Tobi

jenkins test sanity amac jdk19

jenkins compile win jdk19

jenkins test sanity amac jdk19

@JasonFengJ9 Have you seen a recent failure of this. I did two grinders and didnt get any failures: https://openj9-jenkins.osuosl.org/job/Grinder/202/tapResults/ https://openj9-jenkins.osuosl.org/job/Grinder/203/tapResults/

50 job grinder, https://openj9-jenkins.osuosl.org/job/Grinder/220/tapResults/

50 job grinder of conccurent load test, https://openj9-jenkins.osuosl.org/job/Grinder/493/ - 2 failures

Here is the Java stacktrace: ``` *** BEGIN STACK WALK, flags = 00400001 walkThread = 0x000000000031BA00 *** ITERATE_O_SLOTS RECORD_BYTECODE_PC_OFFSET Initial values: walkSP = 0x0000000000451D80, PC = 0x0000FFFF49463173, literals = 0x000000000002F558,...

The crash occurs on an invokeVirtual with `O-Slot: p0[0x0000000000451D80] = 0x000000000000FFFF` receiver

Method and bytecodes ``` > !j9method 0x000000000002F558 J9Method at 0x2f558 { Fields for J9Method: 0x0: U8* bytecodes = !j9x 0x0000FFFF49463158 // "*�L" 0x8: struct J9ConstantPool* constantPool = !j9constantpool 0x000000000002BF90 (flags...

There is a `c_jitDecompileOnReturn` on the thread right before the crash. A new bytecode frame is built a 0x451d88, this would mean that the `java/lang/Class.buildAnnotations` bytecode frame is the frame...