Mon, 15 Feb 2016 18:42:45 +0000 Merge
vlivanov [Mon, 15 Feb 2016 18:42:45 +0000] rev 36076
Merge
Mon, 15 Feb 2016 20:26:50 +0300 8148994: Replacing MH::invokeBasic with a direct call breaks LF customization
vlivanov [Mon, 15 Feb 2016 20:26:50 +0300] rev 36075
8148994: Replacing MH::invokeBasic with a direct call breaks LF customization Reviewed-by: jrose, redestad
Mon, 15 Feb 2016 20:26:02 +0300 8138922: StubCodeDesc constructor publishes partially-constructed objects on StubCodeDesc::_list
vlivanov [Mon, 15 Feb 2016 20:26:02 +0300] rev 36074
8138922: StubCodeDesc constructor publishes partially-constructed objects on StubCodeDesc::_list Reviewed-by: kvn, coleenp, dholmes
Mon, 15 Feb 2016 20:02:51 +0300 8149141: Optimized build is broken
vlivanov [Mon, 15 Feb 2016 20:02:51 +0300] rev 36073
8149141: Optimized build is broken Reviewed-by: kvn, kbarrett
Mon, 15 Feb 2016 15:15:56 +0100 8149797: Compilation fails with "assert(in_hash) failed: node should be in igvn hash table"
roland [Mon, 15 Feb 2016 15:15:56 +0100] rev 36072
8149797: Compilation fails with "assert(in_hash) failed: node should be in igvn hash table" Summary: node replaced by dominating dead cast during parsing Reviewed-by: kvn
Mon, 15 Feb 2016 13:26:44 +0100 Merge
kshefov [Mon, 15 Feb 2016 13:26:44 +0100] rev 36071
Merge
Mon, 15 Feb 2016 14:32:49 +0300 8149740: NPEs when executing some HotSpotConstantReflectionProvider with null args
kshefov [Mon, 15 Feb 2016 14:32:49 +0300] rev 36070
8149740: NPEs when executing some HotSpotConstantReflectionProvider with null args Reviewed-by: twisti, dnsimon
Mon, 15 Feb 2016 14:31:21 +0300 8149472: NPE when executing HotSpotConstantReflectionProvider::constantEquals with null first arg
kshefov [Mon, 15 Feb 2016 14:31:21 +0300] rev 36069
8149472: NPE when executing HotSpotConstantReflectionProvider::constantEquals with null first arg Reviewed-by: twisti, kvn, dnsimon
Mon, 15 Feb 2016 11:52:51 +0100 Merge
thartmann [Mon, 15 Feb 2016 11:52:51 +0100] rev 36068
Merge
Thu, 11 Feb 2016 12:42:12 +0100 8149543: range check CastII nodes should not be split through Phi
roland [Thu, 11 Feb 2016 12:42:12 +0100] rev 36067
8149543: range check CastII nodes should not be split through Phi Summary: splitting range check CastIIs through loop induction Phi prevents further optimizations Reviewed-by: kvn, mcberg
(0) -30000 -10000 -3000 -1000 -300 -100 -10 +10 +100 +300 +1000 +3000 +10000 tip