Add back ARMv7 NEON optimization for TransformationMatrix::multiply
Reported by
ossy.sze...@gmail.com,
Nov 29 2016
|
|||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 5.1; rv:50.0) Gecko/20100101 Firefox/50.0 Steps to reproduce the problem: https://codereview.chromium.org/1666613002 removed the ARMv7 optimization, because it was guarded with CPU(APPLE_ARMV7S) as a leftover from WebKit era. But this optimization isn't Apple specific at all, we only need an ARMv7 hardware with NEON support. What is the expected behavior? What went wrong? Removing the optimization made multiply slower on ARMv7. Did this work before? N/A Chrome version: Channel: n/a OS Version: Flash Version:
,
Nov 29 2016
Tagging with current canary milestone.Please change accordingly if needed.
,
Nov 30 2016
,
Dec 1 2016
,
Dec 1 2016
,
Dec 2 2016
Marking this as Untriaged as CL is under review.
,
Jun 13 2017
,
Mar 8 2018
,
Mar 9 2018
ossy.szeged@: is this still applicable?
,
Mar 9 2018
No idea, I'm not working actively on Chromium for a long time. Feel free to pick it up, if somebody else is interested in it. |
|||||||||
►
Sign in to add a comment |
|||||||||
Comment 1 by ossy.sze...@gmail.com
, Nov 29 2016