fare type. Otherwise, you have to pay the normal
price, but you avoid queues at the ticket counter.
Looking forward to seeing many of you in Brussels,
--
Rémi Denis-Courmont
http://www.remlab.net/
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
Le lauantaina 2. syyskuuta 2023, 23.03.32 EEST Michael Niedermayer a écrit :
> > And yes, UB is a very well known accronym in this context. In my
> > experience, people who don't know the accronym don't know the concept
> > either, so spelling it out wouldn't even help.
>
> "Undefined behavior"
Le torstaina 31. elokuuta 2023, 18.28.48 EEST Stefano Sabatini a écrit :
> On date Tuesday 2023-08-29 10:34:45 +0200, Anton Khirnov wrote:
> > Quoting Stefano Sabatini (2023-08-27 14:38:44)
> >
> > > Il sab 26 ago 2023, 20:08 Anton Khirnov ha scritto:
> > > > Document our longstanding de facto
Le perjantaina 8. syyskuuta 2023, 16.09.49 EEST Michael Niedermayer a écrit :
> In the past, most developers in FFmpeg where primarly FFmpeg developers. But
> as FFmpeg is used by almost everyone now, that has changed and many
> developers in FFmpeg today are primarly developer of 3rd party
Le keskiviikkona 6. syyskuuta 2023, 2.26.29 EEST Kacper Michajłow a écrit :
> Other C++ standard libraries exist. Also, this is not a proper way to
> link the standard library anyway. Instead when a C++ dependency is
> detected, switch to the C++ compiler driver to properly link everything.
>
>
Hi,
I am not on the GA, but there are probably people with my locale on the GA. And
it seems that the voting system hits a Perl syntax error if your browser locale
is set to French.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
Hi,
Le 27 octobre 2023 14:10:15 GMT+03:00, Thilo Borgmann via ffmpeg-devel
a écrit :
>> Le 26 octobre 2023 18:45:23 GMT+03:00, Michael Niedermayer
>> a écrit :
>>> This is financial sustainability Plan A (SPI)
>>> ATM SPI has like 150k $, we do not activly seek donations, we do not
>>>
Le 26 octobre 2023 18:45:23 GMT+03:00, Michael Niedermayer
a écrit :
>This is financial sustainability Plan A (SPI)
>ATM SPI has like 150k $, we do not activly seek donations, we do not currently
>use SPI money to fund any development. SPI money is ultimately controlled by
>the FFmpeg
Hi,
Le 25 octobre 2023 18:52:31 GMT+03:00, Thilo Borgmann via ffmpeg-devel
a écrit :
>Am 25.10.23 um 16:23 schrieb Thilo Borgmann via ffmpeg-devel:
>> Am 25.10.23 um 16:22 schrieb Rémi Denis-Courmont:
>>> Hi,
>>>
>>> I am not on the GA, but there are prob
Hi,
Le perjantaina 27. lokakuuta 2023, 15.24.38 EEST Thilo Borgmann via ffmpeg-
devel a écrit :
> >>> Why should it be via SPI? What's the benefit of that hypothetical future
additional funding going via SPI, as opposed to:
> >> obviously transparency and community control. None of which is
he wishful thinking that it would kickstart
sustainable financing for FFmpeg development.
--
Rémi Denis-Courmont
http://www.remlab.net/
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe,
---
libavcodec/riscv/pixblockdsp_init.c | 26 +++---
libavcodec/riscv/pixblockdsp_rvv.S | 6 +++---
2 files changed, 18 insertions(+), 14 deletions(-)
diff --git a/libavcodec/riscv/pixblockdsp_init.c
b/libavcodec/riscv/pixblockdsp_init.c
index aa39a8a665..8f24281217 100644
This will be required for the following changesets.
---
libavcodec/riscv/idctdsp_init.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/riscv/idctdsp_init.c b/libavcodec/riscv/idctdsp_init.c
index e6e616a555..4106d90c55 100644
--- a/libavcodec/riscv/idctdsp_init.c
If the scan lines are aligned, we can load each row as a 64-bit value,
thus avoiding segmentation. And then we can factor the conversion or
subtraction.
In principle, the same optimisation should be possible for high depth,
but would require 128-bit elements, for which no FFmpeg CPU flag
exists.
s contracts have business secrecy.
