Possible Performance Regression with Mesa

[João Paulo Gonçalves]

Hello everybody,

On Thu, Apr 25, 2024 at 02:08:06PM +0200, Lucas Stach wrote:
> 1. weston switches to the scanout tranche, as it would like to put the
> surface on a plane
> 2. the client reallocates as linear but does so on the render node
> 3. weston still isn't able to put the buffer on the plane, as it's
> still scanout incompatible due to being non-contig, so needs to fall
> back to rendering
> 4. now we are stuck at a linear buffer being used for rendering, which
> is very non-optimal
> ...
>
> Regards,
> Lucas

Could you please clarify a question I have about this issue? In the end,
is the problem Lucas described caused by the Mesa implementation (and
should be fixed in Mesa) or by the Weston implementation? From what I
understand, this is an issue with the DRM backend implementation in
Weston, and it is Weston that needs to be fixed.

Thanks!

Best Regards,
João Paulo Gonçalves