Summary: | commit 'i965: Fix BRW_WM_MAX_INSN to reflect current limits.' causes performance regressions | ||
---|---|---|---|
Product: | Mesa | Reporter: | maximlevitsky |
Component: | Drivers/DRI/i965 | Assignee: | Eric Anholt <eric> |
Status: | RESOLVED FIXED | QA Contact: | |
Severity: | major | ||
Priority: | high | CC: | eric, shuang.he |
Version: | git | Keywords: | regression |
Hardware: | Other | ||
OS: | All | ||
Whiteboard: | |||
i915 platform: | i915 features: |
Description
maximlevitsky
2009-11-07 11:59:07 UTC
(moving to mesa category) This seems similar to bug#24847 Thanks for the bisect -- I had a hard time believing it caused a regression, but sure enough: commit c5413839b3e99c7b162f1260142f3c175502b0ce Author: Eric Anholt <eric@anholt.net> Date: Tue Nov 10 15:51:29 2009 -0800 i965: avoid memsetting all the BRW_WM_MAX_INSN arrays for every compile. For an app that's blowing out the state cache, like sauerbraten, the memset of the giant arrays ended up taking 11% of the CPU even when only a "few" of the entries got used. With this, the WM program compile drops back down to 1% of CPU time. Bug #24981 (bisected to BRW_WM_MAX_INSN increase). Of course, the real problem here is that sauer's blowing out the state cache. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.