Re: [FFmpeg-devel] [PATCH]Fix overwrite when reading invalid Quickdraw images

2015-05-17 Thread Carl Eugen Hoyos
Michael Niedermayer michaelni at gmx.at writes: +if (pos = offset) +return AVERROR_INVALIDDATA; ok but theres a 2nd case below which i think needs a similar change Merged by you with the similar change. Thank you, Carl Eugen

Re: [FFmpeg-devel] [PATCH]Force colour range for vc1 gray output

2015-05-17 Thread Carl Eugen Hoyos
Michael Niedermayer michaelni at gmx.at writes: +else { avctx-pix_fmt = AV_PIX_FMT_GRAY8; +if (avctx-color_range == AVCOL_RANGE_UNSPECIFIED) +avctx-color_range = AVCOL_RANGE_MPEG; this variant LGTM as well This variant was merged. Carl Eugen

Re: [FFmpeg-devel] [PATCH]Print an error for invalid wav files and write no incorrect file length

2015-05-17 Thread Carl Eugen Hoyos
Michael Niedermayer michaelni at gmx.at writes: wavenc.c |8 ++-- 1 file changed, 6 insertions(+), 2 deletions(-) 7e2f589f1d543161f5b7d3682d7cd02cf76ff74e patchwav4genc.diff probably ok Merged by you. Thank you, Carl Eugen ___

Re: [FFmpeg-devel] [PATCH]Tell users about -use_absolute_path

2015-05-17 Thread Carl Eugen Hoyos
Carl Eugen Hoyos cehoyos at ag.or.at writes: +av_log(fc, AV_LOG_ERROR, + Absolute path %s not tried for security reasons, + pass -use_absolute_path 1 to force using absolute paths\n, + ref-path); A variant was merged by Michael, wording improvements

Re: [FFmpeg-devel] [PATCH]Fix flac with high lpc precision

2015-05-17 Thread Carl Eugen Hoyos
Carl Eugen Hoyos cehoyos at ag.or.at writes: Implementation of the option to force buggy decoding was less straightforward than expected, two patches attached. The patches were merged. Carl Eugen ___ ffmpeg-devel mailing list

Re: [FFmpeg-devel] [PATCH]lavf/wavdec: Allow decoding of invalid large wav files

2015-05-17 Thread Carl Eugen Hoyos
Michael Niedermayer michaelni at gmx.at writes: wavdec.c |5 - 1 file changed, 4 insertions(+), 1 deletion(-) 227da1c1acd2affa88538476148a1c6bbf43979b patchwav4gdec.diff maybe ok i think a warning or something should be printed for such files though Merged with a warning

Re: [FFmpeg-devel] [PATCH]Print a warning if gray decoding was requested but not enabled

2015-05-17 Thread Carl Eugen Hoyos
Andreas Cadhalpun andreas.cadhalpun at googlemail.com writes: On 08.05.2015 15:01, Carl Eugen Hoyos wrote: +av_log(avctx, AV_LOG_WARNING, + gray decoding requested but not ensabled at configuration time\n); ^

[FFmpeg-devel] [PATCH]Never decode vc1 as gray if gray decoding was not enabled

2015-05-17 Thread Carl Eugen Hoyos
Hi! Currently, when compiling FFmpeg with --disable-gray (the default) and decoding vc1 with CODEC_FLAG_GRAY set, the output is gray with artefacts. The attached patch makes the output coloured in this case, but maybe the current behaviour is intended? Carl Eugen diff --git

Re: [FFmpeg-devel] [PATCH]Never decode vc1 as gray if gray decoding was not enabled

2015-05-17 Thread Michael Niedermayer
On Sun, May 17, 2015 at 10:57:10AM +0200, Carl Eugen Hoyos wrote: Hi! Currently, when compiling FFmpeg with --disable-gray (the default) and decoding vc1 with CODEC_FLAG_GRAY set, the output is gray with artefacts. The attached patch makes the output coloured in this case, but maybe the

Re: [FFmpeg-devel] Fwd: Help with codecs

2015-05-17 Thread compn
On Thu, 14 May 2015 10:31:37 +0200 Fernandon Bautista fernandobautista...@gmail.com wrote: Hi all, thanks for your support :) I have been reading the posts and well ... we need the codec for linux not windows. Mi question is: have you i posted a quick dirty patch to make the codec work in

[FFmpeg-devel] [PATCH] Fix 'while' loop condition to prevent movtext crashes by tracking packet size

2015-05-17 Thread Niklesh Lalwani
From: Niklesh niklesh.lalw...@iitb.ac.in Hi all, This patch fixes some movtext crashes caused due to incorrect 'while' loop condition. I will post several other patches to improve upon the code and null pointer dereferences once this is applied. Signed-off-by: Niklesh

Re: [FFmpeg-devel] [PATCH] Fix 'while' loop condition to prevent movtext crashes by tracking packet size

2015-05-17 Thread Philip Langdale
On Sun, 17 May 2015 21:36:42 +0530 Niklesh Lalwani niklesh.lalw...@iitb.ac.in wrote: From: Niklesh niklesh.lalw...@iitb.ac.in Hi all, This patch fixes some movtext crashes caused due to incorrect 'while' loop condition. I will post several other patches to improve upon the code and null

