ffmpeg.git
4 years agonetwork: Do not leave context locked on error
Vittorio Giovara [Wed, 22 Apr 2015 13:59:55 +0000 (14:59 +0100)]
network: Do not leave context locked on error

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 45340f9fc77e254924f129c764b685fe682e380c)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agovp9: remove one optimization branch in iadst16 which causes overflows.
Ronald S. Bultje [Wed, 22 Apr 2015 18:53:01 +0000 (14:53 -0400)]
vp9: remove one optimization branch in iadst16 which causes overflows.

See sample vp90-2-14-resize-fp-tiles-16-8-4-2-1.webm from the vp9 test
vector set which reproduces the issue. This probably costs a few cycles,
but I don't think there's an easy way to workaround that.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit d02d04a18f300ebe97319ca6e91fc943cb14f58b)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agofate: Include branch information in the payload header
Timothy Gu [Sun, 12 Apr 2015 15:28:42 +0000 (08:28 -0700)]
fate: Include branch information in the payload header

The server is properly equiped not to choke on that now.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 7ad27f1221562b037178d1fb605510ba349f04cc)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavformat/utils: Ensure that AVFMT_FLAG_CUSTOM_IO is set before use
Michael Niedermayer [Mon, 20 Apr 2015 20:22:31 +0000 (22:22 +0200)]
avformat/utils: Ensure that AVFMT_FLAG_CUSTOM_IO is set before use

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit ba631b791435c395361e2026fc7419b341e57813)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavformat/img2dec: do not rewind custom io buffers
Michael Niedermayer [Mon, 20 Apr 2015 21:59:53 +0000 (23:59 +0200)]
avformat/img2dec: do not rewind custom io buffers

Fixes double free with some applications

Fixes vlc ticket14121
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit e6e8cc8ce9c2a398fbb51254a5067f4bd3c4fa8a)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/alsdec: Use av_mallocz_array() for chan_data to ensure the arrays never conta...
Michael Niedermayer [Sat, 18 Apr 2015 18:50:23 +0000 (20:50 +0200)]
avcodec/alsdec: Use av_mallocz_array() for chan_data to ensure the arrays never contain random data

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 7e104647a3556fc61a139483cee1cb7dfa2dc5bd)

Conflicts:

libavcodec/alsdec.c

4 years agoavcodec/atrac3plusdsp: fix on stack alignment
Michael Niedermayer [Fri, 17 Apr 2015 09:58:41 +0000 (11:58 +0200)]
avcodec/atrac3plusdsp: fix on stack alignment

Fixes fate failure on ARM
(cherry picked from commit 38f67260684aec8a02d87ab4056b1a1fbf964c03)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoswresample/swresample-test: Randomly wipe out channel counts
Michael Niedermayer [Sun, 12 Apr 2015 20:08:52 +0000 (22:08 +0200)]
swresample/swresample-test: Randomly wipe out channel counts

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit ff50b1b13be965c93a9a7169edb62631a928e308)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoswresample: Check channel layouts and channels against each other and print human...
Michael Niedermayer [Sun, 12 Apr 2015 19:08:09 +0000 (21:08 +0200)]
swresample: Check channel layouts and channels against each other and print human readable error messages

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 3c77bb5f23b2e149495c814759beab7eedeede6c)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoswresample: Allow reinitialization without ever setting channel layouts
Michael Niedermayer [Sun, 12 Apr 2015 18:50:53 +0000 (20:50 +0200)]
swresample: Allow reinitialization without ever setting channel layouts
(cherry picked from commit 80a28c7509a11114e1aea5b208d56c6646d69c07)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoswresample: Allow reinitialization without ever setting channel counts
Michael Niedermayer [Sun, 12 Apr 2015 17:43:08 +0000 (19:43 +0200)]
swresample: Allow reinitialization without ever setting channel counts

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit d7b9cb2f7a51351586791e65fa22e6536ee58c9f)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agodashenc: replace attribute id with contentType for the AdaptationSet element
James Almer [Sun, 10 May 2015 06:31:44 +0000 (03:31 -0300)]
dashenc: replace attribute id with contentType for the AdaptationSet element

id should be an integer, not a string. It is also optional, so use
contentType instead which is the proper attribute for these values.

This addresses ticket #4545, fixing an MPD validation error.

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 32a4177a627ddce984a5a9ed7023d9a63dcbdf85)

4 years agoavformat/matroskaenc: Use avoid_negative_ts_use_pts if no stream writes dts
Michael Niedermayer [Tue, 5 May 2015 10:57:24 +0000 (12:57 +0200)]
avformat/matroskaenc: Use avoid_negative_ts_use_pts if no stream writes dts

This reduces the number of cases where timestamps need to be shifted

Fixes Ticket4487

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 9d4fdfe24c731d1880797dee65365154b41c1dea)

4 years agoavformat/mux: Add avoid_negative_ts_use_pts
Michael Niedermayer [Tue, 5 May 2015 10:44:20 +0000 (12:44 +0200)]
avformat/mux: Add avoid_negative_ts_use_pts

This allows using pts instead of dts for negative TS avoidance

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 7ac5c38ec5baeea6ad647ccc9bb7e97564c50ec2)

4 years agotests/fate-run: do not attempt to parse tiny_psnrs output if it failed
Michael Niedermayer [Thu, 23 Apr 2015 02:27:56 +0000 (04:27 +0200)]
tests/fate-run: do not attempt to parse tiny_psnrs output if it failed

This avoids confusing syntax errors with awk later

Likely fixes awk errors at:
http://buildd.debian-ports.org/status/fetch.php?pkg=ffmpeg&arch=sparc64&ver=7%3A2.6.2-1&stamp=1428928967

Reviewed-by: Timothy Gu <timothygu99@gmail.com>
Thanks-to: Andreas Cadhalpun <andreas.cadhalpun@googlemail.com> for the link
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit c0d847e457c1ef72843a63853f1135d52b74131e)

4 years agocafdec: free extradata before allocating it
Andreas Cadhalpun [Tue, 12 May 2015 22:22:03 +0000 (00:22 +0200)]
cafdec: free extradata before allocating it

This fixes a memleak if read_kuki_chunk is executed more than once.

Reviewed-by: Carl Eugen Hoyos <cehoyos@ag.or.at>
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
(cherry picked from commit cb7c4f73e5e3debe2646279eaa7cfb493573118b)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agocafdec: check avio_read return value
Andreas Cadhalpun [Tue, 12 May 2015 21:49:45 +0000 (23:49 +0200)]
cafdec: check avio_read return value

If avio_read fails, the buffer can contain uninitialized values.

Reviewed-by: Carl Eugen Hoyos <cehoyos@ag.or.at>
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
(cherry picked from commit a3ede6b742f37d511253ab4c2fd98c13203f1cd3)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoimgutils: initialize palette padding bytes in av_image_alloc
Andreas Cadhalpun [Tue, 12 May 2015 19:45:42 +0000 (21:45 +0200)]
imgutils: initialize palette padding bytes in av_image_alloc

av_image_fill_pointers always aligns the palette, but the padding
bytes don't (and can't) get initialized in av_image_copy.

Thus initialize them in av_image_alloc.

This fixes 'Syscall param write(buf) points to uninitialised byte(s)'
valgrind warnings.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 51f64552853e16d72644308db53abee870aecfb9)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoaacdec: don't return frames without data
Andreas Cadhalpun [Tue, 12 May 2015 18:27:21 +0000 (20:27 +0200)]
aacdec: don't return frames without data

Since commit 676a395a aac->frame->data is not necessarily allocated at
the end of aac_decode_frame_int if avctx->channels is 0.

In this case a bogus frame without any data, but non-zero nb_samples is
returned.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit ec38a1ba404b8cb8d71ccee2b8dcd6f3fcbde273)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoid3v2: catch avio_read errors in check_tag
Andreas Cadhalpun [Mon, 11 May 2015 18:07:49 +0000 (20:07 +0200)]
id3v2: catch avio_read errors in check_tag

Since len is an unsigned int, the comparison is currently treated as
unsigned and thus ignores all errors from avio_read.

Thus cast len to int, which is unproblematic, because at that point len
is between 0 and 4.

This fixes 'Conditional jump or move depends on uninitialised value'
valgrind warnings in is_tag.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 0382c94f13b4b20456b7259e90b170dc020419b8)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoavi: Validate sample_size
Andreas Cadhalpun [Wed, 6 May 2015 00:26:57 +0000 (02:26 +0200)]
avi: Validate sample_size

And either error out or set it to 0 if it is negative.

CC: libav-stable@libav.org
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
(cherry picked from commit 4d0ee4962be7e07cdc038a78008ef2e4e47e5f81)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Conflicts:
libavformat/avidec.c

4 years agoaacsbr: break infinite loop in sbr_hf_calc_npatches
Andreas Cadhalpun [Wed, 22 Apr 2015 13:23:24 +0000 (15:23 +0200)]
aacsbr: break infinite loop in sbr_hf_calc_npatches

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 584cc1ade10a3297ef9c107ef3a2081c04024156)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agodiracdec: check that block length is valid
Andreas Cadhalpun [Wed, 6 May 2015 13:34:53 +0000 (15:34 +0200)]
diracdec: check that block length is valid

In init_planes p->xblen and p->yblen are set to:
            p->xblen = s->plane[0].xblen >> s->chroma_x_shift;
            p->yblen = s->plane[0].yblen >> s->chroma_y_shift;

These are later used as block_w and block_h arguments of
s->vdsp.emulated_edge_mc. If one of them is 0 it triggers an av_assert2
in emulated_edge_mc:
    av_assert2(start_x < end_x && block_w > 0);
    av_assert2(start_y < end_y && block_h > 0);

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 75fc81c8318505aa7946e05a9bee08d47241fc66)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agodiracdec: check if reference could not be allocated
Andreas Cadhalpun [Tue, 5 May 2015 21:51:48 +0000 (23:51 +0200)]
diracdec: check if reference could not be allocated

s->ref_pics[i] is later used as ref argument of interpolate_refplane,
where it is dereferenced.

If it is NULL, it causes a segmentation fault.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit d93181ef3eacdb862d93448f31c97765a523d1db)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agodiracdec: avoid overflow of bytes*8 in decode_lowdelay
Andreas Cadhalpun [Tue, 5 May 2015 20:10:44 +0000 (22:10 +0200)]
diracdec: avoid overflow of bytes*8 in decode_lowdelay

If bytes is large enough, bytes*8 can overflow and become negative.

In that case 'bufsize -= bytes*8' causes bufsize to increase instead of
decrease.

This leads to a segmentation fault.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 9e66b39aa87eb653a6e5d15f70b792ccbf719de7)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agodiracdec: prevent overflow in data_unit_size check
Andreas Cadhalpun [Tue, 5 May 2015 19:33:08 +0000 (21:33 +0200)]
diracdec: prevent overflow in data_unit_size check

buf_idx + data_unit_size can overflow, causing the '> buf_size' check to
wrongly fail.

This causes a segmentation fault.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 984f50deb2d48f6844d65e10991b996a6d29e87c)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoavidec: check for valid bit_rate range
Andreas Cadhalpun [Mon, 4 May 2015 19:07:52 +0000 (21:07 +0200)]
avidec: check for valid bit_rate range

If bit_rate is negative, it can trigger an av_assert2 in av_rescale_rnd.

Since av_rescale returns int64_t, but st->codec_bit_rate is int, it can
also overflow into a negative value.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 0eec40b713eee84e2aec8af35ccce059817cad2a)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoavformat/matroskadec: Use tracks[k]->stream instead of s->streams[k]
Michael Niedermayer [Mon, 4 May 2015 13:47:54 +0000 (15:47 +0200)]
avformat/matroskadec: Use tracks[k]->stream instead of s->streams[k]

The later is not correct

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 5d309d309108684f742bbf5fc2393f1c519cda72)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agomatroskadec: check s->streams[k] before using it
Andreas Cadhalpun [Sun, 3 May 2015 21:55:20 +0000 (23:55 +0200)]
matroskadec: check s->streams[k] before using it

This fixes a segmentation fault.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit e54540655f229d06667dc7fa7005f2a20e101e80)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agomatroskadec: use uint64_t instead of int for index_scale
Andreas Cadhalpun [Sun, 3 May 2015 21:07:20 +0000 (23:07 +0200)]
matroskadec: use uint64_t instead of int for index_scale

index_scale is set to matroska->time_scale of type uint64_t.

When index_scale is int, the assignment can overflow and e.g. result
in index_scale = 0. This causes a floating point exception due to the
division by index_scale.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit eb9fb508b0e09d85d234fe694333b2005e1d7a7e)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agopngdec: don't use AV_PIX_FMT_MONOBLACK for apng
Andreas Cadhalpun [Sun, 3 May 2015 18:36:20 +0000 (20:36 +0200)]
pngdec: don't use AV_PIX_FMT_MONOBLACK for apng

AV_PIX_FMT_MONOBLACK has the AV_PIX_FMT_FLAG_BITSTREAM flag, i.e.
linesize can be smaller than width.

Since x_offset is only check against the width, this can lead to
x_offset * bpp >= image_linesize.

In this case ptr could be set to a position outside the image_buf in
png_handle_row, leading to memory corruption and thus crashes.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 372aa0777aaacf726de7cd7dd0e6797026a124ee)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agopngdec: return correct error code from decode_frame_common
Andreas Cadhalpun [Sun, 3 May 2015 15:50:26 +0000 (17:50 +0200)]
pngdec: return correct error code from decode_frame_common

During the loop ret can get changed. Since it is not set on all failure
paths, decode_frame_common can return 0 even though an error occurred.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 8f760be4d312bb6e78f80d39b9d0062253332e08)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agopngdec: check s->last_picture.f->data[0] before using it
Andreas Cadhalpun [Sun, 3 May 2015 14:21:39 +0000 (16:21 +0200)]
pngdec: check s->last_picture.f->data[0] before using it

This check was removed in commit 08aec6f6, but
s->last_picture.f->data[0] is still used in handle_p_frame_apng
unconditionally.

This fixes a segmentation fault.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 287dbb0771d558b336e377d0594e26c0a6291755)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agonutdec: fix illegal count check in decode_main_header
Andreas Cadhalpun [Tue, 28 Apr 2015 20:37:19 +0000 (22:37 +0200)]
nutdec: fix illegal count check in decode_main_header

The existing check has two problems:
 1) i + count can overflow, so that the check '< 256' returns true.
 2) In the (i == 'N') case occurs a j-- so that the loop runs once more.

This can trigger the assertion 'nut->header_len[0] == 0' or cause
segmentation faults or infinite hangs.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 7c24ca1bda2d4df1dc9b2b982941be532d60da21)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agonutdec: fix memleaks on error in nut_read_header
Andreas Cadhalpun [Tue, 28 Apr 2015 18:58:21 +0000 (20:58 +0200)]
nutdec: fix memleaks on error in nut_read_header

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 361702660d2c37a63b7d6381d39e1e1de8405260)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agonutdec: check chapter creation in decode_info_header
Andreas Cadhalpun [Tue, 28 Apr 2015 18:57:59 +0000 (20:57 +0200)]
nutdec: check chapter creation in decode_info_header

This fixes a segmentation fault when accessing the metadata.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 3ff1af2b0db7132d5717be6395227a94c8abab07)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agonutdec: check for negative frame rate in decode_info_header
Andreas Cadhalpun [Tue, 28 Apr 2015 18:31:56 +0000 (20:31 +0200)]
nutdec: check for negative frame rate in decode_info_header

A negative frame rate triggers an av_assert2 in av_rescale_rnd.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 6621105877ce0d65724a8ab60b3a50160adbe65d)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoapedec: prevent out of array writes in decode_array_0000
Andreas Cadhalpun [Tue, 28 Apr 2015 09:13:43 +0000 (11:13 +0200)]
apedec: prevent out of array writes in decode_array_0000

s->decoded_buffer is allocated with a min_size of:
    2 * FFALIGN(blockstodecode, 8) * sizeof(*s->decoded_buffer)

Then it is assigned to s->decoded[0] (and s->decoded_buffer + FFALIGN(blockstodecode, 8)
to s->decoded[1]) and passed as out buffer to decode_array_0000.

