ffmpeg.git
3 years agodoc/filters.texi: ebur128 grammar fix
Kyle Swanson [Mon, 26 Oct 2015 22:44:33 +0000 (17:44 -0500)]
doc/filters.texi: ebur128 grammar fix

Reviewed-by: Lou Logan <lou@lrcd.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agofix: assigning instead of comparing
AppChecker [Mon, 26 Oct 2015 17:21:07 +0000 (20:21 +0300)]
fix: assigning instead of comparing

Signed-off-by: AppChecker <support.appc@cnpo.ru>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec: disallow hwaccel with frame threads
Hendrik Leppkes [Fri, 23 Oct 2015 11:54:25 +0000 (13:54 +0200)]
avcodec: disallow hwaccel with frame threads

HWAccels with frame threads are fundamentally flawed in avcodecs current
design, and there are several known problems ranging from image corruption
to driver crashes.

These problems come down to two design problems in the interaction of
threads and HWAccel decoding:

(1)
While avcodec prevents parallel decoding and as such simultaneous access
to the hardware accelerator from the decoding threads, it cannot account
for the user code and its access to the hardware surfaces and the hardware
itself.

This can result in image corruption or even driver crashes if the
user code locks image surfaces while they are being used by the decoder
threads as reference frames.

The current HWAccel API does not offer any way to ensure exclusive access
to the hardware or the surfaces if frame threading is used.

(2)
Initialization of the HWAccel with frame threads is non-trivial, and many
decoders had and still have issues that cause excess calls to the
get_format callback.

This will potentially cause duplicate HWAccel initialization, which in
extreme cases can even lead to driver crashes if the HWAccel is
re-initialized while the user code is actively accessing the hardware
surfaces associated with it, or lead to image corruption due to lost
reference frames.

While both of these issues are solvable, fixing (1) would at least require
a huge API redesign which would move a lot of complexity into the user
code.

The only reason the combination of frame threads and HWAccel was
considered useful is to allow a seamless fallback to multi-threaded
software decoding if the HWAccel is not available, however the issues
outlined above far outweigh this.

The proper solution for a fallback is to re-open the AVCodecContext with
threading enabled if the HWAccel failed, which is a practice commonly used
by various user applications using avcodec today already.

Reviewed-by: Gwenole Beauchesne <gb.devel@gmail.com>
Reviewed-by: wm4 <nfxjfg@googlemail.com>
Signed-off-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoFATE: Increase FUZZ value on AAC LTP encoding test
Rostislav Pehlivanov [Tue, 27 Oct 2015 07:41:33 +0000 (07:41 +0000)]
FATE: Increase FUZZ value on AAC LTP encoding test

Clang prerelease on Darwin is making the test fail.

3 years agosrtenc: Reindent
Timothy Gu [Tue, 27 Oct 2015 06:24:45 +0000 (23:24 -0700)]
srtenc: Reindent

3 years agodrawutils: Reindent
Timothy Gu [Tue, 27 Oct 2015 06:24:30 +0000 (23:24 -0700)]
drawutils: Reindent

3 years agoavfilter: Reindent
Timothy Gu [Tue, 27 Oct 2015 06:24:18 +0000 (23:24 -0700)]
avfilter: Reindent

3 years agoavcodec: srtdec: Reindent
Timothy Gu [Tue, 27 Oct 2015 06:23:59 +0000 (23:23 -0700)]
avcodec: srtdec: Reindent

3 years agox86/vf_w3fdif: use aligned loads in w3fdif_complex_high
James Almer [Tue, 27 Oct 2015 04:49:22 +0000 (01:49 -0300)]
x86/vf_w3fdif: use aligned loads in w3fdif_complex_high

Found-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
3 years agoFATE: Slightly increase thresholds on prediction AAC encoding tests
Rostislav Pehlivanov [Tue, 27 Oct 2015 03:48:17 +0000 (03:48 +0000)]
FATE: Slightly increase thresholds on prediction AAC encoding tests

They barely fail on some systems by being off by 0.81.

3 years agoopusdec: Don't run vector_fmul_scalar on zero length arrays
Kieran Kunhya [Mon, 26 Oct 2015 23:09:44 +0000 (23:09 +0000)]
opusdec: Don't run vector_fmul_scalar on zero length arrays

Fixes crashes on fuzzed files
Fixes Ticket4969 part2

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/opusdec: Fix extra samples read index
Michael Niedermayer [Tue, 27 Oct 2015 00:18:22 +0000 (01:18 +0100)]
avcodec/opusdec: Fix extra samples read index

Fixes crash
Fixes Ticket4969 part 1

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter: add vibrato filter
Kyle Swanson [Sat, 24 Oct 2015 05:08:19 +0000 (00:08 -0500)]
avfilter: add vibrato filter

