Skip to content

VM crashes on shutdown #23844

Closed
Closed
@jimbeveridge

Description

@jimbeveridge

We are seeing a flakey crash on Dart VM shutdown when our Mojo app exits. The gdb stack backtrace is shown below. This can't be reduced to a simple repro case, so I'm not sure how to proceed.

#0  base::debug::(anonymous namespace)::StackDumpSignalHandler (
    signal=<error reading variable: Cannot access memory at address 0xffffffffffffffd4>, 
    info=<error reading variable: Cannot access memory at address 0xffffffffffffffc8>, 
    void_context=<error reading variable: Cannot access memory at address 0xffffffffffffffc0>)
    at ../../base/debug/stack_trace_posix.cc:208
#1  <signal handler called>
#2  0x00007f680401bb6e in dart::Monitor::Exit (this=<optimised out>) at ../../dart/runtime/vm/os_thread_linux.cc:306
#3  0x00007f68040e4623 in dart::ThreadPool::Worker::Loop (this=<optimised out>) at ../../dart/runtime/vm/thread_pool.cc:277
#4  0x00007f68040e4594 in dart::ThreadPool::Worker::Main (args=<optimised out>) at ../../dart/runtime/vm/thread_pool.cc:320
#5  0x00007f680401b2fb in dart::ThreadStart (data_ptr=<optimised out>) at ../../dart/runtime/vm/os_thread_linux.cc:86
#6  0x00007f6808466182 in start_thread (arg=0x7f680201b700) at pthread_create.c:312
#7  0x00007f680819347d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111

Metadata

Metadata

Assignees

Labels

area-vmUse area-vm for VM related issues, including code coverage, and the AOT and JIT backends.type-bugIncorrect behavior (everything from a crash to more subtle misbehavior)

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions