RFR(XS): 8027622: java.time.Instant.create failing since hs25-b56

Rickard Bäckman rickard.backman at oracle.com
Mon Nov 4 01:52:37 PST 2013


Hi,

can I please have some reviews for this change?
If we have a control edge set on Cmp nodes we mess up in matching. The
MathExact nodes forced all nodes (except Phi) to have a control edge to
the overflow check. This change excludes Cmp nodes as well.

Webrev: http://cr.openjdk.java.net/~rbackman/8027622/
Bug: https://bugs.openjdk.java.net/browse/JDK-8027622

Thanks
/R


More information about the hotspot-compiler-dev mailing list