numatop segfaults on the 4.14 kernel
by Jirka Hladky
Hello,
we experience numatop segfaults on 4.14 kernel. On older kernels, numatop
runs just fine.
HW: DELL server with 4x E5-4627 v2 @ 3.30GHz CPU
We run NAS Parallel Benchmarks (OMP variant)
https://www.nas.nasa.gov/publications/npb.html
OMP_NUM_THREADS=4 NPB_sources/bin/ft.C.x
and in another terminal, we start numatop monitoring with
ulimit -c unlimited (to allow coredump)
echo "N" | numatop -d /tmp/numatop.txt >/dev/null 2>&1 &
In about every 5th run we are getting the segfault:
[37530.811419] numatop[46404]: segfault at 100000038 ip 0000000000405c54 sp
00007f887bca3da0 error 4 in numatop[400000+20000]
$ gdb numatop core.46400
Reading symbols from
/mnt/tests/performance/NAS_parallel_benchmark-test/certification/numatop/numatop...done.
[New LWP 46404]
[New LWP 46406]
[New LWP 46400]
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
Core was generated by `numatop -d /tmp/numatop.txt'.
Program terminated with signal 11, Segmentation fault.
#0 proc_find_nolock (pid=-1) at common/proc.c:210
210 if (proc->pid == pid) {
$ uname -r
4.14.0-0.rc5.1.el7a.x86_64
Please let me know how to proceed. I would really really appreciate if
somebody could look into it.
Thanks a lot!
Jirka
4 years, 9 months
- 2
- 5
- ← Newer
- 1
- Older →