Gustavo Sverzut Barbieri [Fri, 7 Mar 2014 04:52:11 +0000 (01:52 -0300)]
evas/fb: use stride instead of width.
While most framebuffes use stride = width, some may have stride bigger
than width to provide better alignment. Then we must always use stride.
Thanks to Arjan van de Ven, the one that spotted the issue.
@fix
Jean-Philippe Andre [Fri, 7 Mar 2014 02:00:26 +0000 (11:00 +0900)]
Evas: Reduce verbosity of alpha functions
And remove commented-out code.
Jean-Philippe Andre [Fri, 7 Mar 2014 01:59:54 +0000 (10:59 +0900)]
Evas filters: Reduce verbosity (INF --> DBG)
Jean-Philippe Andre [Fri, 7 Mar 2014 01:32:47 +0000 (10:32 +0900)]
Evas filters: Improve detection of the GL engine
According to cedric's horrified comment :)
And add a comment in the code. Yes, this IS a temporary solution,
but the GL engineS being what they are (tons of duplicated code),
I think it's still better for now to just make things work.
Jean-Philippe Andre [Thu, 6 Mar 2014 09:28:46 +0000 (18:28 +0900)]
Evas filters: Complete GL support with scaling function
Directly use the scale functions from scalecache when
running the GL engine.
With this, and glReadPixels support, the GL engine support is now
100% complete. It will be DAMN SLOW, but ALL filters should work
in both OpenGL and Software rendering.
Jean-Philippe Andre [Fri, 7 Mar 2014 00:36:16 +0000 (09:36 +0900)]
Evas filters: Implement proxy rendering for GL
Make use of glReadPixel to access the source's pixel data.
Use all classic CPU functions to blend and use that data.
Save pointer to the GL image and update it with the latest data
during target render.
Use ENFN's surface_lock, read_pixels, unlock.
Also, add some more error checks to make sure the images are valid,
or return an error at runtime.
Jean-Philippe Andre [Thu, 6 Mar 2014 08:52:57 +0000 (17:52 +0900)]
Evas filters: Allow image copy with the GL engine
For now, this simply uses memcpy.
The SW engine equivalent uses image_draw, which wouldn't work
under the GL engine.
Jean-Philippe Andre [Thu, 6 Mar 2014 08:42:24 +0000 (17:42 +0900)]
Evas gl_x11: Add support for glReadPixels
This will be needed by the filters for proxy rendering,
for textures and maps (displacement).
Add new engine functions to unleash the (sluggish) power of glReadPixels.
The idea is to be able to bypass glReadPixels later, so 3 new APIs are
added:
- surface_lock
- surface_read_pixels
- surface_unlock
They must be called in that order.
Note (for history):
glReadPixels was always getting the wrong data during first draw,
but the right data during a redraw...
Why? Well simply because for OpenGL itself, the image had never
been drawn in teh first place! Only the Evas GL context knew
about the image drawing, as it was queued somewhere in the pipe.
One line solution: Call evas_gl_common_context_flush before
doing anything else.
Jean-Philippe Andre [Wed, 5 Mar 2014 09:04:35 +0000 (18:04 +0900)]
Evas filters: Minor change (type is void*)
Cedric BAIL [Fri, 7 Mar 2014 00:36:52 +0000 (09:36 +0900)]
git: ignore generated file.
Cedric BAIL [Fri, 7 Mar 2014 00:35:16 +0000 (09:35 +0900)]
evas: let's still compile when we have no hw accel for DRM.
Chris Michael [Thu, 6 Mar 2014 14:46:46 +0000 (14:46 +0000)]
ecore-drm: Add ecore-drm.pc to gitignore
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Gustavo Sverzut Barbieri [Thu, 6 Mar 2014 14:17:30 +0000 (11:17 -0300)]
evas/fb: implement evas_fb_outbuf_fb_reconfigure()
refactor code used in evas_fb_outbuf_fb_setup_fb() so it can be used
in evas_fb_outbuf_fb_reconfigure().
