RFR (S) 8194246: JVM crashes when calling getStackTrace if stack contains a method that is a member of a very large class

coleen.phillimore at oracle.com coleen.phillimore at oracle.com
Sat Jan 27 14:23:03 UTC 2018



On 1/26/18 6:26 PM, mandy chung wrote:
>
>
> On 1/26/18 2:48 PM, coleen.phillimore at oracle.com wrote:
>> Summary: Use unsigned short to save method_id in stack trace.
>>
>> Tested with tier1 tests and test provided.  jck testing in progress.
>>
>> open webrev at http://cr.openjdk.java.net/~coleenp/8194246.01/webrev
>> bug link https://bugs.openjdk.java.net/browse/JDK-8194246
>>
>
> Looks good to me.
>
> Minor comment on the test:
>
>   29  * @modules java.base/jdk.internal.org.objectweb.asm
>   30  *          java.base/jdk.internal.misc
>   31  *          java.management
>
> I thought the test library dependency to java.management was removed 
> at one point.
> You can verify this by adding a vm option `--limit-modules java.base`.
>
>   32  * @compile -XDignore.symbol.file LargeClassTest.java
>
> This can be simplified to #build LargeClassTest since -XDignore.symbol.file
> is not needed.
> Mandy
Thank you for the code review, Mandy.  I had copied this from another 
test but have removed these and it still passes.

Thanks!
Coleen


More information about the hotspot-runtime-dev mailing list