JDK-8010255 : tools/javac/7142086/T7142086.java times out on slower architectures
  • Type: Bug
  • Component: tools
  • Sub-Component: javac
  • Affected Version: 8
  • Priority: P3
  • Status: Closed
  • Resolution: Won't Fix
  • OS: generic
  • CPU: generic
  • Submitted: 2013-03-18
  • Updated: 2013-06-20
  • Resolved: 2013-06-20
Related Reports
Relates :  
JDK-7142086 - performance problem in Check.checkOverrideClashes(...)
Description
T7142086 times out running on a SPARC machine:

--------------------------------------------------
TEST: tools/javac/7142086/T7142086.java
JDK under test: (/java/re/jdk/8/latest/binaries/solaris-sparc)
java version "1.8.0-ea"
Java(TM) SE Runtime Environment (build 1.8.0-ea-b81)
Java HotSpot(TM) Client VM (build 25.0-b22, mixed mode)

Timeout signaled after 10 seconds

ACTION: build -- Passed. All files up to date
REASON: Named class compiled on demand
TIME:   0.001 seconds
messages:
command: build T7142086
reason: Named class compiled on demand
elapsed time (seconds): 0.001

ACTION: main -- Error. Program `/java/re/jdk/8/latest/binaries/solaris-sparc/bin/java' interrupted! (timed out?)
REASON: User specified action: run main/timeout=10 T7142086 
TIME:   10.239 seconds
messages:
command: main T7142086
reason: User specified action: run main/timeout=10 T7142086 
elapsed time (seconds): 10.239
STDOUT:
STDERR:

TEST RESULT: Error. Program `/java/re/jdk/8/latest/binaries/solaris-sparc/bin/java' interrupted! (timed out?)
--------------------------------------------------

Is the 10 second timeout for real?
Comments
As jon said, I think the timeout defeats the purpose of the test ie. invoking JDK from NFS, as we all know, several things can skew the result, as the test is measuring critical performance operation. I suggest retrying with local JDK, and if the problem continues to persist please reopen or open a new issue.
20-06-2013
Executing directly from /java/re/ seems a particularly unwise thing to do. Running with a 10 second timeout seems unwise, but if you're running on a slow system (with java bits from a server) you should really be running with the -timefactor set to 4 or more. I recommend closing this Not A Problem; the submitter should reconfigure the test run.
17-06-2013
The sparcs we have are notoriously slow, perhaps increase the timeout for these systems OR simply ignore the test under these cases.
17-06-2013
Should the timeout be adjusted for very slow architectures ?
05-04-2013