Gustavo Sverzut Barbieri [Thu, 6 Mar 2014 14:16:27 +0000 (11:16 -0300)]
evas/fb: use a copy of fb_var since ioctl() modifies it and we consider it const
Gustavo Sverzut Barbieri [Thu, 6 Mar 2014 14:15:05 +0000 (11:15 -0300)]
evas/fb: make sure FB_Mode attributes are zeroed.
We may read only geometry or timings, then make sure memory is fully
zeroed by doing a memset on the new array item.
Gustavo Sverzut Barbieri [Thu, 6 Mar 2014 14:12:44 +0000 (11:12 -0300)]
ecore_fb: @fix use of uninitialized values reported by valgrind.
Gustavo Sverzut Barbieri [Thu, 6 Mar 2014 05:18:55 +0000 (02:18 -0300)]
evas/fb: extensive logs to help solve bugs.
FrameBuffer can be tricky with all combinations and it's hard to tell
users to send useful information, then print the information we use so
we can get useful bug reports.
Gustavo Sverzut Barbieri [Thu, 6 Mar 2014 05:16:03 +0000 (02:16 -0300)]
evas/fb: expose and call fb_freemode() and fb_cleanup()
properly clean memory and restore framebuffer on error conditions.
@fix
Gustavo Sverzut Barbieri [Thu, 6 Mar 2014 05:09:44 +0000 (02:09 -0300)]
evas/fb: handle fds using -1 for unset/error.
clean evas_fb_main.c so it returns -1 to indicate invalid fds, same as
open() and what is written in evas_fb.h example.
call fb_cleanup() in all error conditions and always return -1 in
fb_postinit() if we did call fb_cleanup().
Gustavo Sverzut Barbieri [Thu, 6 Mar 2014 04:08:44 +0000 (01:08 -0300)]
evas: @fix convert to 24bits so dst_jump is used properly.
in all other convert functions, dst_jump is provided in pixesl and
multiplied by the number of bytes-per-pixel either explicitly or
implicitly by using a different type for dst pointer (DATA16,
DATA32...).
As in 24 bits we use DATA8 we must explicitly multiply dst_jump by 3.
Gustavo Sverzut Barbieri [Thu, 6 Mar 2014 03:53:08 +0000 (00:53 -0300)]
evas/fb: avoid crashing it outbuf setup failed.
if fb setup fails, then we shouldn't use re->ob as it's NULL.
@fix
Tom Hacohen [Thu, 6 Mar 2014 09:13:38 +0000 (09:13 +0000)]
Fix evas compilation and installation.
Daniel Zaoui [Wed, 5 Mar 2014 11:54:39 +0000 (13:54 +0200)]
Eolian: generation of the first Eo file.
The first object that we generate with Eolian is Evas_Line, as it is a
simple one.
Two files are generated during build:
- the .eo.c contains the APIs definitions invoking Eo, the Eo functions
extracting the parameters and calling the hand written functions and
Eo structures to define the objects. These hand written functions are
located in e.g evas_object_line.c.
- the .eo.h contains the APIs and Eo prototyes.
We will continue with the other objects. If you note something wrong,
please update us asap:
daniel.zaoui@samsung.com
yossi.kantor@samsung.com
Daniel Zaoui [Wed, 5 Mar 2014 11:45:14 +0000 (13:45 +0200)]
Eolian/Generator: Switch set/get functions generation.
We want to generate first set properties and then get properties to be
in accordance with the current op ids order.
Tom Hacohen [Wed, 5 Mar 2014 10:19:08 +0000 (10:19 +0000)]
Eolian helper: Fixed rule dependency to correctly depend on eolian_gen.
Without this change, bootstrapping doesn't work as well.
Daniel Zaoui [Wed, 5 Mar 2014 08:06:05 +0000 (10:06 +0200)]
Eolian/Lexer: fix parsing for functions description.
If the first character of the description was not an alphanumeric, it was
not finalizing the parsing of the description and was not inserting it
in the database.
It was occurring when "@since" was the first word in the description of
the function.
Daniel Zaoui [Wed, 5 Mar 2014 08:04:25 +0000 (10:04 +0200)]
Eolian/Generator: Fix comments generation in headers.
Trailing whitespaces were added in description empty lines.
