to increase or adjust the sample count for more
reliable results.
Signed-off-by: J. Dekker
---
This could also be implemented as a compile time define or a configure
argument.
tests/checkasm/checkasm.c | 12 +++-
tests/checkasm/checkasm.h | 4 ++--
2 files changed, 13 insertions
to increase or adjust the sample count for more
reliable results.
Signed-off-by: J. Dekker
---
Auto-detection can be added later when either a count is omitted or a specific
value or term such as '0' or 'auto' is provided. This is a development tool,
the users will be
developers to increase the sample count for more
reliable results.
Signed-off-by: J. Dekker
---
tests/checkasm/checkasm.c | 16 +++-
tests/checkasm/checkasm.h | 7 ---
2 files changed, 19 insertions(+), 4 deletions(-)
diff --git a/tests/checkasm/checkasm.c b/tests/checkasm
Lynne via ffmpeg-devel writes:
> [[PGP Signed Part:Undecided]]
> On 21/05/2024 15:51, J. Dekker wrote:
>> Some timers on certain device and test combinations can produce noisy
>> results, affecting the reliability of performance measurements. One
>> notable example of
Lynne via ffmpeg-devel writes:
> Helps make sense of the possible noise in the results.
> ---
> tests/checkasm/checkasm.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/tests/checkasm/checkasm.c b/tests/checkasm/checkasm.c
> index 669f2be9c1..b1c175b95d 100644
> --- a/tests/checka
Kacper Michajłow writes:
> It may be invalidated by the time it is used.
>
> Fixes use after free when accessing current segment.
>
> Fixes: #10825
> Signed-off-by: Kacper Michajłow
> ---
> libavformat/hls.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/libavformat/hls.c b/libavfo
"J. Dekker" writes:
> Kacper Michajłow writes:
>
>> It may be invalidated by the time it is used.
>>
>> Fixes use after free when accessing current segment.
>>
>> Fixes: #10825
>> Signed-off-by: Kacper Michajłow
>> ---
>
4: 1.2
h264_idct8_add_dc_12bpp_c: 4.2
h264_idct8_add_dc_12bpp_rvv_i64: 1.2
h264_idct8_add_dc_14bpp_c: 4.2
h264_idct8_add_dc_14bpp_rvv_i64: 1.2
Signed-off-by: J. Dekker
---
libavcodec/riscv/Makefile | 1 +
libavcodec/riscv/h264dsp_init.c | 40 +-
libavcodec/riscv/h264dsp_rvv.S
On Mon, Jun 17, 2024, at 13:10, Martin Storsjö wrote:
> On Wed, 12 Jun 2024, Zhao Zhili wrote:
>
>> From: Zhao Zhili
>>
>> On m1, kpc_get_counter_count(KPC_MASK) return 8. The exact value
>> doesn't matter in our case.
>
> This is somewhat unexpected, I had expected that this API was originally
>
Signed-off-by: J. Dekker
---
libavutil/riscv/cpu.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavutil/riscv/cpu.h b/libavutil/riscv/cpu.h
index bb8e08aa14..191e4478c5 100644
--- a/libavutil/riscv/cpu.h
+++ b/libavutil/riscv/cpu.h
@@ -65,7 +65,7 @@ static inline size_t
On 7 Oct 2021, at 12:40, Arif Driessen wrote:
Hi,
I think this is a typo...
---
README.md | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/README.md b/README.md
index 447347c700..f54299d340 100644
--- a/README.md
+++ b/README.md
@@ -9,7 +9,7 @@ such as audio, video, subtitl
0 ?
5.0
Should it be made in december 2021 ? (as was suggested in jbs release
mail)
This seems reasonable.
