Issue metadata
Sign in to add a comment
|
2.2% regression in system_health.common_desktop at 543193:543293 |
||||||||||||||||||||
Issue descriptionSmall regression, just one one platform/metric/page. Will close if bisect can't repro.
,
Mar 25 2018
📍 Pinpoint job started. https://pinpoint-dot-chromeperf.appspot.com/job/12e7ab53440000
,
Mar 25 2018
📍 Found a significant difference after 1 commit. https://pinpoint-dot-chromeperf.appspot.com/job/12e7ab53440000 Allow HSW stages to detect FMAs opportunistically. by mtklein@chromium.org https://chromium.googlesource.com/chromium/src/+/670c468885c8ca351af506db4f44d07b2f4e2acb Understanding performance regressions: http://g.co/ChromePerformanceRegressions
,
Mar 27 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/811d1a139265b7d8c94fc1690184de2351d9230f commit 811d1a139265b7d8c94fc1690184de2351d9230f Author: Mike Klein <mtklein@chromium.org> Date: Tue Mar 27 00:23:29 2018 Revert "Allow HSW stages to detect FMAs opportunistically." This reverts commit 670c468885c8ca351af506db4f44d07b2f4e2acb. Reason for revert: Mysteriously, this has caused some regressions on a bunch of bots that do not support the Haswell instruction set. Reverting just to see if that's real (and because this wasn't really a noticeable win anywhere when landing the first time). Bug: 825647 , 825586 , 825646 , 825104 Original change's description: > Allow HSW stages to detect FMAs opportunistically. > > We used to build the .S files with -ffp-contract=fast for this, and > adding it here to the HSW opts will have the equivalent effect. None of > the other x86 targets support FMAs. > > In many key places we do call _mm256_fmadd_ps explicitly, but in several > other stages we have been relying on Clang to sniff out opportunities, > especially for fused-multiply-subtracts, which we're just not very good > at seeing in our heads. > > Bug: 821012 > Change-Id: I2767cbdb7151d60ea7553e83f51b28ff9340eb94 > Reviewed-on: https://chromium-review.googlesource.com/963141 > Reviewed-by: Florin Malita <fmalita@chromium.org> > Commit-Queue: Mike Klein <mtklein@chromium.org> > Cr-Commit-Position: refs/heads/master@{#543262} TBR=fmalita@chromium.org,herb@chromium.org,mtklein@chromium.org # Not skipping CQ checks because original CL landed > 1 day ago. Bug: 821012 Change-Id: I61c0627270853b8acdcf6b20fca17ba0fd8bba55 Reviewed-on: https://chromium-review.googlesource.com/980293 Commit-Queue: Mike Klein <mtklein@chromium.org> Reviewed-by: Mike Klein <mtklein@chromium.org> Cr-Commit-Position: refs/heads/master@{#545897} [modify] https://crrev.com/811d1a139265b7d8c94fc1690184de2351d9230f/skia/BUILD.gn
,
Mar 28 2018
|
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by 42576172...@developer.gserviceaccount.com
, Mar 25 2018