[aarch64-port-dev ] RFR 8135018: AARCH64: Missing memory barriers for CMS collector

[Doerr, Martin]

Hi,

I guess the change would apply cleanly on top of
http://hg.openjdk.java.net/jdk9/jdk9/hotspot/rev/4a179f93d775

If it's easy to backport (I haven't tried) doing that one first should be the preferred way.

But if that change is not wanted in 8u, I'll be fine with Ix's version.
Having the UseCondCardMark code only at some places wouldn't break anything. There's no consistency requirement.

Best regards,
Martin


> -----Original Message-----
> From: Andrew Haley <aph at redhat.com>
> Sent: Freitag, 13. Dezember 2019 10:44
> To: Liu Xin <navy.xliu at gmail.com>
> Cc: Doerr, Martin <martin.doerr at sap.com>; aarch64-port-
> dev at openjdk.java.net
> Subject: Re: RFR 8135018: AARCH64: Missing memory barriers for CMS
> collector
> 
> On 12/13/19 8:48 AM, Liu Xin wrote:
> > May I ask to review this backport?
> >
> > bugs: https://bugs.openjdk.java.net/browse/JDK-8135018
> > webrev: https://cr.openjdk.java.net/~xliu/8135018/00/webrev/
> >
> > I can't cleanly apply the original patch, but it isn't hard to adapt it.
> > This patch also brings UseCondCardMark to C1 and template interpreter,
> > which is consistent with C2.
> >
> > For validation part, I passed hotspot:tier1 test and jcstress with
> > UseConcMarkSweepGC on an aarch64 machine.
> 
> That seems like a lot more than simply adding memory barriers. Looks like
> the UseCondCardMark code is missing altogether.
> 
> I can't remember the history of this; I'm concerned that there may be other
> missing hunks.
> 
> --
> 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