components are divided by the spec weights) */
-static const int dv100_weight_shift = 16;
static const int dv_weight_1080[2][64] = {
{ 8192, 65536, 65536, 61681, 61681, 61681, 58254, 58254,
58254, 58254, 58254, 58254, 55188, 58254, 58254, 55188,
--
2.29.2
-- Peter
(A907 E02F A6E5 0CD2 34CD
(AVFilterContext *ctx)
{
WaveformContext *s = ctx->priv;
--
2.29.2
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listi
ring if it makes sense to send in a zero sized packet for muxing
or not. That is, should I try to fix that or should I rather try to compensate
for this in oggenc.c?
Regards,
Peter
> On 15 Dec 2020, at 14:35, Peter Zebühr wrote:
>
> Hello!
>
> I have stumbled on an issue with th
hat
is failing?
I can have a look at that, but not sure if there is anything there. The other
alternative that I can see is to update the reference for the fate test to
expect one more page in the test that is failing.
/ Peter
___
ffmpeg-devel maili
, let me know what you think.
/ Peter
0001-Fix-ogg-page-termination-on-even-last-packets.patch
Description: Binary data
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link
2] = {
> -{ 1, 2 },
> -{ -0, -1 }, // '00', '01'
> -{ -2, -3 }, // '10', '11'
> -};
> -
> static const uint8_t vp8_coeff_band[16] = {
> 0, 1, 2, 3, 6, 4, 5, 6, 6, 6, 6, 6, 6, 6, 6, 7
> };
looks good
-- Peter
(A907
Ping
> On 14 Jan 2021, at 18:13, Peter Zebühr wrote:
>
> Hi,
>
> I had a look at different options on how to fix the regression, though I am
> not sure what I arrived at is what you had in mind.
>
> I tried to read up a bit on opus as well and I assume you are ta
OBJS-$(CONFIG_AVI_MUXER) += avienc.o mpegtsenc.o avlanguage.o
rawutils.o
OBJS-$(CONFIG_AVM2_MUXER)+= swfenc.o swf.o
OBJS-$(CONFIG_AVR_DEMUXER) += avr.o pcm.o
--
2.29.2
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
> pkt->data, pkt->size, pkt->pos);
> --
> 2.25.1
ok
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: PGP signature
___
ffmpeg-devel mailing list
->theora = get_bits(gb, 24);
> av_log(avctx, AV_LOG_DEBUG, "Theora bitstream version %X\n", s->theora);
> --
> 2.17.1
ok
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: PGP signature
0)
> +return AVERROR_INVALIDDATA;
> +st->codecpar->bits_per_coded_sample = bpc;
> } else
> return AVERROR_INVALIDDATA;
ok
could tighten it further. realistically (audio) bits per codeded sample will
never >16 here.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D
6); // Reserved1 .. AutoExpRect
+set_metadata_float(&st->metadata, "wbgain[0].r",
av_int2float(avio_rl32(pb)), 1);
+set_metadata_float(&st->metadata, "wbgain[0].b",
av_int2float(avio_rl32(pb)), 1);
+avio_skip(pb, 36); // WBGain[1].. WBView
32(pb), 1);
-set_metadata_int(&st->metadata, "gamma", avio_rl32(pb), 1);
+set_metadata_float(&st->metadata, "gamma", xlat_gamma(avio_rl32(pb)), 0);
avio_skip(pb, 12 + 16); // Reserved1 .. AutoExpRect
set_metadata_float(&st->metadata, &
On Thu, Nov 13, 2014 at 12:58:07AM +0100, Michael Niedermayer wrote:
> On Sun, Nov 09, 2014 at 12:05:41PM +1100, Peter Ross wrote:
> > For cehoyos:
> >
> > The PhantomCINE file format stores unprocessed sensor data from the
> > camera CFA, and reports image processi
See http://www.itu.int/rec/R-REC-BT.709
Item 1.2, overall opto-electronic transfer characteristics at source
Signed-off-by: Peter Ross
---
This is required for cine playback.
(Yes, the transform still occurs in yuv colorspace, which is a limitation.)
libavfilter/vf_lut.c | 17
Signed-off-by: Peter Ross
---
libavformat/aiffdec.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavformat/aiffdec.c b/libavformat/aiffdec.c
index 91ef2a4..6271c3d 100644
--- a/libavformat/aiffdec.c
+++ b/libavformat/aiffdec.c
@@ -116,13 +116,13 @@ static unsigned
Signed-off-by: Peter Ross
---
Revised patch. Now passes FATE.
libavformat/aiffdec.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/libavformat/aiffdec.c b/libavformat/aiffdec.c
index 91ef2a4..8dbed32 100644
--- a/libavformat/aiffdec.c
+++ b/libavformat/aiffdec.c
Signed-off-by: Peter Ross
---
Thanks Reimar.
libavformat/aiffdec.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/libavformat/aiffdec.c b/libavformat/aiffdec.c
index 8dbed32..301d90f 100644
--- a/libavformat/aiffdec.c
+++ b/libavformat/aiffdec.c
@@ -116,12 +116,12
On Tue, Apr 07, 2015 at 10:22:17AM +1000, Peter Ross wrote:
> On Mon, Apr 06, 2015 at 07:24:35PM +0100, Derek Buitenhuis wrote:
> > On 4/6/2015 4:43 AM, Michael Niedermayer wrote:
> [..]
> > > * Conferences / exhibitions, public stuff
> > > Theres a conferen
.vp6 files may contain two video streams: one for the primary video
stream and another for the alpha mask. The file format uses identical
data structures for both streams.
Signed-off-by: Peter Ross
---
libavformat/electronicarts.c | 136 +--
1 file
This is required for the alpha stream demux patch.
Signed-off-by: Peter Ross
---
libavformat/electronicarts.c | 123 +++
1 file changed, 67 insertions(+), 56 deletions(-)
diff --git a/libavformat/electronicarts.c b/libavformat/electronicarts.c
index
Electronic Arts VP6 files may contain two video streams: one for the
primary video stream and another for the alpha mask. The file format
uses identical data structures for both streams.
Signed-off-by: Peter Ross
---
libavformat/electronicarts.c | 18 +++---
1 file changed, 15
Signed-off-by: Peter Ross
---
The binary file is in /incoming/fate-ea-vp6 (64kb)
-map 0 required because there are two video streams in this file.
tests/fate/demux.mak | 3 ++
tests/ref/fate/d-eavp6-demux | 98
2 files changed, 101
On Fri, Jun 26, 2015 at 05:47:16PM +, Carl Eugen Hoyos wrote:
> Peter Ross xvid.org> writes:
>
> > Electronic Arts VP6 files may contain two video
> > streams: one for the primary video stream and
> > another for the alpha mask. The file format
> > uses i
On Mon, Oct 19, 2015 at 11:35:15AM +0200, Paul B Mahol wrote:
> Such files have empty gaps between chunks.
Paul, where might I find a sample to test?
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: Digital signat
gt; +if (avio_feof(pb))
> +ret = AVERROR_EOF;
> packet_read = 1;
> break;
>
LGTM. Tested with samples from NFS8U2.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: Digital signature
__
Signed-off-by: Peter Ross
http://wiki.multimedia.cx/index.php?title=Electronic_Arts_SCxl
---
libavformat/electronicarts.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavformat/electronicarts.c b/libavformat/electronicarts.c
index 5d21d49..335a635 100644
--- a/libavformat
rts files
'end audio stream tags' (SEAD, 1SNx, SCxLl). but when with this hunk applied,
the loop only ends early
if 'SCHl_TAG' is reached.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: Digital signature
__
->height);
> +return AVERROR_INVALIDDATA;
> +}
> avio_skip(pb, 8 + 16 + 24); // black_level, white_level, xywh,
> active_area, exposure_bias
> if (avio_rl32(pb) != 0x2010100) /* RGGB */
uint64_t position = avio_rl64(pb);
> +while (frame_nb > e->size && e <= e_end) {
> + e->pos = last_position;
> + e++;
^^^ this indent isn't aligned.
ot
ctually gets used when one demuxes/
> operates on the data per wikipedia.
> Thus, I believe previous behavior was correct.
converting between DSD->PCM reduces the sample rate by 8.
at present, the DSD->PCM conversion happens libavcodec/dsddec.c.
however, a 'codec' cannot chan
On 08/03/2015 01:50 PM, Ronald S. Bultje wrote:
> Just different offsets and ranges. Unsigned, n bits (e.g. u8) is in the
> range [0,2^^n> (e.g. for u8, the range is [0,255] or [0,256>), where
> silence is 2^^n-1, e.g. 0x80 (128) for u8. Signed is in the range
> [-2^^n-1,2^^n-1> (e.g. s16, the most
Hello,
I've been working on FATE tests for FFV1 in the past already [1]. My
tests didn't work on all platforms and therefore never made it upstream.
I think it's better if I try to provide these new tests in smaller
chunks now :)
First of all, there are things I find inconsistent or confusing wi
On 08/30/2015 07:33 PM, Paul B Mahol wrote:
> On 8/30/15, Peter B. wrote:
>> This will produce quite a number of tests :(
>> I guess it is desired to keep the number of tests as low as necessary?
> Not at all. Just look at ffv1 coverage at coverage.ffmpeg.org
:)
I'd love t
On 08/30/2015 09:32 PM, Michael Niedermayer wrote:
>> - Target "fate-vsynth%-*" tests default to sws_flags
>> "accurate_rnd+bitexact". FFV1.3 tests have "neighbor+bitexact". Why?
> it makes more cases lossless IIRC
> the default upscaling + default downscaling is not binary identical
Roger tha
Hello,
I've read the documentation on FATE tests, and would like to add new ones.
How can I reset the FATE test data for certain targets, without having
to re-run the whole test suite?
Thank in advance,
Pb
___
ffmpeg-devel mailing list
ffmpeg-devel@ffm
.
Replaced the reference files accordingly, and ran "make fate" to verify
that they still run properly.
What do you think?
Pb
>From be1b2ee092a82fdf70ccb99eb20d70e228205182 Mon Sep 17 00:00:00 2001
From: Peter B.
Date: Mon, 31 Aug 2015 03:02:36 +0200
Subject: [PATCH] Renamed p
Am Mo, 31.08.2015, 08:31 schrieb Carl Eugen Hoyos:
> Peter B. das-werkstatt.com> writes:
>
>> Replaced the reference files accordingly
>
> This cannot be done, we want to run fate (also)
> with old sources.
Hm. Now I'm confused...
I expected old sources to use old
Am Mo, 31.08.2015, 11:22 schrieb Hendrik Leppkes:
> On Mon, Aug 31, 2015 at 11:13 AM, Peter B. wrote:
>>
>> In my case, there are files in "tests/ref/vsynth" that contain the
>> expected reference results from tests.
>>
>> If I only provide the changed
On 08/31/2015 01:44 PM, Tobias Rapp wrote:
> On 30.08.2015 21:32, Michael Niedermayer wrote:
>>> - Target "fate-vsynth%-*" tests default to sws_flags
>>> "accurate_rnd+bitexact". FFV1.3 tests have "neighbor+bitexact". Why?
>>
>> it makes more cases lossless IIRC
>> the default upscaling + defa
On 08/30/2015 09:32 PM, Michael Niedermayer wrote:
> probably, send patches
> and probably better few and small ones at once then wait and see
> in which direction reviewes go before spending too much time in some
> specific direction
For starters, I'd like to extract the FATE tests for FFV1 to it
On 09/01/2015 09:54 PM, Peter B. wrote:
> [...]
> But actually, I think it'd be best to change the FFV1 tests to match the
> ones in "lossless-video.mak", using framecrc rather than the enc_dec
> function used in "vcodec.mak".
>
> Changing that however,
This patch makes ffv1.mak work without requiring (or being included in)
vcodec.mak.
That's another step towards having a better environment for additional
FFV1 tests.
Regards,
Pb
>From 593f1551b999fbb4c4a378d0723c2140e3b469df Mon Sep 17 00:00:00 2001
From: Peter B.
Date: Tue, 1 Sep 2015
On 09/01/2015 11:40 PM, Michael Niedermayer wrote:
>> +include tests/fate/ffv1.mak
> doesnt this belong in tests/Makefile ?
I did so first, but then all the prerequisites set in vcodec.mak are not
applied, causing ffv1.mak to fail.