Even here, I know at least one of my colleagues has applied to have their
taxable income delisted on the basis of GDPR.
--
Rémi Denis-Courmont
http://www.remlab.net/
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg
Then that is nowhere near the level of labour-intensive (for the GA) and
privacy-intrusive (for the consultants) that SPI funding would involve, more
or less making my point.
--
Rémi Denis-Courmont
http://www.remlab.net/
___
ffmpeg-devel mai
This follows the same idea as with pixblockdsp, but applied at the
other end, whilst writing data at the end of the function.
---
libavcodec/riscv/idctdsp_rvv.S | 27 +--
1 file changed, 9 insertions(+), 18 deletions(-)
diff --git a/libavcodec/riscv/idctdsp_rvv.S
---
libavcodec/riscv/idctdsp_rvv.S | 28 ++--
1 file changed, 14 insertions(+), 14 deletions(-)
diff --git a/libavcodec/riscv/idctdsp_rvv.S b/libavcodec/riscv/idctdsp_rvv.S
index 4ff72f48d2..fafdddb174 100644
--- a/libavcodec/riscv/idctdsp_rvv.S
+++
---
libavcodec/riscv/idctdsp_rvv.S | 25 +
1 file changed, 9 insertions(+), 16 deletions(-)
diff --git a/libavcodec/riscv/idctdsp_rvv.S b/libavcodec/riscv/idctdsp_rvv.S
index fafdddb174..e93e6b5e7a 100644
--- a/libavcodec/riscv/idctdsp_rvv.S
+++
P.S.:
It took some additional efforts to get some benchmarks with proto-RVV.
But here they are:
idctdsp.add_pixels_clamped_c: 259.5
idctdsp.add_pixels_clamped_rvv_i64: 90.5
idctdsp.put_pixels_clamped_c: 186.5
idctdsp.put_pixels_clamped_rvv_i64: 65.5
idctdsp.put_signed_pixels_clamped_c: 209.5
ECODER) += riscv/huffyuvdsp_init.o
diff --git a/libavcodec/riscv/jpeg2000dsp_init.c
b/libavcodec/riscv/jpeg2000dsp_init.c
new file mode 100644
index 00..9415a22f79
--- /dev/null
+++ b/libavcodec/riscv/jpeg2000dsp_init.c
@@ -0,0 +1,36 @@
+/*
+ * Copyright © 2023 Rémi Denis-Courmont.
+ *
+ * Thi
jpeg2000_rct_int_c: 2592.2
jpeg2000_rct_int_rvv_i32: 1154.2
---
libavcodec/riscv/jpeg2000dsp_init.c | 8 ++--
libavcodec/riscv/jpeg2000dsp_rvv.S | 23 +++
2 files changed, 29 insertions(+), 2 deletions(-)
diff --git a/libavcodec/riscv/jpeg2000dsp_init.c
This is so that they can be loaded from assembler, rather than
duplicated.
---
libavcodec/jpeg2000dsp.c | 3 ++-
libavcodec/jpeg2000dsp.h | 2 ++
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/libavcodec/jpeg2000dsp.c b/libavcodec/jpeg2000dsp.c
index b1bff6d5b1..50bc1ecee6 100644
hf_g_filt_c: 1552.5
hf_g_filt_rvv_f32: 679.5
---
libavcodec/riscv/sbrdsp_init.c | 3 +++
libavcodec/riscv/sbrdsp_rvv.S | 20
2 files changed, 23 insertions(+)
diff --git a/libavcodec/riscv/sbrdsp_init.c b/libavcodec/riscv/sbrdsp_init.c
index 1b85b2cae9..71de681185
As in the aligned case, we can use VLSE64.V, though the way of doing so
gets more convoluted, so the performance gains are more modest:
get_pixels_unaligned_c: 126.7
get_pixels_unaligned_rvv_i32: 145.5 (before)
get_pixels_unaligned_rvv_i64: 62.2 (after)
For the reference, those are the
Hi,
FWIW, FFmpeg will most probably be granted a free community booth at the next
SCaLE in 21x a month earlier also in South-Western USA. If this unfolds as it
usually does, we will get confirmation in January.
