecoder_h264.c:
31871 * gst-libs/gst/vaapi/gstvaapidecoder_h264.h:
31872 * gst/vaapi/gstvaapidecode.c:
31873 * tests/test-decode.c:
31874 Add initial H.264 decoder.
31876 2012-01-26 15:28:42 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31878 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31879 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
31880 decoder: retain proxy surface until the GstVaapiPicture is destroyed.
31881 Keep a valid reference to the proxy in GstVaapiPicture so that frames
31882 marked as "used for reference" could be kept during the lifetime of the
31883 picture. i.e. don't release them too soon as they could be re-used right
31886 2012-01-26 15:19:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31888 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31889 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31890 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
31891 decoder: properly reference count pictures.
31892 This fixes cases where a GstVaapiPicture would be destroyed whereas
31893 there is still a valid instance of it in either prev, current or
31896 2012-01-26 14:54:31 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31898 * gst-libs/gst/vaapi/gstvaapidecoder.c:
31899 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
31900 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31901 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
31902 decoder: simplify output of decoded frames.
31903 Drop obsolete gst_vaapi_decoder_push_surface() that was no longer used.
31904 Change gst_vaapi_decoder_push_surface_proxy() semantics to assume PTS
31905 is already set correctly and reference count increased, if necessary.
31907 2012-01-26 09:48:11 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31909 * gst-libs/gst/vaapi/Makefile.am:
31910 * gst-libs/gst/vaapi/gstvaapicodec_objects.c:
31911 * gst-libs/gst/vaapi/gstvaapicodec_objects.h:
31912 * gst-libs/gst/vaapi/gstvaapidecoder.c:
31913 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
31914 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
31915 * gst-libs/gst/vaapi/gstvaapidecoder_objects.c:
31916 * gst-libs/gst/vaapi/gstvaapidecoder_objects.h:
31917 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
31918 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
31919 decoder: rework the internal VA objects API.
31920 The new API simplifies a lot reference counting and makes it more
31921 flexible for future additions/changes. The GstVaapiCodecInfo is
31922 also gone. Rather, new helper macros are provided to allocate
31923 picture, slice and quantization matrix parameter buffers.
31925 2012-01-24 10:21:45 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31927 * gst-libs/gst/vaapi/gstvaapidecoder.c:
31928 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
31929 decoder: allow slices to be attached to pictures later.
31931 2011-11-21 18:39:49 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31933 * gst-libs/gst/vaapi/gstvaapidecoder.c:
31934 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
31935 decoder: add ref_count to GstVaapiPicture.
31937 2012-01-23 11:48:42 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31939 * gst/vaapi/gstvaapisink.c:
31940 vaapisink: cap window size to the maximum display size.
31942 2012-01-18 10:23:41 +0100 Zhao Halley <halley.zhao@intel.com>
31944 * gst-libs/gst/vaapi/gstvaapiprofile.c:
31945 * gst-libs/gst/vaapi/gstvaapiworkarounds.h:
31946 profile: match video/x-h263 as H.263 Baseline profile.
31947 HACK: qtdemux does not report profiles for H.263. So, assume plain
31948 "video/x-h263" is H.263 Baseline profile.
31949 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31951 2012-01-18 10:22:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31953 * gst-libs/gst/vaapi/gstvaapidisplay.c:
31954 * gst-libs/gst/vaapi/gstvaapiworkarounds.h:
31955 display: report H.263 Baseline profile.
31956 HACK: expose GST_VAAPI_PROFILE_H263_BASELINE for decoding if MPEG-4:2 Simple
31957 profile (VAProfileMPEG4Simple) is supported.
31959 2012-01-24 10:06:37 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31961 * gst-libs/gst/vaapi/Makefile.am:
31962 * gst-libs/gst/vaapi/gstvaapiworkarounds.h:
31963 Add template for workarounds.
31965 2012-01-18 10:47:56 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31967 * tests/test-decode.c:
31968 tests: error out if FFmpeg|codecparsers are not supported.
31970 2012-01-18 10:42:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31972 * docs/reference/libs/libs-sections.txt:
31973 * gst-libs/gst/vaapi/gstvaapisurface.h:
31974 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
31975 surface: don't expose gst_vaapi_surface_get_parent_context().
31976 gst_vaapi_surface_get_parent_context() was not meant to be exposed globally.
31977 It's just an internal helper function. However, it's still possible to get
31978 the parent context through the "parent-context" property.
31980 2012-01-16 14:19:00 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31982 * tests/Makefile.am:
31983 * tests/test-subpicture.c:
31984 tests: fix build without FFmpeg.
31986 2012-01-16 14:09:57 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31989 Bump version for development.
31991 2012-01-16 11:05:31 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
31997 2012-01-16 11:03:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32000 * debian.upstream/copyright:
32001 * gst-libs/gst/vaapi/gstvaapicontext.c:
32002 * gst-libs/gst/vaapi/gstvaapidecoder.c:
32003 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
32004 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
32005 * gst-libs/gst/vaapi/gstvaapidisplay.c:
32006 * gst-libs/gst/vaapi/gstvaapidisplay.h:
32007 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
32008 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
32009 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
32010 * gst-libs/gst/vaapi/gstvaapiimage.c:
32011 * gst-libs/gst/vaapi/gstvaapiimage.h:
32012 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
32013 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
32014 * gst-libs/gst/vaapi/gstvaapiutils.c:
32015 * gst-libs/gst/vaapi/gstvaapiutils.h:
32016 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
32017 * gst-libs/gst/vaapi/gstvaapiwindow.c:
32018 * gst/vaapi/gstvaapi.c:
32019 * gst/vaapi/gstvaapidecode.c:
32020 * gst/vaapi/gstvaapidownload.c:
32021 * gst/vaapi/gstvaapidownload.h:
32022 * gst/vaapi/gstvaapipluginutil.c:
32023 * gst/vaapi/gstvaapipluginutil.h:
32024 * gst/vaapi/gstvaapisink.c:
32025 * gst/vaapi/gstvaapiupload.c:
32026 * gst/vaapi/gstvaapiupload.h:
32027 * tests/test-decode.c:
32028 legal: fix year for some copyright notices.
32030 2012-01-16 10:42:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32032 * gst-libs/gst/vaapi/gstvaapicontext.c:
32033 * gst-libs/gst/vaapi/gstvaapicontext.h:
32034 * gst-libs/gst/vaapi/gstvaapidecoder.h:
32035 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
32036 * gst-libs/gst/vaapi/gstvaapidisplay.c:
32037 * gst-libs/gst/vaapi/gstvaapidisplay.h:
32038 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
32039 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
32040 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
32041 * gst-libs/gst/vaapi/gstvaapiimage.c:
32042 * gst-libs/gst/vaapi/gstvaapiimage.h:
32043 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
32044 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
32045 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
32046 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
32047 * gst-libs/gst/vaapi/gstvaapisurface.c:
32048 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
32049 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
32050 * gst-libs/gst/vaapi/gstvaapiutils.c:
32051 * gst-libs/gst/vaapi/gstvaapiutils.h:
32052 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
32053 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
32054 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
32055 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
32056 * gst-libs/gst/vaapi/gstvaapiwindow.c:
32057 * gst/vaapi/gstvaapidownload.h:
32058 * gst/vaapi/gstvaapisink.h:
32059 * gst/vaapi/gstvaapiupload.h:
32060 * tests/test-decode.c:
32061 legal: add Intel copyright on modified files.
32063 2012-01-16 10:41:10 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32065 * gst-libs/gst/vaapi/gstvaapi_priv.h:
32066 * gst-libs/gst/vaapi/gstvaapicompat.h:
32067 * gst-libs/gst/vaapi/gstvaapicontext.c:
32068 * gst-libs/gst/vaapi/gstvaapicontext.h:
32069 * gst-libs/gst/vaapi/gstvaapidebug.h:
32070 * gst-libs/gst/vaapi/gstvaapidecoder.c:
32071 * gst-libs/gst/vaapi/gstvaapidecoder.h:
32072 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
32073 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.h:
32074 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
32075 * gst-libs/gst/vaapi/gstvaapidisplay.c:
32076 * gst-libs/gst/vaapi/gstvaapidisplay.h:
32077 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
32078 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
32079 * gst-libs/gst/vaapi/gstvaapidisplay_glx_priv.h:
32080 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
32081 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
32082 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
32083 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
32084 * gst-libs/gst/vaapi/gstvaapiimage.c:
32085 * gst-libs/gst/vaapi/gstvaapiimage.h:
32086 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
32087 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
32088 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
32089 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
32090 * gst-libs/gst/vaapi/gstvaapiobject.c:
32091 * gst-libs/gst/vaapi/gstvaapiobject.h:
32092 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
32093 * gst-libs/gst/vaapi/gstvaapiparamspecs.c:
32094 * gst-libs/gst/vaapi/gstvaapiparamspecs.h:
32095 * gst-libs/gst/vaapi/gstvaapiprofile.c:
32096 * gst-libs/gst/vaapi/gstvaapiprofile.h:
32097 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
32098 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
32099 * gst-libs/gst/vaapi/gstvaapisurface.c:
32100 * gst-libs/gst/vaapi/gstvaapisurface.h:
32101 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
32102 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
32103 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
32104 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
32105 * gst-libs/gst/vaapi/gstvaapitexture.c:
32106 * gst-libs/gst/vaapi/gstvaapitexture.h:
32107 * gst-libs/gst/vaapi/gstvaapitypes.h:
32108 * gst-libs/gst/vaapi/gstvaapiutils.c:
32109 * gst-libs/gst/vaapi/gstvaapiutils.h:
32110 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
32111 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
32112 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
32113 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
32114 * gst-libs/gst/vaapi/gstvaapivalue.c:
32115 * gst-libs/gst/vaapi/gstvaapivalue.h:
32116 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
32117 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
32118 * gst-libs/gst/vaapi/gstvaapivideopool.c:
32119 * gst-libs/gst/vaapi/gstvaapivideopool.h:
32120 * gst-libs/gst/vaapi/gstvaapivideosink.c:
32121 * gst-libs/gst/vaapi/gstvaapivideosink.h:
32122 * gst-libs/gst/vaapi/gstvaapiwindow.c:
32123 * gst-libs/gst/vaapi/gstvaapiwindow.h:
32124 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
32125 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
32126 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
32127 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
32128 * gst/vaapi/gstvaapidecode.c:
32129 * gst/vaapi/gstvaapidecode.h:
32130 * gst/vaapi/gstvaapidownload.c:
32131 * gst/vaapi/gstvaapidownload.h:
32132 * gst/vaapi/gstvaapisink.c:
32133 * gst/vaapi/gstvaapisink.h:
32134 * gst/vaapi/gstvaapiupload.c:
32135 * gst/vaapi/gstvaapiupload.h:
32138 * tests/test-decode.c:
32139 * tests/test-decode.h:
32140 * tests/test-display.c:
32141 * tests/test-h264.c:
32142 * tests/test-h264.h:
32143 * tests/test-mpeg2.c:
32144 * tests/test-mpeg2.h:
32145 * tests/test-surfaces.c:
32146 * tests/test-textures.c:
32147 * tests/test-vc1.c:
32148 * tests/test-vc1.h:
32149 * tests/test-windows.c:
32150 legal: fix copyright notices to include "Copyright" term.
32152 2011-12-09 16:44:03 +0800 Zhao Halley <halley.zhao@intel.com>
32154 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
32155 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.h:
32156 mpeg4: replace GstVaapiTSB with GstAdapter (gst-plugins-base >= 0.10.24).
32157 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32159 2011-12-09 16:28:11 +0800 Zhao Halley <halley.zhao@intel.com>
32161 * gst-libs/gst/vaapi/Makefile.am:
32162 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.c:
32163 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg4.h:
32164 * gst/vaapi/gstvaapidecode.c:
32165 Add initial MPEG-4 decoder.
32166 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32168 2011-11-18 15:41:40 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
32170 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
32171 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.h:
32172 vc1: replace GstVaapiTSB with GstAdapter (gst-plugins-base >= 0.10.24).
32173 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32175 2011-10-07 11:50:20 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32177 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
32178 vc1: fix codec-data decoding for WMV3 format.
32180 2011-10-07 11:12:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32182 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
32183 vc1: fix presentation timestamps.
32185 2011-10-06 15:59:22 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32187 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
32188 vc1: fix MV mode packing.
32190 2011-10-05 16:41:57 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32192 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
32193 vc1: handle codec-data.
32195 2011-10-05 15:56:36 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32197 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
32198 vc1: handle encapsulated bitstreams.
32200 2011-10-04 17:51:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32202 * gst-libs/gst/vaapi/gstvaapidecoder.c:
32203 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
32204 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
32205 vc1: fix bitplanes decoding.
32207 2011-10-04 14:15:55 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32209 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
32210 vc1: fix BFRACTION reconstruction.
32212 2011-09-30 17:16:23 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32214 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
32215 vc1: fix framerate calculation.
32217 2011-09-30 13:40:11 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32219 * gst-libs/gst/vaapi/Makefile.am:
32220 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.c:
32221 * gst-libs/gst/vaapi/gstvaapidecoder_vc1.h:
32222 * gst/vaapi/gstvaapidecode.c:
32223 * tests/test-decode.c:
32224 Add initial VC-1 decoder.
32226 2012-01-09 17:37:34 +0100 Zhao Halley <halley.zhao@intel.com>
32228 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
32229 mpeg2: fix first field detection.
32230 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32232 2012-01-06 16:44:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32234 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
32235 mpeg2: fix quantisation matrix construction.
32237 2011-11-18 15:06:07 +0200 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
32239 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
32240 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.h:
32241 mpeg2: replace GstVaapiTSB API with GstAdapter (gst-plugins-base >= 0.10.24).
32242 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32244 2011-09-14 18:11:57 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32246 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
32247 mpeg2: fix packets spanning over two buffers.
32249 2011-09-12 18:20:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32251 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
32252 mpeg2: ignore system start codes (PES headers).
32254 2011-09-12 18:02:53 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32256 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
32257 mpeg2: handle closed_gop.
32259 2011-08-05 11:55:11 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32261 * gst-libs/gst/vaapi/Makefile.am:
32262 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.c:
32263 * gst-libs/gst/vaapi/gstvaapidecoder_mpeg2.h:
32264 * gst/vaapi/gstvaapidecode.c:
32265 * tests/test-decode.c:
32266 Add initial MPEG-2 decoder.
32268 2011-08-12 10:21:19 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32271 * gst-libs/gst/vaapi/Makefile.am:
32272 * gst/vaapi/gstvaapidecode.c:
32273 * tests/test-decode.c:
32274 Allow conditional build of GStreamer/FFmpeg bitstream parsers.
32276 2011-08-05 11:53:50 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32278 * gst-libs/gst/vaapi/gstvaapidecoder.c:
32279 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
32280 Add VA decoder helpers.
32282 2011-08-05 11:52:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32284 * gst-libs/gst/vaapi/gstvaapidecoder.h:
32285 decoder: add new error codes.
32286 GST_VAAPI_DECODER_STATUS_ERROR_UNSUPPORTED_PROFILE: for unsupported profile
32287 GST_VAAPI_DECODER_STATUS_ERROR_UNSUPPORTED_CHROMA_FORMAT: for unsupported chroma format
32289 2011-09-12 13:00:59 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32291 * gst-libs/gst/vaapi/Makefile.am:
32292 * gst-libs/gst/vaapi/gstvaapiutils_tsb.c:
32293 * gst-libs/gst/vaapi/gstvaapiutils_tsb.h:
32294 Add timestamp buffer store helper utils.
32296 2011-08-04 17:29:41 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32298 * gst-libs/gst/vaapi/gstvaapiutils.c:
32299 * gst-libs/gst/vaapi/gstvaapiutils.h:
32300 Add VA buffer helpers.
32302 2012-01-13 15:03:38 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32304 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
32305 utils: slight improvements to gl_bind_texture().
32307 2012-01-13 14:13:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32309 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
32310 utils: pretty-print output of gl_get_error_string().
32312 2012-01-13 14:03:29 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32314 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
32315 utils: rewrite gl_perspective() as per OpenGL FAQ 9.085.
32317 2012-01-13 12:09:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32319 * gst-libs/gst/vaapi/gstvaapiutils.c:
32320 utils: simplify string of VAProfile/VAEntrypoint.
32322 2012-01-13 11:46:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32324 * gst-libs/gst/vaapi/gstvaapidisplay.c:
32325 * gst-libs/gst/vaapi/gstvaapiutils.c:
32326 * gst-libs/gst/vaapi/gstvaapiutils.h:
32327 utils: drop string_of_FOURCC() in favor of standard GST_FOURCC_* helpers.
32329 2012-01-12 17:18:47 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32334 2012-01-12 15:34:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32336 * tests/test-decode.c:
32337 tests: check for shared VA displays (display cache).
32339 2012-01-12 15:30:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32341 * gst-libs/gst/vaapi/gstvaapidisplay.c:
32342 display: always free VA display cache if it is empty.
32344 2012-01-12 15:03:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32346 * gst-libs/gst/vaapi/gstvaapidisplay.c:
32347 * gst-libs/gst/vaapi/gstvaapidisplay.h:
32348 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
32349 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
32350 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
32351 display: use VA display cache for X11 and GLX winsys.