In this function 64 elements of the out buffer are written
unconditionally and outside the array if blockstodecode is too small.

This causes memory corruption, leading to segmentation faults or other
crashes.

Thus change decode_array_0000 to write at most blockstodecode elements
of the out buffer.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 699341d647f7af785fb8ceed67604467b0b9ab12)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoapedec: set s->samples only when init_frame_decoder succeeded
Andreas Cadhalpun [Mon, 27 Apr 2015 22:30:51 +0000 (00:30 +0200)]
apedec: set s->samples only when init_frame_decoder succeeded

Otherwise range_start_decoding is not necessarily run and thus
ctx->rc.range still 0 in range_dec_normalize leading to an infinite
loop.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 464c49155ce7ffc88ed39eb2511e7a75565c24be)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoswscale/ppc/swscale_altivec.c: POWER LE support in yuv2planeX_8() delete macro GET_VF...
Rong Yan [Mon, 27 Apr 2015 03:19:45 +0000 (03:19 +0000)]
swscale/ppc/swscale_altivec.c: POWER LE support in yuv2planeX_8() delete macro GET_VF() it was wrong

GCC tool had a bug of PPC intrinsic interpret, which has been fixed in GCC 4.9.1. This bug lead to
errors in two of our previous patches. We found this when we update our GCC tools to 4.9.1 and by
reading the related info on GCC website. We fix our previous error in two separate commits

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 603c839398f89d3ef15c47530470fbb0051632c0)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoalac: reject rice_limit 0 if compression is used
Andreas Cadhalpun [Thu, 23 Apr 2015 22:01:43 +0000 (00:01 +0200)]
alac: reject rice_limit 0 if compression is used

If rice_limit is 0, k can be 0 in decode_scalar, which calls show_bits(gb, k).

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 4b657a1b1eedcf38bcf36e89a2f4be6f76b5ce09)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoalsdec: only adapt order for positive max_order
Andreas Cadhalpun [Wed, 22 Apr 2015 14:03:41 +0000 (16:03 +0200)]
alsdec: only adapt order for positive max_order

For max_order = 0 the clipping range is invalid. (amin = 2, amax = 1)

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 58d605ee9b3277289278dc40e022311f8e083833)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agobink: check vst->index_entries before using it
Andreas Cadhalpun [Wed, 22 Apr 2015 15:08:51 +0000 (17:08 +0200)]
bink: check vst->index_entries before using it

This fixes a NULL pointer dereference if vst->duration is 0.

The problem was introduced in commit 0588acaf.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 86d00ede4f9acb02690a0615490173648e1d933c)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agompeg4videodec: only allow a positive length
Andreas Cadhalpun [Wed, 22 Apr 2015 14:32:42 +0000 (16:32 +0200)]
mpeg4videodec: only allow a positive length

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit b3408ae4c64cb674b1d5f0f30171759113ce722a)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoalsdec: check sample pointer range in revert_channel_correlation
Andreas Cadhalpun [Tue, 21 Apr 2015 17:25:50 +0000 (19:25 +0200)]
alsdec: check sample pointer range in revert_channel_correlation

Also change the type of begin, end and smp to ptrdiff_t to make the
comparison well-defined.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Reviewed-by: Thilo Borgmann <thilo.borgmann@mail.de>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit afc7748d1f6abc4b3b1cc957b0fa6941837db3d0)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoaacpsy: correct calculation of minath in psy_3gpp_init
Andreas Cadhalpun [Tue, 21 Apr 2015 16:43:55 +0000 (18:43 +0200)]
aacpsy: correct calculation of minath in psy_3gpp_init

The minimum of the ath(x, ATH_ADD) function depends on ATH_ADD.
This patch uses the first order approximation to determine it.

For ATH_ADD = 4 this results in the value at 3407.06812 (-5.24241638)
not the one at 3410 (-5.24237967).

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Approved-by: Claudio Freire <klaussfreire@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit ca9849eecdf7db91d652c698018a5b096d8b78c7)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoalsdec: validate time diff index
Andreas Cadhalpun [Sat, 18 Apr 2015 18:09:28 +0000 (20:09 +0200)]
alsdec: validate time diff index

If begin is smaller than t, the subtraction 'begin -= t' wraps around,
because begin is unsigned. The same applies for end < t.

This causes segmentation faults.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit faf9fe2c224ea81a98afd53e2f0be0a2e13aeca9)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoalsdec: limit avctx->bits_per_raw_sample to 32
Andreas Cadhalpun [Sat, 18 Apr 2015 18:29:13 +0000 (20:29 +0200)]
alsdec: limit avctx->bits_per_raw_sample to 32

avctx->bits_per_raw_sample is used in get_sbits_long, which only
supports up to 32 bits.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 4c2b88678b436f59132386d9be2fc143e3ee480d)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoalsdec: ensure channel reordering is reversible
Andreas Cadhalpun [Sat, 18 Apr 2015 16:31:36 +0000 (18:31 +0200)]
alsdec: ensure channel reordering is reversible

If the same idx is used for more than one i, at least one entry in
sconf->chan_pos remains uninitialized.

