8281322: C2: always construct strip mined loop initially (even if strip mining is disabled) #7364

rwestrel · 2022-02-07T08:53:41Z

Some of the long range check transformations take advantage of the
safepoint captured by loop strip mining to extract jvm state (in order
to add back empty predicates to the inner loop of a loop nest). As a
consequence, irTests/TestLongRangeChecks.java fails with strip mining
off and users might experience performance anomalies where changing
GCs affect purely computational code.

The strip mined loop nest creation is a 2 step process:

1- when a CountedLoop is created, an OuterStripMinedLoop is also added
but it's not fully constructed

2- at macro expansion time, the OuterStripMinedLoop is turned into an
actual loop by adding Phis and a proper exit condition

I propose always doing 1- whether loop strip mining is enabled or
not. This causes the safepoint to always be captured. Loop strip ming
is not expected to get in the way of loop transformations so this
change in itself should be performance neutral. Then at 2-, if loop
strip mining is not enabled, the OuterStripMinedLoop can be removed
and the safepoint moved back into the loop in case
LoopStripMiningIter=1 or simply removed too.

Progress

Change must not contain extraneous whitespace
Commit message must refer to an issue
Change must be properly reviewed

Issue

JDK-8281322: C2: always construct strip mined loop initially (even if strip mining is disabled)

Reviewers

Christian Hagedorn (@chhagedorn - Reviewer)
Tobias Hartmann (@TobiHartmann - Reviewer) ⚠️ Review applies to 62d8d87

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.java.net/jdk pull/7364/head:pull/7364
$ git checkout pull/7364

Update a local copy of the PR:
$ git checkout pull/7364
$ git pull https://git.openjdk.java.net/jdk pull/7364/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 7364

View PR using the GUI difftool:
$ git pr show -t 7364

Using diff file

Download this PR as a diff file:
https://git.openjdk.java.net/jdk/pull/7364.diff

bridgekeeper · 2022-02-07T08:55:18Z

👋 Welcome back roland! A progress list of the required criteria for merging this PR into master will be added to the body of your pull request. There are additional pull request commands available for use with this pull request.

openjdk · 2022-02-07T08:56:51Z

@rwestrel The following label will be automatically applied to this pull request:

hotspot-compiler

When this pull request is ready to be reviewed, an "RFR" email will be sent to the corresponding mailing list. If you would like to change these labels, use the /label pull request command.

mlbridge · 2022-02-07T09:09:30Z

Webrevs

chhagedorn

That looks reasonable. I'll submit some testing.

chhagedorn · 2022-02-08T08:28:37Z

src/hotspot/share/opto/loopnode.cpp

+    igvn->replace_input_of(inner_cl, LoopBackControl, in(LoopBackControl));
+    // make the outer loop go away
+    igvn->replace_input_of(this, LoopBackControl, igvn->C->top());
+    igvn->C->print_method(PHASE_DEBUG, 2);


Probably a left-over and can be removed.

chhagedorn · 2022-02-08T08:51:44Z

src/hotspot/share/opto/loopnode.cpp

+    Node* outer_test = outer_le->in(1);
+
+    // make counted loop exit test always fail to
+    igvn->replace_input_of(cle, 1, outer_test);


Might be clearer to directly use igvn->intcon(0) instead of outer_test.

openjdk · 2022-02-08T10:06:06Z

@rwestrel 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:

8281322: C2: always construct strip mined loop initially (even if strip mining is disabled)

Reviewed-by: chagedorn, thartmann

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 3 new commits pushed to the master branch:

5f3d403: 8272735: Add missing SubL node transformations
01570ca: 8283017: GHA: Workflows break with update release versions
c0e3d10: 8283008: KRegister documentation out of date

Please see this link for an up-to-date comparison between the source branch of this pull request and the master branch.
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.

➡️ To integrate this PR with the above commit message to the master branch, type /integrate in a new comment.

TobiHartmann · 2022-02-11T10:05:44Z

For the record: Christian's testing found issues (he reported them to Roland already).