--
J. Dekker
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit lin
--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 | 52 +++
2 files changed, 55 insertions(+)
diff --git a
Signed-off-by: J. Dekker
---
libavcodec/aarch64/hevcdsp_init_aarch64.c | 8 ++-
libavcodec/aarch64/hevcdsp_sao_neon.S | 66 +++
2 files changed, 72 insertions(+), 2 deletions(-)
diff --git a/libavcodec/aarch64/hevcdsp_init_aarch64.c
b/libavcodec/aarch64/hevcdsp_init_aarch64
hevc_sao_band_64x64_8_c: 17635.0
hevc_sao_band_64x64_8_neon: 4875.7
Signed-off-by: J. Dekker
---
libavcodec/aarch64/hevcdsp_init_aarch64.c | 6 +-
libavcodec/aarch64/hevcdsp_sao_neon.S | 9 ++---
2 files changed, 11 insertions(+), 4 deletions(-)
diff --git a/libavcodec/aarch64
Signed-off-by: J. Dekker
---
libavcodec/aarch64/hevcdsp_sao_neon.S | 103 +++---
1 file changed, 44 insertions(+), 59 deletions(-)
diff --git a/libavcodec/aarch64/hevcdsp_sao_neon.S
b/libavcodec/aarch64/hevcdsp_sao_neon.S
index 263747149f..c2519da7f5 100644
--- a/libavcodec
ike it would be the most reasonable to test
(unlocked boot-loader, etc).
I am preemptively requesting reimbursement for these devices,
suggestions welcome.
--
J. Dekker
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/
kit is ~1600 EUR, obviously the phone form factor is not ideal
for automated testing but it is significantly cheaper. But yes, if anyone can
find a Snapdragon 888 in a devkit form (i.e. something I can plug into a
network easily) then that would be preferred. Or if someone is able to find one
or mult
This reverts commit 2589060b92eeeb944c6e2b50e38412c0c5fabcf4.
Signed-off-by: J. Dekker
---
libavcodec/arm/hevcdsp_qpel_neon.S | 18 +-
1 file changed, 9 insertions(+), 9 deletions(-)
diff --git a/libavcodec/arm/hevcdsp_qpel_neon.S
b/libavcodec/arm/hevcdsp_qpel_neon.S
index
Signed-off-by: J. Dekker
---
libavcodec/arm/hevcdsp_init_neon.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/libavcodec/arm/hevcdsp_init_neon.c
b/libavcodec/arm/hevcdsp_init_neon.c
index 201a088dac..112edb5edd 100644
--- a/libavcodec/arm/hevcdsp_init_neon.c
+++ b
On 19 Oct 2021, at 10:40, Martin Storsjö wrote:
> On Thu, 7 Oct 2021, J. Dekker wrote:
>
>> Signed-off-by: J. Dekker
>> ---
>> libavcodec/aarch64/hevcdsp_sao_neon.S | 103 +++---
>> 1 file changed, 44 insertions(+), 59 deletions(-)
>>
This relies on the mappings in ff_hevc_pel_weight[].
Signed-off-by: J. Dekker
---
libavcodec/arm/hevcdsp_init_neon.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
Updated commit message.
diff --git a/libavcodec/arm/hevcdsp_init_neon.c
b/libavcodec/arm/hevcdsp_init_neon.c
index 201a08
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
---
libavcodec/arm/hevcdsp_qpel_neon.S | 18 +-
1 file
Signed-off-by: J. Dekker
---
libavcodec/aarch64/hevcdsp_init_aarch64.c | 8 ++-
libavcodec/aarch64/hevcdsp_sao_neon.S | 65 +++
2 files changed, 71 insertions(+), 2 deletions(-)
Used inline shift for ldr & escheduled uxtl/uxtl2 before zip/zip2.
diff --git a/libavc
--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 insertions(+)
Used inline shift for
hevc_sao_band_64x64_8_c: 17635.0
hevc_sao_band_64x64_8_neon: 4875.7
Signed-off-by: J. Dekker
---
libavcodec/aarch64/hevcdsp_init_aarch64.c | 6 +-
libavcodec/aarch64/hevcdsp_sao_neon.S | 9 ++---
2 files changed, 11 insertions(+), 4 deletions(-)
No change since previous patch which was ACK&#
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.
diff --git a/libavcodec/aarch64/hevcdsp_sao_neon.S
b/libavcodec/aarch64/hevcdsp_sao_neon.S
In my eyes, the matter of how we should update the voters list was already
resolved.