So in order to keep the modifications as little as possible, I jus
Previous patch worked with "make fate", but fails to build FFmpeg after
"make clean".
I've had to do fix some unclean parts of ffv1.mak.
Patch attached.
Now everything seems to be fine again.
Regards,
Pb
>From bb0b1dd72dd334d6621221e3c55d53f8cc0485b8 Mon Sep 17 00:00
On 09/02/2015 02:53 PM, Michael Niedermayer wrote:
> the include breaks out of tree builds
> mkdir delthisdir
> cd delthisdir
> ../configure
> make -j12
> /home/michael/ffmpeg/tests/fate/vcodec.mak:71: tests/fate/ffv1.mak: No such
> file or directory
> make: *** No rule to make target `tests/fate/
I've now combined my changes into one patch.
They're basically moving ffv1 tests to ffv1.mak, and making that
independent of vcodec.mak.
Hope it works now :)
Thanks and regards,
Pb
>From 5ba0e0510269e0ee591a9505fcb4654044002743 Mon Sep 17 00:00:00 2001
From: Peter B.
Date: Tue,
On 09/07/2015 09:14 PM, Michael Niedermayer wrote:
> On Thu, Sep 03, 2015 at 10:14:13PM +0200, Peter B. wrote:
>> I've now combined my changes into one patch.
>> They're basically moving ffv1 tests to ffv1.mak, and making that
>> independent of vcodec.mak.
>>
On 09/08/2015 01:46 AM, Michael Niedermayer wrote:
> On Mon, Sep 07, 2015 at 11:18:56PM +0200, Peter B. wrote:
>> Should be fixed.
>> (Getting errors with "make fate" for some fate-dxv*, with git HEAD but
>> that seems unrelated to my changes)
> this seems to loo
MINGW64_NT-6.0
The attached patch seems to fix it.
Regards, Peter
diff -rc ./a/ffmpeg/configure ./b/ffmpeg/configure
*** ./a/ffmpeg/configureFri Sep 11 19:20:03 2015
--- ./b/ffmpeg/configureSat Sep 12 10:16:52 2015
***
*** 3297,3303
exesuf() {
case
On 12.09.2015 11:15, Geek.Song wrote:
diff -u
--- ./a/ffmpeg/configure2015-09-11 19:20:03.0 +0200
+++ ./b/ffmpeg/configure2015-09-12 11:36:32.78060 +0200
@@ -3297,7 +3297,7 @@
exesuf() {
case $1 in
-mingw32*|win32|win64|cygwin*|*-dos|freedos|opendos|o
Hi :)
I'm having difficulties implement FFV1 lossless FATE tests in a way that
seems more appropriate.
I'd be grateful for advice how to deal with this:
I'd rather use "framecrc" instead of the "enc_dec" method used in
vodec.mak. Yet, the usage of framecrc in the tests for other lossless
codecs i
/doxygen/trunk/structAVOption.html
>From d51af4c272d3573f4c4c94dcfe93c7c45e9d7e36 Mon Sep 17 00:00:00 2001
From: Peter B
Date: Sun, 13 Sep 2015 19:04:20 +0200
Subject: [PATCH] Adding ffv1 help reference for "coder" parameter.
---
libavcodec/ffv1enc.c |1 +
1 file changed, 1 insertion(+)
diff --git a/liba
Hello again :)
I've just noticed that the output of "ffprobe -debug 1" for FFV1 files
has different labels for same parameters in different versions.
For example:
== FFV1.1:
[ffv1 @ 0x3a8ab80] ver:1 keyframe:1 coder:0 ec:0 slices:1 bps:8
== FFV1.3:
[ffv1 @ 0x31584e0] global: ver:3.4, coder:0, c
Hello again :)
I've just noticed that by default, FFV1 chooses version "0".
I thought it would be "1".
When adding "-level 1" it properly shows "ver:1".
(and a proper value for "bps")
== For example:
Source video: https://media.xiph.org/video/derf/y4m/football_422_ntsc.y4m
$ ffmpeg -i input.a
On 09/14/2015 12:25 AM, Michael Niedermayer wrote:
> the encoder chooses the lowest version that supports the requested
> features well to maximize compatibility
Thanks for the explanation.
