[Bug libgomp/82555] New: SPECcpu201 Wrf_s deadlock

[kugan at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82555

            Bug ID: 82555
           Summary: SPECcpu201 Wrf_s deadlock
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libgomp
          Assignee: unassigned at gcc dot gnu.org
          Reporter: kugan at gcc dot gnu.org
                CC: jakub at gcc dot gnu.org
  Target Milestone: ---

Wrf_s is hanging or deadlocks when run on 48 threads (cores). It doesnt always
happen and I have to run with --iterations=111 and it will eventually happens.
Sometimes in the 2nd iterations and some times much later.

I attached the process to gdb and the back trace is:
(gdb) bt
#0  0x0000000001019924 in __lll_lock_wait (futex=futex@entry=0x2c3b1e0
<_gfortrani_unit_lock>, private=0) at lowlevellock.c:43
#1  0x0000000001012cbc in __pthread_mutex_lock (mutex=0x2c3b1e0
<_gfortrani_unit_lock>) at pthread_mutex_lock.c:80
#2  0x0000000000fd20ac in __gthread_mutex_lock (__mutex=0x2c3b1e0
<_gfortrani_unit_lock>) at ../libgcc/gthr-default.h:748
#3  _gfortrani_close_units () at ../../../gcc-exp2/libgfortran/io/unit.c:835
#4  0x000000000103950c in __libc_csu_fini ()
#5  0x000000000103f068 in __run_exit_handlers ()
#6  0x000000000103f0b0 in exit ()
#7  0x0000000000fc6e60 in _gfortrani_exit_error (status=1, status@entry=3) at
../../../gcc-exp2/libgfortran/runtime/error.c:196
#8  0x0000000000fc7314 in _gfortrani_internal_error
(cmp=cmp@entry=0xffffcdf23d00, 
    message=message@entry=0x11548a8 "stash_internal_unit(): Stack Size
Exceeded") at ../../../gcc-exp2/libgfortran/runtime/error.c:422
#9  0x0000000000fd1a84 in _gfortrani_stash_internal_unit (dtp=0xffffcdf23d00)
at ../../../gcc-exp2/libgfortran/io/unit.c:549
#10 0x0000000000fd0f6c in _gfortran_st_write_done (dtp=0xffffcdf23d00) at
../../../gcc-exp2/libgfortran/io/transfer.c:4168
#11 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#12 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#13 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#14 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#15 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#16 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#17 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#18 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#19 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#20 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#21 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#22 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#23 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#24 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#25 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#26 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#27 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#28 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#29 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#30 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#31 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#32 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#33 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#34 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#35 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#36 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#37 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#38 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#39 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#40 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#41 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#42 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#43 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#44 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#45 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()
#46 0x0000000000db933c in __module_ra_rrtm_MOD_rrtmlwrad ()

I am running this on AArch64 but I dont think this is an AArch64 specific
issue. Is anyone else seeing this?