[Smeagol-discuss] problem about the parallel using openmpi-1.2.6 and ifort

张广平 284107217 at qq.com
Tue Jan 5 14:12:32 GMT 2010


Dear smeaol user:     
     Now I use openmpi-1.2.6,mkl-10.0.2.018,ifort-10.1.012 to make my parallel smeagol,the compile is no problem,but when i use it in parallel mode using the command:
 mpiexec -n 8 smeagolpara <Auwire.fdf > mx.log &
problems appear:
    ---------------------------------------------------------------
 [1] 32513
[zgp at localhost mx]$ [localhost:32523] *** Process received signal ***
[localhost:32523] Signal: Segmentation fault (11)
[localhost:32523] Signal code: Address not mapped (1)
[localhost:32523] Failing at address: 0xfc
[localhost:32523] [ 0] /lib64/libpthread.so.0 [0x393160e7c0]
[localhost:32523] [ 1] smeagolpara(extrapol_+0x15c) [0x5304d0]
[localhost:32523] [ 2] smeagolpara(MAIN__+0xe650) [0x5ee570]
[localhost:32523] [ 3] smeagolpara(main+0x2a) [0x43f2c2]
[localhost:32523] [ 4] /lib64/libc.so.6(__libc_start_main+0xf4) [0x3930a1d994]
[localhost:32523] [ 5] smeagolpara [0x43f1e9]
[localhost:32523] *** End of error message ***
[localhost:32521] *** Process received signal ***
[localhost:32521] Signal: Segmentation fault (11)
[localhost:32521] Signal code: Address not mapped (1)
[localhost:32521] Failing at address: 0xfc
[localhost:32521] [ 0] /lib64/libpthread.so.0 [0x393160e7c0]
[localhost:32521] [ 1] smeagolpara(extrapol_+0x15c) [0x5304d0]
[localhost:32521] [ 2] smeagolpara(MAIN__+0xe650) [0x5ee570]
[localhost:32521] [ 3] smeagolpara(main+0x2a) [0x43f2c2]
[localhost:32521] [ 4] /lib64/libc.so.6(__libc_start_main+0xf4) [0x3930a1d994]
[localhost:32521] [ 5] smeagolpara [0x43f1e9]
[localhost:32521] *** End of error message ***
[localhost:32522] *** Process received signal ***
[localhost:32522] Signal: Segmentation fault (11)
[localhost:32522] Signal code: Address not mapped (1)
[localhost:32522] Failing at address: 0xfc
[localhost:32522] [ 0] /lib64/libpthread.so.0 [0x393160e7c0]
[localhost:32522] [ 1] smeagolpara(extrapol_+0x15c) [0x5304d0]
[localhost:32522] [ 2] smeagolpara(MAIN__+0xe650) [0x5ee570]
[localhost:32522] [ 3] smeagolpara(main+0x2a) [0x43f2c2]
[localhost:32522] [ 4] /lib64/libc.so.6(__libc_start_main+0xf4) [0x3930a1d994]
[localhost:32522] [ 5] smeagolpara [0x43f1e9]
[localhost:32522] *** End of error message ***
forrtl: error (78): process killed (SIGTERM)
Image              PC                Routine            Line        Source             
mca_btl_sm.so      00002AED7B363A0C  Unknown               Unknown  Unknown
mca_bml_r2.so      00002AED7AF5804A  Unknown               Unknown  Unknown
libopen-pal.so.0   00002AED75AC39BA  Unknown               Unknown  Unknown
libmpi.so.0        00002AED755FDE25  Unknown               Unknown  Unknown
mca_coll_tuned.so  00002AED7BD8803A  Unknown               Unknown  Unknown
libmpi.so.0        00002AED75610857  Unknown               Unknown  Unknown
libmpi_f77.so.0    00002AED753B67A5  Unknown               Unknown  Unknown
smeagolpara        00000000006E3E0E  Unknown               Unknown  Unknown
smeagolpara        00000000005EE919  Unknown               Unknown  Unknown
smeagolpara        000000000043F2C2  Unknown               Unknown  Unknown
libc.so.6          0000003930A1D994  Unknown               Unknown  Unknown
smeagolpara        000000000043F1E9  Unknown               Unknown  Unknown
forrtl: error (78): process killed (SIGTERM)
Image              PC                Routine            Line        Source             
libpthread.so.0    000000393160D610  Unknown               Unknown  Unknown
mca_oob_tcp.so     00002AB23DF65BFF  Unknown               Unknown  Unknown
mca_oob_tcp.so     00002AB23DF64996  Unknown               Unknown  Unknown
mca_oob_tcp.so     00002AB23DF64A65  Unknown               Unknown  Unknown
mca_oob_tcp.so     00002AB23DF66928  Unknown               Unknown  Unknown
libopen-pal.so.0   00002AB23C36DB97  Unknown               Unknown  Unknown
libopen-pal.so.0   00002AB23C3689FB  Unknown               Unknown  Unknown
libmpi.so.0        00002AB23BEA2E25  Unknown               Unknown  Unknown
mca_coll_tuned.so  00002AB24262D03A  Unknown               Unknown  Unknown
libmpi.so.0        00002AB23BEB5857  Unknown               Unknown  Unknown
libmpi_f77.so.0    00002AB23BC5B7A5  Unknown               Unknown  Unknown
