Compile with generic mid_pred function written in the old way on the mips
platform, and objdump its .o file, it will show like this, with more
instructions and branch jump. ```code mid_pred(int, int,
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.or
A gentle reminder for this patch: it's a simple fix that prevents crashing on
iOS.
Thanks
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-req
On Tue, Nov 29, 2022 at 11:47 AM Alessandro Di Nepi <
alessandro.din...@gmail.com> wrote:
> Just to add that this fix, once approved, should be cherry-picked to all
> the release branches where d7f4ad88a0df3c1339e142957bf2c40cd056b8ce has
> been cherry-picked.
> Basically, 4.4, 5.0, and 5.1.
>
> T
From: Zhao Zhili
It's well known that mediacodec encoder requires 16x16 alignment.
Use our bsf to fix the crop info.
---
configure | 2 ++
libavcodec/mediacodecenc.c | 65 +++---
2 files changed, 63 insertions(+), 4 deletions(-)
diff --git a/con
From: Zhao Zhili
Use input PTS as DTS has multiple problems:
1. If there is no reordering, it's better to just use the output
PTS as DTS, since encoder may change the timestamp value (do it
on purpose or rounding error).
2. If there is reordering, input PTS should be shift a few frames
as DTS to
From: Zhao Zhili
---
libavcodec/mediacodecenc.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libavcodec/mediacodecenc.c b/libavcodec/mediacodecenc.c
index 4b0397d1ea..9378563fac 100644
--- a/libavcodec/mediacodecenc.c
+++ b/libavcodec/mediacodecenc.c
@@ -264,6 +264,8 @@ static av_cold i
From: Zhao Zhili
---
libavcodec/mediacodecenc.c | 143 -
libavcodec/version.h | 2 +-
2 files changed, 142 insertions(+), 3 deletions(-)
diff --git a/libavcodec/mediacodecenc.c b/libavcodec/mediacodecenc.c
index c28cce56c6..d098880941 100644
--- a/lib
From: Zhao Zhili
---
libavcodec/mediacodecenc.c | 25 +
libavcodec/version.h | 2 +-
2 files changed, 26 insertions(+), 1 deletion(-)
diff --git a/libavcodec/mediacodecenc.c b/libavcodec/mediacodecenc.c
index 69246ad693..c8d8f84e46 100644
--- a/libavcodec/mediacod
From: Zhao Zhili
---
libavcodec/mediacodecenc.c | 42 ++
1 file changed, 24 insertions(+), 18 deletions(-)
diff --git a/libavcodec/mediacodecenc.c b/libavcodec/mediacodecenc.c
index c8d8f84e46..c28cce56c6 100644
--- a/libavcodec/mediacodecenc.c
+++ b/libavcod
From: Zhao Zhili
It works since most of Android devices don't output B frames by
default. The behavior is documented by Android now, although there
is some exception in history, which should have been fixed now.
---
libavcodec/mediacodecenc.c | 8
libavcodec/version.h | 2 +-
2 fi
On 4 Dec 2022, 17:01 +0200, FFmpeg development discussions and patches
, wrote:
> When this happens, does it continue happening, or is it transient? My main
> concern is log spamming.
Good question: this is just a transient state, so that it won't continue
happening.
To give you some context: whe
Hello,
As discussed at the developer meeting it would be good to have an AVX-512
(ICL) FATE machine and also a development machine.
What I didn't realise until a few weeks ago is that 11th Generation NUCs
have AVX512ICL support.
I suggest buying:
2x
https://www.scan.co.uk/products/intel-nuc-bnuc
xv30be is an obnoxious format that I shouldn't have included in the
first place. xv30 packs 3 10bit channels into 32bits and while our
byte-oriented logic can handle Little Endian correctly, it cannot
handle Big Endian. To avoid that, I marked xv30be as a bitstream
format, but while that didn't pro
On Wed, Nov 23, 2022 at 11:35 AM wrote:
> From: Mark Reid
>
> This patch series adds swscale input/output support for the packed rgb
> float formats.
> A few of the filters also needed support the larger 96/128 bit packed
> pixel sizes.
>
> I also plan to eventually add lossless unscaled convers
This will be needed for the following commit, after which ff_get_buffer() will
stop setting frame->pts to AV_NOPTS_VALUE.
