aqa-tests icon indicating copy to clipboard operation
aqa-tests copied to clipboard

Two jfr tests triggered crash with jdk25 on s390x_linux

Open sophia-guo opened this issue 3 months ago • 2 comments

Found during triage here: https://github.com/adoptium/aqa-tests/issues/6555#issuecomment-3263915085

jdk/jfr/jvm/TestChunkIntegrity.java , jdk/jfr/jvm/TestJFRIntrinsic.java - jvm crashed - no test changes, compared with jdk24, @key jfr is removed (in January, https://github.com/openjdk/jdk25u/commit/99002e4f9d421d08d912187a1f01809d85820427), which shouldn't affect test itself. Also, @required vm.flagless added, then removed (in January, https://github.com/openjdk/jdk25u/commit/8f7e6e2dbc0a3ccf56242bf071e57bfd671de951).

13:47:59  # A fatal error has been detected by the Java Runtime Environment:
13:47:59  #
13:47:59  #  SIGSEGV (0xb) at pc=0x000003ffa84e78de, pid=17570, tid=17587
13:47:59  #
13:47:59  # JRE version: OpenJDK Runtime Environment Temurin-25+36 (25.0+36) (build 25+36-LTS)
13:47:59  # Java VM: OpenJDK 64-Bit Server VM Temurin-25+36 (25+36-LTS, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, g1 gc, linux-s390x)
13:47:59  # Problematic frame:
13:47:59  # V  [libjvm.so+0x9678de]  LibraryCallKit::inline_native_getEventWriter()+0x126e
13:47:59  #
13:47:59  # No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
13:47:59  #
13:47:59  Unsupported internal testing APIs have been used.
13:47:59  
13:47:59  # An error report file with more information is saved as:
13:47:59  # /home/jenkins/workspace/Grinder/aqa-tests/TKG/output_175726618265/jdk_custom_0/work/scratch/hs_err_pid17570.log
13:47:59  #
13:47:59  # Compiler replay data is saved as:
13:47:59  # /home/jenkins/workspace/Grinder/aqa-tests/TKG/output_175726618265/jdk_custom_0/work/scratch/replay_pid17570.log
13:47:59  #
13:47:59  # If you would like to submit a bug report, please visit:
13:47:59  #   https://github.com/adoptium/adoptium-support/issues
13:47:59  # A fatal error has been detected by the Java Runtime Environment:
13:47:59  #
13:47:59  #  SIGSEGV (0xb) at pc=0x000003ff8a8d29c4, pid=16818, tid=16835
13:47:59  #
13:47:59  # JRE version: OpenJDK Runtime Environment Temurin-25+36 (25.0+36) (build 25+36-LTS)
13:47:59  # Java VM: OpenJDK 64-Bit Server VM Temurin-25+36 (25+36-LTS, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, g1 gc, linux-s390x)
13:47:59  # Problematic frame:
13:47:59  # V  [libjvm.so+0x9529c4]  LibraryCallKit::inline_native_jvm_commit()+0x108c
13:47:59  #
13:47:59  # No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
13:47:59  #
13:47:59  # JFR recording file will be written. Location: /home/jenkins/workspace/Grinder/aqa-tests/TKG/output_175726618265/jdk_custom_0/work/scratch/hs_err_pid16818.jfr
13:47:59  #
13:47:59  # An error report file with more information is saved as:
13:47:59  # /home/jenkins/workspace/Grinder/aqa-tests/TKG/output_175726618265/jdk_custom_0/work/scratch/hs_err_pid16818.log
13:47:59  [thread 16846 also had an error]

According to the triage notes, these 2 tests passed using the March 2025 JDK 24 release.

sophia-guo avatar Sep 15 '25 16:09 sophia-guo

Rerun these two testcases to grab log files and attach to issue: https://ci.adoptium.net/job/Grinder/14503/

TestChunkIntegrity - hs_err_pid3820559.log, hs_err_pid3820559.jfr.txt

TestJFRIntrinsic - hs_err_pid3820939.log

smlambert avatar Sep 17 '25 17:09 smlambert

It doesn't seem that there is any available s390x openjdk25 available from other vendors.

sophia-guo avatar Sep 18 '25 15:09 sophia-guo