Details
-
Type:
Bug
-
Status:
Closed
-
Priority:
Blocker
-
Resolution: Fixed
-
Affects Version/s: None
-
Fix Version/s: 2.9.3
-
Component/s: Compiler: Optimizing, MMTk
-
Labels:None
-
Number of attachments :
Description
I think since ~r13904 there is a repeatable failure when run the complete SPEC JVM is run with -m20 -M20 as shown here:
The change was an opt compiler change (to remove unreachable code) but it seems to have exposed a GC bug.
More recent results (r13907) without expression folding suggest that this is a bug somewhere outside of expression folding:
http://jikesrvm.anu.edu.au/cattrack/results/rvmx86lnx32b.anu.edu.au/commit/2040/production/Performance/SPECjvm98/SPECjvm98/default/Output.txt