并行计算浮点报错
-
目前基于fe40下的waves2foam设置的算例报错,模型是一个简单的水槽造波,单核可以运算。并行运算到某个时间步库朗数突然增大,造成浮点溢出,报错信息如下:
集群error文件如下:mkdir: cannot create directory 鈥?鈥? File exists [node03:197162] 3 more processes have sent help message help-mpi-btl-openib.txt / no device params found [node03:197162] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages [node03:197162] 3 more processes have sent help message help-mpi-btl-openib.txt / init-fail-create-q [node03:197201] *** Process received signal *** [node03:197202] *** Process received signal *** [node03:197202] Signal: Floating point exception (8) [node03:197202] Signal code: (-6) [node03:197202] Failing at address: 0x3ef00030252 [node03:197203] *** Process received signal *** [node03:197203] Signal: Floating point exception (8) [node03:197203] Signal code: (-6) [node03:197203] Failing at address: 0x3ef00030253 [node03:197204] *** Process received signal *** [node03:197204] Signal: Floating point exception (8) [node03:197204] Signal code: (-6) [node03:197204] Failing at address: 0x3ef00030254 [node03:197201] Signal: Floating point exception (8) [node03:197201] Signal code: (-6) [node03:197201] Failing at address: 0x3ef00030251 [node03:197201] [ 0] [node03:197204] [ 0] /lib64/libc.so.6(+0x362f0)[0x2b51232f12f0] [node03:197204] [ 1] [node03:197202] [ 0] /lib64/libc.so.6(+0x362f0)[0x2b517e1252f0] [node03:197202] [ 1] [node03:197203] [ 0] /lib64/libc.so.6(+0x362f0)[0x2aecd7a942f0] [node03:197203] [ 1] /lib64/libc.so.6(+0x362f0)[0x2b6690c022f0] [node03:197201] [ 1] /lib64/libc.so.6(gsignal+0x37)[0x2b6690c02277] [node03:197201] [ 2] /lib64/libc.so.6(gsignal+0x37)[0x2aecd7a94277] [node03:197203] [ 2] /lib64/libc.so.6(gsignal+0x37)[0x2b51232f1277] [node03:197204] [ 2] /lib64/libc.so.6(gsignal+0x37)[0x2b517e125277] [node03:197202] [ 2] /lib64/libc.so.6(+0x362f0)[0x2aecd7a942f0] [node03:197203] [ 3] /lib64/libc.so.6(+0x362f0)[0x2b51232f12f0] [node03:197204] [ 3] /lib64/libc.so.6(+0x362f0)[0x2b517e1252f0] [node03:197202] [ 3] /lib64/libc.so.6(+0x362f0)[0x2b6690c022f0] [node03:197201] [ 3] /home/DSQ/foam/foam-extend-4.0/lib/linux64GccDPOpt/libfoam.so(_ZN4Foam4Time12adjustDeltaTEv+0x59)[0x2b5121dcc429] [node03:197204] [ 4] /home/DSQ/foam/foam-extend-4.0/lib/linux64GccDPOpt/libfoam.so(_ZN4Foam4Time12adjustDeltaTEv+0x59)[0x2aecd656f429] [node03:197203] [ 4] /home/DSQ/foam/foam-extend-4.0/lib/linux64GccDPOpt/libfoam.so(_ZN4Foam4Time12adjustDeltaTEv+0x59)[0x2b668f6dd429] [node03:197201] [ 4] /home/DSQ/foam/foam-extend-4.0/lib/linux64GccDPOpt/libfoam.so(_ZN4Foam4Time12adjustDeltaTEv+0x59)[0x2b517cc00429] [node03:197202] [ 4] waveFoam(main+0x23be)[0x42be2e] [node03:197202] [ 5] waveFoam(main+0x23be)[0x42be2e] [node03:197201] [ 5] waveFoam(main+0x23be)[0x42be2e] [node03:197203] [ 5] waveFoam(main+0x23be)[0x42be2e] [node03:197204] [ 5] /lib64/libc.so.6(__libc_start_main+0xf5)[0x2b51232dd445] [node03:197204] [ 6] waveFoam[0x430b8e] [node03:197204] *** End of error message *** /lib64/libc.so.6(__libc_start_main+0xf5)[0x2b517e111445] [node03:197202] [ 6] waveFoam[0x430b8e] [node03:197202] *** End of error message *** /lib64/libc.so.6(__libc_start_main+0xf5)[0x2b6690bee445] [node03:197201] [ 6] waveFoam[0x430b8e] [node03:197201] *** End of error message *** /lib64/libc.so.6(__libc_start_main+0xf5)[0x2aecd7a80445] [node03:197203] [ 6] waveFoam[0x430b8e] [node03:197203] *** End of error message ***
我用fsifoam计算其他算例也会出现这个问题,有人遇到过类似的问题嘛?