8259231: Epsilon: improve performance under contention during virtual space expansion #1794
Conversation
|
Hi @lhtin, welcome to this OpenJDK project and thanks for contributing! We do not recognize you as Contributor and need to ensure you have signed the Oracle Contributor Agreement (OCA). If you have not signed the OCA, please follow the instructions. Please fill in your GitHub username in the "Username" field of the application. Once you have signed the OCA, please let us know by writing If you already are an OpenJDK Author, Committer or Reviewer, please click here to open a new issue so that we can record that fact. Please use "Add GitHub user lhtin" as summary for the issue. If you are contributing this work on behalf of your employer and your employer has signed the OCA, please let us know by writing |
|
/covered |
|
Thank you! Please allow for a few business days to verify that your employer has signed the OCA. Also, please note that pull requests that are pending an OCA check will not usually be evaluated, so your patience is appreciated! |
bridgekeeper
bot
removed
oca
|
/test |
lhtin
changed the title
EpsilonHeap::allocate_work
|
/issue add JDK-8259231 |
openjdk
bot
changed the title
EpsilonHeap::allocate_work
|
@lhtin The primary solved issue for a PR is set through the PR title. Since the current title does not contain an issue reference, it will now be updated. |
Webrevs
|
| @@ -64,7 +64,7 @@ | |||
| "smaller TLABs until policy catches up.") \ | |||
| \ | |||
| product(bool, EpsilonElasticTLABDecay, true, EXPERIMENTAL, \ | |||
| "Use timed decays to shrik TLAB sizes. This conserves memory " \ | |||
| "Use timed decays to shrink TLAB sizes. This conserves memory " \ | |||
There was a problem hiding this comment.
Let's not do the typo fixes in this PR. Maybe there are other spelling problems elsewhere in gc/epsilon that we could fix wholesale in another PR?
There was a problem hiding this comment.
OK, I will revert the typo fixes change.
| // No space left: | ||
| return NULL; | ||
| } | ||
| { |
There was a problem hiding this comment.
I see what you are trying to do, and it makes sense. I believe this form would be cleaner:
HeapWord* res = NULL;
while (true) {
// Try to allocate, assume space is available
res = par_allocate(size);
if (res != NULL) {
break;
}
MutexLocker ml(Heap_Lock);
// Try to allocate under the lock, assume another thread was able to expand
res = par_allocate(size);
if (res != NULL) {
break;
}
// Expand and loop back if space is available
size_t space_left = max_capacity() - capacity();
size_t want_space = MAX2(size, EpsilonMinHeapExpand);
...
}
There was a problem hiding this comment.
Yes, the new form is cleaner very. And I think it would be a little cleaner if wrap the lock scope in curly braces. like this:
HeapWord* res = NULL;
while (true) {
// Try to allocate, assume space is available
res = par_allocate(size);
if (res != NULL) {
break;
}
{
MutexLocker ml(Heap_Lock);
// Try to allocate under the lock, assume another thread was able to expand
res = par_allocate(size);
if (res != NULL) {
break;
}
// Expand and loop back if space is available
size_t space_left = max_capacity() - capacity();
size_t want_space = MAX2(size, EpsilonMinHeapExpand);
...
}
}
lhtin
changed the title
|
@lhtin This change now passes all automated pre-integration checks. ℹ️ This project also has non-automated pre-integration requirements. Please see the file CONTRIBUTING.md for details. After integration, the commit message for the final commit will be: You can use pull request commands such as /summary, /contributor and /issue to adjust it as needed. At the time when this comment was updated there had been 183 new commits pushed to the
As there are no conflicts, your changes will automatically be rebased on top of these commits when integrating. If you prefer to avoid this automatic rebasing, please check the documentation for the /integrate command for further details. As you do not have Committer status in this project an existing Committer must agree to sponsor your change. Possible candidates are the reviewers of this PR (@shipilev) but any other Committer may sponsor as well. ➡️ To flag this PR as ready for integration with the above commit message, type |
|
Tests of gc/epsilon have passed on {macosx,linux}-x86_64-server-{release,fastdebug,slowdebug}. |
|
/integrate |
|
/sponsor |
openjdk
bot
added
integrated
|
@shipilev @lhtin Since your change was applied there have been 183 commits pushed to the
Your commit was automatically rebased without conflicts. Pushed as commit 722f236. 💡 You may see a message that your pull request was closed with unmerged commits. This can be safely ignored. |
|
Thank you @shipilev for reviewing and sponsoring. |
Hi all,
The
EpsilonHeap::allocate_workmethod maybe can be fixed and improved by this:_space->par_allocatebefore expanding virtual space, when there not enough virtual space but another thread expanding succeeded just and has enough space.res = _space->par_allocate(size);out of lock scope.Test on macosx-x86_64-server-{release, fastdebug, slowdebug} with current test case.
Progress
Issue
Reviewers
Download
$ git fetch https://git.openjdk.java.net/jdk pull/1794/head:pull/1794$ git checkout pull/1794