JDK-8331575 : C2: crash when ConvL2I is split thru phi at LongCountedLoop
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 17,21,22,23
  • Priority: P3
  • Status: Open
  • Resolution: Unresolved
  • Submitted: 2024-05-02
  • Updated: 2024-05-06
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
JDK 23
23Unresolved
Related Reports
Duplicate :  
Relates :  
Relates :  
Relates :  
Description
Attached test case crashes with:

#  SIGFPE (0x8) at pc=0x00007f903c54e3d0, pid=3372325, tid=3372344
#
# JRE version: OpenJDK Runtime Environment (23.0) (fastdebug build 23-internal-adhoc.roland.jdk-jdk)
# Java VM: OpenJDK 64-Bit Server VM (fastdebug 23-internal-adhoc.roland.jdk-jdk, mixed mode, sharing, compressed oops, compressed class ptrs, g1 gc, linux-amd64)
# Problematic frame:
# J 41 c2 TestLongCountedLoopConvL2I.test1(I)I (61 bytes) @ 0x00007f903c54e3d0 [0x00007f903c54e360+0x0000000000000070]

Likely exists since LongCountedLoop exists.
Comments
ILW = Same as JDK-8298851 = P3
03-05-2024

Ok seems like you've just sent a PR out in this very second. Then I suggest to close the other one as dup. Maybe you want to add the tests from JDK-8298851 as additional regression tests.
03-05-2024

Seems like a duplicate of JDK-8298851. I've had a patch ready for quite some time but never gotten to it. Do you want to take this over [~roland]?
03-05-2024

A pull request was submitted for review. URL: https://git.openjdk.org/jdk/pull/19086 Date: 2024-05-03 12:33:43 +0000
03-05-2024