> https://vote.ffmpeg.org/cgi-bin/civs/vote.pl?id=E_07e9c717f7820201&key=240812a0c2e373ac
> This is your private URL. Do not give it to anyone else, because they could
> use it to vote for you.
This vote seems
On Sun, Nov 12, 2023, at 18:31, Michael Niedermayer wrote:
> On Sun, Nov 12, 2023 at 11:03:21AM -0300, James Almer wrote:
>> On 11/12/2023 10:59 AM, Thilo Borgmann via ffmpeg-devel wrote:
>> > I will also start the repeat vote now and everybody can hold their
>> > horses before going to flamewar. D
Derek Buitenhuis writes:
> On 11/20/2023 3:44 PM, Thilo Borgmann via ffmpeg-devel wrote:
>> the results are available at [1]. As they confirm the just updated GA list
>> as
>> well as the update procedure twice a year on Jan 1st & Jul 1st, I think the
>> upcoming votes (extra GA members, TC/
Michael Niedermayer writes:
> Signed-off-by: Michael Niedermayer
> ---
> doc/git-howto.texi | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
Most people probably use git@ already, seems to be more common practice
in general. Both users map to the same id on the server so would make
s
Signed-off-by: J. Dekker
---
tests/checkasm/hevc_deblock.c | 110 --
1 file changed, 106 insertions(+), 4 deletions(-)
Yes, this only supports 8bit. 10/12bit should be trivial, will add if this
looks reasonable (I checked code paths using gdb, and as far as I
"Jean-Baptiste Kempf" writes:
> $subject
>
> See attachment.
Pushed with slight commit rewording for clarity.
--
jd
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link a
> On Mon, 25 Mar 2024, Martin Storsjö wrote:
>
>> Since some time, we have pretty complete AArch64 NEON coverage
>> for the hevc decoder.
>>
>> However, some of these functions require the I8MM instruction set
>> extension, and many of them (but not all) lack a plain NEON
>> version.
>>
>> Thi
Signed-off-by: J. Dekker
---
configure | 3 ++-
doc/texidep.pl| 2 +-
ffbuild/libversion.sh | 1 +
tests/fate-run.sh | 2 +-
tests/fate.sh | 2 +-
5 files changed, 6 insertions(+), 4 deletions(-)
diff --git a/configure b/configure
index 2d46ef0b9c..d332f18e3d
Signed-off-by: J. Dekker
---
configure | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/configure b/configure
index d332f18e3d..0ff1ff0335 100755
--- a/configure
+++ b/configure
@@ -6108,11 +6108,7 @@ extern_prefix=${sym%%ff_extern*}
check_cc pragma_deprecated
In some cases, these scripts can be called directly by packagers, and
some systems require the interpreter to be explicit.
Signed-off-by: J. Dekker
---
configure | 3 ++-
doc/texidep.pl| 2 +-
ffbuild/libversion.sh | 1 +
tests/fate-run.sh | 2 +-
tests/fate.sh
fail silently.
This patch changes the check to a more commonly used define for
big-endian systems. More checks may need to be added in the future to
cover legacy machines.
Signed-off-by: J. Dekker
---
configure | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/configure b
formats can fail silently.
This patch changes the check to a more commonly used define for GCC
style compilers. More checks may be needed to cover other potential
compilers that don't use the __BYTE_ORDER__ define.
Signed-off-by: J. Dekker
---
configure | 6 +-
1 file changed, 1 insertion(
ses we add the interpreter explicitly.