32353 2012-01-12 12:46:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32355 * gst-libs/gst/vaapi/Makefile.am:
32356 * gst-libs/gst/vaapi/gstvaapidisplay.h:
32357 * gst-libs/gst/vaapi/gstvaapidisplaycache.c:
32358 * gst-libs/gst/vaapi/gstvaapidisplaycache.h:
32359 display: implement a VA display cache.
32361 2012-01-11 14:13:06 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32363 * gst/vaapi/gstvaapipluginutil.c:
32364 vaapiplugin: fix gst_vaapi_ensure_display() to use system defaults.
32365 This ensures the display name provided to gst_vaapi_display_*_new()
32366 maps to the system defaults, instead of forcing "" that could be different
32367 from the current DISPLAY name.
32369 2011-08-26 15:44:25 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
32371 * gst/vaapi/gstvaapiupload.c:
32372 vaapiupload: only set caps on newly created buffers.
32373 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32375 2012-01-11 14:11:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32377 * gst/vaapi/gstvaapisink.c:
32378 vaapisink: ensure VA display in GstBaseSink::start() hook.
32379 This ensures a VA display is ready by the time upstream elements request
32382 2011-08-26 15:44:46 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
32384 * gst/vaapi/gstvaapisink.c:
32385 vaapisink: don't leak GL texture.
32386 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32388 2012-01-09 16:51:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32390 * gst/vaapi/gstvaapisink.c:
32391 vaapisink: fix calculation of render region.
32393 2012-01-09 11:23:39 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32395 * gst/vaapi/gstvaapisink.c:
32396 vaapisink: automatically fit video to window.
32398 2012-01-09 10:37:30 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32400 * gst/vaapi/gstvaapisink.c:
32401 vaapisink: implement GstXOverlay::set_render_rectangle().
32403 2012-01-09 11:04:21 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32405 * gst-libs/gst/vaapi/gstvaapiwindow.c:
32406 window: always check geometry when the window is mapped.
32408 2012-01-06 17:51:59 +0100 Zhao Halley <halley.zhao@intel.com>
32410 * gst-libs/gst/vaapi/gstvaapiutils.c:
32411 Add missing profiles from VA-API 0.32.0.
32412 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32414 2012-01-06 16:48:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32417 .gitignore: add test-subpicture.
32419 2012-01-06 11:23:21 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32422 Bump version for development.
32424 2012-01-06 11:20:48 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32430 2012-01-06 11:18:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32432 * tests/Makefile.am:
32433 tests: fix make dist (ship with test-subpicture-data.h).
32435 2012-01-05 17:35:12 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32440 2012-01-05 17:09:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32442 * gst-libs/gst/vaapi/gstvaapidecoder.c:
32443 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
32444 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
32445 decoder: fix possible leak of VA surfaces.
32446 Under some circumstances, we could have leaked a surface, thus not
32447 releasing it to the pool of available surfaces in the VA context.
32448 The strategy is now to use a proxy earlier and automatically ref/unref
32449 whenever necessary. In particular, during the lifetime needed for FFmpeg.
32451 2012-01-05 16:59:57 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32453 * docs/reference/libs/libs-sections.txt:
32454 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
32455 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
32456 surfaceproxy: add helper to retrieve the VA surface ID.
32458 2012-01-05 16:44:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32460 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
32461 surfaceproxy: simplify destruction.
32462 Also make sure to always make sure to release the surface back to the
32463 pool of surfaces in the associated VA context, if any.
32465 2012-01-05 16:26:49 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32467 * gst/vaapi/gstvaapidecode.c:
32468 vaapidecode: fix deinitialization order.
32470 2012-01-05 14:50:26 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32472 * gst-libs/gst/vaapi/gstvaapicontext.c:
32473 context: avoid self reference loops with surfaces.
32475 2012-01-05 11:23:01 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32477 * debian.upstream/control.in:
32478 debian: update control.in description for new plugins.
32480 2012-01-05 11:01:56 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32482 * gst/vaapi/gstvaapiupload.c:
32483 vaapiupload: use new gst_vaapi_append_surface_caps() helper.
32484 This also fixes extra structures, beyond the one at index 0, to hold
32485 the right additional values.
32487 2012-01-05 10:55:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32489 * gst/vaapi/gstvaapiupload.c:
32490 vaapiupload: fix sink (YUV) caps to not report type and opengl fields.
32492 2012-01-05 10:50:59 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32495 * docs/reference/plugins/plugins-docs.xml.in:
32496 * docs/reference/plugins/plugins-sections.txt:
32497 * docs/reference/plugins/plugins.types:
32498 * gst/vaapi/Makefile.am:
32499 * gst/vaapi/gstvaapi.c:
32500 * gst/vaapi/gstvaapidownload.c:
32501 * gst/vaapi/gstvaapidownload.h:
32502 vaapidownload: add new plugin to download pixels from VA surfaces.
32504 2012-01-05 11:00:39 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32506 * gst/vaapi/gstvaapipluginutil.c:
32507 * gst/vaapi/gstvaapipluginutil.h:
32508 vaapipluingutils: add helper to append surface caps to YUV caps.
32510 2012-01-05 10:29:48 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32512 * docs/reference/libs/libs-sections.txt:
32513 * gst-libs/gst/vaapi/gstvaapiimage.c:
32514 * gst-libs/gst/vaapi/gstvaapiimage.h:
32515 image: add helpers to extract pixels to user buffers.
32517 2012-01-04 11:34:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32519 * gst-libs/gst/vaapi/gstvaapiimage.c:
32520 image: simplify initialization of raw images from video buffers.
32522 2012-01-04 11:29:11 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32524 * gst-libs/gst/vaapi/gstvaapiimage.c:
32525 image: fix update from NV12 buffers.
32527 2012-01-03 18:16:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32529 * gst/vaapi/gstvaapiupload.c:
32530 vaapiupload: fix memory leak in _init() function.
32532 2012-01-03 14:34:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32536 * debian.upstream/control.in:
32537 * docs/reference/plugins/plugins-docs.xml.in:
32538 * docs/reference/plugins/plugins-sections.txt:
32539 * docs/reference/plugins/plugins.types:
32540 * gst/vaapi/Makefile.am:
32541 * gst/vaapi/gstvaapi.c:
32542 * gst/vaapi/gstvaapiupload.c:
32543 * gst/vaapi/gstvaapiupload.h:
32544 Rename vaapiconvert element to vaapiupload.
32546 2012-01-03 13:54:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32549 Bump version for development.
32551 2012-01-03 13:42:12 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32557 2011-12-14 15:22:24 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32562 2011-12-14 14:40:37 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32564 * gst-libs/gst/vaapi/gstvaapisurface.c:
32565 * gst-libs/gst/vaapi/gstvaapisurface.h:
32566 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
32567 * gst/vaapi/gstvaapisink.c:
32568 surface: apply composition to the parent context, if requested.
32570 2011-12-14 14:35:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32572 * docs/reference/libs/libs-sections.txt:
32573 * gst-libs/gst/vaapi/gstvaapicontext.c:
32574 * gst-libs/gst/vaapi/gstvaapicontext.h:
32575 context: make it possible to apply composition globally.
32577 2011-12-14 14:13:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32579 * gst-libs/gst/vaapi/gstvaapisurface.c:
32580 surface: fix associate subpicture to not report deassociation errors.
32582 2011-12-14 13:46:26 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32584 * gst-libs/gst/vaapi/gstvaapisurface.c:
32585 surface: fix typo in debug message.
32587 2011-12-14 13:16:21 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32589 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
32590 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
32591 * gst-libs/gst/vaapi/gstvaapisurface.c:
32592 subpicture: add helper to create subpicture from GstVideoOverlayRectangle.
32594 2011-12-13 16:53:15 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32596 * docs/reference/libs/libs-sections.txt:
32597 * gst-libs/gst/vaapi/Makefile.am:
32598 * gst-libs/gst/vaapi/gstvaapicontext.c:
32599 * gst-libs/gst/vaapi/gstvaapisurface.c:
32600 * gst-libs/gst/vaapi/gstvaapisurface.h:
32601 * gst-libs/gst/vaapi/gstvaapisurface_priv.h:
32602 surface: record parent context.
32604 2011-12-13 15:59:02 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32606 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
32607 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
32608 * gst/vaapi/gstvaapidecode.c:
32609 * gst/vaapi/gstvaapisink.c:
32612 2011-12-13 15:51:58 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32614 * gst-libs/gst/vaapi/gstvaapisurface.c:
32615 * gst-libs/gst/vaapi/gstvaapisurface.h:
32616 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
32617 * gst/vaapi/gstvaapisink.c:
32618 Rename gst_vaapi_surface_update_composition() to gst_vaapi_surface_set_subpictures_from_composition().
32620 2011-12-13 13:40:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32623 configure: check for GstVideoOverlayComposition.
32625 2011-12-12 18:42:44 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32630 2011-11-25 15:00:25 -0500 Nicolas Dufresne <nicolas.dufresne@collabora.co.uk>
32632 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
32633 converter: add support for GstVideoOverlayComposition planes.
32634 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32636 2011-11-23 16:45:46 -0300 Thibault Saunier <thibault.saunier@collabora.com>
32638 * gst/vaapi/gstvaapisink.c:
32639 vaapisink: handle GstVideoOverlayComposition planes.
32640 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32642 2011-12-12 18:27:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32644 * gst-libs/gst/vaapi/gstvaapisurface.c:
32645 surface: use unscaled overlay rectangle for blending.
32647 2011-12-12 18:37:13 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32649 * gst-libs/gst/vaapi/gstvaapisurface.c:
32650 surface: fix VA image leak when an error occurred.
32652 2011-11-25 14:59:56 -0500 Nicolas Dufresne <nicolas.dufresne@collabora.co.uk>
32654 * gst-libs/gst/vaapi/gstvaapisurface.c:
32655 * gst-libs/gst/vaapi/gstvaapisurface.h:
32656 surface: add helper to handle GstVideoOverlayComposition.
32657 This helper resets the subpictures to reflect the current composition
32658 layers provided with the buffers.
32659 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32661 2011-12-12 18:13:19 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32663 * docs/reference/libs/libs-sections.txt:
32664 * gst-libs/gst/vaapi/Makefile.am:
32665 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
32666 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
32667 * gst/vaapi/gstvaapiconvert.c:
32668 image: add gst_vaapi_image_format_from_video() helper.
32670 2011-12-12 16:34:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32672 * gst-libs/gst/vaapi/gstvaapiimage.c:
32673 * gst-libs/gst/vaapi/gstvaapiimage.h:
32674 image: allow updates from GstVaapiImageRaw.
32676 2011-12-12 14:34:03 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32678 * gst-libs/gst/vaapi/gstvaapiimage.c:
32679 * gst-libs/gst/vaapi/gstvaapiimage.h:
32680 * gst/vaapi/gstvaapiconvert.c:
32681 * tests/test-subpicture.c:
32682 image: allow partial updates.
32684 2011-12-12 15:31:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32686 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
32687 subpicture: fix doc for gst_vaapi_subpicture_set_image().
32689 2011-12-12 13:39:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32691 * gst-libs/gst/vaapi/gstvaapidisplay.c:
32692 display: fix has_image_format() to check against subpicture formats.
32694 2011-10-17 18:43:15 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32696 * tests/test-subpicture.c:
32697 tests: fix subpicture test.
32699 2011-10-14 13:00:12 -0300 Thibault Saunier <thibault.saunier@collabora.com>
32701 * tests/Makefile.am:
32702 * tests/test-subpicture-data.c:
32703 * tests/test-subpicture-data.h:
32704 * tests/test-subpicture.c:
32705 tests: add test for subpictures.
32706 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32708 2011-11-25 12:28:04 -0500 Nicolas Dufresne <nicolas.dufresne@collabora.co.uk>
32710 * gst/vaapi/gstvaapiconvert.c:
32711 * gst/vaapi/gstvaapidecode.c:
32712 Add missing video context queries.
32713 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32715 2011-12-12 13:22:07 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32718 Bump version for development.
32720 2011-12-12 10:04:32 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32722 * gst-libs/gst/video/Makefile.am:
32723 * gst-libs/gst/video/gstbasevideocodec.c:
32724 * gst-libs/gst/video/gstbasevideocodec.h:
32725 * gst-libs/gst/video/gstbasevideodecoder.c:
32726 * gst-libs/gst/video/gstbasevideodecoder.h:
32727 * gst-libs/gst/video/gstbasevideoutils.c:
32728 * gst-libs/gst/video/gstbasevideoutils.h:
32729 Drop unused copy of GstBaseVideoDecoder.
32731 2011-12-09 11:46:45 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32737 2011-12-09 11:38:43 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32740 README: update dependencies.
32742 2011-12-09 11:38:34 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32747 2011-12-09 11:20:04 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32750 configure: check for GstBaseSink 'query' vfunc.
32752 2011-12-09 10:45:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32754 * gst/vaapi/Makefile.am:
32755 vaapiplugin: include local build dir to CFLAGS for generated files.
32757 2011-12-09 10:44:52 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32760 autogen: don't configure if NO_CONFIGURE variable is set.
32762 2011-12-08 11:54:59 +0100 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
32764 * gst/vaapi/gstvaapidecode.c:
32765 * gst/vaapi/gstvaapidecode.h:
32766 vaapidecode: return sink caps template if decoder is in NULL state.
32767 Otherwise, the decoder would always create its own X display instead
32768 of probing it from the downstream element, which is not reliable.
32769 e.g. DISPLAY is not :0 or when running on Wayland.
32770 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32772 2011-12-08 15:44:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32774 * gst/vaapi/gstvaapiconvert.c:
32775 * gst/vaapi/gstvaapidecode.c:
32776 vaapiplugin: properly set surface type to "vaapi" in caps.
32778 2011-12-08 15:16:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32780 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
32781 decoder: drop unused headers.
32783 2011-11-04 19:47:25 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.co.uk>
32785 * gst/vaapi/gstvaapiconvert.c:
32786 * gst/vaapi/gstvaapidecode.c:
32787 vaapiplugin: properly set opengl support in caps.
32788 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32790 2011-11-04 20:07:52 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.co.uk>
32792 * gst/vaapi/gstvaapiconvert.c:
32793 * gst/vaapi/gstvaapidecode.c:
32794 vaapiplugin: allocate GLX buffers when supported.
32795 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32797 2011-11-04 19:47:09 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.co.uk>
32799 * gst-libs/gst/vaapi/Makefile.am:
32800 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
32801 * gst-libs/gst/vaapi/gstvaapivideobuffer_glx.c:
32802 * gst-libs/gst/vaapi/gstvaapivideobuffer_glx.h:
32803 * gst-libs/gst/vaapi/gstvaapivideobuffer_priv.h:
32804 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.c:
32805 * gst-libs/gst/vaapi/gstvaapivideoconverter_glx.h:
32806 videobuffer: add GLX buffer support.
32807 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32809 2011-10-06 16:06:15 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
32811 * gst-libs/gst/vaapi/Makefile.am:
32812 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
32813 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
32814 Port to GstSurfaceBuffer interface.
32815 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32817 2011-10-06 16:04:37 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
32819 * gst/vaapi/gstvaapidecode.c:
32820 * gst/vaapi/gstvaapisink.c:
32821 Don't use downstream buffer allocation.
32822 With the new video/x-surface abstraction, we can't rely on having a VA
32823 specific sink downstream. Also, there was no particular reason to do that.
32824 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32826 2011-11-04 17:16:23 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.co.uk>
32828 * gst-libs/gst/vaapi/Makefile.am:
32829 * gst-libs/gst/vaapi/gstvaapicontext.c:
32830 * gst-libs/gst/vaapi/gstvaapisurface.h:
32831 * gst/vaapi/gstvaapiconvert.c:
32832 * gst/vaapi/gstvaapisink.c:
32833 * tests/Makefile.am:
32834 * tests/test-surfaces.c:
32835 Change caps to use new video/x-surface generic type.
32836 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32838 2011-11-04 16:50:15 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.co.uk>
32841 * gst-libs/gst/vaapi/Makefile.am:
32842 * gst-libs/gst/vaapi/gstvaapiutils_gst.c:
32843 * gst/vaapi/Makefile.am:
32844 * gst/vaapi/gstvaapiconvert.c:
32845 * gst/vaapi/gstvaapidecode.c:
32846 * gst/vaapi/gstvaapipluginutil.c:
32847 * gst/vaapi/gstvaapipluginutil.h:
32848 * gst/vaapi/gstvaapisink.c:
32849 * gst/vaapi/gstvaapisink.h:
32850 Port to GstVideoContext interface.
32851 This new interface allows for upstream and downstream display sharing
32852 that works in both static and dynamic pipelines.
32853 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32855 2011-12-08 14:57:36 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32857 * gst/vaapi/gstvaapisink.c:
32858 vaapisink: use GST_ERROR to print error messages.
32860 2011-12-08 13:30:51 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32862 * gst/vaapi/Makefile.am:
32863 vaapiplugin: link against VA/GLX when enabled.
32865 2011-12-07 19:09:55 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32867 * gst/vaapi/gstvaapiconvert.c:
32868 * gst/vaapi/gstvaapisink.c:
32869 Add Intel copyright information.
32871 2011-12-07 19:04:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32873 * gst/vaapi/gstvaapisink.c:
32874 vaapisink: allow compatibility with gst-plugins-base < 0.10.31.
32876 2011-12-07 18:40:35 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32880 * gst-libs/gst/Makefile.am:
32881 * gst-libs/gst/gstutils_version.h.in:
32882 Add new GStreamer version check utilities.
32884 2011-07-28 11:14:49 +0300 Sreerenj Balachandran <sreerenj.balachandran@intel.com>
32886 * gst/vaapi/gstvaapisink.c:
32887 vaapisink: replace the deprecated xoverlay API with the new one.