Signed-off-by: James Almer
---
libavcodec/binkaudio.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/libavcodec/binkaudio.c b/libavcodec/binkaudio.c
The idea behind last_pkt_props was to store the properties of the last packet
fed to the decoder. Any sort of queueing required by decoders that consume
several packets before they start outputting frames should be done by the
decoders in question. An example of this is in the libdav1d wrapper.
Th
Use the opaque field instead to store this value.
No functional change, but removes the hack that made the packet technically
invalid, allowing the safe usage of functions like av_packet_ref() on it
if required.
Signed-off-by: James Almer
---
libavcodec/decode.c | 7 ---
1 file changed, 4 i
Making it point to the input packet results in different behavior during flush,
where its contents will be that of an empty packet instead of containing the
props from the last input packet fed to the decoder.
After this change, decoding with more than one thread will shield the same
results as usi
This ensures the video stream duration is not lost after decoding.
Signed-off-by: James Almer
---
libavcodec/h263dec.c | 13 +
libavcodec/mpegvideo.h | 1 +
2 files changed, 14 insertions(+)
diff --git a/libavcodec/h263dec.c b/libavcodec/h263dec.c
index ac7a8521e5..0a2d7487a8 100
Signed-off-by: Marton Balint
---
libavformat/mov.c | 4
1 file changed, 4 insertions(+)
diff --git a/libavformat/mov.c b/libavformat/mov.c
index 29bd3103e3..935b2f8d9f 100644
--- a/libavformat/mov.c
+++ b/libavformat/mov.c
@@ -8772,6 +8772,10 @@ static int mov_read_packet(AVFormatContext *s
Patch 03d81a044ad587ea83567f75dc36bc3d64278199 disallowed zero sample sizes,
but there are some files in the wild which have zero sized samples (e.g.
no audio in some part of a live recording).
Fix this by only disallowing zero sized samples if the size is coming from the
default sample size and n
On Wed, Nov 23, 2022 at 11:35:40AM -0800, mindm...@gmail.com wrote:
> From: Mark Reid
>
> ---
> libswscale/output.c | 92
> libswscale/swscale_unscaled.c| 4 +-
> libswscale/tests/floatimg_cmp.c | 4 +-
> libswscale/utils.c
On Fri, 2 Dec 2022, Gyan Doshi wrote:
On 2022-12-02 06:16 am, Chris Ribble wrote:
On Thu, Dec 1, 2022 at 4:51 PM Marton Balint wrote:
Can you explain why those files are considered valid, or why it makes
sense to generate such files?
Thanks,
Marton
As far as I can tell, the file
> More strict enforcement of sample size was introduced to avoid DOS/Timeout
> with crafted (fuzzed) files and disallow emitting zero sized packets.
>
> Invalid file support is not something that is always worth doing, there
> are other, more important factors, like limiting code complexity or
> im
Xavier Laffargue 于2022年12月4日周日 02:38写道：
>
> Signed-off-by: Xavier Laffargue
> ---
> libavformat/hlsenc.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c
> index a86fc8907f..92276084aa 100644
> --- a/libavformat/hlsenc.c
> +++ b/l
Basel Sayeh 于2022年11月29日周二 08:08写道：
>
> Please ignore this version and use the other one, as it was sent using
> the wrong email client and the patch is missed up.
Hi Basel,
Could you send a new version patchset name lead by "v2" ?
git format -patch -v 2
Thanks
Steven
> On Nov 19, 2022, at 02:48, Zhao Zhili wrote:
>
> From: Zhao Zhili
>
> Unlike the pipe protocol, fd protocol has seek support if it
> corresponding to a regular file.
> ---
> Sometimes it's the only way to access files via file descriptor, e.g.,
> requesting a shared file on Android:
> http
Zhao Zhili 于2022年12月5日周一 01:13写道：
>
> From: Zhao Zhili
>
> ---
> libavcodec/mediacodecenc.c | 42 ++
> 1 file changed, 24 insertions(+), 18 deletions(-)
>
> diff --git a/libavcodec/mediacodecenc.c b/libavcodec/mediacodecenc.c
> index c8d8f84e46..c28cce56c6 100
On Wed, 2022-11-23 at 13:54 -0800, Brian Norris wrote:
> There can be more than one available render node, and it's not
> guaranteed the first node we come across is the correct one. In
> particular, 'vgem' devices are common, and are
> never VAAPI-enabled and thus not valid here.