Chris Michael [Thu, 6 Mar 2014 12:38:06 +0000 (12:38 +0000)]
ecore-evas: Include Ecore_Drm header
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 12:36:53 +0000 (12:36 +0000)]
efl: Fix build with drm disabled
@bugfix: Drm is optional, so ecore-evas should not internally depend
on it unless ecore-drm is found.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 12:36:22 +0000 (12:36 +0000)]
efl: Fix installing ecore-drm pkgconfig file
@bugfix: If we build with ecore-drm support, let's install the
pkgconfig file yes ? :)
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 11:23:57 +0000 (11:23 +0000)]
Merge branch 'devs/devilhorns/ecore_drm'
Chris Michael [Thu, 6 Mar 2014 11:17:26 +0000 (11:17 +0000)]
ecore-drm: Add ecore-drm Makefile
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 11:17:01 +0000 (11:17 +0000)]
ecore-drm: Add Ecore_Drm to main Makefile
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 11:15:29 +0000 (11:15 +0000)]
ecore-drm: Add code for SPARTACUS !! ;)
@ferature: This adds code for the ecore-drm auth process to open
restricted inputs/cards/etc by the user.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 11:12:23 +0000 (11:12 +0000)]
ecore-drm: Add missing EINA_UNUSED for unused function paramaters.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 11:10:26 +0000 (11:10 +0000)]
ecore-drm: Comment out Gbm Format
@bugfix: This fixes ecore-drm building since we commented out all the
hardware accel support (for now).
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 11:08:55 +0000 (11:08 +0000)]
ecore-drm: Add missing define
@bugfix: Add missing define for number of framebuffers
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 11:02:25 +0000 (11:02 +0000)]
evas-drm: Add headers for hardware acceleration
@bugfix: Add missing headers for egl/gl support in evas-drm
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 10:59:44 +0000 (10:59 +0000)]
evas-drm: Remove hardware acceleration fields from engine structure
@bugfix: Removed hardware acceleration fields from engine structure.
These are now located inside the buffer management code itself, so no
need for them here.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 10:03:21 +0000 (10:03 +0000)]
ecore-drm: Add configure.ac autofoo for ecore-drm
@feature: This adds necessary dependencies for ecore_drm library to
build.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 10:03:00 +0000 (10:03 +0000)]
ecore-drm: Add package config file
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 09:43:48 +0000 (09:43 +0000)]
ecore-drm: Add Ecore_Drm code
@feature: Add Ecore_Drm library code for setting up drm card, outputs,
virtual terminals, framebuffers, and input for use with ecore_evas drm
code.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Joel Klinghed [Thu, 6 Mar 2014 10:18:50 +0000 (19:18 +0900)]
efl: Use actual destination dirs in warning about using non-standard prefix
Summary:
Instead of relying on ${prefix}/share to be the default for ${datadir} or
${dbusservicedir} use the actual paths where the files will be placed.
Somewhat tricky because of autoconfs desire to keep prefix changeable but
continuously evaluating the variables until they don't change work.
Reviewers: raster, englebass
Reviewed By: raster
CC: cedric, stefan_schmidt
Differential Revision: https://phab.enlightenment.org/D581
se.osadchy [Thu, 6 Mar 2014 09:26:51 +0000 (18:26 +0900)]
edje/edje_edit: Check for NULL
Summary: Add check for NULL to the edje_edit_state_external_param_set function
@fix
Reviewers: cedric, raster, seoz
Reviewed By: cedric
CC: cedric, reutskiy.v.v
Differential Revision: https://phab.enlightenment.org/D584
Signed-off-by: Cedric BAIL <cedric.bail@samsung.com>
Vyacheslav Reutskiy [Thu, 6 Mar 2014 09:22:15 +0000 (18:22 +0900)]
edje/edje_edit: fix part add bug
Summary:
Invert the check of the ed->file. If the ed->file is NULL then return
EINA_FALSE.
@fix
Reviewers: raster, cedric, seoz
CC: cedric
Differential Revision: https://phab.enlightenment.org/D603
Signed-off-by: Cedric BAIL <cedric.bail@samsung.com>
Chris Michael [Thu, 6 Mar 2014 07:36:39 +0000 (07:36 +0000)]
evas-drm: Add buffer management file
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 07:36:15 +0000 (07:36 +0000)]
evas-drm: Add buffer manager file to build order
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Thu, 6 Mar 2014 07:35:39 +0000 (07:35 +0000)]
evas-drm: Fix requirements when building for drm hw accel
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Cedric BAIL [Thu, 6 Mar 2014 07:33:06 +0000 (16:33 +0900)]
edje: add AUTHORS and more than one license file to Edje.