There are no hidden costs *there*. But of course it's a very different crowd of
ugh would probably not be newsworthy. And it seems unlikely that
major ones like Kodi, mpv, VLC, etc, would let this slip through in the first
place.
> All these news articles are free amplification of the message ;)
That most probably will not happen, and if it does, it will most proba
Le torstaina 2. marraskuuta 2023, 23.07.03 EET Rémi Denis-Courmont a écrit :
> This uses a more traditional approach allowing up processing of up to
> period minus two elements per iteration. This also allows the algorithm
> to work for all and any vector length.
>
> As the T-H
This uses a more traditional approach allowing up processing of up to
period minus two elements per iteration. This also allows the algorithm
to work for all and any vector length.
As the T-Head C908 device under test can load 16 elements loop, there is
unsurprisingly a little performance drop
/libavcodec/riscv/sbrdsp_init.c
@@ -0,0 +1,37 @@
+/*
+ * Copyright © 2023 Rémi Denis-Courmont.
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software
sum_square_c: 803.5
sum_square_rvv_f32: 283.2
---
libavcodec/riscv/sbrdsp_init.c | 2 ++
libavcodec/riscv/sbrdsp_rvv.S | 19 +++
2 files changed, 21 insertions(+)
diff --git a/libavcodec/riscv/sbrdsp_init.c b/libavcodec/riscv/sbrdsp_init.c
index 837f24e1e0..e0e62278b0
With 128-bit vectors, this is mostly pointless but also harmless.
Performance gains should be more noticeable with larger vector sizes.
neg_odd_64_c: 76.2
neg_odd_64_rvv_i64: 74.7
---
libavcodec/riscv/sbrdsp_init.c | 5 +
libavcodec/riscv/sbrdsp_rvv.S | 17 +
2 files
Le sunnuntaina 29. lokakuuta 2023, 18.12.58 EET Michael Niedermayer a écrit :
> On Sun, Oct 29, 2023 at 04:35:35PM +0200, Rémi Denis-Courmont wrote:
> > Hi,
> >
> > Le 28 octobre 2023 21:01:57 GMT+03:00, Michael Niedermayer
a écrit :
> > >On Sat, Oct 28, 202
Le sunnuntaina 29. lokakuuta 2023, 18.47.34 EET Nicolas George a écrit :
> Rémi Denis-Courmont (12023-10-29):
> > And unfortunately, I do believe that Ronald is correct in pointing out
> > that big companies will want oversight in exchange for money.
>
> And this is why the o
Hi,
Le 28 octobre 2023 21:01:57 GMT+03:00, Michael Niedermayer
a écrit :
>On Sat, Oct 28, 2023 at 07:21:03PM +0200, Michael Niedermayer wrote:
>> Hi ronald
>>
>> On Sat, Oct 28, 2023 at 12:43:15PM -0400, Ronald S. Bultje wrote:
>> > Hi Thilo,
>> >
>> > On Sat, Oct 28, 2023 at 11:31 AM Thilo
In the aligned case, the existing RVI assembler is actually much
faster. In the unaligned case, there is nothing much to gain over C.
---
libavcodec/riscv/pixblockdsp_init.c | 7 +--
libavcodec/riscv/pixblockdsp_rvv.S | 7 ---
2 files changed, 1 insertion(+), 13 deletions(-)
diff --git
---
libavutil/riscv/cpu.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavutil/riscv/cpu.c b/libavutil/riscv/cpu.c
index fa45c0ad83..460d3e9f91 100644
--- a/libavutil/riscv/cpu.c
+++ b/libavutil/riscv/cpu.c
@@ -67,7 +67,7 @@ int ff_get_cpu_flags_riscv(void)
#endif
Le maanantaina 6. marraskuuta 2023, 17.36.18 EET Kieran Kunhya a écrit :
> On Mon, 6 Nov 2023 at 15:19, Michael Riedl
>
> wrote:
> > Whitespaces after semicolon breaks some servers
>
> Are you sure this patch doesn't break other servers? SDP is a painfully
> fragile format.
