GitLab

upstream status: imx specific

upstream status: i.Mx specific

when check whether video buffer is read only, the gst_buffer_get_memory call
should make sure idx don't exceed the total memory block number

upstream status: i.mx specific
Signed-off-by: Haihua Hu <jared.hu@nxp.com>

upstream status: Pending
https://bugzilla.gnome.org/show_bug.cgi?id=789952

set fd memory to keep mapped.

Upstream Status: Pending

https://bugzilla.gnome.org/show_bug.cgi?id=768794

This reverts commit c08618de.

This reverts commit 6065fd41.

This reverts commit 822c9f25.

upstream status: I.mx specific

As heap id and flags is platform dependent. Changed to set those
value by property.

Upstream Status: Pending

https://bugzilla.gnome.org/show_bug.cgi?id=768794

DMA Buf allocator based on ion.

Upstream Status: Pending

https://bugzilla.gnome.org/show_bug.cgi?id=768794

Add free function to free subclass and fd memory resource.

Upstream Status: Pending

https://bugzilla.gnome.org/show_bug.cgi?id=768794

- the orc optimization for lib video in plugins base may
cause segmentation fault
- disalbe orc optimization for lib video and just use the c source

package:  gstreamer1.0-plugins-base
Community ticket: https://bugzilla.gnome.org/show_bug.cgi?id=759286


Upstream status: pending
Signed-off-by: Lyon Wang <lyon.wang@freescale.com>

Add physical memory allocator

Upstream Status: Inappropriate [i.MX specific]
Signed-off-by: Haihua Hu <jared.hu@nxp.com>

image" item, preview, find the time can not display completely. 100%

As IPU need 8 pixels alignment, add one workaround in base text overlay
to generate 8 pixels alignment text video buffer. The side effect should
cause all text a little smaller.

Upstream Status: Inappropriate [i.MX specific]

Signed-off-by: Song Bing b06498@freescale.com

For those clips with corrupt audio track,
there might be no output from audio decoder
and thus the audio track have no chance to negotiate.
We can just print error warning instead of return ERROR,
so that other track can be played normally

https://bugzilla.gnome.org/show_bug.cgi?id=758215



Upstream Status:  Pending
Signed-off-by: Lyon Wang <lyon.wang@freescale.com>

When there is input data and no output data to the end of the stream, it will
send GST_ELEMENT_ERROR, So the clips playing will quit.
However, if only one of the tracks is corrupt, there is no need to quit other
tracks playing.

The patch comments the GST_ELEMENT_ERROR() and just add GST_ERROR_OBJECT()
information instead.

https://bugzilla.gnome.org/show_bug.cgi?id=741542



Upstream Status: Pending
Signed-off-by: Lyon Wang <lyon.wang@freescale.com>

https://bugzilla.gnome.org/show_bug.cgi?id=737246



Upstream status: Pending
Signed-off-by: Jian Li <lj.qfy.sh@gmail.com>

- remove default deinterlace flag in playbin for i.MX SoCs

Upstream Status: Inappropriate [platform specific]
Signed-off-by: Jian Li <jian.li@freescale.com>

https://bugzilla.gnome.org/show_bug.cgi?id=728312



Upstream Status: Pending
Signed-off-by: zhouming <b42586@freescale.com>

subparse works in push mode, chain funciton will be called once
up stream element finished the seeking and flushing.
if set need_segment flag in src pad event handler, the segment
event will be pushed earlier, result in the subtitle text will
be send out to down stream from the beginning.

Upstream Status: Inappropriate [i.MX specific]
Signed-off-by: Mingke Wang <mingke.wang@freescale.com>

some parser will pass in the original ssa text line which starts with "Dialog:"
and there's are maybe multiple Dialog lines in one input buffer.

Upstream Status: Inappropriate [i.MX specific]
Signed-off-by: Mingke Wang <mingke.wang@freescale.com>

set async to false lead to A/V sync problem when seeking.
the preroll need use GAP event instead of set async to false.

Upstream Status: Inappropriate [i.MX specific]
Signed-off-by: Mingke Wang <mingke.wang@freescale.com>

1. since gst_buffer_make_writable just lookup the refcount to determine if
   a buffer is writable, and it will use _gst_buffer_copy() which don't
   perform a deep memory copy even if the flag of a memory is set to
   GST_MEMORY_FLAG_READONLY. So, we detect the memory flag and use
   gst_buffer_copy_region with GST_BUFFER_COPY_DEEP parameter to perform
   deep memory copy. if the allocator of a memory don't support mem_copy
   interface, the it will return NULL, if this case, we can use
   gst_buffer_make_writable() to get a shared memory buffer or the orignal
   buffer if the buffer's refcount is 1.
Signed-off-by: Mingke Wang <mingke.wang@freescale.com>

https://bugzilla.gnome.org/show_bug.cgi?id=784639

If we are using a downstream bufferpool we need to set the size of the
buffer to our output size.

The missing annotations prevented proper usage from introspected
bindings like python.

https://bugzilla.gnome.org/show_bug.cgi?id=783330

And only set low-percent/high-percent if not using downloadbuffer, just
like in old uridecodebin. using the watermark based buffering causes
playback to hang never finish buffering with downloadbuffer.

Otherwise for RTSP streams for example, the pipeline will never go to
PLAYING as it will be missing an ASYNC_DONE message.

https://bugzilla.gnome.org/show_bug.cgi?id=780099

The base class is trying to align the processed data, but it endup
removing the GstVideoMeta. That caused wrong result. Instead, just copy
from the process function with the appropriate alignment.

https://bugzilla.gnome.org/show_bug.cgi?id=781204