Signed-off-by: Kyle Swanson <k@ylo.ph>
3 years agoavcodec/options: remove a few more redundant "default" information
Clément Bœsch [Mon, 26 Oct 2015 10:18:37 +0000 (11:18 +0100)]
avcodec/options: remove a few more redundant "default" information

3 years agoavcodec/options: remove redundant and wrong default information for skipcmp option
Clément Bœsch [Mon, 26 Oct 2015 10:17:03 +0000 (11:17 +0100)]
avcodec/options: remove redundant and wrong default information for skipcmp option

3 years agoavcodec/avdct: remove redundant "default" information in options
Clément Bœsch [Mon, 26 Oct 2015 10:13:31 +0000 (11:13 +0100)]
avcodec/avdct: remove redundant "default" information in options

3 years agotests/aac: Add bitexact flags to AAC LTP Encode test
Derek Buitenhuis [Mon, 26 Oct 2015 14:31:21 +0000 (14:31 +0000)]
tests/aac: Add bitexact flags to AAC LTP Encode test

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
3 years agomov: Add support parsing QuickTime Metadata Keys.
Tinglin Liu [Thu, 22 Oct 2015 22:04:46 +0000 (15:04 -0700)]
mov: Add support parsing QuickTime Metadata Keys.

The Apple dev specification:
    https://developer.apple.com/library/mac/documentation/QuickTime/QTFF/Metadata/Metadata.html

Basically the structure is like:
    |--meta
    |----hdlr
    |----keys
    |----ilst

1) The handler type in the metadata handler atom is ‘mdta’.
2) The key and value are stored separately for each key-value pair.
   The 'keys' atom stores the key table, while 'ilst' atom stores the
   values corresponding to the indices in the key table.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
3 years agotimecode: Support HFR values
Vittorio Giovara [Mon, 26 Oct 2015 13:52:40 +0000 (14:52 +0100)]
timecode: Support HFR values

3 years agotimecode: Do not fail for non-standard framerates
Vittorio Giovara [Mon, 26 Oct 2015 13:44:43 +0000 (14:44 +0100)]
timecode: Do not fail for non-standard framerates

Instead just warn, and use the parse fps normally.

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>
3 years agotests/fate/aac: Add bitexact flags to fate-aac-pns-encode
Michael Niedermayer [Mon, 26 Oct 2015 12:14:24 +0000 (13:14 +0100)]
tests/fate/aac: Add bitexact flags to fate-aac-pns-encode

This fixes a fate failure after bumping the minor version
Its unknown why this is not needed for the other aac tests,
more investigation needed but for now i dont want to leave
it broken while its investigated

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter/vf_removegrain: replace qsort with AV_QSORT
Ganesh Ajjanagadde [Sun, 25 Oct 2015 19:06:01 +0000 (15:06 -0400)]
avfilter/vf_removegrain: replace qsort with AV_QSORT

filter_slice calls qsort, so qsort is in a performance critical
position. AV_QSORT is substantially faster due to the inlining of the
comparison callback. Thus, the increase in performance is worth the
increase in binary size.

Sample benchmark (x86-64, Haswell, GNU/Linux),
filter-removegrain-mode-02 (from FATE)
new:
  24060 decicycles in qsort,       1 runs,      0 skips
  15690 decicycles in qsort,       2 runs,      0 skips
   9307 decicycles in qsort,       4 runs,      0 skips
   5572 decicycles in qsort,       8 runs,      0 skips
   3485 decicycles in qsort,      16 runs,      0 skips
   2517 decicycles in qsort,      32 runs,      0 skips
   1979 decicycles in qsort,      64 runs,      0 skips
   1911 decicycles in qsort,     128 runs,      0 skips
   1568 decicycles in qsort,     256 runs,      0 skips
   1596 decicycles in qsort,     512 runs,      0 skips
   1614 decicycles in qsort,    1024 runs,      0 skips
   1874 decicycles in qsort,    2046 runs,      2 skips
   2186 decicycles in qsort,    4094 runs,      2 skips

old:
 246960 decicycles in qsort,       1 runs,      0 skips
 135765 decicycles in qsort,       2 runs,      0 skips
  70920 decicycles in qsort,       4 runs,      0 skips
  37710 decicycles in qsort,       8 runs,      0 skips
  20831 decicycles in qsort,      16 runs,      0 skips
  12225 decicycles in qsort,      32 runs,      0 skips
   8083 decicycles in qsort,      64 runs,      0 skips
   6270 decicycles in qsort,     128 runs,      0 skips
   5321 decicycles in qsort,     256 runs,      0 skips
   4860 decicycles in qsort,     512 runs,      0 skips
   4424 decicycles in qsort,    1024 runs,      0 skips
   4191 decicycles in qsort,    2046 runs,      2 skips
   4934 decicycles in qsort,    4094 runs,      2 skips

