[FFmpeg-devel] [PATCH 00/72] Implement support for Vulkan multiplane images and video decoding

Lynne dev at lynne.ee
Fri Feb 17 12:45:44 EET 2023

Feb 17, 2023, 10:45 by h.leppkes at gmail.com:

> On Fri, Feb 17, 2023 at 10:09 AM Jean-Baptiste Kempf <jb at videolan.org> wrote:
>> Hello,
>> On Fri, 17 Feb 2023, at 04:43, Lynne wrote:
>> > This small patchset mostly rewrites Vulkan to enable using multiplane images,
>> This is not small. We're talking about thousands of lines of code;
>> And this changes numerous h264 and hevc headers, and adds callback to the make AVHWAccel.
>> And that is besides the full rewrite of some Vulkan files...
>> This will take a long time to review.
> I would agree, this set is too large to try to skirt it in just under
> the deadline of a new release, with the potential of needing to fix
> stuff later.
> Just have it land on master after the branch, and there is no time
> pressure to review it, or make numerous fixes on the release branch.

The code's been tested by a lot of folks already, and importantly,
passes the validation layers (with some warnings due to spec issues
that the workgroup agreed to fix).
The output is also conformant on all implementations.
All the code needs is some looking over to make sure I haven't done
something silly. Valgrind output is clean too.
It is a lot of code in terms of diff, but rather than reviewing each patch
individually, you should just try looking at the final files directly, that way,
you only have a few thousand lines to look at, rather than a few
thousand times twenty.
I've abstracted all the low-level Vulkan stuff from the decoding patchset,
so the output is high-level enough that anyone who's written hwaccels can
understand it.

> PS:
> Can you please send it as a proper patchset? Reviewing 72 patches off
> of one mail is not really fitting the workflow at all.

I'll do that.

More information about the ffmpeg-devel mailing list