8231627: ThreadsListHandleInErrorHandlingTest.java fails in printing all threads #1891

dcubed-ojdk · 2020-12-24T17:33:21Z

A small robustness fix in ThreadsSMRSupport::print_info_on() to reduce the
likelihood of crashes during error reporting. Uses Threads_lock->try_lock()
for safety and restricts some reporting to when the Threads_lock has been
acquired (depends on JDK-8256383). Uses a ThreadsListHandle to make
the current ThreadsList safe for reporting (depends on JDK-8258284). Also
detects when the system ThreadsList (_java_thread_list) has changed and
will warn that some of the reported info may now be stale.

Two existing tests have been updated to reflect the use of a ThreadsListHandle
in ThreadsSMRSupport::print_info_on(). Mach5 Tier[1-6] testing has no regressions.

Progress

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

Issue

JDK-8231627: ThreadsListHandleInErrorHandlingTest.java fails in printing all threads

Reviewers

Erik Österlund (@fisk - Reviewer) ⚠️ Review applies to 334bdd2
Coleen Phillimore (@coleenp - Reviewer)
Patricio Chilano Mateo (@pchilano - Committer) ⚠️ Review applies to 334bdd2
Serguei Spitsyn (@sspitsyn - Reviewer)

Download

$ git fetch https://git.openjdk.java.net/jdk pull/1891/head:pull/1891
$ git checkout pull/1891

…ava fails because error occurred during printing all threads

dcubed-ojdk · 2020-12-24T17:33:40Z

/label add hotspot-runtime serviceability

bridgekeeper · 2020-12-24T17:34:17Z

👋 Welcome back dcubed! 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 · 2020-12-24T17:35:12Z

@dcubed-ojdk
The hotspot-runtime label was successfully added.

The serviceability label was successfully added.

dcubed-ojdk · 2020-12-24T17:39:53Z

@dholmes-ora, @fisk and @robehn - You folks might be interested in this
Thread-SMR related robustness fix. It should help during error reporting.

mlbridge · 2020-12-24T17:43:03Z

Webrevs

01: Full - Incremental (c9f4c4e)
00: Full (334bdd2)

fisk

Looks good. We have something similar in the precious GC log code during error reporting.

openjdk · 2020-12-24T22:02:17Z

@dcubed-ojdk 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:

8231627: ThreadsListHandleInErrorHandlingTest.java fails in printing all threads

Reviewed-by: eosterlund, coleenp, pchilanomate, sspitsyn

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 50 new commits pushed to the master branch:

8b45497: 8259037: livenmethods cannot find hsdis library
7d76966: 8255757: Javac emits duplicate pool entries on array::clone
cf9908b: 8258937: Remove JVM IgnoreRewrites flag
4d3d599: 8259223: Simplify boolean expression in the SunJSSE provider
1b60acd: 8259252: Shenandoah: Shenandoah build failed on AArch64 after JDK-8258459
7ddc2b5: 8258852: Arrays.asList() for single item could be replaced with List.of()
85bac8c: 8259021: SharedSecrets should avoid double racy reads from non-volatile fields
d5aa49d: 8259236: C2 compilation fails with assert(is_power_of_2(value)) failed: value must be a power of 2: 8000000000000000
82bdbfd: 8258857: Zero: non-PCH release build fails after JDK-8258074
f4122d6: 8258896: Remove the JVM ForceFloatExceptions option
... and 40 more: https://git.openjdk.java.net/jdk/compare/91244cc738e92163d99b8951c5d95b546447f341...master

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.

dcubed-ojdk · 2020-12-24T22:07:25Z

@fisk - Thanks for the review! And Merry Christmas Eve!!

fisk · 2020-12-24T22:14:43Z

@fisk - Thanks for the review! And Merry Christmas Eve!!

Merry Christmas to you too Dan!

dcubed-ojdk · 2021-01-04T15:53:08Z

Ping! I could use a second review here...

coleenp

Looks good. One comment. Also, ha ha, Harold caught me today: you need to update the copyrights!

coleenp · 2021-01-04T16:28:42Z

src/hotspot/share/runtime/threadSMR.cpp

+    }
    return;
  }
  st->print_cr("_java_thread_list_alloc_cnt=" UINT64_FORMAT ", "


Are these statistics stable if you don't have the Threads_lock? Seems like a good place to return unconditionally to me, but it's up to you whether this is wrong and it matters. It doesn't follow any pointers so doesn't look like it'll crash.