AFAIK, you're not
/riscv/huffyuvdsp_init.c
new file mode 100644
index 00..0f7bc4d692
--- /dev/null
+++ b/libavcodec/riscv/huffyuvdsp_init.c
@@ -0,0 +1,37 @@
+/*
+ * Copyright © 2023 Rémi Denis-Courmont.
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it
Le lauantaina 28. lokakuuta 2023, 16.56.40 EEST Rémi Denis-Courmont a écrit :
> +#include "config.h"
> +#include "libavutil/attributes.h"
> +#include "libavutil/cpu.h"
> +#include "libavcodec/huffyuvdsp.h"
> +
> +void ff_add_int16_r
/utvideodsp_init.c
b/libavcodec/riscv/utvideodsp_init.c
new file mode 100644
index 00..dfaa16692a
--- /dev/null
+++ b/libavcodec/riscv/utvideodsp_init.c
@@ -0,0 +1,38 @@
+/*
+ * Copyright © 2023 Rémi Denis-Courmont.
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute
restore_rgb_planes10_c: 185852.2
restore_rgb_planes10_rvv_i32: 90130.5
---
libavcodec/riscv/utvideodsp_init.c | 9 +++-
libavcodec/riscv/utvideodsp_rvv.S | 35 ++
2 files changed, 43 insertions(+), 1 deletion(-)
diff --git
art of the GA, and I have neither the expertise and
credibility nor the time and motivation to take up this project, so that's
just my free advice.
--
Rémi Denis-Courmont
http://www.remlab.net/
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
h
Le keskiviikkona 27. syyskuuta 2023, 23.18.29 EEST Michael Niedermayer a écrit
:
> And you can repeat it as often as you want, iam not interrested.
> Why do you not understand this ?
You can repeat the contrary as much as you want, we do not believe that your
SDR code fits in FFmpeg. Why do you
SDR-outside-FFmpeg. He can stick to spending his "time on
boring things" if/when he gets paid.
That will add incentives for big tech corps to actually fund FFmpeg. This
community, and presumably Michael in particular, are way too good at doing
those "boring things" for
Le torstaina 28. syyskuuta 2023, 18.28.50 EEST Nicolas George a écrit :
> Rémi Denis-Courmont (12023-09-28):
> > You can repeat the contrary as much as you want, we do not believe that
> > your SDR code fits in FFmpeg. Why do you not understand this?
>
> We understand that
Le torstaina 28. syyskuuta 2023, 19.43.57 EEST Nicolas George a écrit :
> Rémi Denis-Courmont (12023-09-28):
> > Calling BBB, Kieran and myself dishonest is.
>
> I call your arguments dishonest.
You almost had me doubting my recollection for a minute. But:
Michael wrote:
> &g
Le torstaina 28. syyskuuta 2023, 19.32.44 EEST Nicolas George a écrit :
> Rémi Denis-Courmont (12023-09-28):
> > Err, it is very much an issue w.r.t. "catching on".
>
> Moving the goalpost much.
I think obviously not, considering the original quote (EMPHASIS ADDED):
&g
ber of cases _should_ be low enough that
year-on-year statistics should not be very meaningful. Also that seems like
useless paperwork added on top of what is already free and not particularly
fun community service.
--
Rémi Denis-Courmont
rvv_f32: 4.2
> - af_afir.fcmul_add [OK]
>fcmul_add_c: 4.7
>fcmul_add_rvv_f32: 3.5
>
>
>Rémi Denis-Courmont 于2023年9月28日周四 00:41写道:
>
>> Le tiistaina 26. syyskuuta 2023, 12.24.58 EEST flow gg a écrit :
>> > benchmark:
>> > fcmul_add_c: 19.7
>> &g
Le torstaina 28. syyskuuta 2023, 19.41.06 EEST Nicolas George a écrit :
> (1) There was no violations in my message. Calling your arguments
> bullshit is not a personal attack, by definition.
Calling BBB, Kieran and myself dishonest is.
> (2) Even if there were any violation, there is nobody to
Le torstaina 28. syyskuuta 2023, 18.33.33 EEST Nicolas George a écrit :
> Rémi Denis-Courmont (12023-09-28):
> > Strange, I thought FFmpeg really became popular as a back-end library for
> > mplayer, before it was picked up by all other OSS multimedia at the time
> > (gstr
Le torstaina 28. syyskuuta 2023, 19.13.37 EEST Nicolas George a écrit :
> Rémi Denis-Courmont (12023-09-28):
> > Thanks for making my point.