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoavformat/thp: set duration for audio stream too
Paul B Mahol [Sun, 25 Oct 2015 22:07:21 +0000 (23:07 +0100)]
avformat/thp: set duration for audio stream too

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavcodec: add SDX2 DPCM decoder
Paul B Mahol [Sat, 24 Oct 2015 18:41:32 +0000 (20:41 +0200)]
avcodec: add SDX2 DPCM decoder

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoMAINTAINERS: add key fingerprint
Ganesh Ajjanagadde [Sun, 25 Oct 2015 01:59:56 +0000 (21:59 -0400)]
MAINTAINERS: add key fingerprint

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoavutil/opt: print more meaningful default flags values
Clément Bœsch [Sat, 24 Oct 2015 13:14:33 +0000 (15:14 +0200)]
avutil/opt: print more meaningful default flags values

Example:
% ./ffmpeg -h encoder=gif
[...]
GIF encoder AVOptions:
  -gifflags          <flags>      E..V.... set GIF flags (default offsetting+transdiff)
     offsetting                   E..V.... enable picture offsetting
     transdiff                    E..V.... enable transparency detection between frames

3 years agolavfi/drawutils: add const to blending mask.
Nicolas George [Sun, 25 Oct 2015 10:27:15 +0000 (11:27 +0100)]
lavfi/drawutils: add const to blending mask.

3 years agoavutil/tree: clean up pointer incompatibility warnings
Ganesh Ajjanagadde [Sun, 25 Oct 2015 16:41:48 +0000 (12:41 -0400)]
avutil/tree: clean up pointer incompatibility warnings

Commit 7c8fcbbde3a299096974f9061c8b5be0e526f4c2 introduced some warnings
that get triggered on the test build. This should fix them.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agolavf/xwma: Support wmapro.
Carl Eugen Hoyos [Sun, 25 Oct 2015 16:37:17 +0000 (17:37 +0100)]
lavf/xwma: Support wmapro.

Fixes ticket #4963.

3 years agoavutil/tree: Document the guaranteed ordering of compare arguments for av_tree_find()
Michael Niedermayer [Sun, 25 Oct 2015 16:13:42 +0000 (17:13 +0100)]
avutil/tree: Document the guaranteed ordering of compare arguments for av_tree_find()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/huffman: replace qsort with AV_QSORT
Ganesh Ajjanagadde [Thu, 22 Oct 2015 23:54:31 +0000 (19:54 -0400)]
avcodec/huffman: replace qsort with AV_QSORT

ff_huff_build_tree uses qsort underneath. AV_QSORT is substantially
faster due to the inlining of the comparison callback. Furthermore, this
code is reasonably performance critical, since in e.g the fraps codec,
ff_huff_build_tree is called on every frame. This routine is also called
in vp6 on every frame in some circumstances.

Sample benchmark (x86-64, Haswell, GNU/Linux), vp6 from FATE:
vp6 (old):
  78930 decicycles in qsort,       1 runs,      0 skips
  45330 decicycles in qsort,       2 runs,      0 skips
  27825 decicycles in qsort,       4 runs,      0 skips
  17471 decicycles in qsort,       8 runs,      0 skips
  12296 decicycles in qsort,      16 runs,      0 skips
   9554 decicycles in qsort,      32 runs,      0 skips
   8404 decicycles in qsort,      64 runs,      0 skips
   7405 decicycles in qsort,     128 runs,      0 skips
   6740 decicycles in qsort,     256 runs,      0 skips
   7540 decicycles in qsort,     512 runs,      0 skips
   9498 decicycles in qsort,    1024 runs,      0 skips
   9938 decicycles in qsort,    2048 runs,      0 skips
   8043 decicycles in qsort,    4095 runs,      1 skips

vp6 (new):
  15880 decicycles in qsort,       1 runs,      0 skips
  10730 decicycles in qsort,       2 runs,      0 skips
  10155 decicycles in qsort,       4 runs,      0 skips
   7805 decicycles in qsort,       8 runs,      0 skips
   6883 decicycles in qsort,      16 runs,      0 skips
   6305 decicycles in qsort,      32 runs,      0 skips
   5854 decicycles in qsort,      64 runs,      0 skips
   5152 decicycles in qsort,     128 runs,      0 skips
   4452 decicycles in qsort,     256 runs,      0 skips
   4161 decicycles in qsort,     511 runs,      1 skips
   4081 decicycles in qsort,    1023 runs,      1 skips
   4072 decicycles in qsort,    2047 runs,      1 skips
   4004 decicycles in qsort,    4095 runs,      1 skips

Reviewed-by: Timothy Gu <timothygu99@gmail.com>
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoversion.sh: add note that ffversion.h is auto-generated
Ganesh Ajjanagadde [Thu, 15 Oct 2015 23:03:25 +0000 (19:03 -0400)]
version.sh: add note that ffversion.h is auto-generated

Reviewed-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoall: add const-correctness to qsort comparators
Ganesh Ajjanagadde [Sun, 25 Oct 2015 00:59:05 +0000 (20:59 -0400)]
all: add const-correctness to qsort comparators

This adds const-correctness when needed for the comparators.

