RFR: 8272788: Nonleaf ranked locks should not be safepoint_check_never [v3]

Coleen Phillimore coleenp at openjdk.java.net
Wed Sep 1 18:42:30 UTC 2021


On Wed, 1 Sep 2021 17:52:49 GMT, Coleen Phillimore <coleenp at openjdk.org> wrote:

>> src/hotspot/share/runtime/mutexLocker.cpp line 270:
>> 
>>> 268: 
>>> 269:   def(Threads_lock                 , PaddedMonitor, barrier,     true,  _safepoint_check_always);  // Used for safepoint protocol.
>>> 270:   def(NonJavaThreadsList_lock      , PaddedMutex,   leaf+1,      true,  _safepoint_check_never);
>> 
>> Why do we need to change this rank? We now assert a lock should be _safepoint_check_always if the rank is >= nonleaf, but this is barrier so it should be good.
>
> This was leftover from a different change where I was trying to move all the _safepoint_check_never locks to leaf level.  In any case, I reverted this change and am retesting tier1-3.

Tier1-3 passes by reverting that change.

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

PR: https://git.openjdk.java.net/jdk/pull/5203


More information about the serviceability-dev mailing list