[Bug 109216] 4-27% performance drop in Vulkan benchmarks

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Mon Jan 7 15:43:29 UTC 2019


--- Comment #3 from Lionel Landwerlin <lionel.g.landwerlin at linux.intel.com> ---
(In reply to Eero Tamminen from comment #2)
> (In reply to Lionel Landwerlin from comment #1)
> > Ian: I wonder if that could be a result of this series :
> > https://gitlab.freedesktop.org/mesa/mesa/merge_requests/9
> This series doesn't seem Vulkan specific.
> There was no drop in GL or GLES version of Aztec Ruins, only in Vulkan
> version of AztecRuins (*).  And no drop in other GL benchmarks, so I think
> the regression is Vulkan specific.
> (*) While Vulkan version runs windowed (because GfxBench doesn't seem to
> support fullscreen for Vulkan), and GL & GLES versions are fullscreen, I
> don't think that test being windowed is relevant because:
> - the other regressing Vulkan tests are fullscreen
> - AztecRuins has low FPS
> Were you able to reproduce e.g. Deferred Multisampling Vulkan demo drop?

Yep, bisecting now just to verify.

You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-3d-bugs/attachments/20190107/699cb2f7/attachment.html>

More information about the intel-3d-bugs mailing list