Reviewed-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoavformat/aiffdec: give friendly message if compressed codec tag is unsupported
Paul B Mahol [Sun, 25 Oct 2015 11:48:36 +0000 (12:48 +0100)]
avformat/aiffdec: give friendly message if compressed codec tag is unsupported

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agodoc/general: update after recent additions
Paul B Mahol [Sun, 25 Oct 2015 10:51:15 +0000 (11:51 +0100)]
doc/general: update after recent additions

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavformat: add xvag demuxer
Paul B Mahol [Fri, 23 Oct 2015 10:03:28 +0000 (12:03 +0200)]
avformat: add xvag demuxer

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavutil/tree: add additional const qualifier to the comparator
Ganesh Ajjanagadde [Sat, 24 Oct 2015 21:53:21 +0000 (17:53 -0400)]
avutil/tree: add additional const qualifier to the comparator

libc's qsort comparator has a const qualifier on both arguments. This
adds a missing const qualifier to exactly match the comparator API.

Existing usages of av_tree_find, av_tree_insert are appropriately
modified: type signature changes of the comparators, and removal of
unnecessary void * casts of function pointers.

Reviewed-by: Henrik Gramner <henrik@gramner.com>
Reviewed-by: wm4 <nfxjfg@googlemail.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoavutil/tree: improve documentation for av_tree_find, av_tree_insert
Ganesh Ajjanagadde [Sat, 24 Oct 2015 21:47:21 +0000 (17:47 -0400)]
avutil/tree: improve documentation for av_tree_find, av_tree_insert

This documents the additional constness, and provides a useful libc
reference for the API specification of the comparator.

Reviewed-by: Henrik Gramner <henrik@gramner.com>
Reviewed-by: wm4 <nfxjfg@googlemail.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoavfilter/vf_deshake: use a void * comparator for consistency
Ganesh Ajjanagadde [Sat, 24 Oct 2015 21:49:57 +0000 (17:49 -0400)]
avfilter/vf_deshake: use a void * comparator for consistency

For generality, qsort uses a comparator whose elements are void *. This
makes the comparator have such a form, and thus makes the void * cast of
the comparator pointer useless. Furthermore, this makes the code more
consistent with other usages of qsort across the codebase.

Reviewed-by: Henrik Gramner <henrik@gramner.com>
Reviewed-by: wm4 <nfxjfg@googlemail.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoall: fix -Wextra-semi reported on clang
Ganesh Ajjanagadde [Fri, 23 Oct 2015 15:23:42 +0000 (11:23 -0400)]
all: fix -Wextra-semi reported on clang

This fixes extra semicolons that clang 3.7 on GNU/Linux warns about.
These were trigggered when built under -Wpedantic, which essentially
checks for strict ISO compliance in numerous ways.

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoall: remove some casts of function pointer to void *
Ganesh Ajjanagadde [Fri, 23 Oct 2015 15:46:41 +0000 (11:46 -0400)]
all: remove some casts of function pointer to void *

These casts are unnecessary, and may safely be removed.
Found by enabling -Wpedantic on clang 3.7.

Tested with FATE.

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agovp9: uses ff_set_dimensions (which sets coded_width/height).
Ronald S. Bultje [Sat, 24 Oct 2015 13:24:41 +0000 (09:24 -0400)]
vp9: uses ff_set_dimensions (which sets coded_width/height).

Fixes ticket 4935.

3 years agovideodsp: don't overread edges in vfix3 emu_edge.
Ronald S. Bultje [Fri, 23 Oct 2015 15:11:53 +0000 (11:11 -0400)]
videodsp: don't overread edges in vfix3 emu_edge.

Fixes trac ticket 3226. Also see Andreas' analysis in
https://bugs.debian.org/801745, which was very helpful.

3 years agoavfilter: avoid zero arguments to variadic macro
Ganesh Ajjanagadde [Fri, 23 Oct 2015 16:36:52 +0000 (12:36 -0400)]
avfilter: avoid zero arguments to variadic macro

ISO C requires at least one argument in the place of the ellipsis in a
variadic macro. In particular, under -pedantic, this triggers the
warning -Wgnu-zero-variadic-macro-arguments on clang 3.7.

Reviewed-by: Nicolas George <george@nsup.org>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agolibzvbi-teletextdec: fix AVSubtitleRect pict compatiblity code
Marton Balint [Fri, 23 Oct 2015 18:40:23 +0000 (20:40 +0200)]
libzvbi-teletextdec: fix AVSubtitleRect pict compatiblity code

Only set pict if we got a valid rect.

Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agoffplay: use a separate struct for the rescaled YUVA AVSubtitle rectangles
Marton Balint [Fri, 23 Oct 2015 18:32:51 +0000 (20:32 +0200)]
ffplay: use a separate struct for the rescaled YUVA AVSubtitle rectangles

Current code segfaults since the deprecation of AVSubtitleRect.pict because it
freed/realloced AVSubtitleRect.pict.data by itself.

The new code stores the generated YUVA AVSubtitle rectangles in their own
struct and keeps the original AVSubtitle structure untouched, because
overwriting it is considered invalid API usage.

Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agolavf/ingenientdec: Add a probe function.
Carl Eugen Hoyos [Sat, 24 Oct 2015 11:28:32 +0000 (13:28 +0200)]
lavf/ingenientdec: Add a probe function.

3 years agolavc/hapdec: Use correct no-transparency colour space.
Carl Eugen Hoyos [Sat, 24 Oct 2015 11:09:58 +0000 (13:09 +0200)]
lavc/hapdec: Use correct no-transparency colour space.

Reviewed-by: Tom Butterworth
3 years agoavcodec/dpxenc: Fix "libavcodec/dpxenc.c:250:44: warning: passing argument 3 of av_im...
Michael Niedermayer [Sat, 24 Oct 2015 10:06:22 +0000 (12:06 +0200)]
avcodec/dpxenc: Fix "libavcodec/dpxenc.c:250:44: warning: passing argument 3 of av_image_copy_to_buffer from incompatible pointer type"

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavformat/genh: add ADPCM IMA DVI support
Paul B Mahol [Fri, 23 Oct 2015 19:09:49 +0000 (21:09 +0200)]
avformat/genh: add ADPCM IMA DVI support

Our decoder have wrong name.

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavformat/ads: extend format long description
Paul B Mahol [Fri, 23 Oct 2015 17:16:17 +0000 (19:16 +0200)]
avformat/ads: extend format long description

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavformat/svag: extend format long description
Paul B Mahol [Fri, 23 Oct 2015 17:12:32 +0000 (19:12 +0200)]
avformat/svag: extend format long description

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavcodec: extend long decription for adpcm psx codec
Paul B Mahol [Fri, 23 Oct 2015 17:07:37 +0000 (19:07 +0200)]
avcodec: extend long decription for adpcm psx codec

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavformat/msf: extend format long description
Paul B Mahol [Fri, 23 Oct 2015 17:06:14 +0000 (19:06 +0200)]
avformat/msf: extend format long description

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoAdd myself as maintainer for Hap
Tom Butterworth [Fri, 23 Oct 2015 11:04:56 +0000 (12:04 +0100)]
Add myself as maintainer for Hap

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoRevert "configure: add -Wstrict-prototypes when available"
Ganesh Ajjanagadde [Fri, 23 Oct 2015 13:10:01 +0000 (09:10 -0400)]
Revert "configure: add -Wstrict-prototypes when available"

This reverts commit e6a93e59adeea1c140f2ef63e4055a15823bfeda,
Wstrict-prototypes is already enabled.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoconfigure: add -Wstrict-prototypes when available
Ganesh Ajjanagadde [Thu, 22 Oct 2015 18:37:31 +0000 (14:37 -0400)]
configure: add -Wstrict-prototypes when available

GCC (and Clang) have this useful warning that is not enabled by -Wall or
-Wextra. This will ensure that issues like those fixed in
4da52e3630343e8d3a79aef2cafcb6bf0b71e8da
will trigger warnings.

Reviewed-by: Hendrik Leppkes <h.leppkes@gmail.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoavutil/qsort: use the do while form for AV_QSORT, AV_MSORT
Ganesh Ajjanagadde [Fri, 23 Oct 2015 02:52:11 +0000 (22:52 -0400)]
avutil/qsort: use the do while form for AV_QSORT, AV_MSORT

Reviewed-by: Clément Bœsch <u@pkh.me>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agovf_psnr/ssim: don't crash if stats_file is NULL.
Ronald S. Bultje [Thu, 22 Oct 2015 22:00:40 +0000 (18:00 -0400)]
vf_psnr/ssim: don't crash if stats_file is NULL.

3 years agoavcodec/hap: set bits_per_coded_sample
Tom Butterworth [Thu, 22 Oct 2015 22:40:41 +0000 (23:40 +0100)]
avcodec/hap: set bits_per_coded_sample

fixes issue where alpha is ignored in some players

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter: add shuffleframes filter
Paul B Mahol [Tue, 20 Oct 2015 08:26:54 +0000 (10:26 +0200)]
avfilter: add shuffleframes filter

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavformat/mov: Autodetect mp3s which need parsing
Michael Niedermayer [Fri, 23 Oct 2015 01:22:43 +0000 (03:22 +0200)]
avformat/mov: Autodetect mp3s which need parsing

mp3 packets all have the same duration and number of samples
if their duration indicated in the container varies then thats an
indication that they are not 1 mp3 packet each.
If this autodetection fails for some case then please contact us
and provide a testcase.

