summary |
shortlog | log |
commit |
commitdiff |
tree
first ⋅ prev ⋅ next
Sebastian Dröge [Tue, 12 Jul 2011 09:36:42 +0000 (11:36 +0200)]
omxvideodec: Make sink/src pad template caps configurable
Sebastian Dröge [Tue, 12 Jul 2011 09:13:50 +0000 (11:13 +0200)]
omx: Allow to set a preferred configuration directory with the GST_OMX_CONFIG_DIR environment variable
Sebastian Dröge [Tue, 12 Jul 2011 08:55:57 +0000 (10:55 +0200)]
omxvideodec: Make core/component-name and in/outport index configurable
Sebastian Dröge [Tue, 12 Jul 2011 08:05:31 +0000 (10:05 +0200)]
omx: Add initial version of configuration system
This now only registers elements that are specified in the
configuration file.
The configuration file is a keyfile in the first XDG configuration
directory with the name gstomx.conf.
Sebastian Dröge [Tue, 12 Jul 2011 06:53:15 +0000 (08:53 +0200)]
omxh264dec: Rename from omxh264videodec to omxh264dec
Sebastian Dröge [Tue, 12 Jul 2011 06:40:48 +0000 (08:40 +0200)]
omxh264videodec: Require alignment=au and stream-format={avc,bytestream}
Sebastian Dröge [Mon, 11 Jul 2011 10:59:07 +0000 (12:59 +0200)]
basevideodecoder: First inform subclass about resetting before resetting/freeing all internal state
The subclass might want to access the old state.
Sebastian Dröge [Mon, 11 Jul 2011 10:36:42 +0000 (12:36 +0200)]
basevideodecoder: Track present position on discont before resetting it
Sebastian Dröge [Mon, 11 Jul 2011 09:52:33 +0000 (11:52 +0200)]
basevideodecoder: Also protect the list of pending frames from concurrent access when pushing all pendings events
Sebastian Dröge [Mon, 11 Jul 2011 09:28:40 +0000 (11:28 +0200)]
basevideocodec: Protect access to the list of pending frames with the object lock
This is required if ::finish_frame() and all buffer output happens
on a different thread than the sinkpad streaming thread.
Sebastian Dröge [Mon, 11 Jul 2011 07:35:25 +0000 (09:35 +0200)]
basevideodecoder: Set the correct lists to NULL after freeing
Sebastian Dröge [Mon, 11 Jul 2011 06:54:53 +0000 (08:54 +0200)]
basevideodecoder: Work with a copy of the GstVideoState in setcaps until the caps are accepted
Also fix a refcount problem with the codec_data.
Sebastian Dröge [Tue, 12 Jul 2011 06:34:44 +0000 (08:34 +0200)]
omxh264videodec: Add h.264 video decoder
Sebastian Dröge [Tue, 12 Jul 2011 06:29:15 +0000 (08:29 +0200)]
omxmpeg4videodec: Fix debug category name
Sebastian Dröge [Sat, 9 Jul 2011 09:41:42 +0000 (11:41 +0200)]
basevideo: Move the utils from the codec header to its own header
Sebastian Dröge [Sat, 9 Jul 2011 09:32:06 +0000 (11:32 +0200)]
basevideo: Use GSlice for allocating GstVideoFrame and don't duplicate code in the decoder base class
Sebastian Dröge [Sat, 9 Jul 2011 09:05:37 +0000 (11:05 +0200)]
omxvideodec: Use the destroy notify to free the coder_hook
Sebastian Dröge [Sat, 9 Jul 2011 08:57:52 +0000 (10:57 +0200)]
basevideo: Add destroy notify for the coder_hook to prevent memory leaks
Fixes bug #654293.
Sebastian Dröge [Sat, 9 Jul 2011 08:44:16 +0000 (10:44 +0200)]
basevideo: Fix GType names to not conflict with the public video base classes
It's still not possible to include headers of both in the same file
or compile/link both into the same plugin but that shouldn't be
necessary anyway.
Sebastian Dröge [Fri, 8 Jul 2011 13:42:56 +0000 (15:42 +0200)]
omxvideodec: Fix some minor memory leaks
Sebastian Dröge [Fri, 8 Jul 2011 13:25:07 +0000 (15:25 +0200)]
omx: Rework port reconfiguration
We always reconfigure all ports now if the settings of one
port changes to prevent lots of race conditions, dropped
frames and similar issues.
Sebastian Dröge [Fri, 8 Jul 2011 11:16:45 +0000 (13:16 +0200)]
omxvideodec: Use the frames storage of the base class instead of implementing our own
They could get out of sync and we could store already destroyed frames.