32888 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32890 2011-12-07 17:31:09 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32893 configure: allow for pre-releases.
32895 2011-10-13 17:08:13 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
32897 * gst-libs/gst/vaapi/gstvaapidisplay.c:
32898 display: destroy display on creation failure.
32899 This allows element to detect that the display creation has actually
32901 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32903 2011-10-13 17:07:35 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
32905 * gst-libs/gst/vaapi/gstvaapidisplay.c:
32906 display: don't crash when config is empty.
32907 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32909 2011-12-07 14:42:14 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32912 * debian.upstream/copyright:
32913 doc: mention Collabora copyrights.
32915 2011-12-07 14:40:20 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32918 .gitignore: refine for generated docs.
32920 2011-09-14 15:12:41 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
32923 * docs/reference/plugins/Makefile.am:
32925 * gst/vaapi/Makefile.am:
32926 * gst/vaapi/gstvaapi.c:
32927 * gst/vaapi/gstvaapiconvert.c:
32928 * gst/vaapi/gstvaapiconvert.h:
32929 * gst/vaapi/gstvaapidecode.c:
32930 * gst/vaapi/gstvaapidecode.h:
32931 * gst/vaapi/gstvaapisink.c:
32932 * gst/vaapi/gstvaapisink.h:
32933 * gst/vaapiconvert/Makefile.am:
32934 * gst/vaapidecode/Makefile.am:
32935 Group all plugins into the same bundle
32936 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32938 2011-07-21 14:31:30 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32941 Bump version for development.
32943 2011-12-07 14:17:32 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32946 * debian.upstream/control.in:
32947 debian: build against upstream libva packages.
32949 2011-12-07 13:52:17 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32951 * docs/reference/libs/Makefile.am:
32952 * docs/reference/plugins/Makefile.am:
32953 * tests/Makefile.am:
32954 Fix build on Ubuntu 11.10 (Oneric).
32956 2011-12-07 13:14:28 +0100 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32961 2011-09-12 16:20:16 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
32965 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32967 2011-10-24 16:18:16 -0400 Nicolas Dufresne <nicolas.dufresne@collabora.com>
32969 * gst/vaapiconvert/gstvaapiconvert.c:
32970 * gst/vaapidecode/gstvaapidecode.c:
32971 * gst/vaapisink/gstvaapisink.c:
32972 Update license in plugin definition
32973 Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32975 2011-10-12 14:00:50 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32978 * gst-libs/gst/vaapi/gstvaapidecoder.c:
32979 decoder: fix use of invalid data at the end-of-stream.
32981 2011-10-19 14:47:31 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32983 * gst/vaapiconvert/gstvaapiconvert.c:
32984 vaapiconvert: fix some warnings.
32986 2011-10-19 14:43:56 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
32989 * debian.upstream/changelog.in:
32990 * debian.upstream/control.in:
32991 * debian.upstream/copyright:
32992 * gst/vaapiconvert/gstvaapiconvert.c:
32993 * gst/vaapidecode/gstvaapidecode.c:
32994 * gst/vaapisink/gstvaapisink.c:
32995 Update with my current e-mail address.
32997 2011-10-19 14:39:21 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33001 Splitted-Desktop systems relicensed plugins and tests to LGPL v2.1+.
33003 2011-10-18 09:18:20 +0200 warly <warly@warly.org>
33007 * tests/test-decode.c:
33008 * tests/test-decode.h:
33009 * tests/test-display.c:
33010 * tests/test-h264.c:
33011 * tests/test-h264.h:
33012 * tests/test-mpeg2.c:
33013 * tests/test-mpeg2.h:
33014 * tests/test-surfaces.c:
33015 * tests/test-textures.c:
33016 * tests/test-vc1.c:
33017 * tests/test-vc1.h:
33018 * tests/test-windows.c:
33019 switch tests licence to LGPL v2.1+
33021 2011-10-18 09:06:52 +0200 warly <warly@warly.org>
33025 * debian.upstream/copyright:
33026 * gst/vaapiconvert/gstvaapiconvert.c:
33027 * gst/vaapiconvert/gstvaapiconvert.h:
33028 * gst/vaapidecode/gstvaapidecode.c:
33029 * gst/vaapidecode/gstvaapidecode.h:
33030 * gst/vaapisink/gstvaapisink.c:
33031 * gst/vaapisink/gstvaapisink.h:
33032 move plugins to LGPL v2.1+
33034 2011-09-14 13:07:18 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33036 * gst/vaapidecode/gstvaapidecode.c:
33037 vaapidecode: fix sink caps to not expose size information.
33038 This fixes this particular issue:
33039 GStreamer-WARNING **: pad vaapidecode0:sink returned caps which are not
33040 a real subset of its template caps
33042 2011-09-14 11:34:05 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33045 * gst/vaapidecode/gstvaapidecode.c:
33046 vaapidecode: fix decoding of MPEG-2 PS files.
33048 2011-09-12 13:00:04 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33050 * gst-libs/gst/vaapi/Makefile.am:
33051 Cosmetics (sort source files).
33053 2011-09-08 14:50:24 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33055 * gst/vaapiconvert/gstvaapiconvert.c:
33056 * gst/vaapiconvert/gstvaapiconvert.h:
33057 vaapiconvert: fix direct-rendering caps detection.
33059 2011-09-08 14:40:08 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33061 * gst-libs/gst/vaapi/gstvaapiimage.c:
33062 Fix gst_vaapi_image_new_with_image().
33064 2011-09-08 13:09:17 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33066 * gst/vaapiconvert/gstvaapiconvert.c:
33067 vaapiconvert: warn when surface failed to be updated with image.
33069 2011-09-06 18:34:33 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33071 * gst/vaapiconvert/gstvaapiconvert.c:
33072 vaapiconvert: fix autodetection for vaDeriveImage() support.
33074 2011-09-06 17:47:10 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33076 * gst/vaapiconvert/gstvaapiconvert.c:
33077 vaapiconvert: fix memory leak (VA surface image).
33079 2011-09-05 16:20:20 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33082 * gst/vaapiconvert/gstvaapiconvert.c:
33083 vaapiconvert: fix direct-rendering mode.
33085 2011-09-06 16:49:43 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33087 * docs/reference/libs/libs-sections.txt:
33088 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
33089 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
33090 Add gst_vaapi_video_buffer_new_from_buffer().
33091 Add helper function to bind a foreign buffer into a GstVaapiVideoBuffer.
33092 Any image, surface or surface proxy will be inherited from the source buffer
33093 if it is a GstVaapiVideoBuffer.
33095 2011-09-05 17:23:05 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33097 * gst/vaapiconvert/gstvaapiconvert.c:
33098 vaapiconvert: protect access to direct_rendering.
33100 2011-09-05 16:18:14 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33102 * gst/vaapiconvert/gstvaapiconvert.c:
33103 vaapiconvert: use gst_vaapi_display_lookup_downstream() helper to get a VA display.
33105 2011-08-01 14:15:39 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33109 * debian.upstream/copyright:
33110 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33111 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33112 * gst-libs/gst/vaapi/gstvaapisurface.h:
33113 * gst/vaapidecode/gstvaapidecode.c:
33114 * gst/vaapidecode/gstvaapidecode.h:
33115 Add Intel copyright information.
33117 2011-07-22 15:59:00 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33122 2011-07-22 15:55:47 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33124 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33125 Fix build with newer FFmpeg versions.
33127 2011-07-22 15:39:51 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33129 * gst/vaapidecode/gstvaapidecode.c:
33130 * gst/vaapidecode/gstvaapidecode.h:
33131 Fix decoding of MPEG-2 TS files.
33133 2011-07-22 15:34:48 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33135 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33136 Report caps update only once per video resolution change.
33138 2011-07-22 15:33:13 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33140 * gst-libs/gst/vaapi/gstvaapisurface.h:
33141 Add canonical form (type name) of VA surface caps.
33143 2011-07-22 15:42:16 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33146 Bump version for development.
33148 2011-07-19 17:38:40 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33151 Use pretty build output with automake >= 1.11.
33153 2011-07-15 16:08:08 +0200 Gwenole Beauchesne <gwenole.beauchesne@intel.com>
33155 * gst/vaapiconvert/Makefile.am:
33156 * gst/vaapidecode/Makefile.am:
33157 * gst/vaapisink/Makefile.am:
33158 * tests/Makefile.am:
33159 Fix build with libva headers not in a standard include dir.
33161 2011-06-14 15:59:08 +0200 Gwenole Beauchesne <gbeauchesne@splitted-desktop.com>
33166 2011-06-14 13:52:56 +0200 Gwenole Beauchesne <gbeauchesne@splitted-desktop.com>
33168 * gst-libs/gst/vaapi/gstvaapi_priv.h:
33169 * gst-libs/gst/vaapi/gstvaapicompat.h:
33170 * gst-libs/gst/vaapi/gstvaapicontext.c:
33171 * gst-libs/gst/vaapi/gstvaapicontext.h:
33172 * gst-libs/gst/vaapi/gstvaapidebug.h:
33173 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33174 * gst-libs/gst/vaapi/gstvaapidecoder.h:
33175 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33176 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.h:
33177 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
33178 * gst-libs/gst/vaapi/gstvaapidisplay.c:
33179 * gst-libs/gst/vaapi/gstvaapidisplay.h:
33180 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
33181 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
33182 * gst-libs/gst/vaapi/gstvaapidisplay_glx_priv.h:
33183 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
33184 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
33185 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
33186 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
33187 * gst-libs/gst/vaapi/gstvaapiimage.c:
33188 * gst-libs/gst/vaapi/gstvaapiimage.h:
33189 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
33190 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
33191 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
33192 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
33193 * gst-libs/gst/vaapi/gstvaapiobject.c:
33194 * gst-libs/gst/vaapi/gstvaapiobject.h:
33195 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
33196 * gst-libs/gst/vaapi/gstvaapiparamspecs.c:
33197 * gst-libs/gst/vaapi/gstvaapiparamspecs.h:
33198 * gst-libs/gst/vaapi/gstvaapiprofile.c:
33199 * gst-libs/gst/vaapi/gstvaapiprofile.h:
33200 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
33201 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
33202 * gst-libs/gst/vaapi/gstvaapisurface.c:
33203 * gst-libs/gst/vaapi/gstvaapisurface.h:
33204 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
33205 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
33206 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
33207 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
33208 * gst-libs/gst/vaapi/gstvaapitexture.c:
33209 * gst-libs/gst/vaapi/gstvaapitexture.h:
33210 * gst-libs/gst/vaapi/gstvaapitypes.h:
33211 * gst-libs/gst/vaapi/gstvaapiutils.c:
33212 * gst-libs/gst/vaapi/gstvaapiutils.h:
33213 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
33214 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
33215 * gst-libs/gst/vaapi/gstvaapiutils_gst.c:
33216 * gst-libs/gst/vaapi/gstvaapiutils_gst.h:
33217 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
33218 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
33219 * gst-libs/gst/vaapi/gstvaapivalue.c:
33220 * gst-libs/gst/vaapi/gstvaapivalue.h:
33221 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
33222 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
33223 * gst-libs/gst/vaapi/gstvaapivideopool.c:
33224 * gst-libs/gst/vaapi/gstvaapivideopool.h:
33225 * gst-libs/gst/vaapi/gstvaapivideosink.c:
33226 * gst-libs/gst/vaapi/gstvaapivideosink.h:
33227 * gst-libs/gst/vaapi/gstvaapiwindow.c:
33228 * gst-libs/gst/vaapi/gstvaapiwindow.h:
33229 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
33230 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
33231 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
33232 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
33233 * gst/vaapiconvert/gstvaapiconvert.c:
33234 * gst/vaapiconvert/gstvaapiconvert.h:
33235 * gst/vaapidecode/gstvaapidecode.c:
33236 * gst/vaapidecode/gstvaapidecode.h:
33237 * gst/vaapisink/gstvaapisink.c:
33238 * gst/vaapisink/gstvaapisink.h:
33241 * tests/test-decode.c:
33242 * tests/test-decode.h:
33243 * tests/test-display.c:
33244 * tests/test-h264.c:
33245 * tests/test-h264.h:
33246 * tests/test-mpeg2.c:
33247 * tests/test-mpeg2.h:
33248 * tests/test-surfaces.c:
33249 * tests/test-textures.c:
33250 * tests/test-vc1.c:
33251 * tests/test-vc1.h:
33252 * tests/test-windows.c:
33253 Update copyright notice.
33255 2011-06-14 13:51:41 +0200 Gwenole Beauchesne <gbeauchesne@splitted-desktop.com>
33258 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
33259 * gst-libs/gst/vaapi/gstvaapisurface.c:
33260 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
33261 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
33262 * gst-libs/gst/vaapi/gstvaapitexture.c:
33263 * gst-libs/gst/vaapi/gstvaapiutils.c:
33264 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
33265 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
33266 * gst-libs/gst/vaapi/gstvaapivalue.c:
33267 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
33268 * gst-libs/gst/vaapi/gstvaapivideopool.c:
33269 * gst-libs/gst/vaapi/gstvaapivideosink.c:
33270 * gst-libs/gst/vaapi/gstvaapiwindow.c:
33271 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
33272 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
33273 Fix licensing terms.
33275 2010-07-20 11:23:16 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33280 2010-07-20 11:21:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33282 * debian.upstream/copyright:
33283 Fix license terms...
33285 2010-07-01 13:19:29 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33287 * gst/vaapisink/gstvaapisink.c:
33288 Render pretty background only in use-reflection=true mode.
33290 2010-07-01 11:43:22 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33295 2010-07-01 11:41:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33297 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
33298 Drop the GLX 1.3 requirement.
33300 2010-07-01 11:38:28 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33302 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
33303 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
33304 Call the GLX/Pixmap related functions through the vtable.
33306 2010-07-01 11:11:18 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33310 * gst-libs/gst/vaapi/Makefile.am:
33311 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33312 Drop dependency on libavformat.
33314 2010-06-22 15:15:46 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33316 * gst-libs/gst/video/gstbasevideodecoder.c:
33317 * gst-libs/gst/video/gstbasevideodecoder.h:
33318 Add gst_base_video_decoder_update_src_caps(). Don't forcibly set "interlaced" field if upstream elements did not have any.
33320 2010-06-22 14:06:25 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33322 * gst-libs/gst/video/gstbasevideodecoder.c:
33323 * gst-libs/gst/video/gstbasevideoutils.c:
33324 * gst-libs/gst/video/gstbasevideoutils.h:
33325 Drop superfluous functions.
33327 2010-06-22 13:57:33 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33329 * gst-libs/gst/video/gstbasevideodecoder.c:
33330 Really drop any dependency on libgstvideo. i.e. inline the helpers.
33332 2010-06-22 13:48:30 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33334 * gst-libs/gst/video/gstbasevideodecoder.c:
33335 Further drop dependency on libgstvideo.
33337 2010-06-22 12:57:06 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33339 * gst-libs/gst/video/Makefile.am:
33340 * gst-libs/gst/video/gstbasevideocodec.c:
33341 * gst-libs/gst/video/gstbasevideocodec.h:
33342 * gst-libs/gst/video/gstbasevideodecoder.c:
33343 * gst-libs/gst/video/gstbasevideodecoder.h:
33344 * gst-libs/gst/video/gstbasevideoutils.c:
33345 * gst-libs/gst/video/gstbasevideoutils.h:
33346 Add GstBaseVideoDecoder from gst-plugins-bad git.
33348 2010-06-15 12:36:16 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33350 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
33351 Fix GLX version check.
33353 2010-06-14 14:46:41 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33357 Bump version for development.
33359 2010-06-14 14:14:42 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33362 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33363 Fix decoding of some H.264 streams. e.g. Ice Age 2 trailer.
33365 2010-06-14 12:58:22 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33370 2010-06-14 09:20:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33372 * gst-libs/gst/vaapi/gstvaapicompat.h:
33373 Fix build with older VA-API 0.29-sds.
33375 2010-05-18 11:22:54 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33377 * gst/vaapisink/gstvaapisink.c:
33378 * gst/vaapisink/gstvaapisink.h:
33379 Fix upscaling in foreign window (Totem).
33381 2010-05-17 12:32:34 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33383 * gst/vaapisink/gstvaapisink.c:
33384 Soft validate caps since we only care about video/x-vaapi-surface as input. _setcaps() will check for other fields.
33386 2010-05-17 08:55:51 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33389 * gst/vaapisink/gstvaapisink.c:
33390 Fix video rendering rect within an embedder window (Totem).
33392 2010-05-17 08:28:28 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33395 * gst/vaapisink/gstvaapisink.c:
33396 Disable GLX rendering when vaapisink uses a foreign X window.
33398 2010-05-17 08:24:42 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33400 * gst/vaapisink/gstvaapisink.c:
33401 Simplify GLX rendering code.
33403 2010-05-17 07:32:10 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33406 Bump version for development.
33408 2010-05-16 21:44:17 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33413 2010-05-16 21:35:14 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33415 * gst/vaapidecode/gstvaapidecode.c:
33416 Wait for at most one second for a VA surface to become available.
33418 2010-05-16 21:18:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33422 Build-Requires: gstreamer0.10 >= 0.10.10 for gst_caps_merge().
