[Bug 107510] [GEN8+] up to 10% perf drop on several 3D benchmarks

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Wed Aug 15 16:00:00 UTC 2018


--- Comment #4 from Eero Tamminen <eero.t.tamminen at intel.com> ---
Jason has posted a revert for the regressing commit:

"Commit 4434591bf56a6b0 caused substantially more URB messages in
geometry and tessellation shaders.  Before we can really enable this
sort of optimization,  We either need some way of combining them back
together into vectors or we need to do cross-stage vector element
elimination without splitting everything into scalars."

This explains GfxBench tessellation, SynMark GSCloth and potentially also
Unigine Heaven+tessellation regressions.

However, GfxBench ALU2, GpuTest Julia FP64 and SynMark PSPom don't use those
features, so I wonder whether their (small) regressions come from some other
commit... Mark?

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/20180815/3fe0473e/attachment.html>

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