Sebastian Dröge [Thu, 7 Jul 2011 10:51:03 +0000 (12:51 +0200)]
omx: Clarify GQueue/GPtrArray element types
Sebastian Dröge [Thu, 7 Jul 2011 10:23:24 +0000 (12:23 +0200)]
omx: Add more checks to acquire_buffer() and return the current state additional to the buffer
Also refactor the code flow in the video decoder for this. This makes
the usage of acquire_buffer() easier and more atomic.
Sebastian Dröge [Thu, 7 Jul 2011 10:22:57 +0000 (12:22 +0200)]
omxvideodec: Also flush/unflush the input port when changing the state PAUSED<->READY
Sebastian Dröge [Thu, 7 Jul 2011 10:21:31 +0000 (12:21 +0200)]
omx: Don't broadcast port->port_cond after allocating buffers successfully
Allocating buffers must happen while no thread is waiting for the
cond and especially must happen from the thread that would acquire
buffers from the port.
Sebastian Dröge [Thu, 7 Jul 2011 09:27:15 +0000 (11:27 +0200)]
omxvideodec: Don't leak the codec_data after sending it
Sebastian Dröge [Thu, 7 Jul 2011 08:27:31 +0000 (10:27 +0200)]
omx: Always check if the component is in an error state before waiting for a condition variable to be signalled
Otherwise we might wait forever because nothing is going to signal
the condition variable anymore.
Sebastian Dröge [Thu, 7 Jul 2011 08:22:12 +0000 (10:22 +0200)]
omx: Always hold port->port_lock before signalling port->port_cond when notifying about errors
Otherwise a port might be in the critical section, has checked the error state
already but waits after port->port_cond is signalled, which will lead
to a deadlock.
Sebastian Dröge [Thu, 7 Jul 2011 08:07:43 +0000 (10:07 +0200)]
omxvideodec: Remove reconfiguration test hack
Sebastian Dröge [Wed, 6 Jul 2011 11:27:12 +0000 (13:27 +0200)]
omx: Improve debug output a bit
Sebastian Dröge [Wed, 6 Jul 2011 11:26:51 +0000 (13:26 +0200)]
omx: Always try to deallocate buffers, even if there's a component error
Sebastian Dröge [Wed, 6 Jul 2011 11:26:01 +0000 (13:26 +0200)]
omx: Use G_USEC_PER_SEC for clarity instead of 1000000
Sebastian Dröge [Wed, 6 Jul 2011 11:19:15 +0000 (13:19 +0200)]
omxvideodec: Error out if the GStreamer allocated buffer is smaller than the OpenMAX output buffer
Usually this must never happen but currently it happens during reconfigurations
because of a race condition. Still it's better than crashing.
Sebastian Dröge [Wed, 6 Jul 2011 08:40:13 +0000 (10:40 +0200)]
omx: Don't use port_def.bEnabled to check if the Enable/Disable command is finished
bEnabled should be set immediately after sending the command, it's only
Bellagio that waits until the command is finished before setting it.
Sebastian Dröge [Wed, 6 Jul 2011 08:30:11 +0000 (10:30 +0200)]
omxvideodec: Remove obsolete FIXME comment
Sebastian Dröge [Wed, 6 Jul 2011 08:29:54 +0000 (10:29 +0200)]
omx: Improve error handling and reporting
Sebastian Dröge [Wed, 6 Jul 2011 06:48:37 +0000 (08:48 +0200)]
omxvideodec: Make the inport and outport index configurable by the subclass
Sebastian Dröge [Tue, 28 Jun 2011 06:51:23 +0000 (08:51 +0200)]
omx: Add initial version of OpenMAX framework, video decoder base class and MPEG4 video decoder
This currently hardcodes a lot of stuff but works at least.
Also adds a generic framework for handling OpenMAX cores, components
and ports.
Sebastian Dröge [Tue, 28 Jun 2011 09:47:25 +0000 (11:47 +0200)]
basevideodecoder: Don't reorder serialized src events
And allow to drop EOS by the subclass if ::finish returns
DROPPED.
Fixes bug #653544.
Sebastian Dröge [Mon, 27 Jun 2011 07:41:40 +0000 (09:41 +0200)]
basevideo: Add the caps to the GstVideoState and clean up caps/codec_data properly
Sebastian Dröge [Mon, 27 Jun 2011 07:37:03 +0000 (09:37 +0200)]
basevideo: Add video encoder/decoder base classes from gst-plugins-bad
Sebastian Dröge [Tue, 21 Jun 2011 09:17:35 +0000 (11:17 +0200)]
openmax: Add OpenMAX IL 1.1.2 headers
Sebastian Dröge [Tue, 21 Jun 2011 08:52:13 +0000 (10:52 +0200)]
Initial commit with build system