Signed-off-by: J. Dekker
---
ffbuild/libversion.sh | 1 +
1 file changed, 1 insertion(+)
diff --git a/ffbuild/libversion.sh b/ffbuild/libversion.sh
index a94ab58057..ecaa90cde6 100755
--- a/ffbuild/libversion.sh
+++ b/ffbuild/libversion.sh
@@ -1,3 +1,4 @@
Note that the config.sh file is left without a shebang, this file is
supposed to be sourced into the current environment.
This commit is purely cosmetic.
Signed-off-by: J. Dekker
---
configure | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure b/configure
index
This commit is purely cosmetic.
Signed-off-by: J. Dekker
---
tests/fate-run.sh | 2 +-
tests/fate.sh | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/tests/fate-run.sh b/tests/fate-run.sh
index 9863e4f2d9..6ae0320c60 100755
--- a/tests/fate-run.sh
+++ b/tests/fate
This commit is purely cosmetic.
Signed-off-by: J. Dekker
---
doc/texidep.pl | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/doc/texidep.pl b/doc/texidep.pl
index 099690378e..33e6c7c53e 100644
--- a/doc/texidep.pl
+++ b/doc/texidep.pl
@@ -1,4 +1,4 @@
-#! /usr/bin/env perl
Martin Storsjö writes:
> On Tue, 9 Apr 2024, J. Dekker wrote:
>
>> Note that the config.sh file is left without a shebang, this file is
>> supposed to be sourced into the current environment.
>>
>> This commit is purely cosmetic.
>>
>> Signed-off-by:
exclude_guest is currently only supported on x86. However, not
specifying 'exclude_guest' implies that you can count guest events
should you run one. This creates an ABI issue whereby some non-x86
kernels require specifying exclude_guest = 1 explicitly.
Signed-off-by: J. Dekker
Martin Storsjö writes:
> Before: Cortex A53 A72 A78
> ac3_sum_square_bufferfly_float_neon: 1005.7 516.5 224.5
> After:
> ac3_sum_square_bufferfly_float_neon: 981.7 504.5 223.2
> ---
> libavcodec/aarch64/ac3dsp_neon.S | 16
> 1 file c
to 1, i.e. the only behaviour that makes sense when counting
guest events isn't supported.
Signed-off-by: J. Dekker
---
Made commit message clearer, no functional change since v1.
tests/checkasm/checkasm.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/tests/checkasm/checkasm.c b
Martin Storsjö writes:
> On Wed, 10 Apr 2024, J. Dekker wrote:
>
>> The exclude_guest option only has an effect on x86. Omitting
>> 'exclude_guest' defaults to zero which implies that you can count guest
>> events should you run one. Some non-x86 kernels ju
Signed-off-by: J. Dekker
---
libavfilter/riscv/Makefile | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavfilter/riscv/Makefile b/libavfilter/riscv/Makefile
index 0b968a9c0d..277dde2aed 100644
--- a/libavfilter/riscv/Makefile
+++ b/libavfilter/riscv/Makefile
@@ -1,2
Signed-off-by: J. Dekker
---
tests/checkasm/h264dsp.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/tests/checkasm/h264dsp.c b/tests/checkasm/h264dsp.c
index 0f484e3f43..5cb646ae49 100644
--- a/tests/checkasm/h264dsp.c
+++ b/tests/checkasm/h264dsp.c
@@ -173,6 +173,7
Devin Heitmueller writes:
> On Wed, Apr 24, 2024 at 10:10 AM J. Dekker wrote:
>>
>> Signed-off-by: J. Dekker
>> ---
>> tests/checkasm/h264dsp.c | 6 --
>> 1 file changed, 4 insertions(+), 2 deletions(-)
>>
>> diff --git a/tests/checkasm/
Thilo Borgmann via ffmpeg-devel writes:
> ---
> This text including the link is also meant to be published via our socal
> media.
