HotSpot crashes on an Intel VM created with Colima

[mikehearn]

Description

I'm not sure this is a Colima bug. It might be a qemu issue or something else, but I'm not sure where else to start. Running javac from a regular OpenJDK build segfaults on a semi-regular basis.

Version

colima version 0.9.1
git commit: 0cbf719f5409ce04b9f0607b681c005d2ff7d94a

runtime: docker
arch: aarch64
client: v25.0.4-rd
server: v28.3.3
limactl version 1.2.1
qemu-img version 10.0.3
Copyright (c) 2003-2025 Fabrice Bellard and the QEMU Project developers

Operating System

• macOS Intel <= 13 (Ventura)
• macOS Intel >= 14 (Sonoma)
• Apple Silicon <= 13 (Ventura)
• Apple Silicon >= 14 (Sonoma)
• Linux

Output of colima status

INFO[0000] colima [profile=intel] is running using macOS Virtualization.Framework
INFO[0000] arch: x86_64
INFO[0000] runtime: docker
INFO[0000] mountType: virtiofs
INFO[0000] docker socket: unix:///Users/mike/.colima/intel/docker.sock
INFO[0000] containerd socket: unix:///Users/mike/.colima/intel/containerd.sock

Reproduction Steps

colima start --cpus=8 --memory=32 intel
colima -p intel ssh
$ sudo apt install openjdk-25-jdk
$ cd /tmp
$ cat <<EOF >HelloWorld.java
class HelloWorld {
    public static void main(String[] args) {}
}
$ for i in $(seq 1 100); do echo $i; javac HelloWorld.java; done

leads to javac segfaulting on about one run in five. The exact location and nature of the crash varies from run to run. Occasionally I've seen a SIGILL, so it looks like it's jumping into an invalid pointer.

Needless to say, javac can compile Hello World just fine on a normal Intel machine. The bug is not upstream.

I've attached the hotspot crash log. It dies trying to access 0x0000000ca8abf7bb which is not even close to being something that's mapped.

What's a good way to start debugging this?

hs_err.log

Expected behaviour

No response

Additional context

No response