[Bug 111090] 3% perf drop in GfxBench Manhattan 3.0, 3.1 and CarChase test-cases

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Fri Oct 4 20:09:49 UTC 2019


https://bugs.freedesktop.org/show_bug.cgi?id=111090

Chris Wilson <chris at chris-wilson.co.uk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |WORKSFORME
             Status|NEW                         |RESOLVED

--- Comment #19 from Chris Wilson <chris at chris-wilson.co.uk> ---
The final improvement we are likely to make for the time being is

commit 2850748ef8763ab46958e43a4d1c445f29eeb37d
Author: Chris Wilson <chris at chris-wilson.co.uk>
Date:   Fri Oct 4 14:39:58 2019 +0100

    drm/i915: Pull i915_vma_pin under the vm->mutex

which removes the global struct_mutex from inside the shrinker -- so this pair
of benchmarks that appeared to be kicking kswapd should be hitting the mutex
far less often and so removing contention from the gfx workload.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20191004/400ed81a/attachment.html>


More information about the intel-gfx-bugs mailing list