smeagolpara        00000000006E3E0E  Unknown               Unknown  Unknown
smeagolpara        00000000005EE919  Unknown               Unknown  Unknown
smeagolpara        000000000043F2C2  Unknown               Unknown  Unknown
libc.so.6          0000003930A1D994  Unknown               Unknown  Unknown
smeagolpara        000000000043F1E9  Unknown               Unknown  Unknown
forrtl: error (78): process killed (SIGTERM)
Image              PC                Routine            Line        Source             
mca_btl_sm.so      00002B5424E78A10  Unknown               Unknown  Unknown
mca_bml_r2.so      00002B5424A6D04A  Unknown               Unknown  Unknown
libopen-pal.so.0   00002B541F5D89BA  Unknown               Unknown  Unknown
libmpi.so.0        00002B541F112E25  Unknown               Unknown  Unknown
mca_coll_tuned.so  00002B542589D03A  Unknown               Unknown  Unknown
libmpi.so.0        00002B541F125857  Unknown               Unknown  Unknown
libmpi_f77.so.0    00002B541EECB7A5  Unknown               Unknown  Unknown
smeagolpara        00000000006E3E0E  Unknown               Unknown  Unknown
smeagolpara        00000000005EE919  Unknown               Unknown  Unknown
smeagolpara        000000000043F2C2  Unknown               Unknown  Unknown
libc.so.6          0000003930A1D994  Unknown               Unknown  Unknown
smeagolpara        000000000043F1E9  Unknown               Unknown  Unknown
forrtl: error (78): process killed (SIGTERM)
Image              PC                Routine            Line        Source             
mca_btl_sm.so      00002AEBDD3B1A0C  Unknown               Unknown  Unknown
mca_bml_r2.so      00002AEBDCFA604A  Unknown               Unknown  Unknown
libopen-pal.so.0   00002AEBD7B119BA  Unknown               Unknown  Unknown
libmpi.so.0        00002AEBD764BE25  Unknown               Unknown  Unknown
mca_coll_tuned.so  00002AEBDDDD603A  Unknown               Unknown  Unknown
libmpi.so.0        00002AEBD765E857  Unknown               Unknown  Unknown
libmpi_f77.so.0    00002AEBD74047A5  Unknown               Unknown  Unknown
smeagolpara        00000000006E3E0E  Unknown               Unknown  Unknown
smeagolpara        00000000005EE919  Unknown               Unknown  Unknown
smeagolpara        000000000043F2C2  Unknown               Unknown  Unknown
libc.so.6          0000003930A1D994  Unknown               Unknown  Unknown
smeagolpara        000000000043F1E9  Unknown               Unknown  Unknown
forrtl: error (78): process killed (SIGTERM)
Image              PC                Routine            Line        Source             
mca_btl_sm.so      00002B956A9F29F4  Unknown               Unknown  Unknown
mca_bml_r2.so      00002B956A5E704A  Unknown               Unknown  Unknown
libopen-pal.so.0   00002B95651529BA  Unknown               Unknown  Unknown
libmpi.so.0        00002B9564C8CE25  Unknown               Unknown  Unknown
mca_coll_tuned.so  00002B956B41703A  Unknown               Unknown  Unknown
libmpi.so.0        00002B9564C9F857  Unknown               Unknown  Unknown
libmpi_f77.so.0    00002B9564A457A5  Unknown               Unknown  Unknown
smeagolpara        00000000006E3E0E  Unknown               Unknown  Unknown
smeagolpara        00000000005EE919  Unknown               Unknown  Unknown
smeagolpara        000000000043F2C2  Unknown               Unknown  Unknown
libc.so.6          0000003930A1D994  Unknown               Unknown  Unknown
smeagolpara        000000000043F1E9  Unknown               Unknown  Unknown
mpiexec noticed that job rank 5 with PID 32521 on node localhost exited on signal 11 (Segmentation fault). 
 -------------------------------------------------------------------
 Can't I use openmpi ? when I use the parallel as serial using command:smeagolpara <Auwire.fdf > mx.log &
it is OK!
 Can any one help me ?
 Any advice is welcome!
 BEST REGARDS!
                                                                  Guangping Zhang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.tchpc.tcd.ie/pipermail/smeagol-discuss/attachments/20100105/e49aa05d/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: arch.make
Type: application/octet-stream
Size: 1236 bytes
Desc: not available
Url : http://lists.tchpc.tcd.ie/pipermail/smeagol-discuss/attachments/20100105/e49aa05d/attachment-0001.obj 


More information about the Smeagol-discuss mailing list