33424 2010-05-16 21:17:49 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33428 * gst-libs/gst/vaapi/gstvaapiprofile.c:
33429 Fix decoder caps to report codec aliases.
33431 2010-05-16 21:04:32 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33434 * gst/vaapidecode/gstvaapidecode.c:
33435 * gst/vaapidecode/gstvaapidecode.h:
33436 Fix VC-1 decoding through the playbin2 pipeline.
33438 2010-05-15 15:33:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33440 * docs/reference/libs/libs-sections.txt:
33441 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33442 * gst-libs/gst/vaapi/gstvaapidecoder.h:
33443 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33444 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
33445 Regularly update and expose decoder caps.
33447 2010-05-15 09:43:28 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33449 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33450 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33451 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
33452 Add mechanism to reinsert buffer leftovers into the queue.
33454 2010-05-15 06:59:54 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33457 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33458 Fix memory leak of encoded buffers.
33460 2010-05-15 05:36:15 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33462 * docs/reference/libs/libs-sections.txt:
33463 * gst-libs/gst/vaapi/gstvaapicontext.c:
33464 * gst-libs/gst/vaapi/gstvaapicontext.h:
33465 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33466 * gst-libs/gst/vaapi/gstvaapidecoder.h:
33467 Check for out-of-free-surfaces condition.
33469 2010-05-15 04:35:00 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33471 * gst-libs/gst/vaapi/gstvaapidecoder.h:
33472 Change GST_VAAPI_DECODER_STATUS_ERROR_UNKNOWN value to something more generic (-1).
33474 2010-05-15 04:25:32 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33476 * gst/vaapisink/gstvaapisink.c:
33477 Improve debug info for gst_vaapisink_ensure_render_rect().
33479 2010-05-14 05:02:05 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33482 Bump version for development.
33484 2010-05-13 21:52:22 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33489 2010-05-13 21:39:58 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33491 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33492 Improve previous fix.
33494 2010-05-13 21:27:43 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33497 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33498 Fix a crash in the FFmpeg decoder on close.
33500 2010-05-13 16:41:55 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33503 Sort platforms by name.
33505 2010-05-13 09:40:52 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33509 Bump version for development.
33511 2010-05-13 09:38:47 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33513 * gst/vaapisink/gstvaapisink.c:
33514 Add debug info for _show_frame().
33516 2010-05-13 07:19:46 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33519 Nuke older build dir.
33521 2010-05-13 07:19:21 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33523 * debian.upstream/control.in:
33524 Fix packaging deps.
33526 2010-05-13 06:12:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33531 2010-05-13 06:11:42 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33534 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
33535 Fix OpenGL texture internal format (Clutter).
33537 2010-05-13 04:40:40 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33542 2010-05-13 04:27:44 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33544 * gst/vaapisink/gstvaapisink.c:
33545 Debug video & display PARs.
33547 2010-05-13 04:22:31 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33549 * gst/vaapisink/gstvaapisink.c:
33550 Use XGetGeometry() to retrieve the window size.
33552 2010-05-12 19:40:30 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33554 * gst/vaapisink/gstvaapisink.c:
33557 2010-05-12 19:35:45 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33559 * gst/vaapisink/gstvaapisink.c:
33560 Ensure VA display is created prior to initializing the window from a specific XID. Also move code down.
33562 2010-05-12 19:18:04 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33565 Drop obsolete comment.
33567 2010-05-12 19:14:59 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33572 2010-05-12 19:14:35 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33574 * gst/vaapidecode/gstvaapidecode.c:
33575 Fix GstVaapiDisplay refcounting in vaapidecode.
33577 2010-05-12 14:10:38 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33579 * gst-libs/gst/vaapi/gstvaapiutils_gst.c:
33582 2010-05-12 12:58:53 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33584 * gst/vaapisink/gstvaapisink.c:
33585 * gst/vaapisink/gstvaapisink.h:
33586 Fix render rect when the foreign window size changes.
33588 2010-05-12 11:43:50 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33592 * gst/vaapisink/Makefile.am:
33593 * gst/vaapisink/gstvaapisink.c:
33594 Add GstXOverlay interface to vaapisink (e.g. for Totem).
33596 2010-05-12 10:51:21 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33599 Update deps to match configure.ac versions.
33601 2010-05-12 09:34:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33603 * gst/vaapiconvert/gstvaapiconvert.c:
33604 * gst/vaapidecode/gstvaapidecode.c:
33605 * gst/vaapisink/gstvaapisink.c:
33608 2010-05-12 09:22:49 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33611 * gst/vaapidecode/Makefile.am:
33612 * gst/vaapidecode/gstvaapidecode.c:
33613 * gst/vaapidecode/gstvaapidecode.h:
33614 Fix vaapidecode to expose the HW supported caps only.
33616 2010-05-12 08:32:34 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33618 * gst/vaapidecode/gstvaapidecode.c:
33619 Initialize decoder earlier.
33621 2010-05-12 08:02:45 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33625 * gst/vaapidecode/gstvaapidecode.c:
33626 Fix integration within the playbin2 pipeline.
33628 2010-05-12 08:02:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33630 * docs/reference/libs/Makefile.am:
33631 Exclude gstvaapiutils_gst.h from docs for now.
33633 2010-05-12 08:00:09 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33635 * gst/vaapiconvert/gstvaapiconvert.c:
33636 * gst/vaapidecode/gstvaapidecode.c:
33637 * gst/vaapisink/gstvaapisink.c:
33638 Raise VA-API plugins ranks.
33640 2010-05-12 07:57:55 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33642 * gst-libs/gst/vaapi/Makefile.am:
33643 * gst-libs/gst/vaapi/gstvaapiutils_gst.c:
33644 * gst-libs/gst/vaapi/gstvaapiutils_gst.h:
33645 Add gst_vaapi_display_lookup_downstream() helper.
33647 2010-05-11 16:23:17 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33649 * gst/vaapidecode/gstvaapidecode.c:
33650 Use fixed caps on the src pad, they are not meant to change from video/x-vaapi-surface.
33652 2010-05-11 16:19:30 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33654 * docs/reference/libs/libs-sections.txt:
33655 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
33656 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
33657 * gst/vaapisink/gstvaapisink.c:
33658 Expose VA display through GstVaapiVideoBuffer.
33660 2010-05-11 16:09:49 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33662 * gst/vaapidecode/gstvaapidecode.c:
33663 Simplify gst_vaapidecode_set_caps() and fix memory leak.
33665 2010-05-11 12:06:59 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33667 * docs/reference/libs/libs-sections.txt:
33668 * gst-libs/gst/vaapi/gstvaapivideopool.c:
33669 * gst-libs/gst/vaapi/gstvaapivideopool.h:
33670 Expose video pool display.
33672 2010-05-11 12:03:13 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33674 * gst-libs/gst/vaapi/gstvaapivideosink.c:
33675 Stop iteration if there is no more element to examine.
33677 2010-05-10 09:32:47 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33679 * gst/vaapiconvert/gstvaapiconvert.c:
33680 * gst/vaapidecode/gstvaapidecode.c:
33681 * gst/vaapisink/gstvaapisink.c:
33682 Improve plugin details.
33684 2010-05-07 06:35:31 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33686 * gst-libs/gst/vaapi/gstvaapisurface.h:
33687 * gst/vaapiconvert/gstvaapiconvert.c:
33688 * gst/vaapidecode/gstvaapidecode.c:
33689 * gst/vaapisink/gstvaapisink.c:
33690 Factor out VA surface caps.
33692 2010-05-05 15:36:25 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33694 * gst/vaapidecode/gstvaapidecode.c:
33695 Add gst_vaapidecode_ensure_display() helper for set-caps.
33697 2010-05-05 12:57:59 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33700 Bump version for development.
33702 2010-05-05 12:29:28 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33705 Really make it 0.2.0.
33707 2010-05-05 12:28:59 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33712 2010-05-05 11:48:31 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33714 * docs/reference/libs/Makefile.am:
33715 Don't exclude GstVaapiParamSpecs.
33717 2010-05-05 11:44:06 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33719 * docs/reference/libs/Makefile.am:
33720 * docs/reference/libs/libs-sections.txt:
33721 * gst-libs/gst/vaapi/gstvaapidecoder.h:
33722 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
33725 2010-05-05 06:06:02 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33727 * gst/vaapiconvert/gstvaapiconvert.c:
33728 * gst/vaapidecode/gstvaapidecode.c:
33729 * gst/vaapisink/gstvaapisink.c:
33730 Lower plugins rank for now since playbin2 auto-plugging is not working properly. User applications will have to create their own pipeline or with some hacks around playbin2.
33732 2010-05-04 15:03:47 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33734 * gst-libs/gst/vaapi/Makefile.am:
33735 Really link all helper libraries with libtool -no-undefined.
33737 2010-05-04 15:02:29 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33739 * gst-libs/gst/vaapi/Makefile.am:
33740 Link helper libraries with libtool -no-undefined.
33742 2010-05-04 14:59:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33745 * debian.upstream/gstreamer-vaapi.install.in:
33746 * gst/vaapiconvert/Makefile.am:
33747 * gst/vaapidecode/Makefile.am:
33748 * gst/vaapisink/Makefile.am:
33749 Don't build plugins with SONAME. Make them plain *.so.
33751 2010-05-04 08:59:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33754 Improve documentation for release.
33756 2010-05-03 22:50:56 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33758 * gst-libs/gst/vaapi/gstvaapiprofile.c:
33759 Fix build with older VA-API 0.29.
33761 2010-05-03 22:43:01 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33766 2010-05-03 22:42:46 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33768 * tests/Makefile.am:
33771 2010-05-03 22:36:34 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33773 * docs/reference/libs/Makefile.am:
33774 * docs/reference/libs/libs-docs.xml.in:
33775 * docs/reference/libs/libs-sections.txt:
33776 * docs/reference/libs/libs.core.types:
33777 * docs/reference/plugins/Makefile.am:
33778 * docs/reference/plugins/plugins-docs.xml.in:
33779 * docs/reference/plugins/plugins-sections.txt:
33780 * docs/reference/plugins/plugins.types:
33783 2010-05-03 22:34:53 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33785 * gst-libs/gst/vaapi/gstvaapiprofile.c:
33788 2010-05-03 22:28:02 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33790 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33791 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.h:
33792 * gst/vaapidecode/gstvaapidecode.c:
33793 * tests/test-decode.c:
33794 Rename gst_vaapi_decoder_ffmpeg_new_from_caps() to plain gst_vaapi_decoder_ffmpeg_new().
33796 2010-05-03 22:02:41 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33798 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33799 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
33800 Extract framerate information from caps.
33802 2010-05-03 21:49:35 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33804 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33805 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33806 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
33807 Move caps initialization to parent class.
33809 2010-05-03 21:25:46 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33811 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33814 2010-05-03 21:25:26 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33816 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33819 2010-05-03 21:14:01 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33821 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33822 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33823 More simplifications.
33825 2010-05-03 20:55:17 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33827 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33828 * gst-libs/gst/vaapi/gstvaapidecoder.h:
33829 * tests/test-decode.c:
33830 Simplify GstVaapiDecoder API.
33832 2010-05-03 20:40:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33834 * gst-libs/gst/vaapi/gstvaapidecoder.h:
33835 Drop obsolete defs.
33837 2010-05-03 20:34:57 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33839 * gst-libs/gst/vaapi/gstvaapidecoder.h:
33840 Drop obsolete decls.
33842 2010-05-03 17:36:01 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33844 * gst-libs/gst/vaapi/gstvaapiprofile.c:
33845 * gst/vaapidecode/gstvaapidecode.c:
33846 Add more aliases for MPEG-4 decoding.
33848 2010-05-03 17:04:00 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33850 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33851 Use avctx->coded_{width,height} info to create the VA context.
33853 2010-05-03 16:54:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33855 * tests/test-decode.c:
33856 * tests/test-decode.h:
33857 Use gst_vaapi_decoder_ffmpeg_new_from_caps().
33859 2010-05-03 16:41:13 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33861 * tests/test-decode.c:
33862 * tests/test-decode.h:
33863 * tests/test-h264.c:
33864 * tests/test-h264.h:
33865 * tests/test-mpeg2.c:
33866 * tests/test-mpeg2.h:
33867 * tests/test-vc1.c:
33868 * tests/test-vc1.h:
33869 Simplify tests info.
33871 2010-05-03 16:17:51 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33873 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33874 Try to improve heuristics to use an AVCodecContextParser.
33876 2010-05-03 15:35:22 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33878 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33879 Fix VC-1 decoding, it does not require any specific parser.
33881 2010-05-03 15:34:22 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33883 * gst-libs/gst/vaapi/gstvaapiprofile.c:
33884 Fix VC-1 detection with older gstreamer libs (no "fourcc" field, but a "format" one).
33886 2010-05-03 15:29:18 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33888 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33889 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
33890 Use size information from the demuxer, whenever available. i.e. fix WMV3 decoding.
33892 2010-05-03 15:11:32 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33894 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33895 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33896 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.h:
33897 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
33898 * gst/vaapidecode/gstvaapidecode.c:
33899 * gst/vaapidecode/gstvaapidecode.h:
33900 Add gst_vaapi_decoder_ffmpeg_new_from_caps() helper.
33902 2010-05-03 14:53:18 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33904 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33905 * gst-libs/gst/vaapi/gstvaapiprofile.c:
33906 * gst-libs/gst/vaapi/gstvaapiprofile.h:
33907 Improve WMV3 detection yet further.
33909 2010-05-03 13:44:41 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33911 * gst-libs/gst/vaapi/gstvaapiprofile.c:
33912 Fix detection of plain old WMV3 contents.
33914 2010-05-03 12:25:07 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33916 * tests/test-vc1.c:
33917 Add End-of-Sequence start code.
33919 2010-05-03 11:44:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33921 * gst-libs/gst/vaapi/gstvaapiprofile.c:
33922 * gst/vaapidecode/gstvaapidecode.c:
33923 Fix VC-1 detection.
33925 2010-05-03 08:51:28 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33927 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33928 Fix build with older gstreamer libs where gst_buffer_unref() is not a plain function.
33930 2010-05-03 08:34:57 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33932 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33933 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
33934 Drop obsolete (and wrong) code.
33936 2010-05-03 08:33:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33938 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33939 Cosmetics (spelling).
33941 2010-05-03 08:32:46 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33943 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33944 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33945 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
33946 Try to fix timestamps (step 1). Looks OK on H55.
33948 2010-05-03 07:10:04 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33950 * debian.upstream/Makefile.am:
33951 Ship with COPYING.LIB.
33953 2010-05-03 07:07:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
33958 * gst-libs/gst/vaapi/gstvaapi_priv.h:
33959 * gst-libs/gst/vaapi/gstvaapicompat.h:
33960 * gst-libs/gst/vaapi/gstvaapicontext.c:
33961 * gst-libs/gst/vaapi/gstvaapicontext.h:
33962 * gst-libs/gst/vaapi/gstvaapidebug.h:
33963 * gst-libs/gst/vaapi/gstvaapidecoder.c:
33964 * gst-libs/gst/vaapi/gstvaapidecoder.h:
33965 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
33966 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.h:
33967 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
33968 * gst-libs/gst/vaapi/gstvaapidisplay.c:
33969 * gst-libs/gst/vaapi/gstvaapidisplay.h:
33970 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
33971 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
33972 * gst-libs/gst/vaapi/gstvaapidisplay_glx_priv.h:
33973 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
33974 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
33975 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
33976 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
33977 * gst-libs/gst/vaapi/gstvaapiimage.c:
33978 * gst-libs/gst/vaapi/gstvaapiimage.h:
33979 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
33980 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
33981 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
33982 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
33983 * gst-libs/gst/vaapi/gstvaapiobject.c:
33984 * gst-libs/gst/vaapi/gstvaapiobject.h:
33985 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
33986 * gst-libs/gst/vaapi/gstvaapiparamspecs.c:
33987 * gst-libs/gst/vaapi/gstvaapiparamspecs.h:
33988 * gst-libs/gst/vaapi/gstvaapiprofile.c:
33989 * gst-libs/gst/vaapi/gstvaapiprofile.h:
33990 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
33991 * gst-libs/gst/vaapi/gstvaapisurface.h:
33992 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
33993 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
33994 * gst-libs/gst/vaapi/gstvaapitexture.h:
33995 * gst-libs/gst/vaapi/gstvaapitypes.h:
33996 * gst-libs/gst/vaapi/gstvaapiutils.h:
33997 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
33998 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
33999 * gst-libs/gst/vaapi/gstvaapivalue.h:
34000 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
34001 * gst-libs/gst/vaapi/gstvaapivideopool.h:
34002 * gst-libs/gst/vaapi/gstvaapivideosink.h:
34003 * gst-libs/gst/vaapi/gstvaapiwindow.h:
34004 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
34005 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
34006 Relicense gst-libs/ code to LGPL v2.1+.
34008 2010-05-03 06:49:43 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34010 * gst-libs/gst/vaapi/gstvaapiprofile.c:
34011 Drop extraneous comma.
34013 2010-05-03 06:49:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34015 * gst-libs/gst/vaapi/gstvaapiprofile.c:
34016 Drop variant=itu field to help codec detection.
34018 2010-04-30 15:50:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34020 * gst/vaapidecode/gstvaapidecode.c:
34021 Decode as many surfaces as possible in gst_vaapidecode_step().
34023 2010-04-30 15:37:28 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34025 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34026 * gst-libs/gst/vaapi/gstvaapidecoder.h:
34027 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
34028 * gst/vaapidecode/gstvaapidecode.c:
34029 * gst/vaapidecode/gstvaapidecode.h:
34030 * tests/test-decode.c:
34031 Drop excessive threading that over-complicates synchronisation. MPEG-2 & H.264 videos now play but there are other problems (timestamps).