This can cause segmentation faults.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit ef16501aebed43e34a3721336e8bee732eca2877)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoac3: validate end in ff_ac3_bit_alloc_calc_mask
Andreas Cadhalpun [Thu, 16 Apr 2015 19:25:26 +0000 (21:25 +0200)]
ac3: validate end in ff_ac3_bit_alloc_calc_mask

This fixes an invalid read if end is 0:
     band_end   = ff_ac3_bin_to_band_tab[end-1] + 1;

Depending on what is before the array, this can cause stack smashing,
when band_end becomes too large.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit bc4fee7f2a51635fa3c0f61d1e5164da1efeded3)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoaacpsy: avoid psy_band->threshold becoming NaN
Andreas Cadhalpun [Thu, 16 Apr 2015 18:04:54 +0000 (20:04 +0200)]
aacpsy: avoid psy_band->threshold becoming NaN

If band->thr is 0.0f, the division is undefined, making norm_fac not a
number or infinity, which causes psy_band->threshold to become NaN.

This is passed on to other variables until it finally reaches
sce->sf_idx and is converted to an integer (-2147483648).

This causes a segmentation fault when it is used as array index.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Reviewed-by: Claudio Freire <klaussfreire@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit e224aa41917454e7b5c23d9f2541425743ce595a)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoaasc: return correct buffer size from aasc_decode_frame
Andreas Cadhalpun [Thu, 16 Apr 2015 17:12:02 +0000 (19:12 +0200)]
aasc: return correct buffer size from aasc_decode_frame

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 0be54ad280cf114c02306b7063147e8379f8ed1e)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agoaacdec: consistently use avctx for logging in decode_eld_specific_config
Andreas Cadhalpun [Thu, 16 Apr 2015 14:58:32 +0000 (16:58 +0200)]
aacdec: consistently use avctx for logging in decode_eld_specific_config

ac may be NULL and then accessing ac->avctx results in a segmentation fault.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 5b75689b987e4c4dd4f34d5c8be389547e9cc701)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agomsrledec: use signed pixel_ptr in msrle_decode_pal4
Andreas Cadhalpun [Thu, 16 Apr 2015 12:49:08 +0000 (14:49 +0200)]
msrledec: use signed pixel_ptr in msrle_decode_pal4

This fixes segmentation faults, when pic->linesize[0] is negative.
In that case 'line * pic->linesize[0] + pixel_ptr' is treated as
unsigned and wraps around.

This reverts commit 7d78a964.
The problem was introduced in commit f7e1367f, which should obsolete
that commit.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit ae6fd7300b4e9f81d3b5ba201096ffe7cccf26fb)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agomatroskadec: export cover art correctly
wm4 [Fri, 3 Apr 2015 14:11:53 +0000 (16:11 +0200)]
matroskadec: export cover art correctly

Generally, libavformat exports cover art pictures as video streams with
1 packet and AV_DISPOSITION_ATTACHED_PIC set. Only matroskadec exported
it as attachment with codec_id set to AV_CODEC_ID_MJPEG.

Obviously, this should be consistent, so change the Matroska demuxer to
export a AV_DISPOSITION_ATTACHED_PIC pseudo video stream.

Matroska muxing is probably incorrect too. I know that it can create
broken files with an audio track and just 1 video frame when e.g.
remuxing mp3 with APIC to mkv. But for now this commit does not change
anything about muxing, and also continues to write attachments with
AV_CODEC_ID_MJPEG should the muxer application have special knowledge
that the Matroska is broken in this way.

Fixes trac #4423.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 511585ce7f7272e5069ef011d6be5f073d268901)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agomxfenc: don't try to write footer without header
Andreas Cadhalpun [Wed, 18 Mar 2015 20:57:58 +0000 (21:57 +0100)]
mxfenc: don't try to write footer without header

This fixes a crash, when trying to mux h264 into mxf_opatom.

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Previous version reviewed-by: tomas.hardin@codemill.se
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit b61cb61ab8f9abca98cc8c4d67cbefdb30f1e82a)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agomxfenc: fix memleaks in mxf_write_footer
Andreas Cadhalpun [Sat, 14 Mar 2015 16:47:53 +0000 (17:47 +0100)]
mxfenc: fix memleaks in mxf_write_footer

Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
Reviewed-by: tomas.hardin@codemill.se
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 39ddda12f17c666bb08abb2493d4adf8835b22c9)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agortpenc_mpegts: Set chain->rtp_ctx only after avformat_write_header succeeded
Martin Storsjö [Mon, 9 Mar 2015 21:14:19 +0000 (23:14 +0200)]
rtpenc_mpegts: Set chain->rtp_ctx only after avformat_write_header succeeded

By making sure we at each time only have one pointer set, either a
local variable or one in the context, we avoid potential double frees
in the cleanup routines. If chain->rtp_ctx is set, it is closed by
calling avformat_write_trailer, but that shouldn't be called unless
avformat_write_header succeeded.

This issue was pointed out by Andreas Cadhalpun.

