8350683: Non-C2 / minimal JVM crashes in the build on ppc64 platforms by MBaesken · Pull Request #23794 · openjdk/jdk

MBaesken · 2025-02-26T09:23:25Z

When building a JVM without C2 (e.g. minimal) on ppc64 platforms , it crashes in the build because of unwanted dependencies to C2.
AIX crash is (linux ppc64le crash is similar) :

# Internal Error (compiledIC_ppc.cpp:141), pid=17695018, tid=258
# Error: ShouldNotReachHere()
#
 
iar: 0x0900000008800c60 libjvm.so::AixNativeCallstack::print_callstack_for_context(outputStream*, ucontext_t const*, bool, char*, unsigned long)+0x4bc (C++ uses_alloca saves_cr saves_lr stores_bc gpr_saved:18 fixedparms:5 parmsonstk:1)
lr: 0x09000000087ea9b8 libjvm.so::fdStream::write(char const*, unsigned long)+0x44 (C++ uses_alloca saves_lr stores_bc gpr_saved:4 fixedparms:3 parmsonstk:1)
sp: 0x000000011023aab0 (base - 0x2DD8)
rtoc: 0x08001000a0088ff0
|---stackaddr----| |----lrsave------|: <function name>
0x000000011023aea0 - 0x0900000008800730 libjvm.so::os::Aix::platform_print_native_stack(outputStream*, void const*, char*, int, unsigned char*&)+0x24 (C++ uses_alloca saves_lr stores_bc gpr_saved:1 fixedparms:5 parmsonstk:1)
0x000000011023af20 - 0x0900000008800644 libjvm.so::NativeStackPrinter::print_stack(outputStream*, char*, int, unsigned char*&, bool, int)+0x60 (C++ uses_alloca saves_cr saves_lr stores_bc gpr_saved:6 fixedparms:7 parmsonstk:1)
0x000000011023afc0 - 0x09000000087f6ff8 libjvm.so::VMError::report(outputStream*, bool)+0x11f0 (C++ fp_present uses_alloca saves_cr saves_lr stores_bc gpr_saved:13 fixedparms:2 parmsonstk:1)
0x000000011023b830 - 0x09000000087e9fdc libjvm.so::VMError::report_and_die(int, char const*, char const*, char*, Thread*, unsigned char*, void const*, void const*, char const*, int, unsigned long)+0x6a0 (C++ uses_alloca saves_lr stores_bc gpr_saved:18 fixedparms:8 parmsonstk:1)
0x000000011023ba10 - 0x09000000087e96a0 libjvm.so::report_vm_error(char const*, int, char const*, char const*, ...)+0xa0 (C++ uses_alloca saves_lr stores_bc gpr_saved:5 fixedparms:4 parmsonstk:1)
0x000000011023bad0 - 0x09000000087e95cc libjvm.so::report_vm_error(char const*, int, char const*)+0x24 (C++ uses_alloca saves_lr stores_bc gpr_saved:1 fixedparms:3 parmsonstk:1)
0x000000011023bb50 - 0x09000000087e956c libjvm.so::report_should_not_reach_here(char const*, int)+0x20 (C++ uses_alloca saves_lr stores_bc gpr_saved:1 fixedparms:2 parmsonstk:1)
