RFR: 8135018: AARCH64: Missing memory barriers for CMS collector

Andrew Haley aph at redhat.com
Fri Sep 4 14:37:18 UTC 2015

On 09/03/2015 06:52 PM, Dean Long wrote:

> In LIRGenerator::do_StoreIndexed, it looks like the barrier is being
> done twice, first in store_check, and again in post_barrier. Doing
> it in post_barrier makes more sense to me than doing it in
> store_check.

I guess I don't really see that.  As far as I can see store_check
doesn't emit any barrier.  I've traced through it to try to understand
what you mean.


More information about the hotspot-compiler-dev mailing list