It *looks* like a run-of-the-mill memory-badness kind of error, but
it's impossible to say without more information.
Are you able to run this through valgrind or some other memory-
checking debugger? It looks like the single process case may be the
simplest to check...?
On Aug 13, 2007, at 5:03 PM, Adams, Samuel D Contr AFRL/HEDR wrote:
I tried to run a code that I have running for a while now this
morning,
but for some reason it is causing segmentation faults. I can't really
think of anything that I have done recently that would be causing
these
errors. Does anyone have any idea?
I get this running it on more than one processor......
[sam@prodnode1 all]$ mpirun -np 2 --prefix
/usr/local/profiles/gcc-openmpi/ /home/sam/code/fdtd/fdtd_0.3/fdtd -t
/home/sam/code/fdtd/fdtd_0.3/test_files/tissue.txt -r
/home/sam/code/fdtd/fdtd_0.3/test_files/tester_x002y002z004.raw -v -f
3000 --pw 90,0,1,0 -l test_log.out -a 1
[prodnode1:04400] *** Process received signal ***
[prodnode1:04400] Signal: Segmentation fault (11)
[prodnode1:04400] Signal code: Invalid permissions (2)
[prodnode1:04400] Failing at address: 0x2aaaab000048
[prodnode1:04399] *** Process received signal ***
[prodnode1:04399] Signal: Segmentation fault (11)
[prodnode1:04399] Signal code: Invalid permissions (2)
[prodnode1:04399] Failing at address: 0x2aaaab0a0a48
[prodnode1:04400] [ 0] /lib64/libpthread.so.0 [0x3aa840dd40]
[prodnode1:04400] [ 1]
/usr/local/profiles/gcc-openmpi/lib/libopen-pal.so.0(_int_malloc
+0x2a5)
[0x2aaaaafda345]
[prodnode1:04400] [ 2]
/usr/local/profiles/gcc-openmpi/lib/libopen-pal.so.0(calloc+0xaa)
[0x2aaaaafdbd8a]
[prodnode1:04400] [ 3]
/home/sam/code/fdtd/fdtd_0.3/fdtd(parseTissues+0x23) [0x40c9d3]
[prodnode1:04400] [ 4]
/home/sam/code/fdtd/fdtd_0.3/fdtd(parseArgs+0x489) [0x404b09]
[prodnode1:04400] [ 5] /home/sam/code/fdtd/fdtd_0.3/fdtd(main+0x41)
[0x404eb1]
[prodnode1:04400] [ 6] /lib64/libc.so.6(__libc_start_main+0xf4)
[0x3aa781d8a4]
[prodnode1:04400] [ 7] /home/sam/code/fdtd/fdtd_0.3/fdtd [0x4034b9]
[prodnode1:04400] *** End of error message ***
[prodnode1:04399] [ 0] /lib64/libpthread.so.0 [0x3aa840dd40]
[prodnode1:04399] [ 1]
/usr/local/profiles/gcc-openmpi/lib/libopen-pal.so.0(_int_malloc
+0x2a5)
[0x2aaaaafda345]
[prodnode1:04399] [ 2]
/usr/local/profiles/gcc-openmpi/lib/libopen-pal.so.0(calloc+0xaa)
[0x2aaaaafdbd8a]
[prodnode1:04399] [ 3]
/home/sam/code/fdtd/fdtd_0.3/fdtd(parseTissues+0x23) [0x40c9d3]
[prodnode1:04399] [ 4]
/home/sam/code/fdtd/fdtd_0.3/fdtd(parseArgs+0x489) [0x404b09]
[prodnode1:04399] [ 5] /home/sam/code/fdtd/fdtd_0.3/fdtd(main+0x41)
[0x404eb1]
[prodnode1:04399] [ 6] /lib64/libc.so.6(__libc_start_main+0xf4)
[0x3aa781d8a4]
[prodnode1:04399] [ 7] /home/sam/code/fdtd/fdtd_0.3/fdtd [0x4034b9]
[prodnode1:04399] *** End of error message ***
mpirun noticed that job rank 0 with PID 4399 on node
prodnode1.brooks.af.mil exited on signal 11 (Segmentation fault).
1 additional process aborted (not shown)
--Or I get this if I run it on just one processor.
[sam@prodnode1 all]$ ./script2.sh [prodnode1:04405] *** Process
received
signal ***
[prodnode1:04405] Signal: Segmentation fault (11)
[prodnode1:04405] Signal code: Address not mapped (1)
[prodnode1:04405] Failing at address: 0x18
[prodnode1:04405] [ 0] /lib64/libpthread.so.0 [0x3aa840dd40]
[prodnode1:04405] [ 1] /home/sam/code/fdtd/fdtd_0.3/fdtd(calcMass
+0xac)
[0x40443c]
[prodnode1:04405] [ 2]
/home/sam/code/fdtd/fdtd_0.3/fdtd(parseArgs+0x5a1) [0x404c21]
[prodnode1:04405] [ 3] /home/sam/code/fdtd/fdtd_0.3/fdtd(main+0x41)
[0x404eb1]
[prodnode1:04405] [ 4] /lib64/libc.so.6(__libc_start_main+0xf4)
[0x3aa781d8a4]
[prodnode1:04405] [ 5] /home/sam/code/fdtd/fdtd_0.3/fdtd [0x4034b9]
[prodnode1:04405] *** End of error message ***
mpirun noticed that job rank 0 with PID 4405 on node
prodnode1.brooks.af.mil exited on signal 11 (Segmentation fault).
[sam@prodnode1 all]$
Sam Adams
General Dynamics Information Technology
Phone: 210.536.5945
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
--
Jeff Squyres
Cisco Systems