0x000000011023bbd0 - 0x0900000008906e5c libjvm.so::CompiledDirectCall::emit_to_interp_stub(MacroAssembler*, unsigned char*)+0x28 (C++ uses_alloca saves_lr stores_bc gpr_saved:1 fixedparms:2 parmsonstk:1)
0x000000011023bc50 - 0x0900000008902848 libjvm.so::SharedRuntime::generate_native_wrapper(MacroAssembler*, methodHandle const&, int, BasicType*, VMRegPair*, BasicType)+0x380 (C++ uses_alloca saves_cr saves_lr stores_bc gpr_saved:18 fixedparms:6 parmsonstk:1)
0x000000011023c000 - 0x0900000008901f7c libjvm.so::AdapterHandlerLibrary::create_native_wrapper(methodHandle const&)+0x464 (C++ fp_present uses_alloca saves_cr saves_lr stores_bc gpr_saved:18 fixedparms:1 parmsonstk:1)
0x000000011023c5d0 - 0x09000000088f0d78 libjvm.so::Method::link_method(methodHandle const&, JavaThread*)+0x17c (C++ uses_alloca saves_lr stores_bc gpr_saved:5 fixedparms:3 parmsonstk:1)
0x000000011023c670 - 0x09000000088f0ab8 libjvm.so::InstanceKlass::link_methods(JavaThread*)+0xcc (C++ uses_alloca saves_lr stores_bc gpr_saved:10 fixedparms:2 parmsonstk:1)
0x000000011023c760 - 0x09000000088d5e48 libjvm.so::InstanceKlass::link_class_impl(JavaThread*)+0x38c (C++ uses_alloca saves_cr saves_lr stores_bc gpr_saved:10 fixedparms:2 parmsonstk:1)
0x000000011023c8f0 - 0x09000000088d4de8 libjvm.so::InstanceKlass::initialize_impl(JavaThread*)+0x94 (C++ uses_alloca saves_cr saves_lr stores_bc gpr_saved:14 fixedparms:2 parmsonstk:1)
0x000000011023cae0 - 0x09000000088d4cf4 libjvm.so::InstanceKlass::initialize(JavaThread*)+0x60 (C++ uses_alloca saves_lr stores_bc gpr_saved:3 fixedparms:2 parmsonstk:1)
0x000000011023cb70 - 0x0900000008c40ea0 libjvm.so::initialize_class(Symbol*, JavaThread*)+0x64 (C++ uses_alloca saves_lr stores_bc gpr_saved:2 fixedparms:2 parmsonstk:1)
0x000000011023cc00 - 0x0900000008c35384 libjvm.so::Threads::create_vm(JavaVMInitArgs*, bool*)+0x6ec (C++ uses_alloca saves_lr stores_bc gpr_saved:8 fixedparms:2 parmsonstk:1)
0x000000011023d540 - 0x0900000008c5a91c libjvm.so::JNI_CreateJavaVM+0xa4 (C++ uses_alloca saves_lr stores_bc gpr_saved:8 fixedparms:3 parmsonstk:1)
0x000000011023d640 - 0x00000001000100f4 javac::JavaMain+0x148 (C++ saves_lr stores_bc gpr_saved:11 fixedparms:1 parmsonstk:1)
0x000000011023d730 - 0x000000010000ff74 javac::ThreadJavaMain+0x10 (C++ saves_lr stores_bc fixedparms:1 parmsonstk:1)
0x000000011023d7a0 - 0x090000000056204c libpthreads.a::_pthread_body+0xec (C saves_lr stores_bc gpr_saved:1 fixedparms:1 )
0x000000011023d820 - 0x0000000000000000

