[8u] RFR (S) 8071374: Native disassembler implementation may be not thread-safe

Aleksey Shipilev shade at redhat.com
Fri Apr 23 09:37:55 UTC 2021

Original bug:

This still bites me with 8u JMH runs. The Stress reproducer from the issue also fails without this 
patch and passes with it.

Unfortunately, the patch has a few conflicts when applied to 8u:
   - at times, the context near print_cr-s is different enough for automatic patch to fail;
   - there are no PrintAssemblyOption in 8u (added later by JEP 165);
   - there is no jvmciCompilerToVM.cpp (added later by JEP 243);

8u webrev:

Testing: reproducer from the test (now passes); adhoc JMH perfasm runs (now reliable); hotspot_tier1


More information about the jdk8u-dev mailing list