Fixes Ticket4938

3 years agoavdevice/pulse_audio_common: add av_warn_unused_result
Ganesh Ajjanagadde [Fri, 16 Oct 2015 01:52:01 +0000 (21:52 -0400)]
avdevice/pulse_audio_common: add av_warn_unused_result

This does not trigger any warnings, but adds robustness.

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agovf_ssim: print per-channel dB values.
Ronald S. Bultje [Thu, 22 Oct 2015 18:51:11 +0000 (14:51 -0400)]
vf_ssim: print per-channel dB values.

3 years agovf_psnr: remove %0.2f format specifiers for stream summary line.
Ronald S. Bultje [Thu, 22 Oct 2015 19:25:38 +0000 (15:25 -0400)]
vf_psnr: remove %0.2f format specifiers for stream summary line.

This makes output equally precise as vf_ssim.

3 years agoavcodec/libzvbi-teletextdec: Add variable to fix build
Michael Niedermayer [Thu, 22 Oct 2015 21:00:02 +0000 (23:00 +0200)]
avcodec/libzvbi-teletextdec: Add variable to fix build

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter,swresample,swscale: use fabs, fabsf instead of FFABS
Ganesh Ajjanagadde [Mon, 12 Oct 2015 05:30:22 +0000 (01:30 -0400)]
avfilter,swresample,swscale: use fabs, fabsf instead of FFABS

It is well known that fabs and fabsf are at least as fast and sometimes
faster than the FFABS macro, at least on the gcc+glibc combination.
For instance, see the reference:
http://patchwork.sourceware.org/patch/6735/.
This was a patch to glibc in order to remove their usages of a macro.

The reason essentially boils down to fabs using the __builtin_fabs of
the compiler, while FFABS needs to infer to not use a branch and to
simply change the sign bit. Usually the inference works, but sometimes
it does not. This may be easily checked by looking at the asm.

This also has the added benefit of reducing macro usage, which has
problems with side-effects.

Note that avcodec is not handled here, as it is huge and
most things there are integer arithmetic anyway.

Tested with FATE.

Reviewed-by: Clément Bœsch <u@pkh.me>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agoMerge commit '3ee2c60cc296eee3f63d7b5fee9b4332eeeac9fa'
Hendrik Leppkes [Thu, 22 Oct 2015 19:46:56 +0000 (21:46 +0200)]
Merge commit '3ee2c60cc296eee3f63d7b5fee9b4332eeeac9fa'

* commit '3ee2c60cc296eee3f63d7b5fee9b4332eeeac9fa':
  utils: Use data buffers directly instead of an AVPicture

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit 'ff7956fcbf8e59b21654b95038de3ed88a850a9e'
Hendrik Leppkes [Thu, 22 Oct 2015 19:45:49 +0000 (21:45 +0200)]
Merge commit 'ff7956fcbf8e59b21654b95038de3ed88a850a9e'

* commit 'ff7956fcbf8e59b21654b95038de3ed88a850a9e':
  avplay: Replace avpicture functions with imgutils

avplay and ffplay have diverged quite a bit, and ffplay should be updated
independently.

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoavcodec/libzvbi: Update for AVSubtitleRect changes
Hendrik Leppkes [Thu, 22 Oct 2015 19:43:15 +0000 (21:43 +0200)]
avcodec/libzvbi: Update for AVSubtitleRect changes

3 years agoMerge commit 'a17a7661906ba295d67afd80ac0770422e1b02b3'
Hendrik Leppkes [Thu, 22 Oct 2015 19:18:03 +0000 (21:18 +0200)]
Merge commit 'a17a7661906ba295d67afd80ac0770422e1b02b3'

* commit 'a17a7661906ba295d67afd80ac0770422e1b02b3':
  lavc: Add data and linesize to AVSubtitleRect

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoavcodec/nuv: Fix 'libavcodec/nuv.c:83:19: warning: passing argument 3 of av_image_cop...
Michael Niedermayer [Thu, 22 Oct 2015 18:56:32 +0000 (20:56 +0200)]
avcodec/nuv: Fix 'libavcodec/nuv.c:83:19: warning: passing argument 3 of av_image_copy from incompatible pointer type'

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/x86/vc1dsp: Remove unused macro
Michael Niedermayer [Thu, 22 Oct 2015 19:13:42 +0000 (21:13 +0200)]
avcodec/x86/vc1dsp: Remove unused macro

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoMerge commit 'f890677d05bc4e8b494a73373ab4cc19791bf884'
Hendrik Leppkes [Thu, 22 Oct 2015 18:42:28 +0000 (20:42 +0200)]
Merge commit 'f890677d05bc4e8b494a73373ab4cc19791bf884'

* commit 'f890677d05bc4e8b494a73373ab4cc19791bf884':
  Replace any remaining avpicture function with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoavfilter/vf_zscale: fix typo
Lou Logan [Thu, 22 Oct 2015 18:39:57 +0000 (10:39 -0800)]
avfilter/vf_zscale: fix typo

Fixes #4958 as found by nicol.

