Segmentation fault in pyexamples/d3dHsm
Closed this issue · 1 comments
sballin commented
On multiple systems, doing python runcase.py in pyexamples/d3dHsm produces a segmentation fault when bbb.exmain() is run at this line. Here are some inconclusive results of debugging it, in case they make more sense to someone else.
Output of script
$ python runcase.py
Forthon edition
UEDGE $Name: V7_08_03 $
Wrote file "gridue" with runid: EFITD 09/07/90 # 66832 ,2384ms
***** Grid generation has been completed
Updating Jacobian, npe = 1
iter= 0 fnrm= 9.216531402973144 nfe= 1
nksol --- iterm = 1.
maxnorm(sf*f(u)) .le. ftol, where maxnorm() is
the maximum norm function. u is probably an
approximate root of f.
Interpolants created; mype = -1
Wrote file "gridue" with runid: EFITD 09/07/90 # 66832 ,2384ms
***** Grid generation has been completed
Updating Jacobian, npe = 1
iter= 0 fnrm= 9.216531400941010 nfe= 1
Updating Jacobian, npe = 2
Fatal Python error: Segmentation fault
Current thread 0x000000010d8f75c0 (most recent call first):
File "runcase.py", line 46 in <module>
[1] 92229 segmentation fault python runcase.py
I modified the script, so ignore the line numbers in the output. The segmentation fault happens at the line linked in the first paragraph.
lldb output
$ lldb /usr/local/bin/python
(lldb) target create "/usr/local/bin/python"
Traceback (most recent call last):
File "<input>", line 1, in <module>
File "/usr/local/Cellar/python@2/2.7.16/Frameworks/Python.framework/Versions/2.7/lib/python2.7/copy.py", line 52, in <module>
import weakref
File "/usr/local/Cellar/python@2/2.7.16/Frameworks/Python.framework/Versions/2.7/lib/python2.7/weakref.py", line 14, in <module>
from _weakref import (
ImportError: cannot import name _remove_dead_weakref
Current executable set to '/usr/local/bin/python' (x86_64).
(lldb) target stop-hook add
Enter your stop hook command(s). Type 'DONE' to end.
> bt
> disassemble --pc
Stop hook #1 added.
(lldb) run runcase.py
* thread #1, stop reason = signal SIGSTOP
* frame #0: 0x0000000100005000 dyld`_dyld_start
dyld`_dyld_start:
-> 0x100005000 <+0>: popq %rdi
0x100005001 <+1>: pushq $0x0
0x100005003 <+3>: movq %rsp, %rbp
0x100005006 <+6>: andq $-0x10, %rsp
Process 89747 launched: '/usr/local/bin/python' (x86_64)
* thread #2, stop reason = exec
* frame #0: 0x0000000100005000 dyld`_dyld_start
dyld`_dyld_start:
-> 0x100005000 <+0>: popq %rdi
0x100005001 <+1>: pushq $0x0
0x100005003 <+3>: movq %rsp, %rbp
0x100005006 <+6>: andq $-0x10, %rsp
Process 89747 stopped
* thread #2, stop reason = exec
frame #0: 0x0000000100005000 dyld`_dyld_start
dyld`_dyld_start:
-> 0x100005000 <+0>: popq %rdi
0x100005001 <+1>: pushq $0x0
0x100005003 <+3>: movq %rsp, %rbp
0x100005006 <+6>: andq $-0x10, %rsp
Target 0: (Python) stopped.
(lldb) c
Process 89747 resuming
Forthon edition
UEDGE $Name: V7_08_03 $
Wrote file "gridue" with runid: EFITD 09/07/90 # 66832 ,2384ms
***** Grid generation has been completed
Updating Jacobian, npe = 1
iter= 0 fnrm= 9.216531402973144 nfe= 1
nksol --- iterm = 1.
maxnorm(sf*f(u)) .le. ftol, where maxnorm() is
the maximum norm function. u is probably an
approximate root of f.