I think it would be good to change the bug title to something more descriptive.

rwestrel · 2022-02-28T09:42:34Z

For the record: Christian's testing found issues (he reported them to Roland already).

8282045 (When loop strip mining fails, safepoints are removed from loop anyway) is an underlying issue that Christian's testing caught.

openjdk · 2022-03-07T16:48:47Z

@rwestrel this pull request can not be integrated into master due to one or more merge conflicts. To resolve these merge conflicts and update this pull request you can run the following commands in the local repository for your personal fork:

git checkout JDK-8281322
git fetch https://git.openjdk.java.net/jdk master
git merge FETCH_HEAD
# resolve conflicts and follow the instructions given by git merge
git commit -m "Merge master"
git push

rwestrel · 2022-03-07T16:54:52Z

I pushed a number of new commits that address 3 issues (with LoopStripMiningIter=1):
1- the previous logic would drop nodes pinned on the backedge when the loop nest is transformed back to a single loop.
2- sunk stores need to be moved back into the loop which requires the inner loop's memory graph to be updated. I reused existing logic for that (OuterStripMinedLoopNode::fix_sunk_stores())
3- long counted loop support has a similar transform from a loop nest back to a single counted loop. That transform didn't have part 2 above. I'm not sure if it can cause actually bugs as I expect the counted loop to be transformed back to a loop nest next. In any case, rather than duplicate the logic, I refactored the code so it calls the new OuterStripMinedLoopNode::transform_to_counted_loop() instead. Having it callable from both igvn and loop opts requires some care.

TobiHartmann

This looks reasonable to me. All tests passed.

chhagedorn

Apart from some minor code style things, the update looks good to me!

chhagedorn · 2022-03-14T09:37:30Z

src/hotspot/share/opto/loopnode.cpp

+      if (u->is_Store()) {
+        int alias_idx = igvn->C->get_alias_index(u->adr_type());
+        Node* first = u;
+        for(;;) {


Missing space after for.

chhagedorn · 2022-03-14T09:37:41Z

src/hotspot/share/opto/loopnode.cpp

+          first = next;
+        }
+        Node* last = u;
+        for(;;) {


Missing space after for.

chhagedorn · 2022-03-14T09:42:08Z

src/hotspot/share/opto/loopnode.cpp

+        }
+#endif
+        if (phi == NULL) {
+          // If the an entire chains was sunk, the


If the an entire chains -> If an entire chain

chhagedorn · 2022-03-14T09:58:03Z

src/hotspot/share/opto/loopnode.cpp

@@ -2623,6 +2564,123 @@ BaseCountedLoopNode* BaseCountedLoopNode::make(Node* entry, Node* backedge, Basi
  return new LongCountedLoopNode(entry, backedge);
 }

+void OuterStripMinedLoopNode::fix_sunk_stores(CountedLoopEndNode* inner_cle, LoopNode* target, PhaseIterGVN* igvn,


I suggest to rename target to inner_cl as we are also using inner_cle already. Maybe you can also add some more comments for the intermediate steps (even though most of the code was just moved in this change).

rwestrel · 2022-03-14T13:00:54Z

Apart from some minor code style things, the update looks good to me!

Thanks. The new commit should take care of your comments.

chhagedorn

Thanks for the updates, looks good!

rwestrel · 2022-03-14T13:58:24Z

@TobiHartmann @chhagedorn thanks for the reviews

rwestrel · 2022-03-14T13:58:33Z

/integrate

openjdk · 2022-03-14T13:59:30Z

Going to push as commit ea9eeea.
Since your change was applied there have been 3 commits pushed to the master branch:

5f3d403: 8272735: Add missing SubL node transformations
01570ca: 8283017: GHA: Workflows break with update release versions
c0e3d10: 8283008: KRegister documentation out of date

Your commit was automatically rebased without conflicts.

openjdk · 2022-03-14T13:59:45Z

@rwestrel Pushed as commit ea9eeea.