>
> Stealing the other person rhetoric device does not make you right.
>
> > That does not change the fact that it won't make
audiodsp.vector_clip_int32_c: 17500.7
audiodsp.vector_clip_int32_rvv_i32: 8404.7 (m1)
audiodsp.vector_clip_int32_rvv_i32: 2689.9 (m8)
audiodsp.vector_clipf_c: 33679.7
audiodsp.vector_clipf_rvf: 7019.7
audiodsp.vector_clipf_rvv_f32: 8328.0 (m1)
audiodsp.vector_clipf_rvv_f32: 2209.4
Le 28 septembre 2023 22:28:35 GMT+03:00, Michael Niedermayer
a écrit :
>On Thu, Sep 28, 2023 at 12:28:22PM +0200, Anton Khirnov wrote:
>> Quoting Michael Niedermayer (2023-09-23 18:43:44)
>> > Is a seperate library really the right choice for SDR in FFmpeg at this
>> > stage ?
>> >
>> > Also
Le 29 septembre 2023 10:03:13 GMT+03:00, Nicolas George a
écrit :
>Well, I think this makes it obvious why I think people representing
>downstream projects should have very little influence on the future of
>FFmpeg.
First, I don't *represent* any downstream. I just happen to be one of several
Le 29 septembre 2023 10:03:13 GMT+03:00, Nicolas George a
écrit :
>Rémi Denis-Courmont (12023-09-29):
>> And as a downstream OSS application developer, I'd really rather
>> postproc be officially and properly dropped, so I can justify dropping
>> it downstream a
Le 29 septembre 2023 14:43:11 GMT+03:00, Nicolas George a
écrit :
>Anton Khirnov (12023-09-29):
>> What does anybody gain from it being in our source tree, as opposed to a
>> separate library?
>
>Not having to maintain yet another packaging and build system.
Talk about flimsy "scraps of the
Le lauantaina 23. syyskuuta 2023, 9.49.47 EEST Neal Gompa a écrit :
> On Fri, Sep 22, 2023 at 12:33 PM Michael Niedermayer
>
> wrote:
> > > What does this mean? Does this mean an FFmpeg release containing code
> > > that
> > > interfaces with your SDR library? Or does it mean the library fully
>
---
libavcodec/riscv/fmtconvert_rvv.S | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/riscv/fmtconvert_rvv.S
b/libavcodec/riscv/fmtconvert_rvv.S
index 51365753d9..d0e2f106d5 100644
--- a/libavcodec/riscv/fmtconvert_rvv.S
+++ b/libavcodec/riscv/fmtconvert_rvv.S
@@
ndif
diff --git a/libavutil/riscv/bswap_rvb.S b/libavutil/riscv/bswap_rvb.S
new file mode 100644
index 00..b391e2f9f1
--- /dev/null
+++ b/libavutil/riscv/bswap_rvb.S
@@ -0,0 +1,65 @@
+/*
+ * Copyright © 2022 Rémi Denis-Courmont.
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is fr
cale/riscv/rgb2rgb_rvb.S
@@ -0,0 +1,30 @@
+/*
+ * Copyright © 2022 Rémi Denis-Courmont.
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Fo
This is slower than the Zbb version on real hardware due to register
strides. Proper support for vector byte-swap requires the Zvbb
extension, but it's much too early for me to worry about it.
---
libswscale/riscv/rgb2rgb.c | 2 --
libswscale/riscv/rgb2rgb_rvv.S | 25
Remove some useless vector splat.
---
libavcodec/riscv/aacpsdsp_rvv.S | 12
1 file changed, 4 insertions(+), 8 deletions(-)
diff --git a/libavcodec/riscv/aacpsdsp_rvv.S b/libavcodec/riscv/aacpsdsp_rvv.S
index b581383f77..b85a5cc92c 100644
--- a/libavcodec/riscv/aacpsdsp_rvv.S
+++
---
libavcodec/riscv/aacpsdsp_rvv.S | 46 -
1 file changed, 23 insertions(+), 23 deletions(-)
diff --git a/libavcodec/riscv/aacpsdsp_rvv.S b/libavcodec/riscv/aacpsdsp_rvv.S
index b85a5cc92c..1a92fed515 100644
--- a/libavcodec/riscv/aacpsdsp_rvv.S
+++
This avoids using relatively slow register strides.