>
> We have a 'k
> On Dec 5, 2022, at 12:23, Steven Liu wrote:
>
> Zhao Zhili 于2022年12月5日周一 01:13写道：
>>
>> From: Zhao Zhili
>>
>> ---
>> libavcodec/mediacodecenc.c | 42 ++
>> 1 file changed, 24 insertions(+), 18 deletions(-)
>>
>> diff --git a/libavcodec/mediacodecenc.c b
From: Mark Reid
There are some places in input.c that could use it too
but they aren't currently being pass the SwsContext
---
libswscale/output.c | 36 +++
libswscale/swscale_internal.h | 3 +++
libswscale/swscale_unscaled.c | 26 +-
From: Mark Reid
Also renamed bswap_buf to bswap32_buf
---
configure | 56 ---
libavcodec/4xm.c | 14 ++---
libavcodec/Makefile | 1 -
libavcodec/ac3dec.c | 4 +-
On Sun, Dec 4, 2022 at 4:05 PM Michael Niedermayer
wrote:
> On Wed, Nov 23, 2022 at 11:35:40AM -0800, mindm...@gmail.com wrote:
> > From: Mark Reid
> >
> > ---
> > libswscale/output.c | 92
> > libswscale/swscale_unscaled.c| 4 +-
> > l
> From dfa35dd22e9178971a47709065ed4e976f57f8ef Mon Sep 17 00:00:00 2001
> From: galinart
> Date: Fri, 4 Nov 2022 11:10:52 +
> Subject: [PATCH] libavcodec/qsvenc: enable Hyper Encode
>
> Hyper Encode uses Intel integrated and discrete graphics on one system
> to accelerate encoding of a sin
Follow spec 7.3.2.3.1.
Signed-off-by: Fei Wang
---
update:
1. fix uninitialized variable which may cause segment fault.
libavcodec/hevc_ps.c | 2 +-
tests/ref/fate/hevc-conformance-PS_A_VIDYO_3 | 50 ++--
2 files changed, 26 insertions(+), 26 deletions(-
From: Linjie Fu
Described in HEVC spec A.3.7.
Signed-off-by: Linjie Fu
Signed-off-by: Fei Wang
---
libavcodec/avcodec.h | 1 +
libavcodec/hevc_ps.c | 2 ++
libavcodec/profiles.c | 1 +
3 files changed, 4 insertions(+)
diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
index 3edd8e263
From: Linjie Fu
1. Add extension syntax according to 7.3.2.2.3/7.3.2.3.3 in T-REC-H.265-201911.
2. Keep using parsed PPS when bitstream overread for compatibility. For
example, the clip PS_A_VIDYO_3.bit in FATE test has incomplete extension
syntax which will be overread and un-decodable if withou
From: Linjie Fu
Signed-off-by: Linjie Fu
Signed-off-by: Fei Wang
---
libavcodec/hevcdec.c | 6 ++
libavcodec/hevcdec.h | 4
2 files changed, 10 insertions(+)
diff --git a/libavcodec/hevcdec.c b/libavcodec/hevcdec.c
index 567e8d81d4..f9a97ac7f5 100644
--- a/libavcodec/hevcdec.c
+++ b/
From: Linjie Fu
According to 7.3.6.1, use_integer_mv_flag should be parsed if
motion_vector_resolution_control_idc equals to 2. If not present, it
equals to motion_vector_resolution_control_idc.
Signed-off-by: Linjie Fu
Signed-off-by: Fei Wang
---
libavcodec/hevcdec.c | 8
libavcodec
From: Linjie Fu
Including sps/pps/slice parameters.
Signed-off-by: Linjie Fu
Signed-off-by: Fei Wang
---
libavcodec/vaapi_hevc.c | 52 +
1 file changed, 47 insertions(+), 5 deletions(-)
diff --git a/libavcodec/vaapi_hevc.c b/libavcodec/vaapi_hevc.c
ind
From: Linjie Fu
uint8_t is big enough and keep consistent with the definition in
cbs_h265.h.
Signed-off-by: Linjie Fu
Signed-off-by: Fei Wang
---
libavcodec/hevcdec.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/hevcdec.h b/libavcodec/hevcdec.h
index a7fc669b
From: Linjie Fu
According to 8.1.3 and 8.3.2.