34033 2010-04-30 13:13:50 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34038 2010-04-30 12:04:12 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34040 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34041 Move VA context reset to AVCodecContext.get_context() as the surface sizes can change.
34043 2010-04-30 09:52:29 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34045 * gst-libs/gst/vaapi/gstvaapidisplay.c:
34046 * gst-libs/gst/vaapi/gstvaapidisplay.h:
34047 Fix gst_vaapi_display_has_{decoder,encoder}() to check for the entrypoint too.
34049 2010-04-30 09:48:41 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34051 * gst-libs/gst/vaapi/gstvaapiprofile.c:
34052 * gst-libs/gst/vaapi/gstvaapiprofile.h:
34053 Add GST_VAAPI_ENTRYPOINT_SLICE_ENCODE.
34055 2010-04-30 08:18:07 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34057 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34058 Document H.264 / AVC1 format case better.
34060 2010-04-29 23:09:07 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34062 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34063 Fix H.264 decoding with AVC1 format bitstreams.
34065 2010-04-29 22:00:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34067 * gst/vaapidecode/gstvaapidecode.c:
34068 Complete initialization of the GstVaapiVideoBuffer. Some frames start to show up.
34070 2010-04-29 21:59:14 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34072 * gst/vaapisink/gstvaapisink.c:
34073 Add missing GstBaseSink::buffer_alloc() override. i.e. make sure to allocate a GstVaapiVideoBuffer instead of a plain GstBuffer from the peer pad.
34075 2010-04-29 21:56:10 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34077 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
34078 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
34079 Export gst_vaapi_video_buffer_new().
34081 2010-04-29 21:12:30 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34083 * gst-libs/gst/vaapi/gstvaapiprofile.c:
34084 Fix gst_vaapi_profile_get_caps() to include the "profile" field.
34086 2010-04-29 17:56:42 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34091 2010-04-29 17:55:58 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34096 * gst/vaapidecode/Makefile.am:
34097 * gst/vaapidecode/gstvaapidecode.c:
34098 * gst/vaapidecode/gstvaapidecode.h:
34099 Add FFmpeg/VAAPI decoder for the new `vaapidecode' element.
34101 2010-04-29 17:51:57 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34103 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34104 * gst-libs/gst/vaapi/gstvaapidecoder.h:
34105 Add gst_vaapi_decoder_pause().
34107 2010-04-29 17:11:32 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34109 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34110 * gst-libs/gst/vaapi/gstvaapidecoder.h:
34111 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
34112 Use a GstTask with start/stop semantics for the decoder thread.
34114 2010-04-29 16:08:46 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34116 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
34117 Drop extraneous var.
34119 2010-04-29 15:45:44 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34121 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
34122 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
34123 Add support for GstVaapiSurfaceProxy to GstVaapiVideoBuffer.
34125 2010-04-29 14:58:45 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34127 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34128 Fix gst_vaapi_decoder_get_surface() status.
34130 2010-04-29 14:28:43 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34132 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34133 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
34134 Try to set correct timestamps to the decoded surface proxy.
34136 2010-04-29 12:52:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34138 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
34139 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
34140 * tests/test-decode.c:
34141 Add timestamps to GstVaapiSurfaceProxy.
34143 2010-04-29 09:43:40 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34145 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34146 Fix GstVaapiDecoder::destroy(): GASyncQueue is not a GObject, likewise for GstBuffer.
34148 2010-04-29 09:40:38 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34150 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34151 Fix destructor, av_parser_close() does destroy the struct already, unliker avcodec_close()...
34153 2010-04-29 09:35:54 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34155 * gst-libs/gst/vaapi/gstvaapiprofile.c:
34156 Prefer profile from codec-data if any was found there.
34158 2010-04-29 09:34:54 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34160 * gst-libs/gst/vaapi/gstvaapiprofile.c:
34161 Improve heuristics to find the best profile. Use the highest one if no explicit match on "profile" field.
34163 2010-04-28 23:09:52 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34165 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34166 Make sure gst_vaapi_decoder_get_surface() gets unblocked on error.
34168 2010-04-28 22:30:50 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34170 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34171 Fix VC-1 codec initialization, it really needs an extradata buffer.
34173 2010-04-28 22:16:10 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34175 * gst-libs/gst/vaapi/gstvaapidecoder.h:
34176 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34177 Split decoder creation with actual resources allocation and codec setup (probe). This fixes a memory leak (avctx, pctx) on destroy and most interestingly makes it possible to detect unsupported codecs.
34179 2010-04-28 21:58:58 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34181 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34182 Move gst_vaapi_decoder_ffmpeg_create() call to object constructor.
34184 2010-04-28 21:50:44 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34186 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34187 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34188 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.h:
34189 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
34190 * tests/test-decode.c:
34191 Add "codec-data" property for additional codec data. e.g. VC-1 sequence headers for elementary streams.
34193 2010-04-28 21:20:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34195 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
34196 Cosmetics (weird indentation).
34198 2010-04-28 21:15:55 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34200 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34201 Cosmetics (extraneous variable, debug message).
34203 2010-04-28 09:07:45 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34206 Fix check for VA-API enabled FFmpeg.
34208 2010-04-27 15:26:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34210 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34211 * gst-libs/gst/vaapi/gstvaapidecoder.h:
34212 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34213 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
34214 - Add PTS and framerate information. - Simplify parsing with an AVCodeParserContext.
34216 2010-04-27 11:59:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34218 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34219 * gst-libs/gst/vaapi/gstvaapidecoder.h:
34220 Add more error codes. Fix documentation.
34222 2010-04-26 13:30:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34224 * gst-libs/gst/vaapi/gstvaapiprofile.c:
34225 Fix gst_vaapi_profile_get_codec(). Improve gst_vaapi_profile_from_caps() for H.264 & caps with "codec-data".
34227 2010-04-26 11:44:32 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34229 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34230 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
34231 * tests/test-decode.c:
34232 Handle user end-of-streams. Add gst_vaapi_decoder_{start,stop}() helpers.
34234 2010-04-26 11:36:12 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34236 * tests/test-vc1.c:
34237 Drop useless End-of-Sequence marker.
34239 2010-04-26 08:53:18 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34241 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34242 Flush stream only if avcodec_decode_video() read something. Otherwise, we might still have to seek into the stream. i.e. keep the data longer.
34244 2010-04-26 08:40:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34246 * gst-libs/gst/vaapi/gstvaapidisplay.c:
34247 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
34248 Use a recursive mutex so that a single thread can lock several times. This fixes decoding of MPEG-2 and H.264 because those created a GstVaapiContext later through avcodec_decode_video() that was a protected call.
34250 2010-04-26 08:15:58 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34252 * tests/test-h264.c:
34253 * tests/test-vc1.c:
34254 Regenerate correct clips.
34256 2010-04-23 16:11:55 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34258 * tests/Makefile.am:
34259 * tests/test-decode.c:
34260 * tests/test-h264.c:
34261 * tests/test-h264.h:
34262 * tests/test-mpeg2.c:
34263 * tests/test-mpeg2.h:
34264 * tests/test-vc1.c:
34265 * tests/test-vc1.h:
34266 Add decoder demos. Use -c (mpeg2|h264|vc1) to select the codec.
34267 XXX: only VC-1 decoding works at this time because of awful
34268 bugs left in GstVaapiDecoderFfmpeg et al.
34270 2010-04-23 16:05:58 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34273 * gst-libs/gst/vaapi/Makefile.am:
34274 * gst-libs/gst/vaapi/gstvaapidecoder.c:
34275 * gst-libs/gst/vaapi/gstvaapidecoder.h:
34276 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.c:
34277 * gst-libs/gst/vaapi/gstvaapidecoder_ffmpeg.h:
34278 * gst-libs/gst/vaapi/gstvaapidecoder_priv.h:
34279 Add initial (multithreaded) decoder based on FFmpeg.
34280 There are tons of bugs left:
34281 - Decoder API not nice enough with error conditions
34282 - FFmpeg parser is sometimes broken
34283 - Packets queue can be lost
34285 2010-04-23 16:00:50 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34287 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.c:
34288 * gst-libs/gst/vaapi/gstvaapisurfaceproxy.h:
34289 Add surface proxy that holds a reference to the parent surface and that returns the surface to that context on destruction.
34291 2010-04-23 15:59:31 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34293 * gst-libs/gst/vaapi/gstvaapicontext.c:
34294 * gst-libs/gst/vaapi/gstvaapicontext.h:
34295 Add VA context abstraction.
34297 2010-04-23 10:58:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34299 * gst-libs/gst/vaapi/gstvaapiprofile.h:
34300 Fix VA profiles definitions for gst_vaapi_profile_get_codec() to work.
34302 2010-04-21 15:03:31 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34304 * gst-libs/gst/vaapi/gstvaapivideopool.c:
34305 * gst-libs/gst/vaapi/gstvaapivideopool.h:
34306 Add a means to cap the number of objects allocated in the pool.
34308 2010-04-21 15:02:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34310 * gst-libs/gst/vaapi/gstvaapiprofile.c:
34311 * gst-libs/gst/vaapi/gstvaapiprofile.h:
34312 Add VA entrypoint abstraction.
34314 2010-04-20 13:36:04 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34316 * docs/reference/libs/libs-docs.xml.in:
34317 * docs/reference/libs/libs-sections.txt:
34318 * gst-libs/gst/vaapi/Makefile.am:
34319 * gst-libs/gst/vaapi/gstvaapidisplay.c:
34320 * gst-libs/gst/vaapi/gstvaapidisplay.h:
34321 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
34322 * gst-libs/gst/vaapi/gstvaapiprofile.c:
34323 * gst-libs/gst/vaapi/gstvaapiprofile.h:
34324 * tests/test-display.c:
34325 Add VA profile abstraction.
34327 2010-04-20 07:51:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34329 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
34330 Fix OpenGL rendering on G45 systems.
34332 2010-04-16 13:47:30 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34334 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
34335 Fix gl_create_context() to find a GLXFBConfig compatible with the parent GL context.
34337 2010-04-02 11:27:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34339 * gst-libs/gst/vaapi/gstvaapitexture.c:
34340 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
34341 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
34342 Fix TFP logic and simplify the FBO model. i.e. it's not necessary to create another texture (and storage) for the TFP, simply a new texture name.
34344 2010-04-01 16:11:54 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34346 * gst-libs/gst/vaapi/gstvaapitexture.c:
34347 Fix get-out conditions.
34349 2010-04-01 15:38:59 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34351 * gst-libs/gst/vaapi/gstvaapitexture.c:
34352 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
34353 The shared GL context in GstVaapiTexture is only useful for cases where TFP+FBO are used, thus avoiding the need fully preserve the states and call into glGet*() functions that need synchronization.
34355 2010-04-01 13:55:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34357 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
34360 2010-04-01 13:41:24 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34362 * docs/reference/libs/libs-sections.txt:
34363 * gst-libs/gst/vaapi/Makefile.am:
34364 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
34365 * gst-libs/gst/vaapi/gstvaapidisplay_glx_priv.h:
34366 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
34367 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
34368 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
34369 * gst-libs/gst/vaapi/gstvaapidisplay_x11_priv.h:
34370 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
34371 * gst-libs/gst/vaapi/gstvaapitexture.c:
34372 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
34373 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34374 * tests/test-windows.c:
34375 Make more helpers internal, thus reducing .text size further. Add gst_vaapi_display_x11_get_screen() helper along the way.
34377 2010-04-01 09:47:59 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34379 * gst-libs/gst/vaapi/Makefile.am:
34380 * gst-libs/gst/vaapi/gstvaapi_priv.h:
34381 * gst-libs/gst/vaapi/gstvaapidisplay.c:
34382 * gst-libs/gst/vaapi/gstvaapidisplay.h:
34383 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
34384 * gst-libs/gst/vaapi/gstvaapidisplay_priv.h:
34385 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
34386 * gst-libs/gst/vaapi/gstvaapiimage.c:
34387 * gst-libs/gst/vaapi/gstvaapiobject.c:
34388 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
34389 * gst-libs/gst/vaapi/gstvaapisurface.c:
34390 * gst-libs/gst/vaapi/gstvaapitexture.c:
34391 * gst-libs/gst/vaapi/gstvaapiwindow.c:
34392 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
34393 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34394 Move GST_VAAPI_DISPLAY_VADISPLAY() and GST_VAAPI_DISPLAY_{LOCK,UNLOCK}() to gstvaapidisplay_priv.h.
34396 2010-03-31 15:25:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34398 * gst-libs/gst/vaapi/gstvaapitexture.c:
34399 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
34400 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
34401 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
34402 Improve handling of GL contexts.
34404 2010-03-30 16:41:21 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34409 2010-03-30 13:33:12 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34412 Bump version for development.
34414 2010-03-30 13:29:34 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34417 * debian.upstream/Makefile.am:
34418 * debian.upstream/control.in:
34419 Rename -dev package to libgstvaapi-dev.
34421 2010-03-30 13:17:12 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34427 2010-03-30 13:05:31 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34429 * gst-libs/gst/vaapi/gstvaapiutils.c:
34430 Fix build with VA-API < 0.30.
34432 2010-03-30 13:01:34 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34435 * gst-libs/gst/vaapi/gstvaapicompat.h:
34436 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
34437 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
34438 * gst-libs/gst/vaapi/gstvaapitexture.c:
34439 * tests/test-display.c:
34440 Enable build without VA/GLX extensions. i.e. fallback to TFP + FBO.
34442 2010-03-30 12:59:15 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34444 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
34445 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
34446 Add TFP and FBO helpers.
34448 2010-03-30 12:55:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34450 * gst-libs/gst/vaapi/gstvaapiutils.c:
34451 * gst-libs/gst/vaapi/gstvaapiutils.h:
34452 Cosmetics. Make vaapi_check_status() use GST_DEBUG() for error messages.
34454 2010-03-30 08:13:34 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34456 * gst-libs/gst/vaapi/gstvaapitexture.c:
34457 * gst-libs/gst/vaapi/gstvaapiutils.c:
34458 * gst-libs/gst/vaapi/gstvaapiutils.h:
34459 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34460 Use a shorter function name.
34462 2010-03-30 08:11:50 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34464 * docs/reference/libs/libs-sections.txt:
34465 * gst-libs/gst/vaapi/gstvaapisurface.c:
34466 * gst-libs/gst/vaapi/gstvaapisurface.h:
34467 * gst-libs/gst/vaapi/gstvaapiutils.c:
34468 * gst-libs/gst/vaapi/gstvaapiutils.h:
34469 Add gst_vaapi_surface_query_status() wrapper.
34471 2010-03-30 07:50:11 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34473 * docs/reference/plugins/Makefile.am:
34474 Fix leftover during migration.
34476 2010-03-30 07:46:47 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34480 * docs/reference/plugins/Makefile.am:
34482 * gst/vaapiconvert/Makefile.am:
34483 * gst/vaapiconvert/gstvaapiconvert.c:
34484 * gst/vaapiconvert/gstvaapiconvert.h:
34485 * gst/vaapisink/Makefile.am:
34486 * gst/vaapisink/gstvaapisink.c:
34487 * gst/vaapisink/gstvaapisink.h:
34488 Rename to gst/ as sys/ was too vague.
34490 2010-03-30 07:39:16 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34492 * gst-libs/gst/vaapi/gstvaapiimage.c:
34493 * gst-libs/gst/vaapi/gstvaapiimage.h:
34494 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
34495 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
34496 * gst-libs/gst/vaapi/gstvaapisurface.c:
34497 * gst-libs/gst/vaapi/gstvaapisurface.h:
34498 Try to not reference VA-API types directly.
34500 2010-03-29 16:24:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34502 * sys/vaapisink/gstvaapisink.c:
34503 Fix reflection code to preserve aspect ratio.
34505 2010-03-29 16:17:38 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34507 * sys/vaapisink/gstvaapisink.c:
34508 * sys/vaapisink/gstvaapisink.h:
34509 Fix fullscreen mode.
34511 2010-03-29 15:59:44 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34513 * sys/vaapisink/gstvaapisink.c:
34514 * sys/vaapisink/gstvaapisink.h:
34515 Add OpenGL reflection effect ("use-reflection").
34517 2010-03-29 15:51:54 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34520 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
34521 Use a projection suitable for rotation around the Y axis.
34523 2010-03-29 15:03:30 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34526 * sys/vaapisink/Makefile.am:
34527 * sys/vaapisink/gstvaapisink.c:
34528 * sys/vaapisink/gstvaapisink.h:
34529 Don't build vaapisink/gl by default. However, if this is enabled, use the GL renderer by default.
34531 2010-03-29 14:50:52 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34534 * debian.upstream/Makefile.am:
34535 * debian.upstream/control.in:
34536 * debian.upstream/libgstvaapi-glx.install.in:
34537 Add libgstvaapi-glx-0 package.
34539 2010-03-29 14:47:49 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34541 * pkgconfig/Makefile.am:
34542 Really fix make distclean.
34544 2010-03-29 14:43:22 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34546 * docs/reference/libs/Makefile.am:
34549 2010-03-29 14:42:57 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34551 * pkgconfig/Makefile.am:
