Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 01/10/2016 11:56 AM, Andreas Cadhalpun wrote: This fixes segmentation faults due to out of bounds writes, when color_start is interpreted as negative number. This regression was introduced in commit 57631f. Signed-off-by: Andreas Cadhalpun --- Seriously,

Re: [FFmpeg-devel] [PATCH v5] lavf/matroskadec: Normalize noncompliant A_QUICKTIME/V_QUICKTIME private data

On 01/11/2016 03:53 AM, Mats Peterson wrote: av_malloc(), and which might use memalign(), of which the pointer is apparently not safe to pass to av_malloc(), according to the FFmpeg Not safe to pass to *av_realloc()*, I mean. Mats ___ ffmpeg-devel

Re: [FFmpeg-devel] [PATCH v5] lavf/matroskadec: Normalize noncompliant A_QUICKTIME/V_QUICKTIME private data

On 01/11/2016 01:46 AM, Michael Niedermayer wrote: On Sun, Jan 10, 2016 at 04:04:47AM +0100, Mats Peterson wrote: How about this one, Michael? Rather than using the obviously dangerous av_realloc() it uses av_malloc() to allocate new memory for the private data, then copies the private data to

Re: [FFmpeg-devel] [PATCH v5] lavf/matroskadec: Normalize noncompliant A_QUICKTIME/V_QUICKTIME private data

On 01/11/2016 01:46 AM, Michael Niedermayer wrote: that documentation is misleading, av_malloc() is if one is really pedantic not safe to be passed to av_free or av_realloc() if one isnt pedantic it safe to be passed to both Is the pointer from av_malloc() not safe to be passed to av_free()?

Re: [FFmpeg-devel] [PATCH v5] lavf/matroskadec: Normalize noncompliant A_QUICKTIME/V_QUICKTIME private data

On 01/11/2016 03:53 AM, Mats Peterson wrote: What do you mean by "separating them" by the way? I haven't separated anything as far as I know. Mats Another thing regarding the av_realloc() function: Even if I had been using av_realloc(), it won't cause any memory leak if it exits the

Re: [FFmpeg-devel] [PATCH v5] lavf/matroskadec: Normalize noncompliant A_QUICKTIME/V_QUICKTIME private data

On 01/11/2016 01:46 AM, Michael Niedermayer wrote: On Sun, Jan 10, 2016 at 04:04:47AM +0100, Mats Peterson wrote: How about this one, Michael? Rather than using the obviously dangerous av_realloc() it uses av_malloc() to allocate new memory for the private data, then copies the private data to

[FFmpeg-devel] [PATCH] lavf/matroskadec: Use av_realloc() in get_qt_codec()

Use av_realloc() rather than av_malloc() when normalizing noncompliant private data in get_qt_codec(). Based on your comments, Michael, it seems to be OK, and is more "elegant" in my book than using av_malloc()/memcpy()/av_free(). Mats -- Mats Peterson http://matsp888.no-ip.org/~mats/ >From

[FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

This fixes segmentation faults due to out of bounds writes, when color_start is interpreted as negative number. This regression was introduced in commit 57631f. Signed-off-by: Andreas Cadhalpun --- Seriously, changing the code behavior when "factoring out" is

Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 01/10/2016 11:56 AM, Andreas Cadhalpun wrote: This fixes segmentation faults due to out of bounds writes, when color_start is interpreted as negative number. dYes Anreas, until my normalization patch for matroskadec.c is applied, of course it's very easy for these variables to be negative

Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 01/10/2016 11:56 AM, Andreas Cadhalpun wrote: This fixes segmentation faults due to out of bounds writes, when color_start is interpreted as negative number. This regression was introduced in commit 57631f. Signed-off-by: Andreas Cadhalpun --- Seriously,

Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 01/10/2016 12:29 PM, Mats Peterson wrote: Good advice to a hysterical person like me. Thanks. At least I'm able to admit that I've been wrong, unlike some other people (no names mentioned). Mats ___ ffmpeg-devel mailing list

Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 10.01.2016 12:13, Mats Peterson wrote: > On 01/10/2016 11:56 AM, Andreas Cadhalpun wrote: >> This fixes segmentation faults due to out of bounds writes, when >> color_start is interpreted as negative number. >> >> This regression was introduced in commit 57631f. >> >> Signed-off-by: Andreas

Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 10.01.2016 12:24, Mats Peterson wrote: > On 01/10/2016 12:23 PM, Andreas Cadhalpun wrote: > >> Please check the facts before writing a reply. Thanks. > > It's human to err;) Thanks, anyway. But it's bad style to send two mails contradicting each other within 5 minutes. Just take the 5

Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 01/10/2016 12:29 PM, Andreas Cadhalpun wrote: Just take the 5 minutes to check the facts before sending the first reply. Best regards, Andreas Good advice to a hysterical person like me. Thanks. Mats ___ ffmpeg-devel mailing list

Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 01/10/2016 01:03 PM, Mats Peterson wrote: dYes Anreas, until my normalization patch for matroskadec.c is applied, of course it's very easy for these variables to be negative when using an int, because of the invalid private data. I stand corrected. Sorry, ANDREAS. This Thunderbird is killing

Re: [FFmpeg-devel] [PATCH]lavc/x264: Add compatibility values for coder options

Michael Niedermayer niedermayer.cc> writes: > > Attached patch fixes ticket #5142. > > libx264.c |2 ++ > > 1 file changed, 2 insertions(+) > > 98178e28de0e96f9e4c9f1f1ddcc4d75055d48bd patchx264coder.diff > > diff --git a/libavcodec/libx264.c b/libavcodec/libx264.c > > LGTM Patch

Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 01/10/2016 12:13 PM, Mats Peterson wrote: As far as I remember, those variables were ints in the original code in mov.c, and they will hardly reach a value where they could be interpreted as ints. But it's of course better to be on the safe side. Thanks for the heads-up. Interpreted as

Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 10.01.2016 12:17, Mats Peterson wrote: > On 01/10/2016 12:14 PM, Mats Peterson wrote: >> On 01/10/2016 12:13 PM, Mats Peterson wrote: >>> As far as I remember, those variables were ints in the original code in >>> mov.c, and they will hardly reach a value where they could be >>> interpreted as

Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 01/10/2016 12:23 PM, Andreas Cadhalpun wrote: Please check the facts before writing a reply. Thanks. It's human to err;) Thanks, anyway. ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH] qtpalette: make the color_* variables unsigned again

On 01/10/2016 12:14 PM, Mats Peterson wrote: On 01/10/2016 12:13 PM, Mats Peterson wrote: As far as I remember, those variables were ints in the original code in mov.c, and they will hardly reach a value where they could be interpreted as ints. But it's of course better to be on the safe side.

Re: [FFmpeg-devel] [PATCH 1/2] avcodec: Add Cineform HD Decoder

On 1/10/2016 1:54 AM, James Almer wrote: >> +AVCodec ff_cfhd_decoder = { >> +.name = "cfhd", >> +.long_name = NULL_IF_CONFIG_SMALL("Cineform HD"), >> +.type = AVMEDIA_TYPE_VIDEO, >> +.id = AV_CODEC_ID_CFHD, >> +.priv_data_size =

[FFmpeg-devel] [PATCH]lavf/avidec: Set display_aspect_ratio if no vprp chunk is present

Hi! Attached patch improves display of the samples from ticket #4535. Please comment, Carl Eugen diff --git a/libavformat/avidec.c b/libavformat/avidec.c index 3859810..a6e3b6a 100644 --- a/libavformat/avidec.c +++ b/libavformat/avidec.c @@ -834,6 +834,8 @@ static int

Re: [FFmpeg-devel] [PATCH 2/2] lavc/WIP: allow subtitle text format to be ASS without timing

On Sat, 9 Jan 2016 14:46:52 +0100 Clément Bœsch wrote: > From: Clément Bœsch > > --- > TODO > - ass_split needs to be able to handle dialog event without "Dialogue: > ..." so our encoding chain can handle the sane decoded subtitles > - zvbi needs testing

Re: [FFmpeg-devel] [PATCH]lavf/avidec: Set display_aspect_ratio if no vprp chunk is present

On 1/10/16, Carl Eugen Hoyos wrote: > Hi! > > Attached patch improves display of the samples from ticket #4535. > > Please comment, Carl Eugen > Pure guessing. ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org

Re: [FFmpeg-devel] [PATCH 2/2] lavc/WIP: allow subtitle text format to be ASS without timing

On Sun, 10 Jan 2016 19:57:17 +0100 Clément Bœsch wrote: > On Sun, Jan 10, 2016 at 02:03:46PM +0100, wm4 wrote: > > On Sat, 9 Jan 2016 14:46:52 +0100 > > Clément Bœsch wrote: > > > > > From: Clément Bœsch > > > > > > --- > > > TODO > > > -

Re: [FFmpeg-devel] [PATCH 2/2] avformat: Add Cineform HD demuxing

On Sun, Jan 10, 2016 at 12:28:50AM +, Kieran Kunhya wrote: > --- > libavformat/riff.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/libavformat/riff.c b/libavformat/riff.c > index f297dd4..faf0f86 100644 > --- a/libavformat/riff.c > +++ b/libavformat/riff.c > @@ -377,6 +377,7 @@

Re: [FFmpeg-devel] [PATCH] rtmpdh: Initialize gcrypt before using it

On Sun, Jan 10, 2016 at 07:34:27PM +, Ricardo Constantino wrote: > Tried with 1.5.4 with the link on the first email and another one from > Crunchyroll and it worked. > Tried compiling 1.5.0 but didn't work (on MinGW). > rtmpdh.c | 11 ++- > 1 file changed, 10 insertions(+), 1