Sounds reasonable.
If I read [1] correcly, FFV1.0 does not store "bits_per_raw_sample".
I'd be interested i
I wanted to check what happens if someone uses an invalid parameter
combination:
"-c:v ffv1 -level 1 -slices 2"
I expected an error message about invalid slice number, but the code
selected FFV1.2:
"[ffv1 @ 0x391b060] Version 2 needed for requested features but version
2 is experimental and not e
On 09/13/2015 11:48 PM, Michael Niedermayer wrote:
>> - What is "keyframe"?
> a random access point
I'm not sure what your answer means.
I understand that "A keyframe" is a random access point. The flag in
ffprobe's output seems to always be "1".
I thought it might be the same as "intra" in FFV1.3
On 09/15/2015 08:47 PM, Michael Niedermayer wrote:
> On Tue, Sep 15, 2015 at 01:31:20AM +0200, Peter B. wrote:
>> I wanted to check what happens if someone uses an invalid parameter
>> combination:
>> "-c:v ffv1 -level 1 -slices 2"
>>
>> I expected an er
On Wed, Feb 04, 2015 at 11:27:46PM +0100, Carl Eugen Hoyos wrote:
> Hi!
>
> Attached patch fixes the crash from ticket #4296 for me.
>
> Please review, Carl Eugen
LGTM.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: Di
On Sat, Feb 7, 2015 at 6:23 AM, arwa arif wrote:
> On Mon, Feb 2, 2015 at 10:23 AM, Michael Niedermayer
> wrote:
>
> > On Sat, Jan 31, 2015 at 02:30:29PM +0530, arwa arif wrote:
> > > On Thu, Jan 29, 2015 at 5:18 PM, Stefano Sabatini
> > > wrote:
> > >
> > > > On date Thursday 2015-01-29 03:46:
it.) For review, I'd suggest looking at the
diff of master:av_dict. The 3 commits after that changing libx264.c and
libx265.c should each stand on their own. (And actually the libx264.c one
should be split into a couple commits.)
Anyway, I wanted to post first
On Wed, Mar 4, 2015 at 9:42 AM, Michael Niedermayer
wrote:
> On Wed, Mar 04, 2015 at 02:24:58PM +0100, Michael Niedermayer wrote:
> > On Wed, Mar 04, 2015 at 02:11:42PM +0100, Michael Niedermayer wrote:
> > > On Tue, Mar 03, 2015 at 10:51:01PM -0400,
ormats.c
+++ b/libavformat/allformats.c
@@ -392,4 +392,5 @@ void av_register_all(void)
REGISTER_PROTOCOL(LIBRTMPTE,librtmpte);
REGISTER_PROTOCOL(LIBSSH, libssh);
REGISTER_PROTOCOL(LIBSMBCLIENT, libsmbclient);
+REGISTER_PROTOCOL(LIBZIP, libzip);
}
diff --gi
On Fri, Mar 27, 2015 at 02:55:37PM +0100, Lukasz Marek wrote:
> On 27.03.2015 14:16, Michael Niedermayer wrote:
> >On Fri, Mar 27, 2015 at 01:51:08PM +0100, Lukasz Marek wrote:
> >>On 27.03.2015 13:07, Peter Ross wrote:
> >>>---
> >>>this was create
we'd need some way of escaping the character
if it is legitimately part of either path.
ffplay /tmp/amovie.zip
ffplay http://subtitles.org/amovie.zip#amovie.srt
the syntax should support nested archives (even if the libzip/archive cannot
handle
that yet). e.g.
ffplay /tmp/amovie.rar#/tmp/amovie.zip#amovie.srt
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: Digital signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
On Sat, Mar 28, 2015 at 10:24:55PM +0100, wm4 wrote:
> On Sun, 29 Mar 2015 08:10:29 +1100
> Peter Ross wrote:
>
> > On Sat, Mar 28, 2015 at 08:38:40PM +0100, Lukasz Marek wrote:
> > > On 28.03.2015 20:13, Nicolas George wrote:
> > > >L'octidi 8 g
July-August.