@feature
Cedric BAIL [Thu, 6 Mar 2014 05:26:30 +0000 (14:26 +0900)]
eina: make Eina_Error thread safe.
@feature
Cedric BAIL [Thu, 6 Mar 2014 05:24:56 +0000 (14:24 +0900)]
ecore_evas: remove warning.
Cedric BAIL [Thu, 6 Mar 2014 05:24:25 +0000 (14:24 +0900)]
tests: remove warnings.
Jihoon Kim [Thu, 6 Mar 2014 02:17:34 +0000 (11:17 +0900)]
Edje entry: fix bug preedit text is committed in the next entry when Tab key is pressed.
Key events which ecore_imf_context_filter_event function returns EINA_TRUE should be not propagated.
@fix
Jérémy Zurcher [Wed, 5 Mar 2014 22:57:39 +0000 (23:57 +0100)]
eo: replace composite_objects Eina_List with an array of Eo_Object*
as we don't support multiple composites of the same class,
and know at class elaboration how many composites we should have,
we can create the composites array and pack it at the end of the object.
Jérémy Zurcher [Wed, 5 Mar 2014 21:44:21 +0000 (22:44 +0100)]
eo: memory waste at mixin class elaboration
@fix
mixins data offsets are stored in Eo_Extension_Data_Offset[],
if the constructed class is a mixin, do not reserve space for its
private data, the class is in mixins list and will be handled at
Eo_Extension_Data_Offset computation.
see _eo_data_scope_get(...) for private data retrieval
Chris Michael [Wed, 5 Mar 2014 14:36:39 +0000 (14:36 +0000)]
ecore-evas-x: Fix resource leak from eina_str_split_full usage
@bugfix: This fixes Coverity CID 1188012
Coverity reported resource leak from eina_str_split_full return which
was only being freed if the number was > 0. This is incorrect as
eina_str_split_full will still allocate the array and thus it should
be freed.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Wed, 5 Mar 2014 14:31:22 +0000 (14:31 +0000)]
ecore-evas-x: Fix formatting
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Wed, 5 Mar 2014 14:11:36 +0000 (14:11 +0000)]
evas-drm: Fix setting cached image alpha flag
@bugfix: Set cached image alpha flag properly
This fixes issue where cached image alpha flag was not set properly.
Set it according to the outbuf's destination alpha flag.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Wed, 5 Mar 2014 07:09:19 +0000 (07:09 +0000)]
evas-drm: Check buffer validity in swap function
@bugfix: Check (and set) buffer validity before calling
framebuffer_send. This fixes an issue where buffer was not valid,
causing next_update_get to do full Copies.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Chris Michael [Wed, 5 Mar 2014 07:07:57 +0000 (07:07 +0000)]
evas-drm: Don't call framebuffer set from within send function
@bugfix: We cannot call framebuffer_set from within the send function
because if we are not vsync'd then framebuffer_set would never be
called and thus the buffer would not be marked as valid, causing full
Copies to happen.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
Gustavo Sverzut Barbieri [Wed, 5 Mar 2014 13:00:29 +0000 (10:00 -0300)]
ecore_evas/fb: implement screen_geometry_get()
Sometimes it is useful to query screen geometry (more often size) to
do some metrics as finding the proper theme in my case. As this call
works for X11 and others, let's make it work for framebuffer as well.
Cedric BAIL [Wed, 5 Mar 2014 06:12:17 +0000 (15:12 +0900)]
edje: let's allow to reset escaped text to NULL.
@fix T821.
Jeff Grimshaw [Wed, 5 Mar 2014 05:11:56 +0000 (14:11 +0900)]
doc: remove deprecated setting SYMBOL_CACHE_SIZE
Summary:
The setting SYMBOL_CACHE_SIZE has been deprecated and will
generate a warning when Doxygen runs. This change comments out that
configuration setting to prevent the warning.