Signed-off-by: Lou Logan <lou@lrcd.com>
3 years agoMerge commit '13bddab7de10aebf6efb98aa6d7ff0c51bb0e364'
Hendrik Leppkes [Thu, 22 Oct 2015 18:32:07 +0000 (20:32 +0200)]
Merge commit '13bddab7de10aebf6efb98aa6d7ff0c51bb0e364'

* commit '13bddab7de10aebf6efb98aa6d7ff0c51bb0e364':
  nuv: Replace avpicture functions with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit '48c06386831604921bdaf4fb77ea02766cd615f4'
Hendrik Leppkes [Thu, 22 Oct 2015 18:31:09 +0000 (20:31 +0200)]
Merge commit '48c06386831604921bdaf4fb77ea02766cd615f4'

* commit '48c06386831604921bdaf4fb77ea02766cd615f4':
  dpx: Replace avpicture functions with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoroqvideodec: use av_frame_copy
Hendrik Leppkes [Thu, 22 Oct 2015 18:18:00 +0000 (20:18 +0200)]
roqvideodec: use av_frame_copy

3 years agoMerge commit 'f0a106578d759de6183eea3c75f8373b6d3153c1'
Hendrik Leppkes [Thu, 22 Oct 2015 18:11:21 +0000 (20:11 +0200)]
Merge commit 'f0a106578d759de6183eea3c75f8373b6d3153c1'

* commit 'f0a106578d759de6183eea3c75f8373b6d3153c1':
  roqvideodec: Replace avpicture functions with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit 'ef3a3519c10620c4206738595bf03fc0bed71802'
Hendrik Leppkes [Thu, 22 Oct 2015 18:08:26 +0000 (20:08 +0200)]
Merge commit 'ef3a3519c10620c4206738595bf03fc0bed71802'

* commit 'ef3a3519c10620c4206738595bf03fc0bed71802':
  rawdec: Replace avpicture functions with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit 'fcc1280acb6e6f682b34c2101b075b82f83d71ba'
Hendrik Leppkes [Thu, 22 Oct 2015 18:02:33 +0000 (20:02 +0200)]
Merge commit 'fcc1280acb6e6f682b34c2101b075b82f83d71ba'

* commit 'fcc1280acb6e6f682b34c2101b075b82f83d71ba':
  rawenc: Replace avpicture functions with imgutils

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit '3496cec43304ac040d6d05f9d500a6f93cc049e7'
Hendrik Leppkes [Thu, 22 Oct 2015 17:42:29 +0000 (19:42 +0200)]
Merge commit '3496cec43304ac040d6d05f9d500a6f93cc049e7'

* commit '3496cec43304ac040d6d05f9d500a6f93cc049e7':
  msrle: Use AVFrame instead of AVPicture

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit 'ac981d16415e7fd99683e10297781c7d9ec1a8cd'
Hendrik Leppkes [Thu, 22 Oct 2015 17:41:52 +0000 (19:41 +0200)]
Merge commit 'ac981d16415e7fd99683e10297781c7d9ec1a8cd'

* commit 'ac981d16415e7fd99683e10297781c7d9ec1a8cd':
  APIchanges: Fill in missing dates and hashes

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoffmpeg: add abort_on option to allow aborting on empty output
Marton Balint [Sat, 17 Oct 2015 21:46:09 +0000 (23:46 +0200)]
ffmpeg: add abort_on option to allow aborting on empty output

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agoffmpeg: fix ffmpeg.h trailing whitespace
Marton Balint [Thu, 22 Oct 2015 03:26:46 +0000 (05:26 +0200)]
ffmpeg: fix ffmpeg.h trailing whitespace

How this passed through the commit hook?