---
libswscale/riscv/rgb2rgb_rvv.S | 24 +++-
1 file changed, 11 insertions(+), 13 deletions(-)
diff --git a/libswscale/riscv/rgb2rgb_rvv.S b/libswscale/riscv/rgb2rgb_rvv.S
index 008f098bfe..3e7988ca01 100644
---
We can make do with callee-clobbered registers only now.
As an added bonus, this makes the code XLEN-independent.
---
libswscale/riscv/rgb2rgb.c | 2 --
libswscale/riscv/rgb2rgb_rvv.S | 10 ++
2 files changed, 2 insertions(+), 10 deletions(-)
diff --git a/libswscale/riscv/rgb2rgb.c
---
libavutil/riscv/float_dsp_rvv.S | 45 ++---
1 file changed, 25 insertions(+), 20 deletions(-)
diff --git a/libavutil/riscv/float_dsp_rvv.S b/libavutil/riscv/float_dsp_rvv.S
index 7e9e84d526..91b70bf148 100644
--- a/libavutil/riscv/float_dsp_rvv.S
+++
This revectors the inner loop to reverse vectors element in vectors,
thus eliminating the negative register stride. Note that RVV does not
have a vector reverse instruction, so this uses a gather.
---
libavutil/riscv/float_dsp_rvv.S | 17 +++--
1 file changed, 11 insertions(+), 6
---
libswscale/riscv/rgb2rgb_rvv.S | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libswscale/riscv/rgb2rgb_rvv.S b/libswscale/riscv/rgb2rgb_rvv.S
index ff02eba9bf..56539690bc 100644
--- a/libswscale/riscv/rgb2rgb_rvv.S
+++ b/libswscale/riscv/rgb2rgb_rvv.S
@@ -82,11 +82,11
---
libswscale/riscv/rgb2rgb_rvv.S | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/libswscale/riscv/rgb2rgb_rvv.S b/libswscale/riscv/rgb2rgb_rvv.S
index 3200370224..ff02eba9bf 100644
--- a/libswscale/riscv/rgb2rgb_rvv.S
+++ b/libswscale/riscv/rgb2rgb_rvv.S
@@ -112,14
There was an additional iteration doing nothing for each line,
due to checking the selected vector length instead of the available
vector length.
---
libswscale/riscv/rgb2rgb_rvv.S | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libswscale/riscv/rgb2rgb_rvv.S
---
libswscale/riscv/rgb2rgb_rvv.S | 26 ++
1 file changed, 26 insertions(+)
diff --git a/libswscale/riscv/rgb2rgb_rvv.S b/libswscale/riscv/rgb2rgb_rvv.S
index 71e8fac6f3..671089c842 100644
--- a/libswscale/riscv/rgb2rgb_rvv.S
+++ b/libswscale/riscv/rgb2rgb_rvv.S
@@
It does not make much sense to me, but GCC somehow optimises the
inline assembler way even when the output are very obviously used.
This reverts commit 09731fbfc3a914ec4f6ffad60aa9062db6a8f6aa.
---
libavutil/riscv/timer.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git
ode could be improved regardless.
--
Rémi Denis-Courmont
http://www.remlab.net/
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org wit
have not tested on C910 recently
so maybe it's just how it is). But in any case, performance metrics from C
code compiled with -O0 are worthless.
--
Rémi Denis-Courmont
http://www.remlab.net/
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https:
Le keskiviikkona 27. syyskuuta 2023, 4.47.26 EEST flow gg a écrit :
> >>> please pad mnemonics to at least 8 columns for consistency
>
> okay, changed
>
> >>> It seems that you could just as well use vlseg2 without register
>
> stride, no?
>
> yes, vlseg will better, changed
>
> >>> Note that
Le tiistaina 3. lokakuuta 2023, 21.50.25 EEST Nicolas George a écrit :
> More precisely, I now strongly believe that democracy is a terrible way
> to run a project like FFmpeg.