Signed-off-by: Martin Storsjö <martin@martin.st>
(cherry picked from commit cf402d6fa88acd647cdff993429583bec8a34fdc)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agortpenc_mpegts: Free the right ->pb in the error path in the init function
Martin Storsjö [Mon, 9 Mar 2015 21:09:10 +0000 (23:09 +0200)]
rtpenc_mpegts: Free the right ->pb in the error path in the init function

This fixes a typo from 8e32b1f096.

Signed-off-by: Martin Storsjö <martin@martin.st>
(cherry picked from commit c83dd2d2a458075a58895c384372f57c1ec26276)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
4 years agodoc: add aarch64 cpuflags to fftools documentation n2.6.2
James Almer [Fri, 10 Apr 2015 23:52:29 +0000 (20:52 -0300)]
doc: add aarch64 cpuflags to fftools documentation

Signed-off-by: James Almer <jamrial@gmail.com>
(cherry picked from commit 9fc45681e0c4736e4475966a70402d22c76a2965)

4 years agodoc: add missing arm cpuflags to fftools documentation
James Almer [Fri, 10 Apr 2015 23:51:35 +0000 (20:51 -0300)]
doc: add missing arm cpuflags to fftools documentation

Signed-off-by: James Almer <jamrial@gmail.com>
(cherry picked from commit 666ec9bd097201eaacb9c04da526cb2710cb9cab)

4 years agodoc: add missing x86 cpuflags to fftools documentation
James Almer [Fri, 10 Apr 2015 23:51:11 +0000 (20:51 -0300)]
doc: add missing x86 cpuflags to fftools documentation

Signed-off-by: James Almer <jamrial@gmail.com>
(cherry picked from commit 410c93cfd5ab509d8c9f907f88ae09a87fb743e6)

4 years agoUpdate for 2.6.2
Michael Niedermayer [Sat, 11 Apr 2015 20:23:22 +0000 (22:23 +0200)]
Update for 2.6.2

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/h264: Do not fail with randomly truncated VUIs
Michael Niedermayer [Fri, 10 Apr 2015 16:35:07 +0000 (18:35 +0200)]
avcodec/h264: Do not fail with randomly truncated VUIs

Fixes Ticket4445

Tested-by: Vittorio Giovara <vittorio.giovara@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit bc48c88918f767e0dffcd138ae8e5c3052e8a92f)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/h264_ps: Move truncation check from VUI to SPS
Michael Niedermayer [Wed, 8 Apr 2015 20:18:35 +0000 (22:18 +0200)]
avcodec/h264_ps: Move truncation check from VUI to SPS

This more completely checks for truncation

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 32e06c485be94a4967bff87022cbb065d3cbfd0d)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/h264: Be more tolerant to changing pps id between slices
Michael Niedermayer [Thu, 9 Apr 2015 11:50:07 +0000 (13:50 +0200)]
avcodec/h264: Be more tolerant to changing pps id between slices

Fixes Ticket4446

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 98d0c4236c7542c87f012228d3bc88aea67bddc2)

Conflicts:

libavcodec/h264.c

4 years agoavcodec/aacdec: Fix storing state before PCE decode
Michael Niedermayer [Wed, 8 Apr 2015 22:04:44 +0000 (00:04 +0200)]
avcodec/aacdec: Fix storing state before PCE decode

Fixes Ticket4460

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit e88b3852aefaa39b2170ef185ad03dda18732821)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/h264: reset the counts in the correct context
Michael Niedermayer [Wed, 8 Apr 2015 11:38:55 +0000 (13:38 +0200)]
avcodec/h264: reset the counts in the correct context

Fixes null pointer dereference

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/h264_slice: Dont reset mb_aff_frame per slice
Michael Niedermayer [Wed, 8 Apr 2015 10:29:47 +0000 (12:29 +0200)]
avcodec/h264_slice: Dont reset mb_aff_frame per slice

Fixes null pointer dereference
Fixes Ticket4440

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 386601286fed2dff5e1955bc21a0256f6f35ab19)

Conflicts:

libavcodec/h264_slice.c

4 years agoavcodec/h264: finish previous slices before switching to single thread mode
Michael Niedermayer [Tue, 7 Apr 2015 00:47:36 +0000 (02:47 +0200)]
avcodec/h264: finish previous slices before switching to single thread mode

Fixes null pointer dereference
Fixes Ticket4438

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit c4b2017ba66e1623da9f527704c61c86a6e74844)

Conflicts:

libavcodec/h264.c

4 years agoavcodec/h264: Fix race between slices where one overwrites data from the next
Michael Niedermayer [Mon, 6 Apr 2015 22:40:21 +0000 (00:40 +0200)]
avcodec/h264: Fix race between slices where one overwrites data from the next

Fixes non deterministic crash in ticket4408/fuzz2.264
Likely fixes other samples as well

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 43b434210e597d484aef57c4139c3126d22b7e2b)

Conflicts:

libavcodec/h264.h
libavcodec/h264_slice.c

4 years agoavformat/utils: avoid discarded streams in av_find_default_stream_index()
Michael Niedermayer [Mon, 6 Apr 2015 02:20:51 +0000 (04:20 +0200)]
avformat/utils: avoid discarded streams in av_find_default_stream_index()

