Bug 512371

Summary: FreeBSD arm64: regtest callgrind threads occasionally hangs
Product: [Developer tools] valgrind Reporter: Paul Floyd <pjfloyd>
Component: callgrindAssignee: Paul Floyd <pjfloyd>
Status: REPORTED ---    
Severity: normal    
Priority: NOR    
Version First Reported In: 3.27 GIT   
Target Milestone: ---   
Platform: FreeBSD Ports   
OS: FreeBSD   
Latest Commit: Version Fixed/Implemented In:
Sentry Crash Report:

Description Paul Floyd 2025-11-20 06:46:16 UTC 
This is with FreeBSD 14.3-RELEASE on a RPi 4b.

vgstack says

paulf> vgstack 41833
Thread 1 (Thread 103373 (tid 1 VgTs_Runnable)):
#0  atomic_cmpset_rel_32_llsc (p=0x4affee0, cmpval=2, newval=1) at /usr/obj/usr/src/arm64.aarch64/tmp/usr/include/machine/atomic.h:273
#1  atomic_cmpset_rel_32 (p=0x4affee0, cmpval=2, newval=1) at /usr/obj/usr/src/arm64.aarch64/tmp/usr/include/machine/atomic.h:273
#2  _thr_once (once_control=0x4affee0, init_routine=0x4a5f1c8 <exit_mutex_init_once>) at /usr/src/lib/libthr/thread/thr_once.c:97
#3  0x000000000499b778 in _once (once_control=0x0, init_routine=0x5012000) at /usr/src/lib/libc/gen/_once_stub.c:61
#4  0x0000000004a5f198 in exit (status=0) at /usr/src/lib/libc/stdlib/exit.c:82
#5  0x000000000498ff68 in __libc_start1 (argc=1, argv=0x1fc00007f8, env=0x1fc0000808, cleanup=<optimized out>, mainX=0x210704 <main>) at /usr/src/lib/libc/csu/libc_start1.c:157
#6  0x00000000002106c4 in _start () at /usr/src/lib/csu/aarch64/crt1_s.S:6

And the above does not seem to change.

And bstack says
Thread 1 (LWP 103373 of process 41833):
#0  0x00000000380abb4c in vgCallgrind_setup_bbcc (bb=0x100617eef0) at bbcc.c:697
#1  0x0000001002e3c784 in ?? ()
#2  0x00000000000193cd in ?? ()

The above does vary with samples.

So it looks like the guest is stuck.

I believe that jemalloc was updated recently. Need to look at that.