No the statistics starting on L1161 are not stable if we don't
have the Threads_lock. That's why we detect a change in the
_java_thread_list on L1194 and print a message:

if (_java_thread_list != saved_threads_list) { st->print_cr("The _java_thread_list has changed from " INTPTR_FORMAT " to " INTPTR_FORMAT " so some of the above information may be stale.", p2i(saved_threads_list), p2i(_java_thread_list)); }

iklam · 2021-01-04T18:32:27Z

Maybe the PR/JBS title should be shortened to "ThreadsListHandleInErrorHandlingTest.java fails in printing all threads" for conciseness and to make GitHub happy?

pchilano

Hi Dan,

Looks good to me!

pchilano · 2021-01-04T18:42:54Z

src/hotspot/share/runtime/threadSMR.cpp

+  }
+
+  if (_to_delete_list != NULL) {
+    if (has_Threads_lock) {


Shouldn't we check for Threads_lock->owned_by_self() instead? (case where thread already owns Threads_lock before calling print_info_on()).

Good idea! That will cover one more "safe" case, but I'll want to rename
the new has_Threads_lock to something else.

pchilano · 2021-01-04T18:43:40Z

src/hotspot/share/runtime/threadSMR.cpp

+      st->print_cr("Skipping _to_delete_list fields and contents for safety.");
    }
  }
  if (!EnableThreadSMRStatistics) {


You could switch to "if(EnableThreadSMRStatistics)" instead and put this code at the end to avoid repetition. Also I think the comparison with _java_thread_list could be done unconditionally at the end since it's already racy anyways (even if the info was printed with the Threads_lock held it could have changed right after it's released and before returning).

I like the refactoring suggestion from Patricio above to switch to "if(EnableThreadSMRStatistics)". The code will be a little more elegant.

sspitsyn

Hi Dan,

It looks good modulo a couple of suggestions from Patricio.

Thanks,
Serguei

dcubed-ojdk · 2021-01-05T21:45:19Z

Made copyright changes based on comments from coleenp,
code changes based on comments from pchilano, and changed
the title based on comments from iklam.
Thanks to coleenp, pchilano and sspitsyn for the code reviews.
Re-testing now...

dcubed-ojdk · 2021-01-05T22:36:18Z

Passed local builds and testing on my MBP13. A Mach5 Tier1 has finished all
of the build tasks and is mostly done with the test tasks.

Please re-review when you get the chance.

coleenp

LGTM!

pchilano · 2021-01-06T00:29:10Z

Looks good, thanks Dan!

dcubed-ojdk · 2021-01-06T02:11:50Z

sspitsyn, coleenp, and pchilano - Thanks for the re-reviews!

My local Linux-X64 build and test went fine. My Mach5 Tier1 passed
without failures. Mach5 Tier[23] are still running.

dcubed-ojdk · 2021-01-06T14:34:16Z

/integrate

openjdk · 2021-01-06T14:35:20Z

@dcubed-ojdk Since your change was applied there have been 55 commits pushed to the master branch:

7e01bc9: 8255264: Support for identifying the full range of IPv4 localhost addresses on Windows
8a05d60: 8259042: Inconsistent use of general primitives loops
e3b9da1: 8259287: AbstractCompiler marks const in wrong position for is_c1/is_c2/is_jvmci
32538b5: 8193942: Regression automated test '/open/test/jdk/javax/swing/JFrame/8175301/ScaledFrameBackgroundTest.java' fails
52d3fee: 8258813: [TESTBUG] Fix incorrect Vector API test output message
8b45497: 8259037: livenmethods cannot find hsdis library
7d76966: 8255757: Javac emits duplicate pool entries on array::clone
cf9908b: 8258937: Remove JVM IgnoreRewrites flag
4d3d599: 8259223: Simplify boolean expression in the SunJSSE provider
1b60acd: 8259252: Shenandoah: Shenandoah build failed on AArch64 after JDK-8258459
... and 45 more: https://git.openjdk.java.net/jdk/compare/91244cc738e92163d99b8951c5d95b546447f341...master

Your commit was automatically rebased without conflicts.

