Loading...

XML

Word

Printable

Type: Enhancement
Resolution: Fixed
Priority: P3
Fix Version/s: 10
Affects Version/s: 10
Component/s: hotspot
Labels:
- safepoint

Subcomponent:
runtime
Resolved In Build:
b21
CPU:

generic
OS:

generic

Some operations in safepoint cleanup have been observed to (sometimes) take significant time. Most notably, idle monitor deflation and nmethod marking stick out in some popular applications and benchmarks.

I propose to:
- parallelize safepoint cleanup processing
- enable to hook up idle monitor deflation and nmethod marking to GC VM ops, if GC can support it (resulting in even more efficient deflation/nmethod marking)

relates to

JDK-8132849 Increased stop time in cleanup phase because of single-threaded walk of thread stacks in NMethodSweeper::mark_active_nmethods()

Resolved

JDK-8184751 Provide thread pool for parallel safepoint cleanup

Closed

JDK-8186837 Memory ordering nmethod, _state and _stack_traversal_mark

Resolved

JDK-8185273 Test8004741.java crashes with SIGSEGV in JDK10-hs nightly

Closed

JDK-8185102 TestSAServer.java fails due to "sun.jvm.hotspot.types.WrongTypeException: field "_stack_traversal_mark"

Closed

JDK-8153224 Monitor deflation prolong safepoints

Resolved

JDK-8185580 Refactor Threads::possibly_parallel_oops_do() to use Threads::parallel_java_threads_do()

Resolved

(2 relates to)

Assignee:: Roman Kennke

Reporter:: Roman Kennke

Votes:: 0 Vote for this issue

Watchers:: 3 Start watching this issue

Created:: 2017-05-24 03:18

Updated:: 2019-05-22 13:11

Resolved:: 2017-07-21 11:25

Details

Description

Attachments

Issue Links

Activity

People

Dates