Test Plan: make doc 2> doxyerr.log
CC: cedric
Differential Revision: https://phab.enlightenment.org/D599
Signed-off-by: Cedric BAIL <cedric.bail@samsung.com>
Cedric BAIL [Wed, 5 Mar 2014 05:06:39 +0000 (14:06 +0900)]
eio: make it possible to limit the amount of memory used by threads.
Eio does need to allocate memory from the thread, transmit it to the
main loop where it is processed and later on freed. If the main loop
is to slow to handle the stream of incoming data, a huge peak of
memory can happen. Some people may want to limit the amount of memory
used at peak time and so this new infrastructure make that possible.
@feature
Cedric BAIL [Wed, 5 Mar 2014 05:06:27 +0000 (14:06 +0900)]
eio: remove warning.
Cedric BAIL [Wed, 5 Mar 2014 03:14:40 +0000 (12:14 +0900)]
edje: fix edje_watch to not pass the '\n' to eio_monitor.
Eina_File line iterator does give the length of the line including the '\n'.
We did previously ignore that and passed the '\n' down to eio_monitor. Obviously
it would fail to monitor a PATH that finished with a '\n' and edje_watch did
stop working. I guess nobody did any real testing with edje watch in the past
year.
@fix
Cedric BAIL [Wed, 5 Mar 2014 02:46:12 +0000 (11:46 +0900)]
eio: let's report the path we didn't found.
Cedric BAIL [Wed, 5 Mar 2014 01:41:37 +0000 (10:41 +0900)]
eio: factorize code to have one place where data are pushed.
Cedric BAIL [Wed, 5 Mar 2014 01:40:55 +0000 (10:40 +0900)]
eio: improve log when monitoring files.
Jean-Philippe ANDRE [Tue, 4 Mar 2014 13:31:02 +0000 (22:31 +0900)]
Ecore suite: Don't fail if PulseAudio is not running
In a previous commit I mentionned that make check would fail
if PulseAudio is not running. Well, this is a bit excessive,
especially for buildbots where it doesn't have to be running.
So let's check that PulseAudio is running before failing.
Other errors should still be caught.
A simple "pulseaudio --check" should do the trick.
Daniel Zaoui [Tue, 4 Mar 2014 09:21:47 +0000 (11:21 +0200)]
Eolian: Fix build for Windows
The definition of EAPI was missing in Eolian.h.
Spank on you, Tom! I expect you to be perfect and to tell me those
things before I push all the stuff ;-)
Jean-Philippe Andre [Tue, 4 Mar 2014 06:56:19 +0000 (15:56 +0900)]
Ecore suite: Fix hangs when PulseAudio is not present
EFL can be built with PulseAudio support even if PA is
not present on the system. In that case, ecore_suite will
hang forever. Let's just catch the error instead.
Note: make check will fail if PA is not available (but not hang)
This is not related to Tom's recent mail (but yeah it reminded me
of this bug^^)
Fixes T732.
Jean-Philippe Andre [Tue, 4 Mar 2014 06:17:11 +0000 (15:17 +0900)]
Evas filters: Add return value in filter callback
This will inform the client whether an asynchronous filter properly
rendered or not.
I actually don't know any case where rendering can fail at runtime.
Jean-Philippe Andre [Tue, 4 Mar 2014 05:46:13 +0000 (14:46 +0900)]
Evas cserve2: Fix excessive warnings during shutdown
The global index for fonts will be freed after cserve2 shutdown
but Evas keeps trying to free all its fonts after that.
Jean-Philippe Andre [Tue, 4 Mar 2014 05:14:15 +0000 (14:14 +0900)]
Evas filters: Fix errors in Evas_Object::destructor
Remove invalid call to source_unset and call the appropriate
Eo evas_obj_text_filter_source_set in case of Evas_Text with
proxy sources.
Jean-Philippe Andre [Tue, 4 Mar 2014 04:58:30 +0000 (13:58 +0900)]
Evas filters: Clean up some duplicate includes
No functional change
Jean-Philippe Andre [Tue, 4 Mar 2014 03:05:44 +0000 (12:05 +0900)]
Evas filters: Rename RGBA_Image::mask.data into image.data8
The structure should not be changed, despite the union modification.
