Ping for the patches 2-4
> From: Wu, Jianhua
> Sent: Thursday, September 30, 2021 4:50 PM
> To: ffmpeg-devel@ffmpeg.org
> Cc: Wu, Jianhua
> Subject: [PATCH v4 2/4] libswscale/x86/rgb2rgb: add shuffle_bytes avx512
>
> Performance data(Less is better):
> shuffle_bytes_avx2 0.94288
> s
Ping.
> From: Wu, Jianhua
> Sent: Friday, October 8, 2021 10:31 AM
> To: ffmpeg-devel@ffmpeg.org
> Cc: Wu, Jianhua
> Subject: [PATCH v2 1/3] libavfilter/x86/vf_hflip: add
> ff_flip_byte/short_avx512()
>
> Performance(Less is better):
> 8bit:
> ff_hflip_byte_ssse3 0.61
> ff_hflip_byte_a
Ping.
> From: Wu, Jianhua
> Sent: Thursday, November 4, 2021 12:19 PM
> To: ffmpeg-devel@ffmpeg.org
> Cc: Wu, Jianhua
> Subject: [PATCH v2 1/3] avfilter/x86/vf_exposure: add x86 SIMD optimization
>
> Performance data(Less is better):
> exposure_c:857394
> exposure_sse: 327589
>
> Signed-of
Hi,
Just a quick ping on the v3 of the avformat/imf demuxer patch set.
Any major issues/blockers?
Best,
-- Pierre
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above,
Nov 18, 2021, 06:46 by d...@lynne.ee:
> This commit moves the common Vulkan framework into libavutil,
> allowing it to be used in libavcodec. Each library that uses it is
> expected to #include the .c/.h files into their own templates,
> that may include extra library-specific functions, and compi
On Thu, 2021-11-18 at 21:55 +, Soft Works wrote:
> > -Original Message-
> > From: ffmpeg-devel On Behalf Of
> > Fei
> > Wang
> > Sent: Tuesday, June 30, 2020 5:58 AM
> > To: ffmpeg-devel@ffmpeg.org
> > Cc: Xinpeng Sun ; Zachary Zhou
> > ; Fei Wang
> > Subject: [FFmpeg-devel] [PATCH v1
On 11/18/2021 8:05 PM, Andreas Rheinhardt wrote:
James Almer:
On 11/18/2021 5:35 AM, Andreas Rheinhardt wrote:
James Almer:
Signed-off-by: James Almer
---
fftools/ffmpeg.c | 3 +++
fftools/ffmpeg.h | 1 +
fftools/ffmpeg_hw.c | 17 ++---
3 files changed, 10 inserti
James Almer:
> On 11/18/2021 5:35 AM, Andreas Rheinhardt wrote:
>> James Almer:
>>> Signed-off-by: James Almer
>>> ---
>>> fftools/ffmpeg.c | 3 +++
>>> fftools/ffmpeg.h | 1 +
>>> fftools/ffmpeg_hw.c | 17 ++---
>>> 3 files changed, 10 insertions(+), 11 deletions(-)
>>>
>
On 11/18/2021 6:07 AM, Andreas Rheinhardt wrote:
Signed-off-by: Andreas Rheinhardt
---
fftools/cmdutils.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fftools/cmdutils.c b/fftools/cmdutils.c
index 594eeef379..f80c361eba 100644
--- a/fftools/cmdutils.c
+++ b/fftools/cmdu
On 11/18/2021 5:35 AM, Andreas Rheinhardt wrote:
James Almer:
Signed-off-by: James Almer
---
fftools/ffmpeg.c| 3 +++
fftools/ffmpeg.h| 1 +
fftools/ffmpeg_hw.c | 17 ++---
3 files changed, 10 insertions(+), 11 deletions(-)
diff --git a/fftools/ffmpeg.c b/fftools/ffm
> -Original Message-
> From: ffmpeg-devel On Behalf Of Fei
> Wang
> Sent: Tuesday, June 30, 2020 5:58 AM
> To: ffmpeg-devel@ffmpeg.org
> Cc: Xinpeng Sun ; Zachary Zhou
> ; Fei Wang
> Subject: [FFmpeg-devel] [PATCH v1] avfilter: add overlay vaapi filter
>
> From: Xinpeng Sun
>
> Over
On Thu, Nov 18, 2021 at 09:15:05PM +0100, Andreas Rheinhardt wrote:
> Michael Niedermayer:
> > On Thu, Nov 18, 2021 at 10:07:36AM +0100, Andreas Rheinhardt wrote:
> >> Signed-off-by: Andreas Rheinhardt
> >> ---
> >> fftools/cmdutils.c | 2 +-
> >> 1 file changed, 1 insertion(+), 1 deletion(-)
> >
Tomas Härdin:
> tis 2021-11-09 klockan 19:01 +0100 skrev Andreas Rheinhardt:
>> The actual frame_size is no longer used since commit
>> 3d38e45eb85c7a2420cb48a9cd45625c28644b2e; and the check for
>> "< 0" is equivalent to the CID being valid. But this is already
>> ensured by mxf_dnxhd_codec_uls co
Tomas Härdin:
> tis 2021-11-09 klockan 22:48 +0100 skrev Andreas Rheinhardt:
>> Tomas Härdin:
>>> tis 2021-11-09 klockan 19:01 +0100 skrev Andreas Rheinhardt:
It is just a flag per supported CID. So there is no reason to use
an avpriv function for this purpose.
