>>I was trying libnpp on Big Bunny video by below command:
>>./ffmpeg -loglevel debug -y -hwaccel cuvid -c:v h264_cuvid -vsync 0 -i
>>/root/root/bunny.mp4 -vf scale_npp=1920:1072 -vcodec h264_nvenc
>>/tmp/tmp0.264 -vf scale_npp=1280:720 -vcodec h264_nvenc /tmp/tmp1.264
>>This command is
>> ffmpeg.c is in the middle of some changes being merged, and once that is
>> done ffmpeg_cuvid.c will need to be adapted to a new system. Until then, it
>> would make everything easier if the code actively >> being modified by these
>> merges is not modified by hackish fixes.
When do we
>>These are merges. Unless you volunteer to do the merges yourself (and
>>properly please) you'll have to live with this.
So were there merges that got left and broke the functionality. When do we get
these merges in ? Please let me know if I can help get this up and fixed. There
have been
The pass through mode is broken, since error is returned when none of the
stages is needed. In such a case only 0 should be returned.
Try command line with i/p resolution = o/p resolution to observe the issue.
Sample Command Line : ffmpeg -y -hwaccel cuvid -c:v h264_cuvid -vsync 0 -i
> I created a simpler patch that avoids modifying the external signature
> of the function, and it still fixes it for me. Can you test and
> confirm? Then we can apply this.
> Just tested this patch, and I can confirm that at least a cuvid hwaccel
> transcode works again.
I also verified the
>> There is BUG in Nvidia NVENC when you use AUD for H264 with B-frames, it
>> will return corrupted stream, because NVIDIA is inserting AUD type 0
>> (I-frame) before B-frames instead of AUD type 7 (any-frame).
Thanks for bringing this to notice. We have added a fix in the Nvidia driver
and
>> Please find attached a CUDA based scale filter. This filter will be a
>> starting point to add other CUDA based filters. The filter supports 420-8,
>> 420-10, 444-8, 444-10 formats and will help to do complete 10-bit HW
>> accelerated transcoding when using >> CUVID.
>> Currently the filter
>> Having to run a .bat file prior to building isn't really nice, and not
>> something I and some other people on IRC would like to see in ffmpeg.
>> Instead, it would be nice if configure/make would learn how to handle .cu
>> files, converting them straight to an object-file.
>> My idea for
Pass through mode (input resolution = output resolution) mode is broken in
scale_npp. This patch fixes the same.
Thanks,
Yogender
---
This email message is for the sole use of the intended recipient(s) and may
The better simplification may be :
ctx->outputs[0]->hw_frames_ctx = s->stages[last_stage > 0] ?
s->stages[last_stage].frames_ctx : ctx->inputs[0]->hw_frames_ctx;
-Original Message-
From: ffmpeg-devel [mailto:ffmpeg-devel-boun...@ffmpeg.org] On Behalf Of Timo
Rothenpieler
Sent: Monday,
>> Applied with a slightly changed logic to achieve the same with fewer lines.
I just tested, the modified logic doesn't work as none of the stages are
initialized.
Thanks,
Yogender
-Original Message-
From: ffmpeg-devel [mailto:ffmpeg-devel-boun...@ffmpeg.org] On Behalf Of Timo
Please help to understand what is broken with a command line, so that we can
address this issue.
Thanks,
Yogender
-Original Message-
From: ffmpeg-devel [mailto:ffmpeg-devel-boun...@ffmpeg.org] On Behalf Of wm4
Sent: Monday, July 03, 2017 4:29 PM
To: ffmpeg-devel@ffmpeg.org
Subject: Re:
Added 10/16bit formats to hwupload_cuda.
Thanks,
Yogender
---
This email message is for the sole use of the intended recipient(s) and may
contain
confidential information. Any unauthorized review, use, disclosure
Add P016 support to NVENC.
Thanks,
Yogender
---
This email message is for the sole use of the intended recipient(s) and may
contain
confidential information. Any unauthorized review, use, disclosure or
Added yuv444p16 to hwcontext_cuda.
Thanks,
Yogender
---
This email message is for the sole use of the intended recipient(s) and may
contain
confidential information. Any unauthorized review, use, disclosure or
>> I can push, but to answer Timo's question: I did not add P016 to nvenc
>> because it is technically not correct to pass 12bit content through a 10bit
>> surface - that will lead to truncation, rather than dithering. If you are
>> telling us that nvenc supports 12bit >> input and will
Currently combining CPU and CUDA filters requires insertion of hwupload and
download filters. I am trying to simply this by auto insertion of these
filters. This patch is for hwupload_cuda, but I want to do this for hwdownload
as well.