My spare time is limited, so to have something ready I need to make a start
soon.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: Digital signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
; case AV_CODEC_ID_DIRAC: maxpixels /= 8192; break;
> -case AV_CODEC_ID_DST: maxsamples /= 8192; break;
> +case AV_CODEC_ID_DST: maxsamples /= 1<<20; break;
approve.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Descri
From: Peter Bennett
The header stdatomic.h causes compile errors in a c++ compiler, even
when extern "C" is used around it. The appropriate c++ include is .
We cannot use #include inside error_resilience.h because that causes
compile errors inside extern "C". Wit
From: Peter Bennett
libavformat.v has url_open, url_close and url_write. These
should be ffurl_ in each case.
---
libavformat/libavformat.v | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/libavformat/libavformat.v b/libavformat/libavformat.v
index cf2055f..7df12d3
On 05/15/2018 08:44 PM, James Almer wrote:
On 5/15/2018 9:25 PM, Peter Bennett wrote:
From: Peter Bennett
The header stdatomic.h causes compile errors in a c++ compiler, even
when extern "C" is used around it. The appropriate c++ include is .
We cannot use #inclu
On 05/15/2018 08:58 PM, James Almer wrote:
On 5/15/2018 9:49 PM, Peter Bennett wrote:
On 05/15/2018 08:44 PM, James Almer wrote:
On 5/15/2018 9:25 PM, Peter Bennett wrote:
From: Peter Bennett
The header stdatomic.h causes compile errors in a c++ compiler, even
when extern "C&quo
On 17/05/18 12:20, Carl Eugen Hoyos wrote:
> 2018-05-10 22:50 GMT+02:00, Carl Eugen Hoyos :
>
>> Peter Bubestinger provided a C210 file where every frame starts with
>> 64 bytes of extradata (24 byte "INFO", 16 byte "RDRT", rest "FIEL").
>>
avcodec_receive_frame consistently causes a seg fault when decoding 1080i mpeg2
on android version oreo. When copying the frame, the second plane in the buffer
follows on immediately after 1080 lines of the first plane, but the code assumes
it is after 1088 lines of the first plane, based on slice_
There is an error in this patch. I will submit a corrected version.
Please disregard this one.
Peter
On 07/19/2018 11:57 AM, Peter Bennett wrote:
avcodec_receive_frame consistently causes a seg fault when decoding 1080i mpeg2
on android version oreo. When copying the frame, the second plane
This is a correction of the earlier submission of this patch.
avcodec_receive_frame consistently causes a seg fault when decoding 1080i mpeg2
on android version oreo. When copying the frame, the second plane in the buffer
follows on immediately after 1080 lines of the first plane, but the code ass
put_lines = put_lines_bytes;
> } else {
> -avpriv_request_sample(avctx, "XIMG planes=%d", planes);
> -return AVERROR_PATCHWELCOME;
> + av_assert0(0);
> }
> } else if (planes == 1) {
> palette[0] =
Signed-off-by: Peter Ross
---
Modified to use codec copy instead of aresample
Samples attached to:
https://lists.ffmpeg.org/pipermail/ffmpeg-devel/2021-December/288976.html
tests/fate/audio.mak | 12
tests/ref/fate/mv-mono16bit | 8
tests/ref/fate/mv-mono8bit
t; (uint64_t)bytes_per_sample);
> }
> } else if (!version && avio_rb16(pb) == 3) {
> avio_skip(pb, 4);
> --
> 2.17.1
please apply.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: PGP signature
__
both patches look good, and do not break tests.
i will apply in a couple of days.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffm
t; if (set_channels(avctx, ast, avio_rb32(pb)) < 0)
> return AVERROR_INVALIDDATA;
please apply.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: PGP signature
___
ffmpeg-devel mailing list
is only about 88 KB.
v2 patchset looks good to me. I will apply in a couple of days. Thanks.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
h
ase? I had to apply it manually when
upgrading our build system to 6.1 recently.