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agoqsvdec: fix get_format and hwaccel_context handling
Hendrik Leppkes [Thu, 22 Oct 2015 15:00:49 +0000 (17:00 +0200)]
qsvdec: fix get_format and hwaccel_context handling

This enables the qsv transcoder to actually get activated

3 years agoMerge commit 'fb472e1a11a4e0caed2c3c91da01ea8e35d9e3f8'
Hendrik Leppkes [Thu, 22 Oct 2015 14:18:02 +0000 (16:18 +0200)]
Merge commit 'fb472e1a11a4e0caed2c3c91da01ea8e35d9e3f8'

* commit 'fb472e1a11a4e0caed2c3c91da01ea8e35d9e3f8':
  avconv: add support for Intel QSV-accelerated transcoding

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit 'dc923bc23b3efd949d0bf67ff1abdb95059e5843'
Hendrik Leppkes [Thu, 22 Oct 2015 13:54:14 +0000 (15:54 +0200)]
Merge commit 'dc923bc23b3efd949d0bf67ff1abdb95059e5843'

* commit 'dc923bc23b3efd949d0bf67ff1abdb95059e5843':
  qsvenc: add an API for allocating opaque surfaces

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit '2ec96b6bd5bf7b22978711bcf2cee702bee89c6f'
Hendrik Leppkes [Thu, 22 Oct 2015 13:49:26 +0000 (15:49 +0200)]
Merge commit '2ec96b6bd5bf7b22978711bcf2cee702bee89c6f'

* commit '2ec96b6bd5bf7b22978711bcf2cee702bee89c6f':
  qsvenc: cosmetics, reindent

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit 'f6f32fc93d39caf329869c1bff8ad024ccab1d42'
Hendrik Leppkes [Thu, 22 Oct 2015 13:48:11 +0000 (15:48 +0200)]
Merge commit 'f6f32fc93d39caf329869c1bff8ad024ccab1d42'

* commit 'f6f32fc93d39caf329869c1bff8ad024ccab1d42':
  qsvenc: set the timestamp for PIX_FMT_QSV frames as well

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit '772c87c5a658f36d7c0612f5da583fc2bfa54f79'
Hendrik Leppkes [Thu, 22 Oct 2015 13:46:52 +0000 (15:46 +0200)]
Merge commit '772c87c5a658f36d7c0612f5da583fc2bfa54f79'

* commit '772c87c5a658f36d7c0612f5da583fc2bfa54f79':
  qsvenc: support passing arbitrary external buffers to the encoder

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoavutil/intmath: fix undefined behavior in ff_ctzll_c()
Michael Niedermayer [Thu, 22 Oct 2015 12:10:10 +0000 (14:10 +0200)]
avutil/intmath: fix undefined behavior in ff_ctzll_c()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoutils: Use data buffers directly instead of an AVPicture
Vittorio Giovara [Wed, 14 Oct 2015 09:33:27 +0000 (11:33 +0200)]
utils: Use data buffers directly instead of an AVPicture

Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com>
3 years agolibavformat/electronicarts: also demux mpeg audio layer 2
Peter Ross [Thu, 22 Oct 2015 02:46:25 +0000 (13:46 +1100)]
libavformat/electronicarts: also demux mpeg audio layer 2

Signed-off-by: Peter Ross <pross@xvid.org>
http://wiki.multimedia.cx/index.php?title=Electronic_Arts_SCxl
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agolavc/x86/vc1dsp_init: Fix compilation with --disable-yasm.
Carl Eugen Hoyos [Thu, 22 Oct 2015 09:37:42 +0000 (11:37 +0200)]
lavc/x86/vc1dsp_init: Fix compilation with --disable-yasm.

3 years agolavf/mpjpegdec: Return 0 if an allocation inside the probe function fails.
Carl Eugen Hoyos [Thu, 22 Oct 2015 09:19:45 +0000 (11:19 +0200)]
lavf/mpjpegdec: Return 0 if an allocation inside the probe function fails.

3 years agoavformat/electronicarts: fix demuxing of certain eam files
Paul B Mahol [Thu, 22 Oct 2015 08:35:09 +0000 (10:35 +0200)]
avformat/electronicarts: fix demuxing of certain eam files

Such files have gaps between header chunks.

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavformat/electronicarts: support ADPCM PSX
Paul B Mahol [Wed, 21 Oct 2015 20:59:08 +0000 (22:59 +0200)]
avformat/electronicarts: support ADPCM PSX

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agommaldec: Add mpeg2 decoding support
Julian Scheel [Wed, 21 Oct 2015 15:10:48 +0000 (17:10 +0200)]
mmaldec: Add mpeg2 decoding support

Register mmaldec as mpeg2 decoder. Supporting mpeg2 in mmaldec is just a
matter of setting the correct MMAL_ENCODING on the input port. To ease the
addition of further supported mmal codecs a macro is introduced to generate
the decoder and decoder class structs.

Signed-off-by: Julian Scheel <julian@jusst.de>
Signed-off-by: wm4 <nfxjfg@googlemail.com>
3 years agox86/Makefile: move decoder/encoder objects out of the subsystems section
James Almer [Thu, 22 Oct 2015 06:55:18 +0000 (03:55 -0300)]
x86/Makefile: move decoder/encoder objects out of the subsystems section

Signed-off-by: James Almer <jamrial@gmail.com>