Pushed as commit c0540ff.

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

8231627: runtime/ErrorHandling/ThreadsListHandleInErrorHandlingTest.j…

334bdd2

…ava fails because error occurred during printing all threads

openjdk bot added hotspot-runtime hotspot-runtime-dev@openjdk.org serviceability serviceability-dev@openjdk.org labels Dec 24, 2020

dcubed-ojdk marked this pull request as ready for review December 24, 2020 17:37

openjdk bot added the rfr Pull request is ready for review label Dec 24, 2020

fisk approved these changes Dec 24, 2020

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Dec 24, 2020

coleenp approved these changes Jan 4, 2021

View reviewed changes

pchilano approved these changes Jan 4, 2021

View reviewed changes

sspitsyn approved these changes Jan 5, 2021

View reviewed changes

dcubed-ojdk changed the title ~~8231627: runtime/ErrorHandling/ThreadsListHandleInErrorHandlingTest.java fails because error occurred during printing all threads~~ 8231627: ThreadsListHandleInErrorHandlingTest.java fails in printing all threads Jan 5, 2021

Address comments from coleenp and pchilano.

c9f4c4e

sspitsyn approved these changes Jan 5, 2021

View reviewed changes

coleenp approved these changes Jan 5, 2021

View reviewed changes

openjdk bot closed this Jan 6, 2021

openjdk bot added integrated Pull request has been integrated and removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Jan 6, 2021

dcubed-ojdk deleted the JDK-8231627 branch January 6, 2021 14:37

8231627: ThreadsListHandleInErrorHandlingTest.java fails in printing all threads #1891

8231627: ThreadsListHandleInErrorHandlingTest.java fails in printing all threads #1891

Uh oh!

Conversation

dcubed-ojdk commented Dec 24, 2020 • edited by openjdk bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Progress

Issue

Reviewers

Download

Uh oh!

dcubed-ojdk commented Dec 24, 2020

Uh oh!

bridgekeeper bot commented Dec 24, 2020

Uh oh!

openjdk bot commented Dec 24, 2020

Uh oh!

dcubed-ojdk commented Dec 24, 2020

Uh oh!

mlbridge bot commented Dec 24, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

fisk left a comment

Choose a reason for hiding this comment

Uh oh!

openjdk bot commented Dec 24, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

dcubed-ojdk commented Dec 24, 2020

Uh oh!

fisk commented Dec 24, 2020

Uh oh!

dcubed-ojdk commented Jan 4, 2021

Uh oh!

coleenp left a comment

Choose a reason for hiding this comment

Uh oh!

coleenp Jan 4, 2021

Choose a reason for hiding this comment

Uh oh!

dcubed-ojdk Jan 5, 2021

Choose a reason for hiding this comment

Uh oh!

iklam commented Jan 4, 2021

Uh oh!

pchilano left a comment

Choose a reason for hiding this comment

Uh oh!

pchilano Jan 4, 2021

Choose a reason for hiding this comment

Uh oh!

dcubed-ojdk Jan 5, 2021

Choose a reason for hiding this comment

Uh oh!

pchilano Jan 4, 2021

Choose a reason for hiding this comment

Uh oh!

sspitsyn Jan 5, 2021

Choose a reason for hiding this comment

Uh oh!

dcubed-ojdk Jan 5, 2021

Choose a reason for hiding this comment

Uh oh!

sspitsyn left a comment

Choose a reason for hiding this comment

Uh oh!

dcubed-ojdk commented Jan 5, 2021

Uh oh!

dcubed-ojdk commented Jan 5, 2021

Uh oh!

coleenp left a comment

Choose a reason for hiding this comment

Uh oh!

pchilano commented Jan 6, 2021

Uh oh!

dcubed-ojdk commented Jan 6, 2021

Uh oh!

dcubed-ojdk commented Jan 6, 2021

Uh oh!

dcubed-ojdk commented Dec 24, 2020 •

edited by openjdk bot

Loading

mlbridge bot commented Dec 24, 2020 •

edited

Loading

openjdk bot commented Dec 24, 2020 •

edited

Loading