8342823: Ubsan: ciEnv.cpp:1614:65: runtime error: member call on null pointer of type 'struct CompileTask'

[MBaesken]

When running with ubsanized binaries on Linux x86_64,
hs jtreg test compiler/startup/StartupOutput.java
showed this issue

jdk/src/hotspot/share/ci/ciEnv.cpp:1614:65: runtime error: member call on null pointer of type 'struct CompileTask'
#0 0x7fcea0810117 in ciEnv::dump_replay_data_helper(outputStream*) src/hotspot/share/ci/ciEnv.cpp:1614
#1 0x7fcea3123577 in VMError::report_and_die(int, char const*, char const*, __va_list_tag*, Thread*, unsigned char*, void*, void*, char const*, int, unsigned long) src/hotspot/share/utilities/vmError.cpp:1872
#2 0x7fcea0c01499 in report_fatal(VMErrorType, char const*, int, char const*, ...) src/hotspot/share/utilities/debug.cpp:214
#3 0x7fcea09e9d85 in RuntimeStub::new_runtime_stub(char const*, CodeBuffer*, short, int, OopMapSet*, bool, bool) src/hotspot/share/code/codeBlob.cpp:413
#4 0x7fcea066da1d in Runtime1::generate_blob(BufferBlob*, C1StubId, char const*, bool, StubAssemblerCodeGenClosure*) src/hotspot/share/c1/c1_Runtime1.cpp:233
#5 0x7fcea066dfb0 in Runtime1::generate_blob_for(BufferBlob*, C1StubId) src/hotspot/share/c1/c1_Runtime1.cpp:262
#6 0x7fcea066dfb0 in Runtime1::initialize(BufferBlob*) src/hotspot/share/c1/c1_Runtime1.cpp:272
#7 0x7fcea03d2be1 in Compiler::init_c1_runtime() src/hotspot/share/c1/c1_Compiler.cpp:53
#8 0x7fcea03d2be1 in Compiler::initialize() src/hotspot/share/c1/c1_Compiler.cpp:74
#9 0x7fcea0acc0c2 in CompileBroker::init_compiler_runtime() src/hotspot/share/compiler/compileBroker.cpp:1771
#10 0x7fcea0ad9a3f in CompileBroker::compiler_thread_loop() src/hotspot/share/compiler/compileBroker.cpp:1913
#11 0x7fcea161264a in JavaThread::thread_main_inner() src/hotspot/share/runtime/javaThread.cpp:759
#12 0x7fcea2ec739a in Thread::call_run() src/hotspot/share/runtime/thread.cpp:234
#13 0x7fcea251e1d2 in thread_native_entry src/hotspot/os/linux/os_linux.cpp:858
#14 0x7fcea7c6c6e9 in start_thread (/lib64/libpthread.so.0+0xa6e9) (BuildId: 1b515766201d47a183932ba0c8c8bd0d9ee8755b)
#15 0x7fcea730f58e in clone (/lib64/libc.so.6+0x11858e) (BuildId: 448a3ddd22596e1adb8fb3dec8921ed5b9d54dc2)

So a nullptr check should be better added .

---

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-8342823: Ubsan: ciEnv.cpp:1614:65: runtime error: member call on null pointer of type 'struct CompileTask' (Bug - P4)

Reviewers

• Vladimir Kozlov (@vnkozlov - Reviewer)
• Martin Doerr (@TheRealMDoerr - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/21684/head:pull/21684
$ git checkout pull/21684

Update a local copy of the PR:
$ git checkout pull/21684
$ git pull https://git.openjdk.org/jdk.git pull/21684/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 21684

View PR using the GUI difftool:
$ git pr show -t 21684

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/21684.diff

Webrev

Link to Webrev Comment

[bridgekeeper]

👋 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]

@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:

8342823: Ubsan: ciEnv.cpp:1614:65: runtime error: member call on null pointer of type 'struct CompileTask'

