[11u]: 8255351: Add detection for Graviton 2 CPUs

[Hohensee, Paul]

Thanks, Andrew. I've posted review requests for 8256488 and 8257436 backports.

Paul

-----Original Message-----
From: Andrew Haley <aph at redhat.com>
Date: Friday, December 4, 2020 at 6:29 AM
To: Severin Gehwolf <sgehwolf at redhat.com>, "Hohensee, Paul" <hohensee at amazon.com>, Volker Simonis <volker.simonis at gmail.com>
Cc: Vladimir Kozlov <vladimir.kozlov at oracle.com>, "jdk-updates-dev at openjdk.java.net" <jdk-updates-dev at openjdk.java.net>
Subject: RE: [11u]: 8255351: Add detection for Graviton 2 CPUs

On 04/12/2020 13:01, Severin Gehwolf wrote:
> On Thu, 2020-12-03 at 21:32 +0000, Hohensee, Paul wrote:
>> In that case: maintainers, what's your sense of whether 8256488 and
>> 8257436 backports would be approved? The two patches apply cleanly
>> net of copyright date and context.
>
> Andrew Haley, thoughts? My thinking is that these are performance
> improvements, so not necessarily candidates for 11u backport. On the
> other hand, if they are low-risk I wouldn't object. They're aarch64-
> only. What do you think?

My first thought is I wish people would stop fiddling with memory copy.
People have been committing improvements for one microarchitecture that
are regressions on another.

But under the circumstances, given that the SIMD copy is a regression, I'd
accept it. But I'd like to see a moratorium on any further backports of
such things.

--
Andrew Haley  (he/him)
Java Platform Lead Engineer
Red Hat UK Ltd. <https://www.redhat.com>
https://keybase.io/andrewhaley
EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671