Progress

Change must be properly reviewed (1 review required, with at least 1 Reviewer)
Change must not contain extraneous whitespace
Commit message must refer to an issue

Issue

JDK-8350683: Non-C2 / minimal JVM crashes in the build on ppc64 platforms (Bug - P4)

Reviewers

Martin Doerr (@TheRealMDoerr - Reviewer)
Amit Kumar (@offamitkumar - Committer)

Contributors

Martin Doerr <mdoerr@openjdk.org>

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/23794/head:pull/23794
$ git checkout pull/23794

Update a local copy of the PR:
$ git checkout pull/23794
$ git pull https://git.openjdk.org/jdk.git pull/23794/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 23794

View PR using the GUI difftool:
$ git pr show -t 23794

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/23794.diff

Using Webrev

Link to Webrev Comment

bridgekeeper · 2025-02-26T09:24:06Z

👋 Welcome back mbaesken! 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 · 2025-02-26T09:24:34Z

@MBaesken 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:

8350683: Non-C2 / minimal JVM crashes in the build on ppc64 platforms

Co-authored-by: Martin Doerr <mdoerr@openjdk.org>
Reviewed-by: mdoerr, amitkumar

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 35 new commits pushed to the master branch:

ab4b0ef: 8350889: GenShen: Break out of infinite loop of old GC cycles
0a4c5a8: 8347804: GenShen: Crash with small GCCardSizeInBytes and small Java heap
f1398ec: 8350701: [JMH] test foreign.AllocFromSliceTest failed with Exception for size>1024
3ae80bf: 8349766: GenShen: Bad progress after degen does not always need full gc
2fd7156: 8347426: Invalid value used for enum Cell in iTypeFlow::StateVector::meet_exception
939815f: 8347040: C2: assert(!loop->_body.contains(in)) failed
8323ddf: 8346659: SnippetTaglet should report an error if provided ambiguous links
3c9d64e: 8350858: [IR Framework] Some tests failed on Cascade Lake
d6d9447: 8350786: Some java/lang jtreg tests miss requires vm.hasJFR
e80b76b: 8276995: Bug in jdk.jfr.event.gc.collection.TestSystemGC
... and 25 more: https://git.openjdk.org/jdk/compare/f529bf712d8946584999dfc98abea60c22c97167...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.

openjdk · 2025-02-26T09:25:10Z

@MBaesken 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.

MBaesken · 2025-02-26T09:26:32Z

/contributor add mdoerr

openjdk · 2025-02-26T09:26:55Z

@MBaesken
Contributor Martin Doerr <mdoerr@openjdk.org> successfully added.

mlbridge · 2025-02-26T09:30:08Z

Webrevs

00: Full (2c3aa1bc)

MBaesken · 2025-02-26T13:05:41Z

Hi @offamitkumar , s390x might have the same problem we face on ppc64, see here

https://github.com/openjdk/jdk/blob/master/src/hotspot/cpu/s390/compiledIC_s390.cpp#L58

Maybe you want to check this. On ppc64, the previously C2-only coding in CompiledDirectCall::emit_to_interp_stub was needed for non-C2 JVMs like minimal JVM (where only C1 is present).

offamitkumar · 2025-02-26T16:46:54Z

Hi Matthias,

You mentioned that the crash happened during the build ? I just checked and at least the build is fine on s390x.

This is what I am using for configuration:

bash configure \
	--with-boot-jdk=$HOME/jjboot_jdk_23 \
	--with-jtreg=$HOME/jtreg \
	--with-gtest=$HOME/googletest \
	--with-jmh=build/jmh/jars \
	--with-debug-level=fastdebug \
         --with-jvm-variants=minimal \
	--with-native-debug-symbols=internal \
	--disable-precompiled-headers

MBaesken · 2025-02-27T08:27:12Z

You mentioned that the crash happened during the build ? I just checked and at least the build is fine on s390x.

Yes during the minimal build ( --with-jvm-features=minimal --with-jvm-variants=minimal ).
Seems you miss --with-jvm-features=minimal .

offamitkumar · 2025-02-27T09:19:50Z

Ok I ran another build with this configuration and still build was successful :

bash configure \
	--with-boot-jdk=$HOME/boot_jdk_23 \
	--with-jtreg=$HOME/jtreg \
	--with-gtest=$HOME/googletest \
	--with-jmh=build/jmh/jars \
  --with-jvm-variants=minimal \
  --with-jvm-features=minimal \
	--with-debug-level=fastdebug \
	--with-native-debug-symbols=internal \
	--disable-precompiled-headers

MBaesken · 2025-02-27T09:52:40Z

Hi @offamitkumar this is a bit surprising because the matcher header is only included if COMPILER2 is defined
https://github.com/openjdk/jdk/blob/master/src/hotspot/cpu/s390/compiledIC_s390.cpp#L32
and this should not be the case in minimal JVM , but if this somehow works on your side, then fine :-) !

TheRealMDoerr

LGTM. Thanks for fixing it!

offamitkumar · 2025-02-28T02:56:50Z

Hi @offamitkumar this is a bit surprising because the matcher header is only included if COMPILER2 is defined https://github.com/openjdk/jdk/blob/master/src/hotspot/cpu/s390/compiledIC_s390.cpp#L32 and this should not be the case in minimal JVM , but if this somehow works on your side, then fine :-) !