Signed-off-by: Linjie Fu
Signed-off-by: Fei Wang
---
libavcodec/vaapi_hevc.c | 4
1 file changed, 4 insertions(+)
diff --git a/libavcodec/vaapi_hevc.c b/libavcodec/vaapi_hevc.c
index 6ce1e17fa8..005d782819 100644
--- a/libavcodec/vaapi_hevc.
From: Linjie Fu
Screen Content Coding allows non-intra slice in an IRAP frame which can
reference the frame itself, and would mark the current decoded picture
as "used for long-term reference", no matter TwoVersionsOfCurrDecPicFlag(8.1.3),
hence some previous restricts are not suitable any more.
From: Linjie Fu
Allow current picture as the reference picture.
Signed-off-by: Linjie Fu
Signed-off-by: Fei Wang
---
libavcodec/vaapi_hevc.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/libavcodec/vaapi_hevc.c b/libavcodec/vaapi_hevc.c
index 005d782819..ca14052d56
Keep same style with IS_IDR()/IS_BLA().
Signed-off-by: Fei Wang
---
libavcodec/hevcdec.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/hevcdec.h b/libavcodec/hevcdec.h
index aab816791e..94609e4699 100644
--- a/libavcodec/hevcdec.h
+++ b/libavcodec/hevcdec.h
@@ -7
Signed-off-by: Fei Wang
---
libavcodec/vaapi_hevc.c | 6 --
1 file changed, 6 deletions(-)
diff --git a/libavcodec/vaapi_hevc.c b/libavcodec/vaapi_hevc.c
index ca14052d56..b3ff2f7344 100644
--- a/libavcodec/vaapi_hevc.c
+++ b/libavcodec/vaapi_hevc.c
@@ -538,12 +538,6 @@ static int vaapi_hevc
From: Linjie Fu
Note that Screen-Extended Main 4:4:4 and 4:4:4 10 supports
chroma_format_idc from 0, 1 or 3, hence both 420 and 444 are
supported.
Signed-off-by: Linjie Fu
Signed-off-by: Fei Wang
---
libavcodec/vaapi_decode.c | 4 +++-
libavcodec/vaapi_hevc.c | 14 --
libavcode
On Do, 2022-12-01 at 10:48 +0800, wenbin.chen-at-intel@ffmpeg.org wrote:
> From: Wenbin Chen
>
> When process yuv420 frames, FFmpeg uses same alignment on Y/U/V
> planes. VPL and MSDK use Y plane's pitch / 2 as U/V planes's
> pitch, which makes U/V planes 16-bytes aligned. We need to set
> a
On Thu, 2022-12-01 at 22:52 -0300, James Almer wrote:
> On 12/1/2022 5:13 AM, Fei Wang wrote:
> > +static void colour_mapping_octants(GetBitContext *gb, HEVCPPS
> > *pps, int inp_depth,
> > + int idx_y, int idx_cb, int
> > idx_cr, int inp_length)
> > +{
> > +ui
> On Do, 2022-12-01 at 10:48 +0800, wenbin.chen-at-intel@ffmpeg.org
> wrote:
> > From: Wenbin Chen
> >
> > When process yuv420 frames, FFmpeg uses same alignment on Y/U/V
> > planes. VPL and MSDK use Y plane's pitch / 2 as U/V planes's
> > pitch, which makes U/V planes 16-bytes aligned. We nee
Andreas Rheinhardt:
> This commit tests it in a way that automatically checks
> that using av_dict_iterate() is equivalent to using
> av_dict_get() with key "" and AV_DICT_IGNORE_SUFFIX.
>
> Signed-off-by: Andreas Rheinhardt
> ---
> libavutil/tests/dict.c | 30 +-
> 1
From: Haihao Xiang
Make it easy to add new extra mfx parameter buffer. No functional
change.
Signed-off-by: Haihao Xiang
---
libavfilter/vf_vpp_qsv.c | 86
1 file changed, 34 insertions(+), 52 deletions(-)
diff --git a/libavfilter/vf_vpp_qsv.c b/libavf
From: Wenbin Chen
When process yuv420 frames, FFmpeg uses same alignment on Y/U/V
planes. VPL and MSDK use Y plane's pitch / 2 as U/V planes's
pitch, which makes U/V planes 16-bytes aligned. We need to set
a separate alignment to meet runtime's behaviour.
Now alignment is changed to 16 so that t
53 matches
Mail list logo
