From: Jerome Martinez <jerome@mediaarea.net>
To: ffmpeg-devel@ffmpeg.org
Subject: Re: [FFmpeg-devel] [PATCH 01/18] hwcontext_vulkan: disable descriptor buffer extension on Intel
Date: Sun, 13 Apr 2025 15:38:33 +0200
Message-ID: <d4955d93-8f3b-4309-86ab-f8a60bebe16a@mediaarea.net> (raw)
In-Reply-To: <20250412072256.77815-1-dev@lynne.ee>
I tested the patches as a whole, for the moment with a focus on the
performance.
Tested with a RTX 4070 Ti on Linux & a RTX 3050 on Windows, globally no
regression in speed on both test platforms, and I find a +25% (up to
+35% in some cases, especially high resolutions) in average with 10-bit
or 16-bit content for most cases (it is weird sometimes, no speed
improvement, I'll check).
Easy to decode complex 4K 10-bit in real time with the RTX 4070 Ti now,
which is good performance for lossless compressed content.
Additionally, the cards can handle higher resolutions e.g. 12k9k16bit
succeeds now even on the low end RTX 3050 (at 0.5 fps :) ).
Le 12/04/2025 à 09:22, Lynne a écrit :
> Temporary workaround. Will be replaced with a version check once a fix is
> in the works and a known next version for Mesa with a fix is known.
> ---
> libavutil/hwcontext_vulkan.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
> index 319b71ed04..d11c0274d2 100644
> --- a/libavutil/hwcontext_vulkan.c
> +++ b/libavutil/hwcontext_vulkan.c
> @@ -773,6 +773,11 @@ static int check_extensions(AVHWDeviceContext *ctx, int dev, AVDictionary *opts,
> tstr = optional_exts[i].name;
> found = 0;
>
> + /* Intel has had a bad descriptor buffer implementation for a while */
> + if (p->vkctx.driver_props.driverID == VK_DRIVER_ID_INTEL_OPEN_SOURCE_MESA &&
> + !strcmp(tstr, VK_EXT_DESCRIPTOR_BUFFER_EXTENSION_NAME))
> + continue;
> +
> if (dev &&
> ((debug_mode == FF_VULKAN_DEBUG_VALIDATE) ||
> (debug_mode == FF_VULKAN_DEBUG_PRINTF) ||
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
prev parent reply other threads:[~2025-04-13 13:38 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-12 7:22 Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 02/18] vulkan_ffv1: enable acceleration " Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 03/18] vulkan_ffv1: remove unused define Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 04/18] vulkan_ffv1: slightly optimize the range decoder Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 05/18] vulkan_ffv1: optimize symbol reader Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 06/18] vulkan_ffv1: allocate just as much memory for slice state as needed Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 07/18] vulkan_ffv1: init overread/corrupt fields Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 08/18] vulkan_ffv1: fallback to upload if mapping packet fails, fix fallback Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 09/18] vulkan_ffv1: fix reset shader dependencies Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 10/18] vulkan_ffv1: improve buffer barrier correctness for slice state Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 11/18] vulkan_ffv1: fix left-2 sample addressing Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 12/18] vulkan_ffv1: cache only 2 lines when decoding RGB Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 13/18] ffv1/vulkan: redo context count tracking and quant_table_idx management Lynne
2025-04-13 20:39 ` Jerome Martinez
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 14/18] vulkan_ffv1: externalize extended lookup check Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 15/18] vulkan_ffv1: remove need for scratch data during setup Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 16/18] vulkan_ffv1: shortcut +-1 coeffs in symbol reading Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 17/18] vulkan: add support for expect/assume Lynne
2025-04-12 7:22 ` [FFmpeg-devel] [PATCH 18/18] vulkan_ffv1: add cached symbol reader for AMD Lynne
2025-04-13 13:38 ` Jerome Martinez [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=d4955d93-8f3b-4309-86ab-f8a60bebe16a@mediaarea.net \
--to=jerome@mediaarea.net \
--cc=ffmpeg-devel@ffmpeg.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
This inbox may be cloned and mirrored by anyone:
git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git
# If you have public-inbox 1.1+ installed, you may
# initialize and index your mirror using the following commands:
public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \
ffmpegdev@gitmailbox.com
public-inbox-index ffmpegdev
Example config snippet for mirrors.
AGPL code for this site: git clone https://public-inbox.org/public-inbox.git