RFR: 8267926: AsyncLogGtest.java fails on assert with: decorator was not part of the decorator set specified at creation. [v3]

[Liu Xin]

hi, David,

This is awkward. I think I changed too many in a patch.

Do you think it is a good idea that I revert to
https://openjdk.github.io/cr/?repo=jdk&pr=4257&range=00
It is a trivial bugfix.  Does it solve tier1~4 on aarch64?

If it does, we can know that this new crash is caused by the change in my
new patch.
I would recall it and rework it on aarch64.

It seems that it crashed while executing gtest in async mode.   Could you
tell me which gtest it was executing?
"Command Line: -XX:+ExecutingUnitTests -Xlog:async"

thanks,
--lx


On Mon, May 31, 2021 at 10:13 PM David Holmes <dholmes at openjdk.java.net>
wrote:

> On Mon, 31 May 2021 17:09:48 GMT, Xin Liu <xliu at openjdk.org> wrote:
>
> >> The root cause of the intermittent failure is that _decorators in
> LogDecorations
> >> may be inconsistent with LogOutput::_decorators. It could happen when
> gtest disables a
> >> Log output via set_log_config(TestLogFileName, "all=off").
> >>
> >> Since we copy the entire logDecorations, it's reasonable to copy
> _decorator as well.
> >> LogDecorator is a bitmask of uint. It's even smaller than a reference
> on LP64 platforms.
> >>
> >> This patch not only fixes intermittent crash in JDK-8267926, but also
> fixes JDK-8267952.
> >> Two hazards of async logging in `LogConfiguration::configure_output`
> are eliminated.
> >> This patch also adds a storestore barrier to make sure it's safe on
> weak consistency machines.
> >
> > Xin Liu has updated the pull request incrementally with one additional
> commit since the last revision:
> >
> >   Increase reader's counter before creating LogDecoration.
>
> I put the latest patch through our tier 1-4 testing and it crashed in tier
> 3 on Linux-Aarch64:
>
> #
> # A fatal error has been detected by the Java Runtime Environment:
> #
> #  SIGSEGV (0xb) at pc=0x0000ffff40820720, pid=3972473, tid=3972481
> #
> # JRE version: Java(TM) SE Runtime Environment (17.0) (fastdebug build
> 17-internal+0-LTS-2021-05-31-2211099.david.holmes.jdk-dev4.git)
> # Java VM: Java HotSpot(TM) 64-Bit Server VM (fastdebug
> 17-internal+0-LTS-2021-05-31-2211099.david.holmes.jdk-dev4.git, mixed mode,
> tiered, compressed oops, compressed class ptrs, g1 gc, linux-aarch64)
> # Problematic frame:
> # C  [libc.so.6+0x60720]  flockfile+0x0
> #
> # Core dump will be written. Default location: Core dumps may be processed
> with "/opt/core.sh %p" (or dumping to
> /opt/mach5/mesos/work_dir/slaves/a4f8fba9-f017-4328-b286-c66b6a97143d-S808/frameworks/1735e8a2-a1db-478c-8104-60c8b0af87dd-0196/executors/42993029-a18a-48c4-9a4d-108aee7b3811/runs/a38508fe-5d43-48ad-b70a-61f3e712fcd8/testoutput/test-support/jtreg_open_test_hotspot_jtreg_hotspot_misc/scratch/0/core.3972473)
> #
> # If you would like to submit a bug report, please visit:
> #   https://bugreport.java.com/bugreport/crash.jsp
> #
>
> ---------------  S U M M A R Y ------------
>
> Command Line: -XX:+ExecutingUnitTests -Xlog:async
>
> Host: AArch64, 6 cores, 46G, Oracle Linux Server release 8.3
> Time: Mon May 31 22:37:04 2021 UTC elapsed time: 3.758723 seconds (0d 0h
> 0m 3s)
>
> ---------------  T H R E A D  ---------------
>
> Current thread (0x0000aaadf32b1e10):  Thread "AsyncLog Thread" [stack:
> 0x0000fffefdaa0000,0x0000fffefdca0000] [id=3972481]
>
> Stack: [0x0000fffefdaa0000,0x0000fffefdca0000],  sp=0x0000fffefdc9e510,
> free space=2041k
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native
> code)
> C  [libc.so.6+0x60720]  flockfile+0x0
> V  [libjvm.so+0x1508058]  LogFileOutput::write_blocking(LogDecorations
> const&, char const*)+0x38
> V  [libjvm.so+0x1502764]  AsyncLogWriter::run()+0x2c0
> V  [libjvm.so+0x1a49d08]  Thread::call_run()+0xf8
> V  [libjvm.so+0x1766f18]  thread_native_entry(Thread*)+0x108
> C  [libpthread.so.0+0x76e0]  start_thread+0x198
>
> -------------
>
> PR: https://git.openjdk.java.net/jdk/pull/4257
>