>
> src/index | 8
> 1 file changed, 8 insertions(+)
>
> diff --git a/src/index b/src/index
> index d035ffa..83cc9bf 100644
> --- a/src/index
> +++ b/src/
Signed-off-by: J. Dekker
---
These devices are fundamentally broken and usecases should be switched
away from output devices in general. Discussion in the thread tended towards
deprecation rather than immediate removal to give time for users to figure out
the best alternatives for their usecase
Signed-off-by: J. Dekker
---
libavdevice/sdl2.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/libavdevice/sdl2.c b/libavdevice/sdl2.c
index 342a253dc0..6a6751e40f 100644
--- a/libavdevice/sdl2.c
+++ b/libavdevice/sdl2.c
@@ -51,6 +51,7 @@ typedef struct {
SDL_Rect texture_rect
Signed-off-by: J. Dekker
---
tests/checkasm/hevc_deblock.c | 246 +-
1 file changed, 215 insertions(+), 31 deletions(-)
diff --git a/tests/checkasm/hevc_deblock.c b/tests/checkasm/hevc_deblock.c
index 66fc8d5646..91e57f5cf5 100644
--- a/tests/checkasm
Over/underflow in some cases.
Signed-off-by: J. Dekker
---
libavcodec/x86/hevcdsp_init.c | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/libavcodec/x86/hevcdsp_init.c b/libavcodec/x86/hevcdsp_init.c
index 31e81eb11f..11cb1b3bfd 100644
--- a/libavcodec/x86
Benched using single-threaded full decode on an Ampere Altra.
Bpp Before After Speedup
8 73,3s 65,2s 1.124x
10 114,2s 104,0s 1.098x
12 125,8s 115,7s 1.087x
Signed-off-by: J. Dekker
---
libavcodec/aarch64/hevcdsp_deblock_neon.S | 421 ++
libavcodec/aarch64
Signed-off-by: J. Dekker
---
doc/outdevs.texi| 2 +-
libavdevice/opengl_enc.c| 11 +++
libavdevice/version_major.h | 2 ++
3 files changed, 14 insertions(+), 1 deletion(-)
diff --git a/doc/outdevs.texi b/doc/outdevs.texi
index f0484bbf8f..941429a8c8 100644
--- a/doc
Signed-off-by: J. Dekker
---
doc/outdevs.texi| 8 +++-
libavdevice/sdl2.c | 10 ++
libavdevice/version_major.h | 2 ++
3 files changed, 19 insertions(+), 1 deletion(-)
diff --git a/doc/outdevs.texi b/doc/outdevs.texi
index 941429a8c8..9ee857528e 100644
--- a
Signed-off-by: J. Dekker
---
tests/checkasm/hevc_deblock.c | 246 +-
1 file changed, 215 insertions(+), 31 deletions(-)
diff --git a/tests/checkasm/hevc_deblock.c b/tests/checkasm/hevc_deblock.c
index 66fc8d5646..91e57f5cf5 100644
--- a/tests/checkasm
Over/underflow in some cases.
Signed-off-by: J. Dekker
---
libavcodec/x86/hevcdsp_init.c | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/libavcodec/x86/hevcdsp_init.c b/libavcodec/x86/hevcdsp_init.c
index f5bc342cd5..8831ba957a 100644
--- a/libavcodec/x86
Benched using single-threaded full decode on an Ampere Altra.
Bpp Before After Speedup
8 73,3s 65,2s 1.124x
10 114,2s 104,0s 1.098x
12 125,8s 115,7s 1.087x
Signed-off-by: J. Dekker
---
libavcodec/aarch64/hevcdsp_deblock_neon.S | 421 ++
libavcodec/aarch64
Martin Storsjö writes:
> This fixes building FFmpeg's libavcodec/aarch64/h264idct_neon.S
> for a Linux target. (It's not necessary to use gas-preprocessor for
> such a target for a very long time, but it can be useful to be able
> to test gas-preprocessor there still.)