The attached patch automatically inserts hwupload_cuda
P016 support
Am 18.07.2017 um 13:24 schrieb Yogender Gupta:
> Add P016 support to NVENC.
>
> Thanks,
> Yogender
All three patches look fine to me in general.
I vaguely remember there being some reason for P016 being missing so far, but
it might be long fixed.
I'm not at home and can't
>>+cuda_sdk
>> libnpp
IMO, Libnpp is part of the CUDA SDK, and we can remove all the libnpp related
changes and just replace it by cuda-sdk in the configure/make files.
Thanks,
Yogender
---
This email
>> I'm wondering what the difference between this and constqp mode is, does
>> this set a constant quality in otherwise vbr mode?
Yes, something similar.
Thanks,
Yogender
---
This email message is for the sole use
Please find attached a patch for CUDA Scale Filter. I have modified the
makefiles as suggested during the previous reviews to have all PTX generated
during compile time.
Common.mak has changes for PTX generation
Ptx2c.sh has been added in ffbuild folder. This is needed to covert PTX to C
Users
devel] [PATCH] nvenc : Support Video Codec SDK 8.0
Am 09.05.2017 um 08:18 schrieb Yogender Gupta:
> Video Codec SDK 8.0 has been released today. Please find the patch that
> updates the interface. Also, attached the latest SDK 8.0 interface file for
> nvenc.
ffmpeg master is now updated
CQ rate control in NVENC now supports fractional values (SDK 8.0). Please find
patch attached for the same.
Thanks,
Yogender
---
This email message is for the sole use of the intended recipient(s) and may
contain
Weighted Prediction Support for NVENC (SDK 8.0). Please find patch attached for
the same.
Thanks,
Yogender
---
This email message is for the sole use of the intended recipient(s) and may
contain
confidential
Attached is a CUDA version of the thumbnail filter, this helps accelerate
thumbnails generations significantly, when using the GPU pipeline.
Regards,
Yogender
---
This email message is for the sole use of the
I believe there were concerns on pushing the CUDA thumbnail filter and that is
possible to get similar performance using the normal thumbnail filter. The CUDA
thumbnail filter is useful for generating thumbnails on the hwaccel cuvid
pipeline, as it can directly operate on the video memory and
05:59, Yogender Gupta <ygu...@nvidia.com> wrote:
> I believe there were concerns on pushing the CUDA thumbnail filter and
> that is possible to get similar performance using the normal thumbnail
> filter. The CUDA thumbnail filter is useful for generating thumbnails
> on th
Cc: Ganapathy Raman Kasi; Yogender Gupta; Konda Raju; Srinath K R; Ben Chang
Subject: [FFmpeg-devel] nvenc: missing slice offsets in
NV_ENC_LOCK_BITSTREAM::sliceOffsets
* PGP - S/MIME Signed by an unverified key: 09/02/2017 at 11:57:35 AM
Hello,
I am currently dealing with an issue regarding
Taken care of all comments except the documentation.
Will send out a separate patch for both the CUDA filters documentation.
Regards,
Yogender
---
This email message is for the sole use of the intended recipient(s)
I would like to correct a couple of lines on the following ffmpeg web page. Can
anyone please guide me on how to do it (push patch or raise ticket or edit page)
It is definitely possible to do a complete end to end 10-bit transcode on the
GPU pipeline without passing to system memory. I had
One of the users had recently complained about encoding problems with HEVC
encoding with WP (Weighted Prediction). This is a driver issue that has been
identified, and shows up when using HEVC + WP + (PictureTimingSEI messages and
or buffering period SEI messages).
Recently the PictureTiming
Please find attached a patch for setting key frames.
"-force_key_frames" can then use this option.
Thanks,
Yogender
---
This email message is for the sole use of the intended recipient(s) and may
contain
Improved the patch by dynamic allocation.
Thanks,
Yogender
---
This email message is for the sole use of the intended recipient(s) and may
contain
confidential information. Any unauthorized review, use, disclosure
Yes, this looks better and should have been this way.
Thanks,
Yogender
-Original Message-
From: Philip Langdale
Sent: Tuesday, May 14, 2019 8:42 AM
To: ffmpeg-devel@ffmpeg.org
Cc: Yogender Gupta ; Philip Langdale
Subject: [PATCH 3/3] avfilter/vf_scale_cuda: Simplify output plane
34 matches
Mail list logo