💡 You may see a message that your pull request was closed with unmerged commits. This can be safely ignored.

fix & test

d715ccb

openjdk bot added the hotspot-compiler hotspot-compiler-dev@openjdk.org label Feb 7, 2022

test bug/summary

9cd1a8e

openjdk bot added the rfr Pull request is ready for review label Feb 7, 2022

chhagedorn approved these changes Feb 8, 2022

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Feb 8, 2022

rwestrel added 5 commits February 8, 2022 16:55

test failure

ba5f3f9

Merge branch 'master' into JDK-8281322

e1f7614

more

5f58d04

more

2b95bb1

more

9229f01

rwestrel added 6 commits March 1, 2022 16:19

Merge branch 'master' into JDK-8281322

9b80ffd

merge fix

360bdd7

cleanup

1e4f988

more

757a704

Merge branch 'master' into JDK-8281322

c5d7e36

more

a3876c4

openjdk bot added merge-conflict Pull request has merge conflict with target branch and removed ready Pull request is ready to be integrated labels Mar 7, 2022

rwestrel changed the title ~~8281322: irTests/TestLongRangeChecks.java fails with strip mining off~~ 8281322: C2: always construct strip mined loop initially (even if strip mining is disabled) Mar 7, 2022

Merge branch 'master' into JDK-8281322

e277b7d

openjdk bot added the ready Pull request is ready to be integrated label Mar 7, 2022

openjdk bot removed the merge-conflict Pull request has merge conflict with target branch label Mar 7, 2022

build fix

62d8d87

TobiHartmann approved these changes Mar 14, 2022

View reviewed changes

chhagedorn approved these changes Mar 14, 2022

View reviewed changes

rwestrel added 2 commits March 14, 2022 13:27

Merge branch 'master' into JDK-8281322

4076f70

review

8e257d4

chhagedorn approved these changes Mar 14, 2022

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Mar 14, 2022

openjdk bot closed this Mar 14, 2022

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Mar 14, 2022

8281322: C2: always construct strip mined loop initially (even if strip mining is disabled) #7364

8281322: C2: always construct strip mined loop initially (even if strip mining is disabled) #7364

Uh oh!

Conversation

rwestrel commented Feb 7, 2022 • edited by openjdk bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Progress

Issue

Reviewers

Reviewing

Uh oh!

bridgekeeper bot commented Feb 7, 2022

Uh oh!

openjdk bot commented Feb 7, 2022

Uh oh!

mlbridge bot commented Feb 7, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

chhagedorn left a comment

Choose a reason for hiding this comment

Uh oh!

chhagedorn Feb 8, 2022

Choose a reason for hiding this comment

Uh oh!

chhagedorn Feb 8, 2022

Choose a reason for hiding this comment

Uh oh!

openjdk bot commented Feb 8, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

TobiHartmann commented Feb 11, 2022

Uh oh!

rwestrel commented Feb 28, 2022

Uh oh!

openjdk bot commented Mar 7, 2022

Uh oh!

rwestrel commented Mar 7, 2022

Uh oh!

TobiHartmann left a comment

Choose a reason for hiding this comment

Uh oh!

chhagedorn left a comment

Choose a reason for hiding this comment

Uh oh!

chhagedorn Mar 14, 2022

Choose a reason for hiding this comment

Uh oh!

chhagedorn Mar 14, 2022

Choose a reason for hiding this comment

Uh oh!

chhagedorn Mar 14, 2022

Choose a reason for hiding this comment

Uh oh!

chhagedorn Mar 14, 2022

Choose a reason for hiding this comment

Uh oh!

rwestrel commented Mar 14, 2022

Uh oh!

chhagedorn left a comment

Choose a reason for hiding this comment

Uh oh!

rwestrel commented Mar 14, 2022

Uh oh!

rwestrel commented Mar 14, 2022

Uh oh!

openjdk bot commented Mar 14, 2022

Uh oh!

openjdk bot commented Mar 14, 2022

Uh oh!

Uh oh!

rwestrel commented Feb 7, 2022 •

edited by openjdk bot

Loading

mlbridge bot commented Feb 7, 2022 •

edited

Loading

openjdk bot commented Feb 8, 2022 •

edited

Loading