> ---
> gas-preprocessor.p
Hi,
I attended FOSDEM for ffmpeg and would like to request travel reimbursement.
Flight | 162,46 EUR
===
Total | 162,46 EUR
--
jd
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To u
Nuo Mi writes:
> On Wed, Feb 21, 2024 at 7:10 PM J. Dekker wrote:
>
>> Over/underflow in some cases.
>>
>> Signed-off-by: J. Dekker
>> ---
>> libavcodec/x86/hevcdsp_init.c | 9 +
>> 1 file changed, 5 insertions(+), 4 deletions(-)
>>
Andreas Rheinhardt writes:
> J. Dekker:
>> SAO_BAND_INIT(12, sse2);
>> SAO_EDGE_INIT(12, sse2);
>>
>
> If you disable them here, you should also ensure that they are not
> assembled at all.
>
> - Andreas
Sure, will do on push
Martin Storsjö writes:
> [...]
>
> Are you sure the input is within valid range? It's always possible that
> checkasm produces inputs that the real decoder wouldn't - but it's also
> possible that this is a real decoder bug that just hasn't been triggered by
> any
> other test yet.
>
> // Martin
Weak filter can overflow in delta0 calculation before >> 4 in int16.
Signed-off-by: J. Dekker
---
I do not know x86 simd at all, so this is just an attempt to fix
the implementation rather than write extremely performant code.
Suggestions welcome.
libavcodec/x86/hevc_deblock.as
"Ronald S. Bultje" writes:
>
> New version attached.
>
Thanks for the help, pushed this along with the test.
--
jd
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link abov
"J. Dekker" writes:
> Signed-off-by: J. Dekker
> ---
> tests/checkasm/hevc_deblock.c | 246 +-
> 1 file changed, 215 insertions(+), 31 deletions(-)
Pushed this along with Ronald's patch to fix x86. Have some minor
improvements
Benched using single-threaded full decode on an Ampere Altra.
Bpp Before After Speedup
8 73,3s 65,2s 1.124x
10 114,2s 104,0s 1.098x
12 125,8s 115,7s 1.087x
Signed-off-by: J. Dekker
---
Slightly improved 12bit version.
libavcodec/aarch64/hevcdsp_deblock_neon.S | 417
Martin Storsjö writes:
> On Tue, 27 Feb 2024, J. Dekker wrote:
>
>> Benched using single-threaded full decode on an Ampere Altra.
>>
>> Bpp Before After Speedup
>> 8 73,3s 65,2s 1.124x
>> 10 114,2s 104,0s 1.098x
>> 12 125,8s 11
Martin Storsjö writes:
> On Wed, 28 Feb 2024, J. Dekker wrote:
>
>>
>> Martin Storsjö writes:
>>
>>> On Tue, 27 Feb 2024, J. Dekker wrote:
>>>
>>>> Benched using single-threaded full decode on an Ampere Altra.
>>>>
>>>
Martin Storsjö writes:
> On Wed, 28 Feb 2024, J. Dekker wrote:
>
>>
>> Martin Storsjö writes:
>>
>>> On Wed, 28 Feb 2024, J. Dekker wrote:
>>>
>>>>
>>>> Martin Storsjö writes:
>>>>
>>>>> On Tue, 27 F
Anton Khirnov writes:
> Quoting Lynne (2024-02-21 13:32:19)
>> Feb 21, 2024, 12:34 by j...@itanimul.li:
>>
>> > Signed-off-by: J. Dekker
>> > ---
>> > doc/outdevs.texi| 2 +-
>> > libavdevice/opengl_enc.c| 11 ++
Inactive since 2007.
Signed-off-by: J. Dekker
---
MAINTAINERS | 7 ++-
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index cc7f564f48..ace58f02a1 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -148,7 +148,6 @@ Codecs:
cinepak.c
This was an experimental/research codec of which ffmpeg is the only
encoder and decoder, development has stalled and these files don't exist
in the wild.
