8277444: Data race between JvmtiClassFileReconstituter::copy_bytecodes and class linking#3096
8277444: Data race between JvmtiClassFileReconstituter::copy_bytecodes and class linking#3096chadrako wants to merge 1 commit into
Conversation
|
Waiting to request maintainer approval until backport is merged in newer JDKs |
|
👋 Welcome back chadrako! A progress list of the required criteria for merging this PR into |
|
❗ This change is not yet ready to be integrated. |
|
This backport pull request has now been updated with issue from the original commit. |
|
@chadrako This pull request has been inactive for more than 4 weeks and will be automatically closed if another 4 weeks passes without any activity. To avoid this, simply issue a |
|
/keepalive |
|
@chadrako The pull request is being re-evaluated and the inactivity timeout has been reset. |
|
@chadrako This pull request has been inactive for more than 4 weeks and will be automatically closed if another 4 weeks passes without any activity. To avoid this, simply issue a |
|
@chadrako This pull request has been inactive for more than 8 weeks and will now be automatically closed. If you would like to continue working on this pull request in the future, feel free to reopen it! This can be done using the |
|
/open |
|
@chadrako This pull request is now open |
|
/approval request An almost clean backport (trivial naming differences) that fixes a race condition that occurs when a class is simultaneously retransformed via the Instrumentation API and linked. This can lead to invalid bytecode which will cause a VerifyError. Even worse, if class verifying is turned off, which it is by default for java.* classes, this will cause the JVM to crash/segfault. The included test test/jdk/java/lang/instrument/RetransformBigClassTest.java fails before the patch and passes after. This patch has been in jdk25 and Corretto 11 for 6 months. |
|
/approval request Re-requesting now that this has been backported to jdk17. An almost clean backport (trivial naming differences) that fixes a race condition that occurs when a class is simultaneously retransformed via the Instrumentation API and linked. This can lead to invalid bytecode which will cause a VerifyError. Even worse, if class verifying is turned off, which it is by default for java.* classes, this will cause the JVM to crash/segfault. The included test test/jdk/java/lang/instrument/RetransformBigClassTest.java fails before the patch and passes after. This patch has been in jdk25 and Corretto 11 for 6 months. |
|
@chadrako This pull request has been inactive for more than 4 weeks and will be automatically closed if another 4 weeks passes without any activity. To avoid this, simply issue a |
|
/touch |
|
@chadrako The pull request is being re-evaluated and the inactivity timeout has been reset. |
|
/template |
|
@chadrako Missing command 'append', usage: |
|
/template append |
|
@chadrako The pull request template has been appended to the pull request body |
gnu-andrew
left a comment
There was a problem hiding this comment.
This looks good to me. 11u lacks JDK-8233913 which added the earlier Thread::current() in getBytecodes so we don't have the second hunk in jvmtiEnv.cpp and have to call Thread::current() in the third. But otherwise the change is the same.
Test shows the patch works too.
|
@chadrako This pull request has been inactive for more than 4 weeks and will be automatically closed if another 4 weeks passes without any activity. To avoid this, simply issue a |
Backport of https://bugs.openjdk.org/browse/JDK-8277444. The backport was not clean but the fixes were trivial naming differences
This fixes a race condition that occurs when a class is simultaneously retransformed via the Instrumentation API and linked. This can lead to invalid bytecode which will cause a VerifyError. Even worse, if class verifying is turned off, which it is by default for java.* classes, this will cause the JVM to crash/segfault.
The included test
test/jdk/java/lang/instrument/RetransformBigClassTest.javafails before the patch and passes afterProgress
Issue
Reviewers
Reviewing
Using
gitCheckout this PR locally:
$ git fetch https://git.openjdk.org/jdk11u-dev.git pull/3096/head:pull/3096$ git checkout pull/3096Update a local copy of the PR:
$ git checkout pull/3096$ git pull https://git.openjdk.org/jdk11u-dev.git pull/3096/headUsing Skara CLI tools
Checkout this PR locally:
$ git pr checkout 3096View PR using the GUI difftool:
$ git pr show -t 3096Using diff file
Download this PR as a diff file:
https://git.openjdk.org/jdk11u-dev/pull/3096.diff
Using Webrev
Link to Webrev Comment