>>>
>>> This is data duplic
Michael Niedermayer:
> On Thu, Nov 18, 2021 at 10:07:36AM +0100, Andreas Rheinhardt wrote:
>> Signed-off-by: Andreas Rheinhardt
>> ---
>> fftools/cmdutils.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/fftools/cmdutils.c b/fftools/cmdutils.c
>> index 594eeef379..f80
On Thu, Nov 18, 2021 at 10:07:36AM +0100, Andreas Rheinhardt wrote:
> Signed-off-by: Andreas Rheinhardt
> ---
> fftools/cmdutils.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fftools/cmdutils.c b/fftools/cmdutils.c
> index 594eeef379..f80c361eba 100644
> --- a/fftool
On Wed, 17 Nov 2021, Gijs Peskens wrote:
Introduce fifo_size and overrun_nonfatal params to configure fifo buffer
behavior.
Fifo size is used to left shift 2, since libRIST only accepts powers of 2.
Can you please make fifo_size simply mean the number of packets? User
facing options should
On Wed, 17 Nov 2021, Gijs Peskens wrote:
Correct solution as suggested by Martin Balint on ffmpeg-devel
---
libavformat/librist.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavformat/librist.c b/libavformat/librist.c
index 6eae90cc2f..378b635ea7 100644
--- a/libavformat/librist.c
++
On Sun, 14 Nov 2021, Soft Works wrote:
Signed-off-by: softworkz
---
v3: fixed patch, removed comment about 0 line size
libavutil/frame.h | 18 +++---
1 file changed, 15 insertions(+), 3 deletions(-)
diff --git a/libavutil/frame.h b/libavutil/frame.h
index ff2540a20f..31def23d8b 1
Andreas Rheinhardt:
> It avoids branches lateron and will allow to easily avoid the overhead
> of the linked list currently in use in case there is only one stream.
>
> Signed-off-by: Andreas Rheinhardt
> ---
> libavformat/internal.h | 6 ++
> libavformat/mux.c | 17 +++--
>
On Fri, Nov 12, 2021 at 06:22:06PM +0800, lance.lmw...@gmail.com wrote:
> From: Limin Wang
>
> Signed-off-by: Limin Wang
> ---
> libavformat/rtpdec_rfc4175.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/libavformat/rtpdec_rfc4175.c b/libavformat/rtpdec_rfc4175.c
From: Limin Wang
Signed-off-by: Limin Wang
---
libavfilter/vf_gblur_vulkan.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/libavfilter/vf_gblur_vulkan.c b/libavfilter/vf_gblur_vulkan.c
index 16c8bbb..8407bd1 100644
--- a/libavfilter/vf_gblur_vulkan.c
+++ b/libavfilter/vf
On Thu, Nov 18, 2021 at 05:30:50AM +0100, Andreas Rheinhardt wrote:
> lance.lmw...@gmail.com:
> > From: Limin Wang
> >
> > Signed-off-by: Limin Wang
> > ---
> > libavfilter/vf_gblur_vulkan.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/libavfilter/vf_gblur_vul
On Thu, Nov 18, 2021 at 4:38 AM Andreas Rheinhardt <
andreas.rheinha...@outlook.com> wrote:
> Thilo Borgmann:
> > Hi,
> >
> > shared code with vf_blurriness.c in patch 2/2.
> > RFC: is there a better place to put these?
> >
>
> If you only need this in libavfilter, then why do you put it into
> li
On Thu, Nov 18, 2021 at 6:27 AM Andreas Rheinhardt <
andreas.rheinha...@outlook.com> wrote:
> Wu, Jianhua:
> > Andreas Rheinhardt wrote:
> >> Wu Jianhua:
> >>> The following command is on how to apply hflip_vulkan filter:
> >>>
> >>> ffmpeg -init_hw_device vulkan -i input.264 -vf
> >>>
> >> hwuplo
On Thu, Nov 18, 2021 at 6:59 AM Wu, Jianhua wrote:
> > Andreas Rheinhardt:
> > Wu, Jianhua:
> > > Andreas Rheinhardt wrote:
> > >> Wu Jianhua:
> > >>> The following command is on how to apply hflip_vulkan filter:
> > >>>
> > >>> ffmpeg -init_hw_device vulkan -i input.264 -vf
> > >>>
> > >>
> > hw
Andreas Rheinhardt 于2021年11月18日周四 下午5:14写道：
>
> Fixes Coverity issue #1493345.