Interpolants created; mype = -1
Wrote file "gridue" with runid: EFITD 09/07/90 # 66832 ,2384ms
***** Grid generation has been completed
Updating Jacobian, npe = 1
iter= 0 fnrm= 9.216531400941010 nfe= 1
Updating Jacobian, npe = 2
* thread #2, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=1, address=0x114b09010)
* frame #0: 0x000000010d84a7ba uedgeC.so`daxpy_ + 298
frame #1: 0x000000010d84b6dc uedgeC.so`dgbfa_ + 1036
frame #2: 0x000000010d84cd3f uedgeC.so`dgbco_ + 383
frame #3: 0x000000010d6812e3 uedgeC.so`jac_lu_decomp_ + 467
frame #4: 0x000000010d707909 uedgeC.so`psetnk_ + 2153
frame #5: 0x000000010d8219ec uedgeC.so`model_ + 1580
frame #6: 0x000000010d822c9b uedgeC.so`nksol_ + 4091
frame #7: 0x000000010d75107c uedgeC.so`uedriv_ + 13932
frame #8: 0x000000010d742685 uedgeC.so`exmain_ + 405
frame #9: 0x000000010d57d437 uedgeC.so`bbb_exmain + 55
frame #10: 0x0000000100155f66 Python`PyEval_EvalFrameEx + 19200
frame #11: 0x000000010015124c Python`PyEval_EvalCodeEx + 1540
frame #12: 0x0000000100150c42 Python`PyEval_EvalCode + 32
frame #13: 0x0000000100172633 Python`run_mod + 49
frame #14: 0x00000001001726da Python`PyRun_FileExFlags + 130
frame #15: 0x0000000100172259 Python`PyRun_SimpleFileExFlags + 719
frame #16: 0x0000000100183cd4 Python`Py_Main + 3136
frame #17: 0x00007fff6d9fa3d5 libdyld.dylib`start + 1
frame #18: 0x00007fff6d9fa3d5 libdyld.dylib`start + 1
uedgeC.so`daxpy_:
-> 0x10d84a7ba <+298>: movsd 0x10(%rcx,%rdx), %xmm2 ; xmm2 = mem[0],zero
0x10d84a7c0 <+304>: movhpd 0x18(%rsi,%rdx), %xmm0 ; xmm0 = xmm0[0],mem[0]
0x10d84a7c6 <+310>: movsd (%rcx,%rdx), %xmm3 ; xmm3 = mem[0],zero
0x10d84a7cb <+315>: mulpd %xmm1, %xmm0
Process 89747 stopped
* thread #2, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=1, address=0x114b09010)
frame #0: 0x000000010d84a7ba uedgeC.so`daxpy_ + 298
uedgeC.so`daxpy_:
-> 0x10d84a7ba <+298>: movsd 0x10(%rcx,%rdx), %xmm2 ; xmm2 = mem[0],zero
0x10d84a7c0 <+304>: movhpd 0x18(%rsi,%rdx), %xmm0 ; xmm0 = xmm0[0],mem[0]
0x10d84a7c6 <+310>: movsd (%rcx,%rdx), %xmm3 ; xmm3 = mem[0],zero
0x10d84a7cb <+315>: mulpd %xmm1, %xmm0
Target 0: (Python) stopped.
pudb output
Segfault happens after line shown in _Forthon.py.
Reproducibility
Mac
- macOS 10.14
- Apple LLVM version 10.0.1 (clang-1001.0.46.4)
- GNU Fortran (GCC) 6.3.0
MIT Engaging cluster
- centOS 7
- GNU Fortran (GCC) 6.2.0
Red Hat virtual machine
- RHEL 7.6
- GNU Fortran (GCC) 4.8.5
Tried this because there is no segfault for Maxim on Red Hat 7.7 with gfortran 4.8.5. The segfault appears on the Red Hat VM I tested, so this bug doesn't seem to depend on the compiler version or operating system. Maybe it depends on certain environment variables being set?
Anaconda python
I've been using regular python and pip --user to install things, but I also tried Anaconda python and encountered the same issue.