SIGBUS on instrumentation on M2 CPUs

[adrianlyjak]

We recently added the pyroscope java agent to the application my company runs. Developers running M2 macs have reported that the instrumentation crashes their JVM (Java 17) with dumps like this

#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGBUS (0xa) at pc=0x0000000147d354e4, pid=18543, tid=215043
#
# JRE version: OpenJDK Runtime Environment Temurin-17.0.7+7 (17.0.7+7) (build 17.0.7+7)
# Java VM: OpenJDK 64-Bit Server VM Temurin-17.0.7+7 (17.0.7+7, mixed mode, tiered, compressed oops, compressed class ptrs, g1 gc, bsd-aarch64)
# Problematic frame:
# v  ~StubRoutines::SafeFetch32
#
# No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /Users/name/blah/blah/hs_err_pid18543.log
Compiled method (n/a)   54372  822     n 0       java.io.UnixFileSystem::getBooleanAttributes0 (native)
 total in heap  [0x00000001482ffb90,0x00000001482ffef8] = 872
 relocation     [0x00000001482ffce8,0x00000001482ffd10] = 40
 main code      [0x00000001482ffd40,0x00000001482ffef0] = 432
 metadata       [0x00000001482ffef0,0x00000001482ffef8] = 8
[thread 266499 also had an error]
Compiled method (n/a)   54374  822     n 0       java.io.UnixFileSystem::getBooleanAttributes0 (native)
 total in heap  [0x00000001482ffb90,0x00000001482ffef8] = 872
 relocation     [0x00000001482ffce8,0x00000001482ffd10] = 40
 main code      [0x00000001482ffd40,0x00000001482ffef0] = 432
 metadata       [0x00000001482ffef0,0x00000001482ffef8] = 8
#
# If you would like to submit a bug report, please visit:
#   https://github.com/adoptium/adoptium-support/issues
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#

Interestingly, I'm running an M1 mac, and it works fine.

Let me know if there's more information I can provide that would be helpful

[korniltsev]

This looks like a jvm bug. More info here https://github.com/async-profiler/async-profiler/issues/747

It was fixed both in async-profiler (but seems not released since then) and jvm 17.0.9

I suggest you update jvm, it may solve the problem

[adrianlyjak]

Thanks, I'll try it out

[adrianlyjak]

Awesome that fixes the issue. Thanks for determining the cause!