I am renaming for consistency with older branches that had a mask
field in RGBA_Image. Also, the mask.data or data8 is really just
a way to avoid casting between DATA8 and DATA32 (and it shows
clearly what kind of data you are dealing with).
Jean-Philippe Andre [Tue, 4 Mar 2014 02:50:35 +0000 (11:50 +0900)]
Evas filters: Fix a typo (wrong value checked)
This was all a bad copy and paste :)
Jean-Philippe Andre [Tue, 4 Mar 2014 02:44:19 +0000 (11:44 +0900)]
Evas filters: Minor fix for consistency (clip_to_target)
Make sure sx and sy are properly set.
No need to backport as the function actually expect these values to
be initialized to 0 in the first place.
Jean-Philippe Andre [Mon, 3 Mar 2014 08:45:15 +0000 (17:45 +0900)]
Evas filters: fix clip to target calculation
It was possible to keep negative values for dx,dy which would
then draw pixels out of bounds (= crash).
Make check crashed after the previous commit.
@fix
Jean-Philippe Andre [Mon, 3 Mar 2014 07:58:18 +0000 (16:58 +0900)]
Evas filters: Fallback to normal rendering in case of error
If the filters fail to render at runtime (that is, parsing went fine
but a command failed to run properly), fallback to normal rendering.
This should prevent text from disappearing when using proxies and
the OpenGL engine (for now).
Jean-Philippe Andre [Mon, 3 Mar 2014 07:22:40 +0000 (16:22 +0900)]
Evas filters: Avoid potential memory leaks
Properly free the buffers when they are not in the current context list.
Jean-Philippe ANDRE [Sun, 2 Mar 2014 09:39:08 +0000 (18:39 +0900)]
Evas filters: fix random cases of 'dancing text'
In some situations, text with filters would be rendered in an invalid
position (somewhere too high).
I am not entirely sure of the reason why the original code with BLEND
doesn't work, but this new version is simpler as GL and SW have more
similar behaviours:
- render text to our 'output' buffer
- draw this buffer as an image onto the set target
Thanks zmike for reporting the issue.
And thanks A LOT for using the filters :D
@fix
Signed-off-by: Jean-Philippe Andre <jp.andre@samsung.com>
Jean-Philippe ANDRE [Sun, 2 Mar 2014 10:08:22 +0000 (19:08 +0900)]
Evas filters: fix black squares with the GL engine
If a text object changes regularily, there might be cases where
the object will be rendered as a simple black rectangle for just
one frame.
It seems that the previous output buffer is deleted before being
actually rendered on screen. This patch will delay the deletion
of the previous buffer until the current one has been rendered
to the target surface.
And again, thanks zmike for reporting.
@fix
Signed-off-by: Jean-Philippe Andre <jp.andre@samsung.com>
Jean-Philippe ANDRE [Sun, 2 Mar 2014 05:55:42 +0000 (14:55 +0900)]
Evas filters: Avoid CRI message when using the GL engine
A CRItical message was always displayed when setting a filter
on a text object, saying that proxy rendering is not supported on GL.
Reduce CRI to ERR and skip proxy rendering altogether if there are
no proxy sources.
This @fix needs to be backported.
Thanks zmike for reporting this.
Signed-off-by: Jean-Philippe Andre <jp.andre@samsung.com>
Jean-Philippe Andre [Mon, 24 Feb 2014 04:54:30 +0000 (13:54 +0900)]
Evas filters: Improve RGBA to alpha conversion (visual quality)
Add weights to the R,G,B components when converting to greyscale.
The weights are those used in the RGB to YCbCr conversion formula.
Jean-Philippe Andre [Fri, 21 Feb 2014 07:46:04 +0000 (16:46 +0900)]
Evas filters: Implement mapped blend for alpha buffers
Allow repeat fillmode in blend() for:
alpha --> alpha
alpha --> rgba
rgba --> alpha
Alpha scaling is not implemented yet, but it is not actually
required. Indeed, only proxies can have a different size and
proxies are RGBA images, not alpha.
Alpha scaling may or may not become a requirement in the future,
or for other purposes, but not yet.
