RFR: JDK-8311261: [AIX] TestAlwaysPreTouchStacks.java fails due to java.lang.RuntimeException: Did not find expected NMT output
Varada M
duke at openjdk.org
Fri Jul 21 11:52:42 UTC 2023
On Tue, 18 Jul 2023 17:07:09 GMT, Varada M <duke at openjdk.org> wrote:
> On AIX we are getting the pattern “Stack: “, but the test is expecting the pattern : ".*stack: reserved=(\\d+), committed=(\\d+).*”. This is due to track_as_vm() is returning false on AIX. The function should return false because on AIX by [default page is not aligned](https://github.com/openjdk/jdk/blob/354c6605e32790ca421869636d8bf5456fc51717/src/hotspot/share/services/threadStackTracker.hpp#L57). Thus disabling the test on AIX.
>
> JBS Issue : [JDK-8311261](https://bugs.openjdk.org/browse/JDK-8311261)
> [1.083s][info][os,thread] Thread "Unknown thread" started (pthread id: 5143, attributes: stacksize: 2112k, guardsize: 0k, detached).
> [1.084s][info][os,thread] Thread is alive (tid: 5143, kernel thread id: 81528083, stack [0x0000000116a10000 - 0x0000000116c1d888 (2102k using 64k pages)).
> [1.291s][info][os,thread] Thread "Unknown thread" started (pthread id: 5656, attributes: stacksize: 4096k, guardsize: 0k, detached).
> [1.291s][info][os,thread] Thread is alive (tid: 5656, kernel thread id: 84280773, stack [0x0000000116c30000 - 0x000000011702d888 (4086k using 64k pages)).
@tstuefe It is already 64K aligned and the test is failing with `-XX:-Use64KPages`
-------------
PR Comment: https://git.openjdk.org/jdk/pull/14923#issuecomment-1645456919
More information about the hotspot-runtime-dev
mailing list