Re: [FFmpeg-devel] [PATCH 1/2] lavf/concatdec: do not access packet if av_read_frame returned error

On Fri, 8 Jan 2016, Nicolas George wrote: Le decadi 10 nivôse, an CCXXIV, Marton Balint a écrit : Signed-off-by: Marton Balint --- libavformat/concatdec.c | 10 +++--- 1 file changed, 7 insertions(+), 3 deletions(-) LGTM, thanks, and sorry to have forgotten. Applied,

Re: [FFmpeg-devel] [PATCH 2/2] lavc/WIP: allow subtitle text format to be ASS without timing

On Sun, Jan 10, 2016 at 02:03:46PM +0100, wm4 wrote: > On Sat, 9 Jan 2016 14:46:52 +0100 > Clément Bœsch wrote: > > > From: Clément Bœsch > > > > --- > > TODO > > - ass_split needs to be able to handle dialog event without "Dialogue: > > ..." so our

Re: [FFmpeg-devel] [PATCH] rtmpdh: Initialize gcrypt before using it

Tried with 1.5.4 with the link on the first email and another one from Crunchyroll and it worked. Tried compiling 1.5.0 but didn't work (on MinGW). 0001-rtmpdh-Initialize-gcrypt-before-using-it.patch Description: Binary data ___ ffmpeg-devel mailing

Re: [FFmpeg-devel] [libav-devel] [PATCH 1/2] avcodec: Add Cineform HD Decoder

On 2016-01-10 00:28:49 +, Kieran Kunhya wrote: > --- > libavcodec/Makefile | 1 + > libavcodec/allcodecs.c | 1 + > libavcodec/avcodec.h| 1 + > libavcodec/cfhd.c | 565 > > libavcodec/cfhd.h | 99 + >

Re: [FFmpeg-devel] [PATCH v5] lavf/matroskadec: Normalize noncompliant A_QUICKTIME/V_QUICKTIME private data

On Sun, Jan 10, 2016 at 04:04:47AM +0100, Mats Peterson wrote: > How about this one, Michael? Rather than using the obviously > dangerous av_realloc() it uses av_malloc() to allocate new memory > for the private data, then copies the private data to it, frees the > pointer to the original private

Re: [FFmpeg-devel] [PATCH 1/2] avcodec: Add Cineform HD Decoder

Hi, Great work, but I found some issues. This I got with fuzzed file: ==13422== Invalid read of size 4 ==13422==at 0x834D003: cfhd_decode (cfhd.c:293) ==13422==by 0x863D051: frame_worker_thread (pthread_frame.c:147) ==13422==by 0x40ECF6F: start_thread (pthread_create.c:312)

Re: [FFmpeg-devel] [PATCH] rtmpdh: Initialize gcrypt before using it

On 10 January 2016 at 19:54, Michael Niedermayer wrote: > > please explain in the commit message why secmem is disabled > > 0001-rtmpdh-Initialize-gcrypt-before-using-it.patch Description: Binary data ___ ffmpeg-devel mailing

Re: [FFmpeg-devel] [PATCH v4] lavf/matroskadec: Normalize noncompliant A_QUICKTIME/V_QUICKTIME private data

On Sun, Jan 10, 2016 at 03:05:49 +0100, Mats Peterson wrote: > OK, but I thought an AVERROR(ENOMEM) eventually causes FFmpeg to exit? Even if so, this code is in libavformat, and may be called by any other application than ffmpeg. Moritz ___

[FFmpeg-devel] [PATCH 2/3] avfilter/avf_showspectrum: reduce number of operations

Signed-off-by: Paul B Mahol --- libavfilter/avf_showspectrum.c | 14 -- 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/libavfilter/avf_showspectrum.c b/libavfilter/avf_showspectrum.c index 7fd0ea9..424d1f2 100644 --- a/libavfilter/avf_showspectrum.c

[FFmpeg-devel] [PATCH 1/3] avfilter/avf_showspectrum: add posibility to display phase

Signed-off-by: Paul B Mahol --- doc/filters.texi | 3 +++ libavfilter/avf_showspectrum.c | 56 +++--- 2 files changed, 55 insertions(+), 4 deletions(-) diff --git a/doc/filters.texi b/doc/filters.texi index efb050d..45d22f4

[FFmpeg-devel] [PATCH 3/3] avfilter: add spectrumsynth filter

Signed-off-by: Paul B Mahol --- Only gray input supported, chroma is ignored. (float pixel format would come handy) Only fullframe slide mode implemented. No overlaping and only rectangular window for now. Only power of 2 height supported and tested. --- libavfilter/Makefile