All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=820469

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=1d03f71fd7500a6c9c9e2cce427ad96483b168219257fdfd9c28233e63b6bb74


Bot(s) for this bug's original alert(s):

chromium-rel-win8-dual

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14b1e0ba440000

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/14b1e0ba440000

make SkJumper stages normal Skia code by mtklein@chromium.org
https://skia.googlesource.com/skia/+/22e536e3a1a09405d1c0e6f071717a726d86e8d4

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

The following revision refers to this bug:
  https://skia.googlesource.com/skia/+/a46623b2f3bbf09822537c63a1d64f2e245c6ff0

commit a46623b2f3bbf09822537c63a1d64f2e245c6ff0
Author: Mike Klein <mtklein@chromium.org>
Date: Sat Mar 10 16:07:18 2018

follow JUMPER_NARROW_STAGES in lowp stages too

Should give dramatically better codegen for all 32-bit builds and 64-bit
Windows builds, bringing it in line with how we make highp float stages.

May help this bug, which is mostly Windows perf regressions.

Bug:  chromium:820469 

Change-Id: I223f7568a09dea28ec614b18555766ea7d8365fa
Reviewed-on: https://skia-review.googlesource.com/113665
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Mike Klein <mtklein@chromium.org>

[modify] https://crrev.com/a46623b2f3bbf09822537c63a1d64f2e245c6ff0/src/opts/SkRasterPipeline_opts.h

Looks like that CL helped most of the charts recover, particularly the very worst regressed ones.