>
> As the project grows, democracy gives an increasing weight to people who
> came late to contribute to the project in
---
libavutil/riscv/fixed_dsp_init.c | 6 +-
libavutil/riscv/fixed_dsp_rvv.S | 22 ++
2 files changed, 27 insertions(+), 1 deletion(-)
diff --git a/libavutil/riscv/fixed_dsp_init.c b/libavutil/riscv/fixed_dsp_init.c
index d4ca2e2064..409b6d7b55 100644
---
vector_fmul_add_fixed_c: 2.7
vector_fmul_add_fixed_rvv_i64: 0.7
---
libavutil/riscv/fixed_dsp_init.c | 6 +-
libavutil/riscv/fixed_dsp_rvv.S | 22 ++
2 files changed, 27 insertions(+), 1 deletion(-)
diff --git a/libavutil/riscv/fixed_dsp_init.c
get merged ?
He is suggesting that if one person *appears* to be alone in wanting to take
(or take back) FFmpeg in the given directions that they advocate, they should
consider starting their own project or fork.
--
Rémi Denis-Courmont
http://www.remlab.net/
_
The gather index vector is only used as double-length (due to register
pressure), so no need to initialise it for quad-length. Basically this
matches the multiplier in the prologue to the the multipler in the loop.
---
libavutil/riscv/float_dsp_rvv.S | 2 +-
1 file changed, 1 insertion(+), 1
vector_fmul_fixed_c: 4.0
vector_fmul_fixed_rvv_i64: 0.5
---
libavutil/riscv/fixed_dsp_init.c | 3 +++
libavutil/riscv/fixed_dsp_rvv.S | 17 +
2 files changed, 20 insertions(+)
diff --git a/libavutil/riscv/fixed_dsp_init.c b/libavutil/riscv/fixed_dsp_init.c
index
---
libavutil/riscv/fixed_dsp_init.c | 6 +
libavutil/riscv/fixed_dsp_rvv.S | 44
2 files changed, 50 insertions(+)
diff --git a/libavutil/riscv/fixed_dsp_init.c b/libavutil/riscv/fixed_dsp_init.c
index 02883b31d4..6469b45374 100644
---
vector_fmul_add_fixed_c: 2.2
vector_fmul_add_fixed_rvv_i64: 0.5
---
libavutil/riscv/fixed_dsp_init.c | 5 +
libavutil/riscv/fixed_dsp_rvv.S | 21 +
2 files changed, 26 insertions(+)
diff --git a/libavutil/riscv/fixed_dsp_init.c b/libavutil/riscv/fixed_dsp_init.c
index
---
libavutil/riscv/fixed_dsp_init.c | 3 +++
libavutil/riscv/fixed_dsp_rvv.S | 24
2 files changed, 27 insertions(+)
diff --git a/libavutil/riscv/fixed_dsp_init.c b/libavutil/riscv/fixed_dsp_init.c
index 4b239829f5..470b35fb63 100644
---
Le tiistaina 26. syyskuuta 2023, 12.24.58 EEST flow gg a écrit :
> benchmark:
> fcmul_add_c: 19.7
> fcmul_add_rvv_f32: 6.7
Nit: please pad mnemonics to at least 8 columns for consistency.
I'm a bit surprised that the performance improves this much, considering that
the C910 is notoriously bad
Le tiistaina 26. syyskuuta 2023, 21.40.12 EEST Paul B Mahol a écrit :
> On Tue, Sep 26, 2023 at 8:35 PM Rémi Denis-Courmont wrote:
> > Le tiistaina 26. syyskuuta 2023, 12.24.58 EEST flow gg a écrit :
> > > benchmark:
> > > fcmul_add_c: 19.7
> > > fcmul_add_r
Le tiistaina 26. syyskuuta 2023, 12.24.58 EEST flow gg a écrit :
> benchmark:
> fcmul_add_c: 19.7
> fcmul_add_rvv_f32: 6.7
+li t1, 4
+vsetvli t0, t1, e32, m1, ta, ma
vsetivli t0, 4, ...
But really, DO NOT use a fixed vector length here. At best, you're wasting half
the vector width.
