RFR (XS) 8223450: Disable Shenandoah C2 barriers verification for x86_32
Aleksey Shipilev
shade at redhat.com
Tue May 7 14:12:59 UTC 2019
Thanks, pushed under triviality rules.
-Aleksey
On 5/7/19 12:34 PM, Roman Kennke wrote:
> Looks good, thanks!
>
> Roman
>
>
>> Bug:
>> https://bugs.openjdk.java.net/browse/JDK-8223450
>>
>> Shenandoah C2 barrier verification is disabled when unusual barrier configuration is requested.
>> However, that only takes care of options provided from the command line (as our tests assert). For
>> the configuration that disables barriers implicitly, e.g. x86_32, this is not enough, and tests fail
>> with false negatives. We need to disable C2 barriers verification explicitly there.
>>
>> Fix:
>>
>> diff -r 1c3292907e4b src/hotspot/share/gc/shenandoah/shenandoahArguments.cpp
>> --- a/src/hotspot/share/gc/shenandoah/shenandoahArguments.cpp Tue May 07 12:19:28 2019 +0200
>> +++ b/src/hotspot/share/gc/shenandoah/shenandoahArguments.cpp Tue May 07 12:23:15 2019 +0200
>> @@ -48,10 +48,12 @@
>> FLAG_SET_DEFAULT(ShenandoahLoadRefBarrier, false);
>> FLAG_SET_DEFAULT(ShenandoahKeepAliveBarrier, false);
>> FLAG_SET_DEFAULT(ShenandoahStoreValEnqueueBarrier, false);
>> FLAG_SET_DEFAULT(ShenandoahCASBarrier, false);
>> FLAG_SET_DEFAULT(ShenandoahCloneBarrier, false);
>> +
>> + FLAG_SET_DEFAULT(ShenandoahVerifyOptoBarriers, false);
>> #endif
>>
>> #ifdef _LP64
>> // The optimized ObjArrayChunkedTask takes some bits away from the full 64 addressable
>> // bits, fail if we ever attempt to address more than we can. Only valid on 64bit.
>>
>> Testing: hotspot_gc_shenandoah
>>
--
Thanks,
-Aleksey
Red Hat GmbH, http://www.de.redhat.com/, Registered seat: Grasbrunn,
Commercial register: Amtsgericht Muenchen, HRB 153243,
Managing Directors: Charles Cachera, Michael O'Neill, Tom Savage, Eric Shander
More information about the shenandoah-dev
mailing list