Fixes Ticket2010

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit ff6841c6bb2d35f8d461419e45d59be7542d03fe)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoffmpeg: Fix extradata allocation
Michael Niedermayer [Fri, 3 Apr 2015 21:44:38 +0000 (23:44 +0200)]
ffmpeg: Fix extradata allocation

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 4d02dfbde475d249916eb19c360e890059aa6aa5)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/h264_refs: Do not set reference to things which dont exist
Michael Niedermayer [Sat, 4 Apr 2015 16:08:23 +0000 (18:08 +0200)]
avcodec/h264_refs: Do not set reference to things which dont exist

Fixes deadlock
Fixes Ticket4428
Fixes Ticket4429

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 429de043202286a2b5bcc082cc02de860b734db2)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/h264: Fail for invalid mixed IDR / non IDR frames in slice threading mode
Michael Niedermayer [Sat, 4 Apr 2015 17:28:33 +0000 (19:28 +0200)]
avcodec/h264: Fail for invalid mixed IDR / non IDR frames in slice threading mode

Fixes Ticket4408

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit fc58d5c43b4c7396fc69081eb0dfe5b6a21cb10d)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoRevert "avcodec/exr: fix memset first arg in reverse_lut()"
Michael Niedermayer [Wed, 1 Apr 2015 17:53:23 +0000 (19:53 +0200)]
Revert "avcodec/exr: fix memset first arg in reverse_lut()"

This reverts commit 586ba24ff29468d2a4ee843a9650feea5b2be6f6.

Fixes Ticket 4386

Found-by: Martin Vignali <martin.vignali@gmail.com>
(cherry picked from commit 5dd5b7d5312d2b7cb6d3dbfde12f1eaef7c7767f)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoh264: avoid unnecessary calls to get_format
Rainer Hochecker [Tue, 31 Mar 2015 06:27:01 +0000 (08:27 +0200)]
h264: avoid unnecessary calls to get_format

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavutil/pca: Check for av_malloc* failures
Michael Niedermayer [Mon, 30 Mar 2015 02:37:42 +0000 (04:37 +0200)]
avutil/pca: Check for av_malloc* failures

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit dadc43eee4d9036aa532665a04720238cc15e922)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavutil/cpu: add missing check for mmxext to av_force_cpu_flags
James Almer [Thu, 26 Mar 2015 05:11:55 +0000 (02:11 -0300)]
avutil/cpu: add missing check for mmxext to av_force_cpu_flags

Reviewed-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: James Almer <jamrial@gmail.com>
(cherry picked from commit 1f5d1eed78fad63f1c80a3766d3dc2421b99104d)

4 years agolavc/dnxhd: Fix pix_fmt change.
Carl Eugen Hoyos [Sat, 28 Mar 2015 16:36:02 +0000 (17:36 +0100)]
lavc/dnxhd: Fix pix_fmt change.

Fixes ticket #4400.
(cherry picked from commit dcac15a84c8fc087ec85b2172db13d2e0d543e50)

4 years agoavformat/http: replace cookies with updated values instead of appending forever
Micah Galizia [Tue, 17 Mar 2015 09:22:59 +0000 (20:22 +1100)]
avformat/http: replace cookies with updated values instead of appending forever

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit c59654d67d1afde3fac24021ef0fd9d18cf38455)
Signed-off-by: Micah Galizia <micahgalizia@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavformat/hls: refactor repeated HLS option updates
Micah Galizia [Mon, 16 Mar 2015 09:26:29 +0000 (20:26 +1100)]
avformat/hls: refactor repeated HLS option updates

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit fca085187940a169b7a43d096009f7dac315f9ac)
Signed-off-by: Micah Galizia <micahgalizia@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavformat/hls: store cookies returned in HLS key response
Micah Galizia [Sat, 14 Mar 2015 22:31:59 +0000 (09:31 +1100)]
avformat/hls: store cookies returned in HLS key response

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 7859618affe574c9de7f240d2ddc016f917c37bd)
Signed-off-by: Micah Galizia <micahgalizia@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavformat/rmdec: fix support for 0 sized mdpr
Michael Niedermayer [Thu, 26 Mar 2015 01:02:18 +0000 (02:02 +0100)]
avformat/rmdec: fix support for 0 sized mdpr

Fixes Ticket4393

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit df43d03731c138bb113a66fd9c1b4d8cf2d879e5)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/msrledec: restructure msrle_decode_pal4() based on the line number instead...
Michael Niedermayer [Tue, 24 Mar 2015 14:50:12 +0000 (15:50 +0100)]
avcodec/msrledec: restructure msrle_decode_pal4() based on the line number instead of the pixel pointer

Fixes out of array access
Fixes: da14e86d8462be6493eab16bc2d40f88/asan_heap-oob_204cfd2_528_cov_340150052_COMPRESS.BMP

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit f7e1367f58263593e6cee3c282f7277d7ee9d553)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/hevc_ps: Check cropping parameters more correctly
Michael Niedermayer [Sat, 21 Mar 2015 11:54:16 +0000 (12:54 +0100)]
avcodec/hevc_ps: Check cropping parameters more correctly

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 06c70d45373dedc600f28e345685b130b60203c1)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agohevc: make the crop sizes unsigned
Anton Khirnov [Fri, 20 Mar 2015 20:28:34 +0000 (21:28 +0100)]
hevc: make the crop sizes unsigned
(cherry picked from commit c929659bdd7d2d5848ea52e685a3164c7b901bb0)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/dnxhddec: Reset is_444 if format is not 444
Michael Niedermayer [Thu, 19 Mar 2015 22:33:36 +0000 (23:33 +0100)]
avcodec/dnxhddec: Reset is_444 if format is not 444

