[virglrenderer-devel] [PATCH 1/1] Fixes an issue with blitting R8G8B8X8_UNORM buffer

[Gurchetan Singh]

This patch regresses the following dEQP tests on a Nvidia GL host:

dEQP-GLES3.functional.fbo.msaa.2_samples.rgb8
dEQP-GLES3.functional.fbo.msaa.4_samples.rgb8
dEQP-GLES3.functional.fbo.msaa.8_samples.rgb8

So maybe fixing these tests can fix WebGL?

On Thu, Apr 19, 2018 at 2:23 AM, Alexandros Frantzis
<alexandros.frantzis at collabora.com> wrote:
> On Fri, Apr 13, 2018 at 06:57:14PM -0700, Robert Tarasov wrote:
>> On Fri, Apr 13, 2018 at 12:21 PM, Lepton Wu <lepton at chromium.org> wrote:
>>
>> > This breaks  webgl fish test,  with this CL,
>> >
>> > https://webglsamples.org/aquarium/aquarium.html  show black screen
>> > with Chrome OS inside qemu+virgl, even it show fps.
>> >
>> > I guess dEQP is running with offline rendering  so it doesn't find
>> > this regression.
>>
>> Looking into it.
>
> Hi all,
>
> just a heads-up that the patchset:
>
> vrend: Take format swizzles into account when blitting with GL
> https://lists.freedesktop.org/archives/virglrenderer-devel/2018-April/000358.html
>
> tackles, among other things, the issue of blitting from rgb8 to *, using
> a more general, swizzle-based method instead of having a special case for
> a particular format. In particular, look at the following patch for the
> rgb8 case:
>
> https://lists.freedesktop.org/archives/virglrenderer-devel/2018-April/000364.html
>
> I don't know if this patchset also breaks the webgl fish test, though.
>
> Thanks,
> Alexandros
> _______________________________________________
> virglrenderer-devel mailing list
> virglrenderer-devel at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/virglrenderer-devel