Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

In 1bpp mode, interpret skip&0x80 as "start a new line" instead of "go to next line", this is almost the same except for the first line which was always skipped before and caused to try to write an extra line at the end of the frame (ticket #226).
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Reviewed-by: Baptiste Coudurier

This reverts r6689, "Make PNG produce correct 8-bit pictures".

Inspired by a patch from Michael Root
Idea-by: Baptiste Coudurier
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* shariman/wmall:
  Fix lms_update()
  Move num_lms reading out of a loop
  Use correct value for range
  Fix some int / int16_t / int32_t confusion
  Implement revert_mclms() and associated functions
  Fix two more int16_t vs. int confusion
  Init s->cdlms[][].recent to order - 1
  Add a size argument to dump_int_buffer()
  Get rid of logging that are not required anymore
  Fix some int vs. int16_t confusion

Conflicts:
	libavcodec/wmalosslessdec.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

* qatar/master:
  Code cleanup - mpegvideo.c - 500-1000line
  rv40: NEON optimised weighted prediction
  rv40: NEON optimised chroma MC
  ARM: move NEON H264 chroma mc to a separate file
  rv34: NEON optimised inverse transform functions

Conflicts:
	libavcodec/mpegvideo.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Fixes Ticket711
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Reviewed-by: Roberto Togni

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>

The default of 9 gives different results on different FATE systems.
However the zlib test using compression level 6 works, so
try this instead.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>

Its checked a few lines below too.
The only difference is that empty atoms with size=0 will now get parsed too.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

was enough space to read the 32bit size.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

The computed size doesn't contain the header size because it's already
skipped by incrementing total_size, but then it's skipped again in the
last line. The atom comes out 8 bytes short and the function
mov_read_chan() aborts the whole parsing process. I think the computed
size should be atom.size - total_size + 8.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

This is just cosmetic as the if() is always true.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Reviewed-by: Baptiste Coudurier
Tested-with-QT: Baptiste Coudurier
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Address trac ticket #706.

Signed-off-by: Mans Rullgard <mans@mansr.com>

Signed-off-by: Mans Rullgard <mans@mansr.com>

This allows sharing code with the rv40 version of these functions.
Signed-off-by: Mans Rullgard <mans@mansr.com>

Signed-off-by: Mans Rullgard <mans@mansr.com>

* qatar/master:
  mov: Don't av_malloc(0).
  avconv: only allocate 1 AVFrame per input stream
  avconv: fix memleaks due to not freeing the AVFrame for audio
  h264-fate: remove -strict 1 except where necessary (mr4/5-tandberg).
  misc Doxygen markup improvements
  doxygen: eliminate Qt-style doxygen syntax
  g722: Add a regression test for muxing/demuxing in wav
  g722: Change bits per sample to 4
  g722dec: Signal skipping the lower bits via AVOptions instead of bits_per_coded_sample
  api-example: update to use avcodec_decode_audio4()
  avplay: use avcodec_decode_audio4()
  avplay: use a separate buffer for playing silence
  avformat: use avcodec_decode_audio4() in avformat_find_stream_info()
  avconv: use avcodec_decode_audio4() instead of avcodec_decode_audio3()
  mov: Allow empty stts atom.
  doc: document preferred Doxygen syntax and make patcheck detect it

Conflicts:
	avconv.c
	ffplay.c
	libavcodec/mlpdec.c
	libavcodec/version.h
	libavformat/mov.c
	tests/codec-regression.sh
	tests/fate/h264.mak
Merged-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

More specifically, PNG, v210, zlib and zmbv codecs.
zmbv needs vf_scale to be able to produce PAL8.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>

While quality is bad, PAL8 support is needed to allow testing some
encoders that only support PAL8 input.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>

For some of the code e.g. doing timing measurements there is no
real point in running regression testing on it, thus it should
not be counted against coverage.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>

Adds --enable-coverage to configure and a "coverage-html" make target.
The dependency stuff in the Makefile is a bit questionable, but the
best I could think of so far.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>

malloc() is allowed to return NULL when zero is the argument. This
causes us to think malloc has failed and return AVERROR(ENOMEM). In
addition OS X malloc() returns an unfreeable non-NULL pointer for size
zero when alignment is greater than 16.

This avoids allocating/freeing an AVFrame for each packet decode. It also
simplifies error handling in transcode_audio() and transcode_video().