[Bug 107223] [GEN9+] 50% perf drop in SynMark Fill* tests (E2E RBC gets disabled?)

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=107223

--- Comment #21 from Eero Tamminen <eero.t.tamminen at intel.com> ---
(In reply to Eero Tamminen from comment #20)
> FYI: this series seems to have fixed also earlier perf regression (from
> around May) that was visible on GEN6-GEN8 in (unity/compiz) composited
> windowed 3D benchmarks, but only with modifier enabled gfxstack.

Btw. while it improved a lot of onscreen tests on older GENs too, it dropped
perf on older GENs in SynMark ZBuffer & VSInstancing tests which draw a large
amount of geometry on top of each other.  Apparently caching changed so that
(depth?) reads are now more expensive on those older (6-8) GENs?

-- 
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/20181004/1aedf828/attachment.html>