34552 Fix make distclean.
34554 2010-03-29 14:40:26 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34556 * tests/Makefile.am:
34559 2010-03-29 14:31:17 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34561 * docs/reference/libs/Makefile.am:
34562 * docs/reference/libs/libs.core.types:
34563 * docs/reference/libs/libs.glx.types:
34564 * docs/reference/libs/libs.x11.types:
34567 2010-03-29 14:21:51 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34569 * sys/vaapisink/gstvaapisink.h:
34570 Fix build without GLX.
34572 2010-03-29 14:13:55 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34577 2010-03-29 14:13:26 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34579 * sys/vaapisink/Makefile.am:
34580 * sys/vaapisink/gstvaapisink.c:
34581 * sys/vaapisink/gstvaapisink.h:
34582 Add VA/GLX support to vaapisink.
34584 2010-03-29 13:40:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34586 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
34587 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
34588 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
34589 Add glXSwapBuffers() workaround for NVIDIA.
34591 2010-03-29 13:27:16 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34593 * tests/Makefile.am:
34594 * tests/test-textures.c:
34595 Improve VA/GLX textures test.
34597 2010-03-29 12:51:38 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34599 * gst-libs/gst/vaapi/gstvaapitexture.c:
34600 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
34601 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
34602 Fix texture rendering.
34604 2010-03-29 11:25:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34606 * docs/reference/libs/libs-docs.xml.in:
34607 * docs/reference/libs/libs-sections.txt:
34608 * gst-libs/gst/vaapi/gstvaapidisplay.c:
34609 * gst-libs/gst/vaapi/gstvaapitexture.c:
34612 2010-03-29 10:40:26 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34614 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
34615 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
34616 Add gst_vaapi_window_glx_put_texture() helper.
34618 2010-03-29 09:09:30 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34620 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
34623 2010-03-26 17:00:45 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34627 * tests/test-windows.c:
34630 2010-03-26 16:52:07 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34632 * gst-libs/gst/vaapi/Makefile.am:
34633 * gst-libs/gst/vaapi/gstvaapitexture.c:
34634 * gst-libs/gst/vaapi/gstvaapitexture.h:
34635 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
34636 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
34637 Add initial VA/GLX texture abstraction though the API is not good enough yet.
34639 2010-03-26 15:22:00 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34641 * docs/reference/libs/libs-sections.txt:
34642 * gst-libs/gst/vaapi/gstvaapiobject.c:
34643 * gst-libs/gst/vaapi/gstvaapiobject.h:
34644 Add gst_vaapi_object_{,un}lock_display() helpers.
34646 2010-03-26 15:16:01 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34648 * tests/Makefile.am:
34651 * tests/test-windows.c:
34652 Factor out image utilities.
34654 2010-03-26 11:54:43 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34656 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
34657 MT-Safe: lock display.
34659 2010-03-26 11:50:31 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34661 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
34662 Make sure window resize completed prior to resizing the GL viewport.
34664 2010-03-26 11:39:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34666 * gst-libs/gst/vaapi/Makefile.am:
34667 gstvaapicompat.h is a private header, don't install it.
34669 2010-03-26 11:35:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34671 * docs/reference/libs/libs-sections.txt:
34672 * gst-libs/gst/vaapi/gstvaapidisplay.c:
34673 * gst-libs/gst/vaapi/gstvaapidisplay.h:
34674 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
34675 Add gst_vaapi_display_{sync,flush}() helpers.
34677 2010-03-26 11:30:54 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34679 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
34680 * sys/vaapisink/gstvaapisink.c:
34681 * sys/vaapisink/gstvaapisink.h:
34682 Add "synchronous" mode.
34684 2010-03-26 11:02:12 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34686 * gst-libs/gst/vaapi/gstvaapidisplay.c:
34687 * gst-libs/gst/vaapi/gstvaapidisplay.h:
34688 Only add _display suffix to open & close members because they could be #define to some arbitrary value. lock/unlock are safe names.
34690 2010-03-26 10:09:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34692 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
34693 Restore GLX context only if there is one.
34695 2010-03-26 09:41:12 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34697 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
34698 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
34699 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
34700 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
34701 Add gst_vaapi_window_glx_make_current(). Handle X11 window size changes and reset the GL viewport.
34703 2010-03-26 08:35:24 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34705 * gst-libs/gst/vaapi/gstvaapiwindow.c:
34706 Check GstVaapiWindow::render() is available prior to calling it.
34708 2010-03-26 08:10:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34710 * tests/Makefile.am:
34711 * tests/test-display.c:
34712 * tests/test-textures.c:
34713 Add VA/GLX display tests.
34715 2010-03-26 08:00:32 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34718 * gst-libs/gst/vaapi/Makefile.am:
34721 2010-03-25 17:39:06 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34723 * docs/reference/libs/Makefile.am:
34724 * docs/reference/libs/libs-docs.xml.in:
34725 * docs/reference/libs/libs-sections.txt:
34726 * docs/reference/libs/libs.types:
34727 Add missing API documentation.
34729 2010-03-25 17:28:49 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34732 * gst-libs/gst/vaapi/Makefile.am:
34733 * gst-libs/gst/vaapi/gstvaapidisplay_glx.c:
34734 * gst-libs/gst/vaapi/gstvaapidisplay_glx.h:
34735 * gst-libs/gst/vaapi/gstvaapiutils_glx.c:
34736 * gst-libs/gst/vaapi/gstvaapiutils_glx.h:
34737 * gst-libs/gst/vaapi/gstvaapiwindow_glx.c:
34738 * gst-libs/gst/vaapi/gstvaapiwindow_glx.h:
34739 * pkgconfig/Makefile.am:
34740 * pkgconfig/gstreamer-vaapi-glx.pc.in:
34741 * tests/Makefile.am:
34742 * tests/test-textures.c:
34743 Add initial VA/GLX support.
34745 2010-03-25 17:21:56 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34747 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34748 Add missing includes (for vaapi_check_status()).
34750 2010-03-25 17:21:13 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34752 * gst-libs/gst/vaapi/gstvaapidebug.h:
34753 Only enable GST_DEBUG() if DEBUG is defined. Drop old D(bug()) stuff.
34755 2010-03-25 17:18:36 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34757 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34758 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
34759 Add gst_vaapi_window_x11_is_foreign_xid() helper.
34761 2010-03-25 17:18:06 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34763 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
34764 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
34765 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34766 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
34767 Allow derived classes to specify custom Visual and Colormap.
34769 2010-03-25 13:54:06 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34771 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
34772 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
34773 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34774 Allow window creation with a specific visual (e.g. for GLX support).
34776 2010-03-25 13:21:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34778 * gst-libs/gst/vaapi/gstvaapiimage.c:
34779 Fix return value on error (though it's the same in the end).
34781 2010-03-25 12:39:54 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34783 * gst-libs/gst/vaapi/gstvaapidisplay.c:
34784 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
34785 Simplify initialization of VADisplay.
34787 2010-03-25 10:04:39 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34790 Move __attribute__((visibility("hidden"))) check down.
34792 2010-03-25 09:49:17 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34794 * docs/reference/libs/libs-sections.txt:
34795 * gst-libs/gst/vaapi/gstvaapiimage.c:
34796 * gst-libs/gst/vaapi/gstvaapiimage.h:
34797 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
34798 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
34799 * gst-libs/gst/vaapi/gstvaapisurface.c:
34800 * gst-libs/gst/vaapi/gstvaapisurface.h:
34801 * tests/test-surfaces.c:
34802 Restore the gst_vaapi_{surface,image,subpicture}_get_id() interfaces.
34804 2010-03-25 09:39:17 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34806 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34807 Use the parent display object, no need to maintain another one.
34808 In the end, libgstvaapi-x11 reduced by 1 KB in .text vs. 0.1.1.
34810 2010-03-25 09:37:40 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34812 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
34813 Add more internal helpers.
34815 2010-03-24 17:40:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34817 * gst-libs/gst/vaapi/gstvaapiimage.c:
34818 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
34819 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
34820 * gst-libs/gst/vaapi/gstvaapisurface.c:
34821 * gst-libs/gst/vaapi/gstvaapiwindow.c:
34822 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34823 Rename to GST_VAAPI_OBJECT_DISPLAY().
34825 2010-03-24 17:38:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34827 * docs/reference/libs/libs-sections.txt:
34828 * gst-libs/gst/vaapi/gstvaapiimage.c:
34829 * gst-libs/gst/vaapi/gstvaapiobject.c:
34830 * gst-libs/gst/vaapi/gstvaapiparamspecs.c:
34831 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
34832 * gst-libs/gst/vaapi/gstvaapisurface.c:
34833 * gst-libs/gst/vaapi/gstvaapitypes.h:
34834 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34835 All GstVaapiID are initialized to GST_VAAPI_ID_NONE by default. Besides, all GstVaapiObject derived class shall initialize "id" to a valid value.
34837 2010-03-24 17:22:18 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34839 * gst-libs/gst/vaapi/gstvaapiwindow.c:
34840 * gst-libs/gst/vaapi/gstvaapiwindow.h:
34841 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34842 Make GstVaapiWindow* derive from GstVaapiObject.
34844 2010-03-24 16:37:35 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34847 Factor out use gstreamer-vaapi (PACKAGE name).
34849 2010-03-24 16:35:36 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34852 Improve versioning summary.
34854 2010-03-24 16:27:36 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34856 * gst-libs/gst/vaapi/Makefile.am:
34857 gstvaapicompat.h is now a private header (not installed).
34859 2010-03-24 16:25:56 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34861 * gst-libs/gst/vaapi/gstvaapiimage.c:
34862 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
34863 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
34864 * gst-libs/gst/vaapi/gstvaapiobject.c:
34865 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
34866 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34867 Include gstvaapicompat.h in source files only, not headers.
34869 2010-03-24 16:21:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34872 * gst-libs/gst/vaapi/Makefile.am:
34873 * sys/vaapiconvert/Makefile.am:
34874 * sys/vaapisink/Makefile.am:
34875 * tests/Makefile.am:
34876 Drop tedious LIBVA_EXTRA_{CFLAGS,LIBS} definitions in Makefile.am. Override CFLAGS & LIBS instead.
34878 2010-03-24 16:17:49 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34882 * gst-libs/gst/vaapi/Makefile.am:
34883 * gst-libs/gst/vaapi/gstvaapicompat.h:
34884 * gst-libs/gst/vaapi/gstvaapidisplay.h:
34885 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
34886 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
34887 * gst-libs/gst/vaapi/gstvaapiparamspecs.c:
34888 * gst-libs/gst/vaapi/gstvaapisurface.c:
34889 * gst-libs/gst/vaapi/gstvaapiutils.c:
34890 * gst-libs/gst/vaapi/gstvaapiutils.h:
34891 * pkgconfig/gstreamer-vaapi-x11.pc.in:
34892 * pkgconfig/gstreamer-vaapi.pc.in:
34893 * sys/vaapiconvert/Makefile.am:
34894 * sys/vaapisink/Makefile.am:
34895 * tests/Makefile.am:
34896 Add compatibility with the original VA-API 0.29.
34898 2010-03-24 15:18:33 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34900 * gst-libs/gst/vaapi/gstvaapivalue.h:
34901 Add missing file (gstvaapivalue.h).
34903 2010-03-24 15:12:56 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34905 * gst-libs/gst/vaapi/gstvaapiimage.c:
34906 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
34907 * gst-libs/gst/vaapi/gstvaapisurface.c:
34908 Fix g_warning() invocation.
34910 2010-03-24 15:11:26 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34912 * gst-libs/gst/vaapi/gstvaapisurface.c:
34913 Deassociate subpictures while destroying the surface.
34915 2010-03-24 14:57:33 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34917 * gst-libs/gst/vaapi/gstvaapisurface.c:
34918 Fix destruction order of subpictures. They should be destroyed first.
34920 2010-03-24 14:46:33 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34923 * gst-libs/gst/vaapi/gstvaapiimage.c:
34924 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
34925 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
34926 * tests/test-windows.c:
34927 Add support for AYUV format.
34929 2010-03-24 14:36:39 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34931 * tests/test-windows.c:
34932 Simplify upload process and fallback to subpictures.
34934 2010-03-24 13:44:01 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34936 * docs/reference/libs/libs-docs.xml.in:
34937 * docs/reference/libs/libs-sections.txt:
34938 * gst-libs/gst/vaapi/gstvaapiparamspecs.c:
34941 2010-03-24 13:37:38 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34943 * gst-libs/gst/vaapi/Makefile.am:
34944 * gst-libs/gst/vaapi/gstvaapiobject.c:
34945 * gst-libs/gst/vaapi/gstvaapiparamspecs.c:
34946 * gst-libs/gst/vaapi/gstvaapitypes.h:
34947 * gst-libs/gst/vaapi/gstvaapivalue.c:
34948 Move GValue specific stuff to a dedicated file.
34950 2010-03-24 13:22:25 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34952 * tests/test-surfaces.c:
34953 Cosmetics (lowercase for consistency).
34955 2010-03-24 13:21:54 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34957 * gst-libs/gst/vaapi/gstvaapiobject.c:
34958 Cosmetics (vertical alignment).
34960 2010-03-24 13:20:34 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34962 * gst-libs/gst/vaapi/gstvaapiobject.c:
34963 Fix return value on error.
34965 2010-03-24 13:19:58 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34967 * docs/reference/libs/libs-sections.txt:
34968 * gst-libs/gst/vaapi/gstvaapiimage.c:
34969 * gst-libs/gst/vaapi/gstvaapiimage.h:
34970 * gst-libs/gst/vaapi/gstvaapiobject.c:
34971 * gst-libs/gst/vaapi/gstvaapiobject.h:
34972 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
34973 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
34974 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
34975 * gst-libs/gst/vaapi/gstvaapisurface.c:
34976 * gst-libs/gst/vaapi/gstvaapisurface.h:
34977 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
34978 * tests/test-surfaces.c:
34979 Move "id" down to the GstVaapiObject base.
34981 2010-03-24 12:59:22 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34983 * gst-libs/gst/vaapi/gstvaapitypes.c:
34984 Cosmetics (drop extraneous empty line).
34986 2010-03-24 12:57:54 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34988 * gst-libs/gst/vaapi/Makefile.am:
34989 * gst-libs/gst/vaapi/gstvaapiparamspecs.c:
34990 * gst-libs/gst/vaapi/gstvaapiparamspecs.h:
34991 Add GParamSpecs for GstVaapiID.
34993 2010-03-24 12:54:52 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
34995 * docs/reference/libs/libs.types:
34996 Drop gst_vaapi_id_get_type().
34998 2010-03-24 12:38:40 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35000 * docs/reference/libs/libs-sections.txt:
35001 * gst-libs/gst/vaapi/gstvaapitypes.h:
35002 Add GST_VAAPI_ID_FORMAT() and GST_VAAPI_ID_ARGS() helpers.
35004 2010-03-24 09:52:43 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35006 * docs/reference/libs/libs-sections.txt:
35007 * docs/reference/libs/libs.types:
35008 * gst-libs/gst/vaapi/Makefile.am:
35009 * gst-libs/gst/vaapi/gstvaapitypes.c:
35010 * gst-libs/gst/vaapi/gstvaapitypes.h:
35011 Add GstVaapiID abstraction.
35013 2010-03-24 09:22:00 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35015 * docs/reference/libs/libs.types:
35018 2010-03-24 08:35:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35020 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35021 * gst-libs/gst/vaapi/gstvaapiimage.c:
35022 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35023 * gst-libs/gst/vaapi/gstvaapisurface.c:
35024 Drop useless include (<va/va_backend.h>).
35026 2010-03-24 08:34:11 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35028 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
35029 Optimize GST_VAAPI_OBJECT_GET_DISPLAY to avoid a run-time check.
35031 2010-03-24 08:32:12 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35033 * gst-libs/gst/vaapi/Makefile.am:
35034 * gst-libs/gst/vaapi/gstvaapiimage.c:
35035 * gst-libs/gst/vaapi/gstvaapiobject.c:
35036 * gst-libs/gst/vaapi/gstvaapiobject.h:
35037 * gst-libs/gst/vaapi/gstvaapiobject_priv.h:
35038 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35039 * gst-libs/gst/vaapi/gstvaapisurface.c:
35040 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35041 Move private definitions and accessors to gstvaapiobject_priv.h.
35043 2010-03-24 08:16:32 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35045 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35046 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
35047 * gst-libs/gst/vaapi/gstvaapiimage.c:
35048 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
35049 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
35050 * gst-libs/gst/vaapi/gstvaapiobject.c:
35051 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35052 * gst-libs/gst/vaapi/gstvaapisurface.c:
35053 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
35054 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
35055 * gst-libs/gst/vaapi/gstvaapivideopool.c:
35056 * gst-libs/gst/vaapi/gstvaapivideosink.c:
35057 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35058 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35059 Fix short descriptions.
35061 2010-03-23 18:45:09 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35065 * debian.upstream/Makefile.am:
35066 * debian.upstream/changelog.in:
35067 * debian.upstream/compat:
35068 * debian.upstream/control.in:
35069 * debian.upstream/copyright:
35070 * debian.upstream/gstreamer-vaapi-doc.install.in:
35071 * debian.upstream/gstreamer-vaapi.install.in:
35072 * debian.upstream/libgstvaapi-dev.install.in:
35073 * debian.upstream/libgstvaapi-x11.install.in:
35074 * debian.upstream/libgstvaapi.install.in:
35075 * debian.upstream/rules:
35076 Generate upstream packages through make deb.upstream.
35078 2010-03-23 17:40:03 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35081 Bump version for development.