Reviewed-by: kvn, mdoerr

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 43 new commits pushed to the master branch:

• d49f210: 8342040: Further improve entry lookup performance for multi-release JARs
• d2e716e: 8331958: Update PC/SC Lite for Suse Linux to 2.3.0
• a95374f: 8343101: Rework BasicTest.testTemp test cases
• 00fe9f7: 8343100: Consolidate EmptyFolderTest and EmptyFolderPackageTest jpackage tests into single java file
• 9f6d5b4: 8343020: (fs) Add support for SecureDirectoryStream on macOS
• 1341b81: 8341666: FileInputStream doesn't support readAllBytes() or readNBytes(int) on pseudo devices
• 52382e2: 8338021: Support new unsigned and saturating vector operators in VectorAPI
• e659d9d: 8342975: C2: Micro-optimize PhaseIdealLoop::Dominators()
• 9f6211b: 8341371: CDS cannot load archived heap objects with -XX:+UseSerialGC -XX:-UseCompressedOops
• 120a935: 8342561: Metaspace for generated reflection classes is no longer needed
• ... and 33 more: https://git.openjdk.org/jdk/compare/85774b713edf8782f162ac25b61ce99a77e116f4...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]

@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.

[mlbridge]

Webrevs

• 02: Full - Incremental (1d240ea1)
• 01: Full - Incremental (f019b47f)
• 00: Full (d28c366f)

[vnkozlov]

Good.

[dean-long]

There's not much point in generating an incomplete replay file. How about moving the checks for this->task() to the top of the function? And we could avoid the call altogether by changing VMError::report_and_die() from:

1864      ciEnv* env = ciEnv::current();
1865      if (env != nullptr) {

to

1864      ciEnv* env = ciEnv::current();
1865      if (env != nullptr && env->task() != nullptr) {

[MBaesken]

How about moving the checks for this->task() to the top of the function?

Makes sense, better check just once .

[vnkozlov]

I think we should do the check in VMError::report_and_die() to avoid creating empty replay file.
Note, dump_replay_data_unsafe() is called only in that one place.
An other path through dump_replay_data() call required Compilation ID which is set only when we have task.
We can use assert instead of check in ciEnv::dump_replay_data_helper().

[MBaesken]

I think we should do the check in VMError::report_and_die() to avoid creating empty replay file. Note, dump_replay_data_unsafe() is called only in that one place. An other path through dump_replay_data() call required Compilation ID which is set only when we have task. We can use assert instead of check in ciEnv::dump_replay_data_helper().

I moved the check and added an assert .

[vnkozlov]

Looks good.

[TheRealMDoerr]

LGTM.

[MBaesken]

Hi Vladimir and Martin, thanks for the reviews !

/integrate

[openjdk]

Going to push as commit beff8bf.
Since your change was applied there have been 50 commits pushed to the master branch:

• e389f82: 8343137: C2: VerifyLoopOptimizations fails with "Was reachable in only one"
• 0abfa3b: 8304824: NMT should not use ThreadCritical
• 88dc655: 8342988: GHA: Build JTReg in single step
• df3473e: 8343178: Test BasicTest.java javac compile fails cannot find symbol
• 54327bc: 8342962: [s390x] TestOSRLotsOfLocals.java crashes
• f0075d5: 8343115: SkipIfEqual class is not used after JDK-8335946
• 90bd544: 8342958: Use jvmArgs consistently in microbenchmarks
• d49f210: 8342040: Further improve entry lookup performance for multi-release JARs
• d2e716e: 8331958: Update PC/SC Lite for Suse Linux to 2.3.0
• a95374f: 8343101: Rework BasicTest.testTemp test cases
• ... and 40 more: https://git.openjdk.org/jdk/compare/85774b713edf8782f162ac25b61ce99a77e116f4...master

Your commit was automatically rebased without conflicts.

[openjdk]

@MBaesken Pushed as commit beff8bf.

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