Signed-off-by: J. Dekker
---
Changelog |1 +
configure |3 -
libavcodec/Makefile |3 -
libav
Michael Niedermayer writes:
> [[PGP Signed Part:Undecided]]
> On Wed, Feb 28, 2024 at 01:56:10PM +0100, J. Dekker wrote:
>> This was an experimental/research codec of which ffmpeg is the only
>> encoder and decoder,
>
>
>> development has stalled
>
> Thats n
Vittorio Giovara writes:
> On Wed, Feb 28, 2024 at 2:38 PM J. Dekker wrote:
>
>>
>> Michael Niedermayer writes:
>>
>> > [[PGP Signed Part:Undecided]]
>> > On Wed, Feb 28, 2024 at 01:56:10PM +0100, J. Dekker wrote:
>> >> This was an e
e the encoders by default
in configure, the decoders should be disabled by default next bump.
Signed-off-by: J. Dekker
---
Since this codec was added in 2004, there has only been one major
non-maintenance, non-bugfix commit in 2013 (6026a5ad4f). That's 1 out of 107
total commits touching this fi
On Thu, Feb 29, 2024, at 16:27, Zhao Zhili wrote:
> Hi, I can't pull from main developer git server. Patchwork is slow to loading.
> I have tried two ISP. Anything happened?
All seem okay for me.
--
jd
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.
Martin Storsjö writes:
> The first 32 elements of each row were correct, while the
> last 16 were scrambled.
>
> This hasn't been noticed, because the checkasm test erroneously
> only checked half of the output (for 8 bit functions), and
> apparently none of the samples as part of "fate-hevc" se
: 42958.7
put_hevc_qpel_uni_h64_8_neon: 8366.5
Signed-off-by: J. Dekker
---
Summary of changes since last iteration:
- Interleaved stores
- Changed tiling to loop more naturally
- Increased code reuse (.text reduction by ~60%)
- Simplified function variations through .req
libavcodec/aarch64
On 25 Oct 2022, at 14:48, Andreas Rheinhardt wrote:
> GCC 11 has a bug: When it creates clones of recursive functions
> (to inline some parameters), it clones a recursive function
> eight times by default, even when this exceeds the recursion
> depth. This happens with encode_block() in libavcodec
On 24 Oct 2022, at 14:01, Martin Storsjö wrote:
> On Tue, 11 Oct 2022, J. Dekker wrote:
>
>> [...]
>> diff --git a/libavcodec/aarch64/hevcdsp_init_aarch64.c
>> b/libavcodec/aarch64/hevcdsp_init_aarch64.c
>> index 644cc17715..44399b05d8 100644
>> --- a/libav
4: 1.2
h264_idct8_add_dc_12bpp_c: 4.2
h264_idct8_add_dc_12bpp_rvv_i64: 1.2
h264_idct8_add_dc_14bpp_c: 4.2
h264_idct8_add_dc_14bpp_rvv_i64: 1.2
Signed-off-by: J. Dekker
---
rdcycle always returns 0 on my board, clock_gettime() seems as noisy as
rdtime (just with bigger numbers).
libavcodec/riscv/Makefile
4: 1.2
h264_idct8_add_dc_12bpp_c: 4.2
h264_idct8_add_dc_12bpp_rvv_i64: 1.2
h264_idct8_add_dc_14bpp_c: 4.2
h264_idct8_add_dc_14bpp_rvv_i64: 1.2
Signed-off-by: J. Dekker
---
libavcodec/riscv/Makefile | 1 +
libavcodec/riscv/h264dsp_init.c | 44 ++-
libavcodec/riscv/h264dsp_rvv.S
When collecting performance information from checkasm it is common
to parse the output for use in graphs to compare vs different
architectures.