35083 2010-03-23 17:29:47 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35085 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
35086 Use a black background for new windows.
35088 2010-03-23 17:18:35 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35093 2010-03-23 17:12:40 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35096 * docs/reference/libs/libs.types:
35097 * gst-libs/gst/vaapi/Makefile.am:
35098 * gst-libs/gst/vaapi/gstvaapimarshal.list:
35099 * gst-libs/gst/vaapi/gstvaapiobject.c:
35100 * gst-libs/gst/vaapi/gstvaapiobject.h:
35101 * tests/test-surfaces.c:
35102 Add "destroy" signal.
35104 2010-03-23 16:25:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35106 * docs/reference/libs/libs-docs.xml.in:
35107 Improve gst-plugins-vaapi Library reference template.
35109 2010-03-23 16:21:28 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35111 * docs/reference/libs/libs-docs.xml.in:
35112 * docs/reference/libs/libs-sections.txt:
35113 * gst-libs/gst/vaapi/Makefile.am:
35114 * gst-libs/gst/vaapi/gstvaapiimage.c:
35115 * gst-libs/gst/vaapi/gstvaapiimage.h:
35116 * gst-libs/gst/vaapi/gstvaapiobject.c:
35117 * gst-libs/gst/vaapi/gstvaapiobject.h:
35118 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35119 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
35120 * gst-libs/gst/vaapi/gstvaapisurface.c:
35121 * gst-libs/gst/vaapi/gstvaapisurface.h:
35122 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35123 Make GstVaapi{Surface,Image,Subpicture} derive from a GstVaapiObject.
35125 2010-03-23 16:11:21 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35127 * sys/vaapisink/gstvaapisink.c:
35130 2010-03-23 15:34:51 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35132 * sys/vaapisink/gstvaapisink.c:
35133 Make sure VA display is valid when created with an explicit "display" name.
35135 2010-03-23 15:28:50 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35137 * sys/vaapisink/gstvaapisink.c:
35138 Use plain "display" property for the X11 display name.
35140 2010-03-23 15:22:47 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35142 * sys/vaapiconvert/gstvaapiconvert.c:
35143 * sys/vaapisink/gstvaapisink.c:
35144 Document vaapiconvert & vaapisink plugins.
35146 2010-03-23 14:19:21 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35149 * docs/reference/Makefile.am:
35150 * docs/reference/plugins/Makefile.am:
35151 * docs/reference/plugins/plugins-docs.xml.in:
35152 * docs/reference/plugins/plugins-overrides.txt:
35153 * docs/reference/plugins/plugins-sections.txt:
35154 * docs/reference/plugins/plugins.types:
35155 Add plugins documentation template.
35157 2010-03-23 14:06:42 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35159 * sys/vaapisink/gstvaapisink.c:
35160 * sys/vaapisink/gstvaapisink.h:
35161 Don't export gst_vaapisink_get_display().
35163 2010-03-23 13:32:36 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35166 * docs/reference/libs/libs-docs.xml.in:
35167 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35168 * gst-libs/gst/vaapi/gstvaapiimage.c:
35169 * gst-libs/gst/vaapi/gstvaapisurface.c:
35170 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35171 Drop introspection annotations since they require gtk-doc >= 1.12.
35173 2010-03-23 10:51:35 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35175 * gst-libs/gst/vaapi/gstvaapisurface.c:
35176 Add note about the fact that the surface holds an extra reference to the subpicture.
35178 2010-03-23 10:49:33 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35180 * gst-libs/gst/vaapi/gstvaapiimage.c:
35181 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35182 * gst-libs/gst/vaapi/gstvaapisurface.c:
35183 Improve debugging messages.
35185 2010-03-23 10:48:58 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35187 * tests/test-windows.c:
35188 Unref subpicture earlier as the surface is supposed to hold a reference to it.
35190 2010-03-23 10:36:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35192 * docs/reference/libs/libs-sections.txt:
35193 * gst-libs/gst/vaapi/gstvaapisurface.c:
35194 * gst-libs/gst/vaapi/gstvaapisurface.h:
35195 * tests/test-windows.c:
35196 Add gst_vaapi_surface_{,de}associate_subpicture() API.
35198 2010-03-23 08:13:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35200 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35201 Wait for MapNotify or UnmapNotify events on foreign windows too.
35203 2010-03-23 07:42:05 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35205 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35206 Check whether the foreign XID is mapped at binding time.
35208 2010-03-23 07:34:15 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35210 * sys/vaapisink/gstvaapisink.c:
35211 Add missing includes.
35213 2010-03-23 07:31:04 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35216 * debian/Makefile.am:
35217 * debian/control.in:
35218 * debian/gstreamer-vaapi-doc.install.in:
35222 2010-03-23 06:41:29 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35224 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35225 Fix warnings (drop extraneous var).
35227 2010-03-23 06:40:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35229 * docs/reference/libs/libs-sections.txt:
35230 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
35231 Add GST_VAAPI_WINDOW_XWINDOW() helper macro.
35233 2010-03-22 16:59:29 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35235 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35238 2010-03-22 16:57:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35240 * gst-libs/gst/vaapi/Makefile.am:
35241 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35242 * gst-libs/gst/vaapi/gstvaapiwindow.h:
35243 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
35244 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35245 Try to improve switch to fullscreen mode.
35247 2010-03-22 16:01:34 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35249 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
35250 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
35251 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35252 Improve display locking and rework X event wait functions.
35254 2010-03-22 13:06:41 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35256 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35257 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
35258 Move _GstVaapiWindowPrivate declaration to gstvaapiwindow_priv.h.
35260 2010-03-22 13:05:05 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35262 * gst-libs/gst/vaapi/Makefile.am:
35263 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35264 * gst-libs/gst/vaapi/gstvaapiwindow_priv.h:
35265 Add private API to set window size & fullscreen modes without triggering any notification or virtual functions. This is useful for derived class to fix up sizes whenever appropriate.
35267 2010-03-22 12:47:13 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35269 * docs/reference/libs/libs-sections.txt:
35270 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35271 * gst-libs/gst/vaapi/gstvaapiwindow.h:
35272 Add gst_vaapi_window_get_fullscreen() helper and "fullscreen" property.
35274 2010-03-22 12:39:02 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35276 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35277 * gst-libs/gst/vaapi/gstvaapiwindow.h:
35278 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35279 Add gst_vaapi_window_get_display() to base.
35281 2010-03-22 12:16:47 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35283 * docs/reference/libs/libs-docs.xml.in:
35284 * docs/reference/libs/libs-sections.txt:
35285 * gst-libs/gst/vaapi/Makefile.am:
35286 * gst-libs/gst/vaapi/gstvaapitypes.h:
35287 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35288 * gst-libs/gst/vaapi/gstvaapiwindow.h:
35289 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35290 * sys/vaapisink/gstvaapisink.c:
35291 * sys/vaapisink/gstvaapisink.h:
35292 Add GstVaapiPoint & GstVaapiRectangle data structures.
35294 2010-03-22 12:05:11 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35299 2010-03-22 12:03:26 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35302 * sys/vaapisink/gstvaapisink.c:
35303 * sys/vaapisink/gstvaapisink.h:
35304 Allow `vaapisink` to render videos in fullscreen mode.
35306 2010-03-22 10:51:49 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35308 * docs/reference/libs/libs-sections.txt:
35309 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35310 * gst-libs/gst/vaapi/gstvaapiwindow.h:
35311 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35312 Add gst_vaapi_window_set_fullscreen() API.
35314 2010-03-22 10:03:24 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35316 * sys/vaapisink/gstvaapisink.c:
35317 Size window so that to respect the video and pixel aspect ratio.
35319 2010-03-22 09:32:01 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35321 * docs/reference/libs/libs-sections.txt:
35322 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35323 * gst-libs/gst/vaapi/gstvaapidisplay.h:
35324 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
35325 * tests/test-display.c:
35326 Add gst_vaapi_display_get_pixel_aspect_ratio().
35328 2010-03-22 08:45:03 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35330 * docs/reference/libs/libs-sections.txt:
35333 2010-03-22 08:44:38 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35335 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35336 * gst-libs/gst/vaapi/gstvaapidisplay.h:
35337 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
35338 * tests/test-display.c:
35339 Add display size accessors.
35341 2010-03-22 08:03:12 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35344 * debian/control.in:
35345 Build-Requires: gstreamer-plugins-base >= 0.10.16.
35347 2010-03-21 08:45:09 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35349 * gst-libs/gst/vaapi/gstvaapidisplay.h:
35350 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
35351 * gst-libs/gst/vaapi/gstvaapiimage.h:
35352 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
35353 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
35354 * gst-libs/gst/vaapi/gstvaapisurface.h:
35355 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
35356 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
35357 * gst-libs/gst/vaapi/gstvaapivideopool.h:
35358 * gst-libs/gst/vaapi/gstvaapiwindow.h:
35359 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
35360 Fix documentation of *Class'es.
35362 2010-03-21 08:38:17 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35364 * gst-libs/gst/vaapi/gstvaapiutils.c:
35365 * gst-libs/gst/vaapi/gstvaapiutils.h:
35366 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35367 Move GstVaapiSurfaceRenderFlags conversion to get_PutSurface_flags_from_GstVaapiSurfaceRenderFlags().
35369 2010-03-21 08:22:46 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35371 * docs/reference/libs/libs-sections.txt:
35372 * gst-libs/gst/vaapi/gstvaapisurface.h:
35373 * gst-libs/gst/vaapi/gstvaapiwindow.h:
35374 Move GstVaapiSurfaceRenderFlags to gstvaapisurface.h since this will also be useful for e.g. a gstvaapitexture.h.
35376 2010-03-21 08:12:52 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35378 * docs/reference/libs/libs-sections.txt:
35379 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35380 * gst-libs/gst/vaapi/gstvaapiwindow.h:
35381 * sys/vaapisink/gstvaapisink.c:
35382 * tests/test-windows.c:
35383 Rename gst_vaapi_window_put_surface_full() to plain gst_vaapi_window_put_surface().
35385 2010-03-19 17:15:16 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35387 * docs/reference/libs/Makefile.am:
35388 Fix make dist for --enable-gtk-doc builds.
35390 2010-03-19 17:13:59 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35393 Improve autogen.sh.
35395 2010-03-19 17:11:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35399 Generate gtk-doc.make from gtkdocize.
35401 2010-03-19 17:04:51 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35406 * docs/Makefile.am:
35407 * docs/reference/Makefile.am:
35408 * docs/reference/libs/Makefile.am:
35409 * docs/reference/libs/libs-docs.xml.in:
35410 * docs/reference/libs/libs-overrides.txt:
35411 * docs/reference/libs/libs-sections.txt:
35412 * docs/reference/libs/libs.types:
35413 Document public API for libgstvaapi-*.so.*.
35415 2010-03-19 16:41:52 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35417 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
35418 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
35419 Document GstVaapiVideoBuffer.
35421 2010-03-19 16:08:48 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35423 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
35424 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
35425 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
35426 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
35427 * gst-libs/gst/vaapi/gstvaapivideopool.c:
35428 * gst-libs/gst/vaapi/gstvaapivideopool.h:
35429 Document surface & image pools. Drop obsolete gst_vaapi_video_pool_new() function.
35431 2010-03-19 15:45:21 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35433 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35434 * gst-libs/gst/vaapi/gstvaapidisplay.h:
35435 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
35436 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
35437 * gst-libs/gst/vaapi/gstvaapiimage.c:
35438 * gst-libs/gst/vaapi/gstvaapiimage.h:
35439 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
35440 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
35441 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35442 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
35443 * gst-libs/gst/vaapi/gstvaapisurface.c:
35444 * gst-libs/gst/vaapi/gstvaapisurface.h:
35445 * gst-libs/gst/vaapi/gstvaapivideosink.c:
35446 * gst-libs/gst/vaapi/gstvaapivideosink.h:
35447 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35448 * gst-libs/gst/vaapi/gstvaapiwindow.h:
35449 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35450 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
35451 Add tedious documentation.
35453 2010-03-19 10:42:11 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35455 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35456 Beautify append_formats().
35458 2010-03-19 10:38:45 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35460 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35461 Simplify GstVaapiDisplay (use GArray).
35463 2010-03-19 08:42:51 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35466 * sys/vaapiconvert/gstvaapiconvert.c:
35467 * sys/vaapiconvert/gstvaapiconvert.h:
35468 Factor out direct-rendering infrastructure.
35470 2010-03-18 16:18:17 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35472 * sys/vaapiconvert/gstvaapiconvert.c:
35473 * sys/vaapiconvert/gstvaapiconvert.h:
35474 Allow user to specify inout-buffers & derive-image optimizations.
35476 2010-03-18 15:58:28 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35478 * sys/vaapiconvert/gstvaapiconvert.c:
35479 Reduce number of debug messaged printed out.
35481 2010-03-18 15:53:50 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35483 * sys/vaapiconvert/gstvaapiconvert.c:
35484 * sys/vaapiconvert/gstvaapiconvert.h:
35485 Add vaDeriveImage() optimization.
35487 2010-03-18 15:52:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35489 * gst-libs/gst/vaapi/gstvaapiimage.c:
35490 Fix gst_vaapi_image_create() from a foreign VA image.
35492 2010-03-18 15:28:59 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35494 * gst-libs/gst/vaapi/gstvaapiimage.c:
35495 * gst-libs/gst/vaapi/gstvaapiimage.h:
35496 * gst-libs/gst/vaapi/gstvaapisurface.c:
35497 * gst-libs/gst/vaapi/gstvaapisurface.h:
35498 Add gst_vaapi_surface_derive_image() API.
35500 2010-03-18 13:49:50 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35502 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35503 * gst-libs/gst/vaapi/gstvaapiwindow.h:
35504 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35505 Make it possible to bin an X11 window to GstVaapiWindowX11 with plain g_object_new() and "xid" property. i.e. get foreign window size in gst_vaapi_window_x11_create().
35507 2010-03-18 13:08:17 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35509 * tests/test-windows.c:
35510 Try YV12 & I420 image formats too.
35512 2010-03-18 12:59:55 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35514 * gst-libs/gst/vaapi/gstvaapiimage.c:
35515 Split map/unmap functions into internal functions that don't check preconditions.
35517 2010-03-18 12:56:53 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35519 * gst-libs/gst/vaapi/gstvaapiimage.c:
35520 Improve gst_vaapi_image_new() sanity checks.
35522 2010-03-18 12:52:58 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35524 * tests/test-windows.c:
35527 2010-03-18 08:45:57 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35529 * sys/vaapiconvert/gstvaapiconvert.c:
35530 Check if our inout buffer is still alive or default to a separate output buffer.
35532 2010-03-18 08:16:59 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35534 * sys/vaapiconvert/gstvaapiconvert.c:
35535 * sys/vaapiconvert/gstvaapiconvert.h:
35536 Factor out buffers negotiation and optimization checks.
35538 2010-03-18 08:02:25 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35540 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35541 * gst-libs/gst/vaapi/gstvaapiimage.c:
35542 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
35543 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
35544 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35545 * sys/vaapiconvert/gstvaapiconvert.h:
35548 2010-03-17 10:43:02 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35550 * gst-libs/gst/vaapi/gstvaapiimage.c:
35551 Optimize gst_vaapi_image_is_linear() and simplify gst_vaapi_image_update_from_buffer().
35553 2010-03-17 07:59:31 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35555 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35556 * gst-libs/gst/vaapi/gstvaapidisplay.h:
35557 * gst-libs/gst/vaapi/gstvaapiimage.c:
35558 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35559 * gst-libs/gst/vaapi/gstvaapisurface.c:
35560 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35561 Add VA display locking utilities.
35563 2010-03-17 07:20:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35565 * sys/vaapisink/gstvaapisink.c:
35566 Initialize the X window in a ::set_caps() handler. Also fix build with GStreamer < 0.10.25. i.e. use preroll/render hooks.
35568 2010-03-17 07:17:17 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35570 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35571 * tests/test-windows.c:
35572 Don't show window by default during creation.
35574 2010-03-17 06:49:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35576 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35577 Fix gst_vaapi_window_x11_destroy().
35579 2010-03-16 17:57:57 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35581 * sys/vaapiconvert/gstvaapiconvert.c:
35582 * sys/vaapiconvert/gstvaapiconvert.h:
35583 Alias sink & src pad buffers whenever possible.
35585 2010-03-16 17:57:23 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35587 * gst-libs/gst/vaapi/gstvaapiimage.c:
35588 * gst-libs/gst/vaapi/gstvaapiimage.h:
35589 Extend GstVaapiImage API with *_get_image(), *_is_linear(), *_get_data_size().
35591 2010-03-16 17:10:02 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35593 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
35594 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
35595 Make GstVaapiVideoBuffer handle two pools. i.e. both image & surface at the same time.
35597 2010-03-16 14:37:47 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35599 * sys/vaapiconvert/gstvaapiconvert.c:
35600 Fix image & surface size cache.
35602 2010-03-16 14:12:40 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35605 Move gstreamer-vaapi package versioning to the top.
35607 2010-03-16 14:11:46 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35610 Bump version for development.
35612 2010-03-16 14:07:53 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35615 Cosmetics (shorten lines).