--
\\// Peter - http://www.softwolves.pp.se/
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visi
ret = init_get_bits8(&s->gb, buf, buf_size);
> +if (ret < 0)
> +return ret;
> } else {
> ret = ff_vp56_init_range_decoder(&s->cc, buf, buf_size);
> if (ret < 0)
> --
> 2.17.1
looks good
On Thu, Aug 26, 2021 at 05:08:15AM +, Maryam Ebrahimzadeh wrote:
> Thanks,
> So when will you apply this?
applied.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: PGP signature
___
ffmpeg-devel mailin
}
> av_memcpy_backptr(d+xl, pixels_per_value, s->width - xl);
> run -= s->width / pixels_per_value;
> -xl = s->width;
> +xl = s->width / pixels_per_value * pixels_per_value;
>
{ AV_CODEC_ID_DVAUDIO, 0x0216 },
{ AV_CODEC_ID_ATRAC3, 0x0270 },
+{ AV_CODEC_ID_MSNSIREN,0x028E },
{ AV_CODEC_ID_ADPCM_G722, 0x028F },
{ AV_CODEC_ID_IMC, 0x0401 },
{ AV_CODEC_ID_IAC, 0x0402 },
--
2.33.0
-- Peter
(A907 E02F A6E5 0CD2
in[i] *= -1;
s->tx_fn(s->tx_ctx, s->imdct_out, s->imdct_in, sizeof(float));
--
2.33.0
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-dev
t;number_of_regions, get_bits_left(gb),
+ret = decode_vector(s, s->number_of_regions,
s->decoder_standard_deviation, s->power_categories,
s->imdct_in, s->scale_factor);
if (ret < 0 && !s->microsoft)
--
2.33.0
-- Peter
(A
_error = 1;
+}
if (frame_error) {
memcpy(s->imdct_in, s->backup_frame, number_of_valid_coefs *
sizeof(float));
--
2.33.0
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: PGP signature
___
ffmpeg-
calculated_checksum |= temp1 & 1;
+}
+
+if (checksum != calculated_checksum) {
+if (avctx->err_recognition & AV_EF_EXPLODE)
+return AVERROR_INVALIDDATA;
+frame_error = 1;
+}
+}
if (frame_error) {
mem
On Thu, Sep 09, 2021 at 10:40:06AM -0300, James Almer wrote:
> On 9/9/2021 5:46 AM, Peter Ross wrote:
> > ---
> >
> > Thanks for suggestion. I will apply in a couple days if no objections.
> >
> > libavcodec/siren.c | 32 +++-
&g
if (sample_rate <= 0)
> return AVERROR_INVALIDDATA;
> +st->codecpar->sample_rate = sample_rate;
> avpriv_set_pts_info(st, 33, 1, st->codecpar->sample_rate);
> } else if (!strcmp(name, "SAMPLE_WIDTH")) {
> u
7b763
--- /dev/null
+++ b/libavcodec/gemdec.c
@@ -0,0 +1,343 @@
+/*
+ * GEM Raster image decoder
+ * Copyright (c) 2021 Peter Ross (pr...@xvid.org)
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesse
AGEAUTO_DEMUXER(dds, AV_CODEC_ID_DDS)
IMAGEAUTO_DEMUXER(dpx, AV_CODEC_ID_DPX)
IMAGEAUTO_DEMUXER(exr, AV_CODEC_ID_EXR)
+IMAGEAUTO_DEMUXER(gem, AV_CODEC_ID_GEM)
IMAGEAUTO_DEMUXER(gif, AV_CODEC_ID_GIF)
IMAGEAUTO_DEMUXER(j2k, AV_CODEC_ID_JPEG2000)
IMAGEAUTO_DEMUXER(jpeg,AV_C
;
> +break;
> +}
>
> noise = decoder_standard_deviation[region] * noise_category5[i];
> } else
this fixes the recent msnsiren commit.
please apply
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description
+break;
+}
if (!get_bits1(gb))
decoded_value *=
-decoder_standard_deviation[region];
else
--
2.33.0
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
s
+} else if (get_bits_left(gb) - s->checksum_bits < 0 &&
rate_control + 1 < s->rate_control_possibilities) {
frame_error = 1;
}
--
2.33.0
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
signature.asc
Description: PGP signature
__
+if (avctx->err_recognition & AV_EF_EXPLODE)
+return AVERROR_INVALIDDATA;
+frame_error = 1;
+}
+}
if (frame_error) {
memcpy(s->imdct_in, s->backup_frame, number_of_valid_coefs *
sizeof(float));
-
401 - 500 of 626 matches
Mail list logo
