The open bug URL is:
http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=8004066
Sorry for the noise.
- Kris
On 12/01/2012 02:56 AM, Krystal Mo wrote:
Forgot to mention: the two runs of the test in the last mail were
examples of test failures before the fix. They would run fine after
the fix.
- Kris
On 12/01/2012 02:48 AM, Krystal Mo wrote:
Hi all,
Could I get a couple of review for this small change, please?
And could someone from the JDK side sponsor this change?
Bug: https://jbs.oracle.com/bugs/browse/JDK-8004066
Webrev: http://cr.openjdk.java.net/~kmo/8004066/webrev.00/
The DivModTest introduced in JDK-6282196 [1] checks the contents of
the exception message, but that message isn't specified in JavaDoc
and thus may vary between VM implementations (or even in the same VM).
The issue has affected HotSpot Server VM in nightlies, and the Shark
VM. As OpenJDK library code may receive broader adoption in the
future, this issue may affect other VM implementations in the future
as well.
(Details:
The HotSpot Server Compiler may recompile hot throw sites to throw a
preallocated exception object, with null exception message, leading
to a NPE in the test;
The bytecode interpreter in Zero/Shark VM throws the
ArithmeticException with "/ by long zero" for ldiv, which is
different from normal HotSpot behavior (which is expected by the
test) where the exception message is "/ by zero".)
This change relaxes the test so that it's more lenient and less
sensitive to the error message produced by the VM.
I don't think there's a good/reliable way of verifying whether an
ArithmeticException came from "division by zero", checking the type
should be enough for now.
Tested with the failing nightly test case and jprt.
$ java -Xcomp -showversion -XX:DefaultMaxRAMFraction=8
-XX:-TieredCompilation -XX:CompileThreshold=100
-XX:+UnlockExperimentalVMOptions -XX:+IgnoreUnrecognizedVMOptions
-XX:+AggressiveOpts -XX:+IgnoreUnrecognizedVMOptions
-XX:-UseCompressedOops DivModTests
java version "1.8.0-ea"
Java(TM) SE Runtime Environment (build 1.8.0-ea-b65)
Java HotSpot(TM) 64-Bit Server VM (build 25.0-b11-internal-fastdebug,
compiled mode)
Exception in thread "main" java.lang.NullPointerException
at DivModTests.resultEquals(DivModTests.java:390)
at DivModTests.testIntFloorMod(DivModTests.java:126)
at DivModTests.testIntFloorDivMod(DivModTests.java:103)
at DivModTests.testIntFloorDivMod(DivModTests.java:68)
at DivModTests.main(DivModTests.java:45)
Description from the bug report:
~/jdk/test/java/lang/Math$ java -Xint -showversion DivModTests
java version "1.8.0-ea"
Java(TM) SE Runtime Environment (build 1.8.0-ea-b65)
OpenJDK 64-Bit Shark VM (build 25.0-b11-internal, interpreted mode)
FAIL: long Math.floorDiv(4, 0) = java.lang.ArithmeticException: / by
long zero; expected java.lang.ArithmeticException: / by zero
FAIL: long StrictMath.floorDiv(4, 0) = java.lang.ArithmeticException:
/ by long zero; expected java.lang.ArithmeticException: / by zero
FAIL: long Math.floorMod(4, 0) = java.lang.ArithmeticException: / by
long zero; expected java.lang.ArithmeticException: / by zero
FAIL: long StrictMath.floorMod(4, 0) = java.lang.ArithmeticException:
/ by long zero; expected java.lang.ArithmeticException: / by zero
Exception in thread "main" java.lang.RuntimeException: 4 errors found
in DivMod methods.
at DivModTests.main(DivModTests.java:49)
Regards,
Kris
[1]: https://jbs.oracle.com/bugs/browse/JDK-6282196