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit d3bd943108d321b387f955f3ab1ece4e9977401f)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavcodec/dnxhddec: Check that the frame is interlaced before using cur_field
Michael Niedermayer [Thu, 19 Mar 2015 22:28:39 +0000 (23:28 +0100)]
avcodec/dnxhddec: Check that the frame is interlaced before using cur_field

Fixes Ticket4227

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 2c660e34cf3c2b77cd2bef6f292920334dfd9192)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agomips/float_dsp: fix vector_fmul_window_mips on mips64
James Cowgill [Wed, 18 Mar 2015 13:02:59 +0000 (13:02 +0000)]
mips/float_dsp: fix vector_fmul_window_mips on mips64

Commit dfa920807494 ("mips/float_dsp: fix a bug in vector_fmul_window_mips")
fixed vector_fmul_window_mips by unrolling the loop only 4 times, but also
removed the outer C loop and replaced it with assembly branches and pointer
arithmetic. When submitting my 64-bit porting patch I missed this new
assembly which also needed porting.

This patch fixes a bus error in the fate-float-dsp test when run on 64-bit
mips.

Signed-off-by: James Cowgill <james410@cowgill.org.uk>
Reviewed-by: Nedeljko Babic <nedeljko.babic@imgtec.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit f8323744a0783d5937232a95cd1cc98f6b70a810)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agodoc: Remove non-existing decklink options.
Carl Eugen Hoyos [Tue, 24 Mar 2015 08:53:29 +0000 (09:53 +0100)]
doc: Remove non-existing decklink options.

Reported-by: Bostjan Hamler
(cherry picked from commit c5d0148c7e6d5ded4970d843ae6ef5864958e701)

4 years agodoc: Fix alphabetic ordering for decklink input device.
Carl Eugen Hoyos [Fri, 13 Mar 2015 00:57:42 +0000 (01:57 +0100)]
doc: Fix alphabetic ordering for decklink input device.
(cherry picked from commit a47c2a1d5f9f25b43f302bbf86dee3dc7b2e78e4)

4 years agovp9: make above buffer pointer 32-byte aligned.
Ronald S. Bultje [Sat, 21 Mar 2015 20:12:48 +0000 (17:12 -0300)]
vp9: make above buffer pointer 32-byte aligned.

Fixes ticket #4383

Signed-off-by: James Almer <jamrial@gmail.com>
(cherry picked from commit 1fd1f58bd6a58f2067a8d6b4919e1a0f34eb1f22)

4 years agoChangelog: update for 2.6.1 n2.6.1
Michael Niedermayer [Mon, 16 Mar 2015 19:09:11 +0000 (20:09 +0100)]
Changelog: update for 2.6.1

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoupdate for 2.6.1
Michael Niedermayer [Mon, 16 Mar 2015 18:45:24 +0000 (19:45 +0100)]
update for 2.6.1

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavformat/mov: Disallow ".." in dref unless use_absolute_path is set
Michael Niedermayer [Sat, 14 Mar 2015 20:32:35 +0000 (21:32 +0100)]
avformat/mov: Disallow ".." in dref unless use_absolute_path is set

as this kind of allows to circumvent it to some extend.
We also could add a separate parameter or value to choose this

Found-by: ramiro
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 1e4d0498df6621143da1a550006ddc3526ad51cb)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavfilter/palettegen: make sure at least one frame was sent to the filter
Clément Bœsch [Sun, 15 Mar 2015 18:09:16 +0000 (19:09 +0100)]
avfilter/palettegen: make sure at least one frame was sent to the filter

Fix FPE.

(cherry picked from commit 84da9339c21f679836a024ce86a69c6ea4a6fd11)

4 years agoavformat/mov: Check for string truncation in mov_open_dref()
Michael Niedermayer [Sat, 14 Mar 2015 20:24:54 +0000 (21:24 +0100)]
avformat/mov: Check for string truncation in mov_open_dref()

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 8003816e1619e77d8de051883264aa090e0d78cc)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoavformat/mov: Use sizeof(filename) instead of a literal number
Michael Niedermayer [Sat, 14 Mar 2015 20:23:32 +0000 (21:23 +0100)]
avformat/mov: Use sizeof(filename) instead of a literal number

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 21a53dd08dce7cc5b3fdf9c4826b4b74d8300ea0)

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
4 years agoeac3dec: fix scaling
Christophe Gisquet [Sat, 14 Mar 2015 11:48:59 +0000 (11:48 +0000)]
eac3dec: fix scaling

This is the remaining error, the output on the SPX samples,
respectively csi_miami_stereo_128_spx.eac3 and
csi_miami_5.1_256_spx.eac3, goes from:
stddev:    8.71 PSNR: 77.52 MAXDIFF:  235
stddev:24270.51 PSNR: 22.17 MAXDIFF:47166
to:
stddev:    0.12 PSNR:114.12 MAXDIFF:    1
stddev:    0.12 PSNR:114.73 MAXDIFF:    1

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
(cherry picked from commit 0c3339f4bd7aceebfd74deb437ba2e5c04ef3d0e)

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