Re: [FFmpeg-devel] [PATCH] Fix 'while' loop condition to prevent movtext crashes by tracking packet size

2015-05-17 Thread Philip Langdale
On Sun, 17 May 2015 09:56:10 -0700 Philip Langdale phil...@overt.org wrote: tsmb += 4; tsmb_type = AV_RB32(tsmb); @@ -176,6 +177,7 @@ static int mov_text_decode_frame(AVCodecContext *avctx, av_freep(style_end); av_freep(style_flags);

Re: [FFmpeg-devel] [PATCH] Fix 'while' loop condition to prevent movtext crashes by tracking packet size

2015-05-17 Thread Niklesh Lalwani
Just to be clear, I need to check for avpkt-size after reading box size, box type, style entries because there can be bad packets, otherwise the condition should work, right? Yes, I'll add the support for large boxes after this. I have it in my timeline. Thanks, Niklesh

Re: [FFmpeg-devel] [PATCH 1/2] Support for large boxes(32 bit) for 3gpp timed text

2015-05-17 Thread Philip Langdale
On Mon, 27 Apr 2015 04:53:26 +0530 Niklesh Lalwani niklesh.lalw...@iitb.ac.in wrote: From: Niklesh niklesh.lalw...@iitb.ac.in Signed-off-by: Niklesh niklesh.lalw...@iitb.ac.in --- libavcodec/movtextdec.c | 9 - 1 file changed, 8 insertions(+), 1 deletion(-) diff --git

Re: [FFmpeg-devel] [PATCH 1/2] Support for large boxes(32 bit) for 3gpp timed text

2015-05-17 Thread Niklesh Lalwani
Oh yes. I'll correct that. Thank you. Niklesh. On 17-May-2015 10:40 PM, Philip Langdale phil...@overt.org wrote: On Mon, 27 Apr 2015 04:53:26 +0530 Niklesh Lalwani niklesh.lalw...@iitb.ac.in wrote: From: Niklesh niklesh.lalw...@iitb.ac.in Signed-off-by: Niklesh

Re: [FFmpeg-devel] [PATCH 01/12] vp9: disable all pmulhrsw in 8/16 iadst x86 optimizations.

2015-05-17 Thread Michael Niedermayer
On Fri, May 15, 2015 at 03:43:46PM -0400, Ronald S. Bultje wrote: They all overflow in various samples that are considered valid input. --- libavcodec/x86/vp9itxfm.asm | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) branch which is equivalent to this patchset merged thanks

[FFmpeg-devel] [RFC]Change rematrix_maxval default to 1.0

2015-05-17 Thread Carl Eugen Hoyos
Hi! Attached patch as suggested by Hendrik might fix ticket #4564. Please comment, Carl Eugen diff --git a/doc/resampler.texi b/doc/resampler.texi index f9eef03..c73d402 100644 --- a/doc/resampler.texi +++ b/doc/resampler.texi @@ -66,8 +66,8 @@ Set rematrix volume. Default value is 1.0. @item

Re: [FFmpeg-devel] [PATCH] libavcodec/x265: detect csps in libx265_encode_init_csp()

2015-05-17 Thread Derek Buitenhuis
On 5/16/2015 4:05 AM, Steve Borho wrote: Without this change, if you link with an 8bit libx265 and try to specify a 10bit input color space via: ffmpeg -i in.mov -c:v libx265 -pix_fmt yuv420p10le out.mp4 It will error with: nit: It's not an error, it just changes the filter chain.

[FFmpeg-devel] [PATCH] avcodec/flacdec: Add lpc16m instead of using lpc32 + post analysis and correction

2015-05-17 Thread Michael Niedermayer
This is slower on x86 as the lpc16* functions are not optimized while lpc32 is If someone wants to optimize lpc16m and it would actually be faster then this could be applied and used until then the current solution in git is better as its faster Signed-off-by: Michael Niedermayer michae...@gmx.at

Re: [FFmpeg-devel] [PATCH]Never decode vc1 as gray if gray decoding was not enabled

2015-05-17 Thread Carl Eugen Hoyos
Michael Niedermayer michaelni at gmx.at writes: vc1_block.c | 36 ++-- vc1_loopfilter.c |4 ++-- vc1_mc.c | 12 ++-- vc1dec.c |6 +++--- 4 files changed, 29 insertions(+), 29 deletions(-) LGTM Patch applied.

[FFmpeg-devel] [PATCH] libavcodec/x265: detect csps in libx265_encode_init_csp()

2015-05-17 Thread Steve Borho
Without this change, if you link with an 8bit libx265 and try to specify a 10bit input color space via: ffmpeg -i in.mov -c:v libx265 -pix_fmt yuv420p10le out.mp4 It will error with: Incompatible pixel format 'yuv420p10le' for codec 'libx265', auto-selecting format 'yuv420p' With this

Re: [FFmpeg-devel] [PATCH]Force tv range for mpeg4 gray output

2015-05-17 Thread Carl Eugen Hoyos
Michael Niedermayer michaelni at gmx.at writes: -if (CONFIG_GRAY (avctx-flags CODEC_FLAG_GRAY)) +if (CONFIG_GRAY (avctx-flags CODEC_FLAG_GRAY)) { +if (avctx-color_range == AVCOL_RANGE_UNSPECIFIED) +avctx-color_range = AVCOL_RANGE_MPEG; return