I have finished tier1 on minimal build, will see if this issue gets reproduced by any of the tests.

offamitkumar

Looks Good.

MBaesken · 2025-02-28T07:36:09Z

Thanks for the reviews !

/integrate

openjdk · 2025-02-28T07:36:37Z

Going to push as commit 2af76de.
Since your change was applied there have been 35 commits pushed to the master branch:

ab4b0ef: 8350889: GenShen: Break out of infinite loop of old GC cycles
0a4c5a8: 8347804: GenShen: Crash with small GCCardSizeInBytes and small Java heap
f1398ec: 8350701: [JMH] test foreign.AllocFromSliceTest failed with Exception for size>1024
3ae80bf: 8349766: GenShen: Bad progress after degen does not always need full gc
2fd7156: 8347426: Invalid value used for enum Cell in iTypeFlow::StateVector::meet_exception
939815f: 8347040: C2: assert(!loop->_body.contains(in)) failed
8323ddf: 8346659: SnippetTaglet should report an error if provided ambiguous links
3c9d64e: 8350858: [IR Framework] Some tests failed on Cascade Lake
d6d9447: 8350786: Some java/lang jtreg tests miss requires vm.hasJFR
e80b76b: 8276995: Bug in jdk.jfr.event.gc.collection.TestSystemGC
... and 25 more: https://git.openjdk.org/jdk/compare/f529bf712d8946584999dfc98abea60c22c97167...master

Your commit was automatically rebased without conflicts.

openjdk · 2025-02-28T07:36:44Z

@MBaesken Pushed as commit 2af76de.

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

JDK-8350683

2c3aa1b

openjdk bot changed the title ~~JDK-8350683: Non-C2 / minimal JVM crashes in the build on ppc64 platforms~~ 8350683: Non-C2 / minimal JVM crashes in the build on ppc64 platforms Feb 26, 2025

openjdk bot added the rfr Pull request is ready for review label Feb 26, 2025

openjdk bot added the hotspot-compiler hotspot-compiler-dev@openjdk.org label Feb 26, 2025

TheRealMDoerr approved these changes Feb 27, 2025

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Feb 27, 2025

offamitkumar approved these changes Feb 28, 2025

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Feb 28, 2025

openjdk bot closed this Feb 28, 2025

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Feb 28, 2025

offamitkumar mentioned this pull request Feb 18, 2026

8378149: [s390] Non-C2 / minimal JVM crashes in the build #29777

Closed

3 tasks

Conversation

MBaesken commented Feb 26, 2025 • edited by openjdk bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Progress

Issue

Reviewers

Contributors

Reviewing

Uh oh!

bridgekeeper bot commented Feb 26, 2025

Uh oh!

openjdk bot commented Feb 26, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

openjdk bot commented Feb 26, 2025

Uh oh!

MBaesken commented Feb 26, 2025

Uh oh!

openjdk bot commented Feb 26, 2025

Uh oh!

mlbridge bot commented Feb 26, 2025

Webrevs

Uh oh!

MBaesken commented Feb 26, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

offamitkumar commented Feb 26, 2025

Uh oh!

MBaesken commented Feb 27, 2025

Uh oh!

offamitkumar commented Feb 27, 2025

Uh oh!

MBaesken commented Feb 27, 2025

Uh oh!

TheRealMDoerr left a comment

Choose a reason for hiding this comment

Uh oh!

offamitkumar commented Feb 28, 2025

Uh oh!

offamitkumar left a comment

Choose a reason for hiding this comment

Uh oh!

MBaesken commented Feb 28, 2025

Uh oh!

openjdk bot commented Feb 28, 2025

Uh oh!

openjdk bot commented Feb 28, 2025

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

3 participants

MBaesken commented Feb 26, 2025 •

edited by openjdk bot

Loading

openjdk bot commented Feb 26, 2025 •

edited

Loading

MBaesken commented Feb 26, 2025 •

edited

Loading