RFR: 8258229: Crash in nmethod::reloc_string_for

Manuel Hässig duke at openjdk.org
Thu Apr 24 07:08:00 UTC 2025


On Wed, 23 Apr 2025 20:15:30 GMT, Manuel Hässig <duke at openjdk.org> wrote:

>> src/hotspot/cpu/x86/nativeInst_x86.cpp line 389:
>> 
>>> 387: void NativeJump::patch_verified_entry(address entry, address verified_entry, address dest) {
>>> 388:   // complete jump instruction (to be inserted) is in code_buffer;
>>> 389:   union {
>> 
>> Do you need this change? Meaning, does it add substantially to this fix? Looks like it does not?
>> 
>> I'd omit it here, so the patch is cleanly backportable. We will cleanup the remnants of x86_32 port in due course.
>
> I don't need it. I'll take it out before the RFR. Thanks for pointing it out!

Should I file an RFE or do you already have everything tracked?

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/24831#discussion_r2056814685


More information about the hotspot-compiler-dev mailing list