>
> Signed-off-by: Andreas Rheinhardt
> ---
> The x = y also looks strange. Why does the parameter x exist at all
> if it is immediately overwritten in the function?
>
> libavfilter/vf_huesaturation.c | 1 -
> 1 file
In order to know that the earlier code did not use uninitialized
values one needs to know that the lowest four bits of each used
value of pframe_block4x4_coefficients_tab do not vanish identically.
E.g. Coverity did not get this and warned about it in ticket #1466632.
Fix this by slightly rewriting
Fixes Coverity issue #1493345.
Signed-off-by: Andreas Rheinhardt
---
The x = y also looks strange. Why does the parameter x exist at all
if it is immediately overwritten in the function?
libavfilter/vf_huesaturation.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/libavfilter/vf_huesaturati
Instead reuse stream_options.
Signed-off-by: Andreas Rheinhardt
---
If one used a named variable for the AVOption.unit values,
one could simplify the check in option_is_disposition() to a pointer
comparison.
libavformat/utils.c | 39 +++
1 file changed, 11 in
Signed-off-by: Andreas Rheinhardt
---
fftools/cmdutils.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fftools/cmdutils.c b/fftools/cmdutils.c
index 594eeef379..f80c361eba 100644
--- a/fftools/cmdutils.c
+++ b/fftools/cmdutils.c
@@ -1823,7 +1823,7 @@ int show_sample_fmts(voi
On Wed, 17 Nov 2021, J. Dekker wrote:
Signed-off-by: J. Dekker
---
libavcodec/aarch64/hevcdsp_sao_neon.S | 195 --
1 file changed, 90 insertions(+), 105 deletions(-)
Now matches the 9,25 indentation like other ASM.
No it doesn't.
This changes the indentation of the pr
On Wed, 17 Nov 2021, J. Dekker wrote:
--bench on AWS Graviton:
hevc_sao_band_8x8_8_c: 317.5
hevc_sao_band_8x8_8_neon: 97.5
hevc_sao_band_16x16_8_c: 1115.0
hevc_sao_band_16x16_8_neon: 322.7
hevc_sao_band_32x32_8_c: 4599.2
hevc_sao_band_32x32_8_neon: 1246.2
hevc_sao_band_48x48_8_c: 10021.7
hevc_s
On Wed, 17 Nov 2021, J. Dekker wrote:
--bench on AWS Graviton:
hevc_sao_edge_8x8_8_c: 516.0
hevc_sao_edge_8x8_8_neon: 81.0
Signed-off-by: J. Dekker
---
libavcodec/aarch64/hevcdsp_init_aarch64.c | 3 ++
libavcodec/aarch64/hevcdsp_sao_neon.S | 51 +++
2 files changed, 54
On Wed, 17 Nov 2021, J. Dekker wrote:
--bench on AWS Graviton:
hevc_sao_edge_16x16_8_c: 1857.0
hevc_sao_edge_16x16_8_neon: 211.0
hevc_sao_edge_32x32_8_c: 7802.2
hevc_sao_edge_32x32_8_neon: 808.2
hevc_sao_edge_48x48_8_c: 16764.2
hevc_sao_edge_48x48_8_neon: 1796.5
hevc_sao_edge_64x64_8_c: 32647.5
On Wed, 17 Nov 2021, J. Dekker wrote:
This reverts commit 2589060b92eeeb944c6e2b50e38412c0c5fabcf4 which was
originally to fix the FATE test. The real cause of the test breakage was
fixed in 8dc8f04036eb27c8ad419839d4ed3bc67c44fe7a.
Signed-off-by: J. Dekker
LGTM. (Remember to update the comm
On Wed, 17 Nov 2021, J. Dekker wrote:
The assembly is written assuming that the width is a multiple of 8.
However the real issue is the functions were errorneously assigned to
the 2, 4, 6 & 12 widths. This behaviour never broke the decoder as
samples which trigger the functions for these widths
James Almer:
> Signed-off-by: James Almer
> ---
> fftools/ffmpeg.c| 3 +++
> fftools/ffmpeg.h| 1 +
> fftools/ffmpeg_hw.c | 17 ++---
> 3 files changed, 10 insertions(+), 11 deletions(-)
>
> diff --git a/fftools/ffmpeg.c b/fftools/ffmpeg.c
> index d141f34df9..26030ed25e 100
38 matches
Mail list logo