Daniel Zaoui [Mon, 3 Mar 2014 16:01:53 +0000 (18:01 +0200)]
Eolian: Valgrind fixes
Daniel Zaoui [Mon, 3 Mar 2014 16:00:45 +0000 (18:00 +0200)]
Eolian: remove Ecore_File dependency
Chris Michael [Mon, 3 Mar 2014 16:07:18 +0000 (16:07 +0000)]
evas-drm: Fix evas drawing delay for cedric
@bugfix: Draw to the front buffer first, instead of the back buffer.
Frenchie says this improves the "initial rendering delay" of expedite
tests. Originally, we were drawing to the back buffer first, then
flipping it onto the crtc when drawing was done. This presented a
"perceived" rendering delay when running expedite tests as it would
wait for the back buffer to be drawn before presenting it. Personally,
I think it is not good to draw directly to the front buffer first as
it may get presented "incomplete" ... but cedric says it draws fine so
we'll leave it starting on the front buffer (for now).
Signed-off-by: Chris Michael <devilhorns@comcast.net>
Daniel Juyung Seo [Mon, 3 Mar 2014 15:01:48 +0000 (00:01 +0900)]
eina_list: Optimize eina_list_nth() call in case of index 0.
Check null and return the list data directly if the index 'n' is 0.
Amitesh Singh [Mon, 3 Mar 2014 12:29:34 +0000 (09:29 -0300)]
edje: add support of last input entered in password mode to be always visible in entry.
@feature
Requirement: Last input entered in password mode should be always visible in
entry. e.g. **a -> ***b.
Reviewers: cedric, raster, seoz
Reviewed By: cedric
CC: cedric
Differential Revision: https://phab.enlightenment.org/D589
Signed-off-by: Cedric Bail <cedric.bail@free.fr>
Daniel Zaoui [Mon, 3 Mar 2014 12:28:30 +0000 (14:28 +0200)]
Eolian: fix warnings.
Don't want Stefan to be angry ;-)
Daniel Zaoui [Mon, 3 Mar 2014 12:15:34 +0000 (14:15 +0200)]
Eolian: Merge of Eolian C generator
This patchset includes the parser for the meta-data (.eo) files and
the generator of C and H files.
.eo files are not included and will be added one after the other during
the splitting of the C files.
The APIs supplied by Eolian are beta APIs. Expect changes.
Yossi Kantor [Tue, 25 Feb 2014 15:27:38 +0000 (17:27 +0200)]
Eolian: New features/bugs fixes
1) Include files now have include guards
2) --gh option generates legacy header with --legacy flag and eo header
without --legacy flag
3) EOLIAN keyword is introduced to mark functions used by generated
file.
4) * for comments when comment text is empty
Tom Hacohen [Wed, 26 Feb 2014 15:42:15 +0000 (15:42 +0000)]
Eolian: Updated Eolian_Helper.
In order to use, you need to do as follows:
Add
include /src/Makefile_Eolian_Helper.am
To the Makefile.am that uses it.
Put .eo files in extra dist, e.g:
EXTRA_DIST = elm_code.eo
Add all the files you'd like to generate in BUILT_SOURCES, e.g:
BUILT_SOURCES = elm_code.eo.c elm_code.eo.h elm_code.eo.legacy.h
Yossi Kantor [Thu, 20 Feb 2014 12:47:46 +0000 (14:47 +0200)]
Eolian: New generation feature tested and bugs fixed
1) Underscore for static global names
2) Escape special characters for in-code descriptions
3) Proper spaces for generated lines
Daniel Zaoui [Sun, 23 Feb 2014 09:19:14 +0000 (11:19 +0200)]
Eolian: support non null parameters.
The generation is only needed in legacy headers.
Tom Hacohen [Fri, 21 Feb 2014 13:01:12 +0000 (13:01 +0000)]
Eolian: Fixed pc file to correctly generate.
First of all, it's now correctly generated, unlike before.
Second of all, it now lists eoincludedir, the directory for all
of the .eo include files.
Daniel Zaoui [Thu, 20 Feb 2014 14:46:23 +0000 (16:46 +0200)]
Eolian/Generator: fix data type for implement function