Signed-off-by: J. Dekker
---
tests/checkasm/checkasm.c | 22 --
1 file changed, 20 insertions(+), 2 deletions(-)
diff --git a
Port dav1d's checkasm output format to FFmpeg's checkasm, includes
relative speedups and aligns results.
Signed-off-by: J. Dekker
---
tests/checkasm/checkasm.c | 53 +++
1 file changed, 48 insertions(+), 5 deletions(-)
diff --git a/tests/checkasm/
:1471.5 ( 1.71x)
yuva420p_bgr24_1920_c:4462.6 ( 1.00x)
yuva420p_bgr24_1920_ssse3:2331.1 ( 1.91x)
[...]
Ported from dav1d.
Signed-off-by: J. Dekker
---
tests/checkasm/checkasm.c | 37
From: Niklas Haas
Instead of duplicating these common macros in every file, add them to
the shared utility file. Also add a base case for sanity.
---
libavcodec/riscv/h264addpx_rvv.S | 10 --
libavcodec/riscv/h264idct_rvv.S | 10 --
libavcodec/riscv/startcode_rvb.S | 10
From: Niklas Haas
(Ab)using nested macros to get the number of arguments passed to a
variadic macro. Useful for stack manipulation.
---
libavutil/riscv/asm.S | 17 +
1 file changed, 17 insertions(+)
diff --git a/libavutil/riscv/asm.S b/libavutil/riscv/asm.S
index 175f2a8672..db1
6.57x)
Signed-off-by: Niklas Haas
Signed-off-by: J. Dekker
---
libavcodec/h264qpel.c| 2 +
libavcodec/h264qpel.h| 1 +
libavcodec/riscv/Makefile| 2 +
libavcodec/riscv/h264qpel_init.c | 113 +++
libavcodec/riscv/h264qpel_rvv.S
From: Niklas Haas
Generic helper macros to push/pop multiple registers at once. Expands to
a single `addi` plus a sequence of XLEN-sized stores/loads.
---
libavutil/riscv/asm.S | 37 +
1 file changed, 37 insertions(+)
diff --git a/libavutil/riscv/asm.S b/liba
Lynne via ffmpeg-devel writes:
> On 13/08/2024 16:03, J. Dekker wrote:
>> Port dav1d's checkasm output format to FFmpeg's checkasm, includes
>> relative speedups and aligns results.
>> Signed-off-by: J. Dekker
>> --
Rémi Denis-Courmont writes:
> Le 13 août 2024 17:03:35 GMT+03:00, "J. Dekker" a écrit :
>>From: Niklas Haas
>>
>>Generic helper macros to push/pop multiple registers at once. Expands to
>>a single `addi` plus a sequence of XLEN-sized stores/loads
On Fri, Aug 16, 2024, at 13:03, Anton Khirnov wrote:
> Quoting J. Dekker (2024-08-13 16:03:30)
>> When collecting performance information from checkasm it is common
>> to parse the output for use in graphs to compare vs different
>> architectures.
>>
>> Signed-of
When collecting performance information from checkasm it is common
to parse the output for use in graphs to compare vs different
architectures.
Signed-off-by: J. Dekker
---
tests/checkasm/checkasm.c | 22 --
1 file changed, 20 insertions(+), 2 deletions(-)
diff --git a
Signed-off-by: J. Dekker
---
tests/checkasm/checkasm.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/tests/checkasm/checkasm.c b/tests/checkasm/checkasm.c
index d9b5c94faa..c5c1eeb07a 100644
--- a/tests/checkasm/checkasm.c
+++ b/tests/checkasm/checkasm.c
@@ -775,7
Port dav1d's checkasm output format to FFmpeg's checkasm, includes
relative speedups and aligns results.
Signed-off-by: J. Dekker
---
tests/checkasm/checkasm.c | 53 +++
1 file changed, 48 insertions(+), 5 deletions(-)
diff --git a/tests/checkasm/
1 - 100 of 167 matches
Mail list logo