35617 2010-03-16 13:58:43 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35623 2010-03-16 13:53:54 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35625 * debian/Makefile.am:
35626 * debian/changelog.in:
35628 * debian/control.in:
35629 * debian/copyright:
35630 * debian/gstreamer-vaapi.install.in:
35631 * debian/libgstvaapi-dev.install.in:
35632 * debian/libgstvaapi-x11.install.in:
35633 * debian/libgstvaapi.install.in:
35635 Add debian packaging.
35637 2010-03-16 13:53:09 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35641 Add debian packaging.
35643 2010-03-16 10:13:36 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35645 * pkgconfig/Makefile.am:
35646 Silence GNU make extensions warning.
35648 2010-03-16 09:59:03 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35651 Add AM_PROG_CC_C_O, thus fixing this warning: tests/Makefile.am:16: compiling `test-display.c' with per-target flags requires `AM_PROG_CC_C_O' in `configure.ac'
35653 2010-03-16 09:57:25 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35657 * pkgconfig/Makefile.am:
35658 * pkgconfig/gstreamer-vaapi-x11.pc.in:
35659 * pkgconfig/gstreamer-vaapi.pc.in:
35660 Add pkgconfig files.
35662 2010-03-16 09:39:07 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35665 * gst-libs/gst/vaapi/Makefile.am:
35666 * sys/vaapisink/Makefile.am:
35667 * tests/Makefile.am:
35668 Split X11 support to libgstvaapi-x11-*.so.*
35670 2010-03-16 09:21:15 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35672 * gst-libs/gst/vaapi/Makefile.am:
35673 Don't install private headers.
35675 2010-03-16 09:18:57 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35677 * gst-libs/gst/vaapi/gstvaapidebug.h:
35678 * gst-libs/gst/vaapi/gstvaapiutils.c:
35679 * gst-libs/gst/vaapi/gstvaapiutils.h:
35682 2010-03-16 09:17:41 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35684 * gst-libs/gst/vaapi/Makefile.am:
35685 * gst-libs/gst/vaapi/gstvaapidebug.h:
35686 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35687 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
35688 * gst-libs/gst/vaapi/gstvaapiimage.c:
35689 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
35690 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35691 * gst-libs/gst/vaapi/gstvaapisurface.c:
35692 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
35693 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
35694 * gst-libs/gst/vaapi/gstvaapivideopool.c:
35695 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35696 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35697 Rename vaapi_debug.h to gstvaapidebug.h.
35699 2010-03-16 09:15:48 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35701 * gst-libs/gst/vaapi/Makefile.am:
35702 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35703 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
35704 * gst-libs/gst/vaapi/gstvaapiimage.c:
35705 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35706 * gst-libs/gst/vaapi/gstvaapisurface.c:
35707 * gst-libs/gst/vaapi/gstvaapiutils.c:
35708 * gst-libs/gst/vaapi/gstvaapiutils.h:
35709 * gst-libs/gst/vaapi/vaapi_debug.h:
35710 Move vaapi_utils.* to gstvaapiutils.*
35712 2010-03-16 09:13:16 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35714 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
35715 Cosmetics (remove an extra line).
35717 2010-03-16 09:12:47 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35719 * gst-libs/gst/vaapi/Makefile.am:
35720 * gst-libs/gst/vaapi/gstvaapiutils_x11.c:
35721 * gst-libs/gst/vaapi/gstvaapiutils_x11.h:
35722 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35723 Move X11 utilties to gstvaapiutils_x11.[ch].
35725 2010-03-16 09:03:10 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35727 * gst-libs/gst/vaapi/Makefile.am:
35728 * gst-libs/gst/vaapi/gstvaapivideosink.c:
35729 * gst-libs/gst/vaapi/gstvaapivideosink.h:
35730 * sys/vaapiconvert/gstvaapiconvert.c:
35731 * sys/vaapisink/gstvaapisink.c:
35732 Rename GstVaapiSinkBase to GstVaapiVideoSink.
35734 2010-03-16 08:49:16 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35737 * tests/Makefile.am:
35738 * tests/examples/Makefile.am:
35739 * tests/examples/generic/Makefile.am:
35740 * tests/test-display.c:
35741 * tests/test-surfaces.c:
35742 * tests/test-windows.c:
35743 Move tests to top-level tests/ directory.
35745 2010-03-16 08:43:16 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35747 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35748 * gst-libs/gst/vaapi/gstvaapiimage.c:
35749 * sys/vaapiconvert/gstvaapiconvert.c:
35750 * sys/vaapiconvert/gstvaapiconvert.h:
35751 Handle I420 formats internally in GstVaapiImage.
35753 2010-03-15 17:44:35 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35755 * sys/vaapiconvert/gstvaapiconvert.c:
35756 * sys/vaapiconvert/gstvaapiconvert.h:
35757 Implement I420 (resp. YV12) with YV12 (resp. I420) if the driver does not.
35759 2010-03-15 17:43:29 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35761 * gst-libs/gst/vaapi/gstvaapiimage.c:
35762 Implement I420 and YV12 if the underlying implementation does not.
35764 2010-03-15 17:10:56 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35766 * sys/vaapiconvert/Makefile.am:
35767 * sys/vaapiconvert/gstvaapiconvert.c:
35768 * sys/vaapiconvert/gstvaapiconvert.h:
35769 Add initial vaapiconvert plugin.
35771 2010-03-15 17:09:12 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35773 * sys/vaapisink/gstvaapisink.c:
35774 * sys/vaapisink/gstvaapisink.h:
35777 2010-03-15 16:57:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35779 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35782 2010-03-15 16:57:01 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35784 * tests/examples/generic/test-windows.c:
35785 Generate R/G/B rects.
35787 2010-03-15 16:13:51 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35789 * gst-libs/gst/vaapi/gstvaapisurface.c:
35790 * gst-libs/gst/vaapi/gstvaapisurface.h:
35791 Add gst_vaapi_surface_sync().
35793 2010-03-15 16:13:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35795 * gst-libs/gst/vaapi/gstvaapiimage.c:
35796 Cosmetics (reverse args order).
35798 2010-03-15 15:55:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35800 * gst-libs/gst/vaapi/gstvaapiimage.c:
35801 * gst-libs/gst/vaapi/gstvaapiimage.h:
35802 * gst-libs/gst/vaapi/gstvaapisurface.c:
35805 2010-03-15 15:12:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35807 * gst-libs/gst/vaapi/Makefile.am:
35808 * gst-libs/gst/vaapi/gstvaapiwindow.c:
35809 * gst-libs/gst/vaapi/gstvaapiwindow.h:
35810 * gst-libs/gst/vaapi/gstvaapiwindow_x11.c:
35811 * gst-libs/gst/vaapi/gstvaapiwindow_x11.h:
35812 * tests/examples/generic/Makefile.am:
35813 * tests/examples/generic/test-windows.c:
35814 Add VA/X11 window abstraction.
35816 2010-03-15 14:57:57 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35818 * gst-libs/gst/vaapi/gstvaapidisplay.h:
35819 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
35820 Add VA and X11 display accessors.
35822 2010-03-15 14:57:30 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35824 * gst-libs/gst/vaapi/gstvaapiimage.c:
35827 2010-03-15 13:32:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35829 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35832 2010-03-15 11:49:03 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35834 * gst-libs/gst/vaapi/gstvaapiimage.c:
35835 * gst-libs/gst/vaapi/gstvaapiimage.h:
35836 * gst-libs/gst/vaapi/gstvaapisurface.c:
35837 * gst-libs/gst/vaapi/gstvaapisurface.h:
35838 Add gst_vaapi_{get,put}_image() API.
35840 2010-03-15 10:27:10 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35842 * gst-libs/gst/vaapi/gstvaapiimage.c:
35843 * gst-libs/gst/vaapi/gstvaapiimage.h:
35844 Add gst_vaapi_image_update_from_buffer() helper.
35846 2010-03-12 23:53:48 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35848 * sys/vaapisink/Makefile.am:
35849 * sys/vaapisink/gstvaapisink.c:
35850 * sys/vaapisink/gstvaapisink.h:
35851 Implement GstVaapiSinkBase interface and integrate with GST_DEBUG better.
35853 2010-03-12 23:50:09 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35855 * tests/examples/generic/Makefile.am:
35856 * tests/examples/generic/test-surfaces.c:
35859 2010-03-12 23:48:50 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35861 * gst-libs/gst/vaapi/Makefile.am:
35862 * gst-libs/gst/vaapi/gstvaapivideobuffer.c:
35863 * gst-libs/gst/vaapi/gstvaapivideobuffer.h:
35864 Add basic GstVaapiVideoBuffer.
35866 2010-03-12 23:47:47 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35868 * gst-libs/gst/vaapi/Makefile.am:
35869 * gst-libs/gst/vaapi/gstvaapiimagepool.c:
35870 * gst-libs/gst/vaapi/gstvaapiimagepool.h:
35871 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
35872 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
35873 * gst-libs/gst/vaapi/gstvaapivideopool.c:
35874 * gst-libs/gst/vaapi/gstvaapivideopool.h:
35875 Add GstVaapiImagePool and factor out GstVaapiSurfacePool from a base GstVaapiVideoPool.
35877 2010-03-12 22:32:35 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35879 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
35880 Simplify format conversion code.
35882 2010-03-12 22:28:01 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35884 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
35885 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
35886 Add gst_vaapi_image_format_from_caps() helper.
35888 2010-03-12 17:45:18 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35890 * gst-libs/gst/vaapi/Makefile.am:
35891 * gst-libs/gst/vaapi/gstvaapisurfacepool.c:
35892 * gst-libs/gst/vaapi/gstvaapisurfacepool.h:
35893 Add VA surface pool (lazy allocator).
35895 2010-03-12 17:39:11 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35897 * gst-libs/gst/vaapi/gstvaapisurface.c:
35898 * gst-libs/gst/vaapi/gstvaapisurface.h:
35899 Add gst_vaapi_surface_get_size() helper.
35901 2010-03-12 10:52:08 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35903 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
35904 Avoid use of GstStaticCaps since older gstreamer versions (0.10.22) write to it.
35906 2010-03-11 15:35:43 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35908 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
35909 Reset display-name if the user provided his own X11 display.
35911 2010-03-11 15:21:43 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35913 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
35914 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
35915 * tests/examples/generic/test-display.c:
35916 Add gst_vaapi_display_x11_new_with_display() API.
35918 2010-03-11 15:04:18 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35920 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
35921 * gst-libs/gst/vaapi/gstvaapiimage.h:
35922 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
35923 * gst-libs/gst/vaapi/gstvaapisurface.h:
35924 Fix *_GET_CLASS() definitions...
35926 2010-03-11 15:01:00 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35928 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35929 * gst-libs/gst/vaapi/gstvaapidisplay.h:
35930 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
35931 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
35932 * tests/examples/generic/test-display.c:
35933 API change: gst_vaapi_display_x11_new() now takes an X11 display name.
35935 2010-03-11 13:58:32 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35937 * gst-libs/gst/vaapi/gstvaapisurface.c:
35938 * gst-libs/gst/vaapi/gstvaapisurface.h:
35939 Use GstVaapiChromaType abstraction.
35941 2010-03-11 12:30:12 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35943 * gst-libs/gst/vaapi/gstvaapiimage.c:
35944 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35945 * gst-libs/gst/vaapi/gstvaapisurface.c:
35946 New refcounting policy. All getters return a reference, not a copy. So the user shall reference the object itself, should he wish so.
35948 2010-03-11 12:14:10 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35950 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
35951 Don't warn on failure, just return an appropriate error or value.
35953 2010-03-11 12:11:36 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35955 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35956 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
35957 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
35958 Filter out any format that is not supported by the library (libgstvaapi). Also sort the formats by HW preference.
35960 2010-03-11 10:50:27 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35962 * gst-libs/gst/vaapi/Makefile.am:
35963 * gst-libs/gst/vaapi/gstvaapisinkbase.c:
35964 * gst-libs/gst/vaapi/gstvaapisinkbase.h:
35965 Add helper interface that all VA-API sinks must implement. e.g. vaapisink.
35967 2010-03-10 13:13:51 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35969 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35970 * gst-libs/gst/vaapi/gstvaapiimage.c:
35971 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35972 * gst-libs/gst/vaapi/gstvaapisurface.c:
35973 * gst-libs/gst/vaapi/vaapi_debug.h:
35976 2010-03-10 13:10:59 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35978 * gst-libs/gst/vaapi/gstvaapiimage.c:
35979 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
35980 Fix GstVaapiImage and GstVaapiSubpicture initialization.
35982 2010-03-10 13:02:45 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35984 * gst-libs/gst/vaapi/gstvaapisurface.c:
35985 Fix GstVaapiSurface initialization, override constructed() method, not constructor(). GObject C is awful...
35987 2010-03-10 12:25:38 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35989 * tests/examples/generic/test-display.c:
35992 2010-03-10 12:25:19 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
35994 * gst-libs/gst/vaapi/gstvaapidisplay.c:
35995 * gst-libs/gst/vaapi/gstvaapidisplay.h:
35996 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
35997 Fix GstVaapiDisplay initialization.
35999 2010-03-10 10:43:31 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36001 * gst-libs/gst/vaapi/gstvaapidisplay.c:
36002 * gst-libs/gst/vaapi/gstvaapidisplay.h:
36003 Get VA image & subpicture formats as GstCaps.
36005 2010-03-10 10:41:12 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36007 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
36008 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
36009 Add helper to convert from GstVaapiImageFormat to GstCaps.
36011 2010-03-09 12:00:32 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36013 * gst-libs/gst/vaapi/gstvaapidisplay.c:
36014 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
36015 * gst-libs/gst/vaapi/gstvaapisurface.c:
36016 Cosmetics (drop unused variables).
36018 2010-03-05 17:11:52 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36021 * sys/vaapiconvert/Makefile.am:
36022 * sys/vaapiconvert/gstvaapiconvert.c:
36023 * sys/vaapiconvert/gstvaapiconvert.h:
36024 * sys/vaapisink/Makefile.am:
36025 * sys/vaapisink/gstvaapisink.c:
36026 * sys/vaapisink/gstvaapisink.h:
36027 Add boilerplate for vaapiconvert and vaapisink elements.
36029 2010-03-05 15:29:04 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36033 * sys/vaapiconvert/Makefile.am:
36034 Add vaapiconvert element hierarchy.
36036 2010-03-05 15:26:36 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36038 * sys/vaapisink/Makefile.am:
36039 Rename to vaapisink.
36041 2010-03-05 10:07:22 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36043 * gst-libs/gst/vaapi/gstvaapisurface.c:
36044 Shorter code (and more correct).
36046 2010-03-05 10:04:55 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36048 * gst-libs/gst/vaapi/gstvaapisurface.c:
36049 * gst-libs/gst/vaapi/gstvaapisurface.h:
36050 Add helper to get GstVaapiDisplay from a surface.
36052 2010-03-05 08:52:20 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36054 * gst-libs/gst/vaapi/gstvaapidisplay.c:
36055 Fix subpicture formats list length.
36057 2010-03-04 17:41:34 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36059 * gst-libs/gst/vaapi/gstvaapidisplay.c:
36060 * gst-libs/gst/vaapi/gstvaapidisplay.h:
36061 Add utilities to check whether a VA-API driver supports specific image or subpicture format. Likewise for VA profile.
36063 2010-03-04 17:40:47 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36065 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
36066 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
36067 Cosmetics (more checks, includes).
36069 2010-03-04 17:39:58 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36071 * gst-libs/gst/vaapi/gstvaapisubpicture.c:
36072 * gst-libs/gst/vaapi/gstvaapisubpicture.h:
36073 Really add VA subpicture abstraction.
36075 2010-03-04 17:39:01 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36077 * gst-libs/gst/vaapi/Makefile.am:
36078 * gst-libs/gst/vaapi/gstvaapiimage.c:
36079 * gst-libs/gst/vaapi/gstvaapiimage.h:
36080 * gst-libs/gst/vaapi/gstvaapiimageformat.c:
36081 * gst-libs/gst/vaapi/gstvaapiimageformat.h:
36082 * gst-libs/gst/vaapi/gstvaapisurface.c:
36083 * gst-libs/gst/vaapi/gstvaapisurface.h:
36084 Add VA surface, image, subpicture abstractions. Ported over from Gnash.
36086 2010-01-25 16:15:01 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36089 * gst-libs/gst/vaapi/Makefile.am:
36090 * gst-libs/gst/vaapi/gstvaapidisplay.c:
36091 * gst-libs/gst/vaapi/gstvaapidisplay.h:
36092 * gst-libs/gst/vaapi/gstvaapidisplay_x11.c:
36093 * gst-libs/gst/vaapi/gstvaapidisplay_x11.h:
36094 * gst-libs/gst/vaapi/vaapi_debug.h:
36095 * gst-libs/gst/vaapi/vaapi_utils.c:
36096 * gst-libs/gst/vaapi/vaapi_utils.h:
36097 * tests/examples/generic/Makefile.am:
36098 * tests/examples/generic/test-display.c:
36099 Add initial VA display abstraction.
36101 2010-01-25 15:04:10 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36105 * tests/Makefile.am:
36106 * tests/examples/Makefile.am:
36107 * tests/examples/generic/Makefile.am:
36108 Add tests infrastructure.
36110 2010-01-25 14:59:37 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36113 Clean up VA-API checks.
36115 2010-01-25 13:49:55 +0000 gb <gb@5584edef-b1fe-4b99-b61b-dd2bab72e969>
36118 Check for __attribute__((visibility("hidden"))).