Le torstaina 21. syyskuuta 2023, 21.56.52 EEST Michael Niedermayer a écrit :
> Hi
>
> On Thu, Sep 21, 2023 at 05:33:54PM +0100, Kieran Kunhya via ffmpeg-devel
wrote:
> > On Thu, Sep 21, 2023 at 5:21 PM Michael Niedermayer
> >
> > wrote:
> > > OTOH If a majority of people are against the SDR
---
libavcodec/riscv/aacpsdsp_rvv.S | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavcodec/riscv/aacpsdsp_rvv.S b/libavcodec/riscv/aacpsdsp_rvv.S
index 007002de70..fe250cd83b 100644
--- a/libavcodec/riscv/aacpsdsp_rvv.S
+++ b/libavcodec/riscv/aacpsdsp_rvv.S
@@ -203,12
---
libavcodec/riscv/aacpsdsp_rvv.S | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavcodec/riscv/aacpsdsp_rvv.S b/libavcodec/riscv/aacpsdsp_rvv.S
index f552d50e23..007002de70 100644
--- a/libavcodec/riscv/aacpsdsp_rvv.S
+++ b/libavcodec/riscv/aacpsdsp_rvv.S
@@ -168,10
---
libavcodec/riscv/aacpsdsp_rvv.S | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavcodec/riscv/aacpsdsp_rvv.S b/libavcodec/riscv/aacpsdsp_rvv.S
index 1a92fed515..f552d50e23 100644
--- a/libavcodec/riscv/aacpsdsp_rvv.S
+++ b/libavcodec/riscv/aacpsdsp_rvv.S
@@ -38,14
/audiodsp_init.o \
diff --git a/libavcodec/riscv/ac3dsp_init.c b/libavcodec/riscv/ac3dsp_init.c
new file mode 100644
index 00..20f294f1de
--- /dev/null
+++ b/libavcodec/riscv/ac3dsp_init.c
@@ -0,0 +1,38 @@
+/*
+ * Copyright © 2023 Rémi Denis-Courmont.
+ *
+ * This file is part of FFmpeg
Le keskiviikkona 4. lokakuuta 2023, 22.51.10 EEST Rémi Denis-Courmont a écrit
:
> +vwmul.vv v12, v16, v28
> +vwmul.vv v8, v16, v24
> +// vwnmsac.vv does _not_ exist so multiply & subtract separately
> +vwmul.vv v4, v20, v24
> +vwm
AV_CPU_FLAG_RVB_ADDR)) {
+c->reorder_pixels = ff_reorder_pixels_rvv;
+}
+#endif
+}
diff --git a/libavcodec/riscv/exrdsp_rvv.S b/libavcodec/riscv/exrdsp_rvv.S
new file mode 100644
index 00..f4a35f58ff
--- /dev/null
+++ b/libavcodec/riscv/exrdsp_rvv.S
@@ -0,0 +1,38 @@
+/*
+ *
quot; from some years ago.
If you do not agree with Nicolas on this point, then admonishing Kieran and
Vittorio about alleged CoC violations is not conducive to resolving your
disagreement, nor is preaching to them about an united FFmpeg team.
I understand that you do not like this situatio
Le keskiviikkona 18. lokakuuta 2023, 0.57.45 EEST Michael Niedermayer a écrit
:
> On Tue, Oct 17, 2023 at 09:50:41PM +0300, Rémi Denis-Courmont wrote:
> > Le perjantaina 13. lokakuuta 2023, 22.19.34 EEST Michael Niedermayer a
écrit :
> > > But some goals would proba
tware
> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301
> USA + */
> +
> +#ifndef AVUTIL_EXECUTOR_H
> +#define AVUTIL_EXECUTOR_H
> +
> +typedef struct AVExecutor AVExecutor;
> +typedef struct AVTask AVTask;
> +
> +struct AVTask {
>
No
Le 27 août 2023 17:43:13 GMT+03:00, Tmc Tmc a écrit :
>Hi All,
>I found a bug in ffmpeg's RTSP implementation.
>
>The workflow is as follows:
>
>1. Have a RTSP server that supports Pause.
>2. Have ffmpeg play a video from that server (rtsp://:...).
>3. Pause the video in ffmpeg.
>4. ffmpeg
1 - 100 of 757 matches
Mail list logo