[CIG-SEISMO] error specfem3D_globe 7.0 with cuda5

[Ольга Усольцева]

Dear colleagues.
I have a problem with running of specfem3D_globe 7.0 with cuda5.
I send you texts of my error, file names gpu_device_info.txt and result of utility check_cuda_device.cu.
Could you explain me how to correct error.
Thank you.
 
My error is
 
Error testing memory allocation on device failed
Error rank 1: cudaMalloc failed: unknown error
 
ERROR: CUDA runtime error: cudaMalloc failed
 
 
Error: cudaMemGetInfo fails, unknown error
-------------------------------------------------------
Primary job terminated normally, but 1 process returned
a non-zero exit code.. Per user-direction, the job has been aborted.
-------------------------------------------------------
--------------------------------------------------------------------------
mpirun detected that one or more processes exited with non-zero status, thus causing
the job to be terminated. The first process to do so was:
 
Process name: [[6221,1],0]
Exit code: 1
---------------------------
 
This is information from OUTPUT_FILES/gpu_device_info.txt
 
Device Name = Tesla X2070
memory:
totalGlobalMem (in MB): 5375.437500
totalGlobalMem (in GB): 5.249451
totalConstMem (in bytes): 65536
Maximum 1D texture size (in bytes): 65536
sharedMemPerBlock (in bytes): 49152
regsPerBlock (in bytes): 32768
blocks:
Maximum number of threads per block: 1024
Maximum size of each dimension of a block: 1024 x 1024 x 64
Maximum sizes of each dimension of a grid: 65535 x 65535 x 65535
features:
Compute capability of the device = 2.0
multiProcessorCount: 14
canMapHostMemory: TRUE
deviceOverlap: TRUE
concurrentKernels: TRUE
 
This is a result of check_cuda_device.cu (mpi case).
process 1 found number of CUDA devices = 2
 
process 0 found number of CUDA devices = 2
 
 
GPU device for rank: 0
 
Device Name = Tesla X2070
memory:
totalGlobalMem (in MB): 5375.437500
totalGlobalMem (in GB): 5.249451
sharedMemPerBlock (in bytes): 49152
blocks:
Maximum number of threads per block: 1024
Maximum size of each dimension of a block: 1024 x 1024 x 64
Maximum sizes of each dimension of a grid: 65535 x 65535 x 65535
features:
Compute capability of the device = 2.0
multiProcessorCount: 14
canMapHostMemory: TRUE
deviceOverlap: TRUE
 
 
GPU device for rank: 1
 
Device Name = Tesla X2070
memory:
totalGlobalMem (in MB): 5375.437500
totalGlobalMem (in GB): 5.249451
sharedMemPerBlock (in bytes): 49152
blocks:
Maximum number of threads per block: 1024
Maximum size of each dimension of a block: 1024 x 1024 x 64
Maximum sizes of each dimension of a grid: 65535 x 65535 x 65535
features:
Compute capability of the device = 2.0
multiProcessorCount: 14
canMapHostMemory: TRUE
deviceOverlap: TRUE
 
0: GPU memory usage: used = 59.343750 MB, free = 5316.093750 MB, total = 5375.437500 MB
 
 
1: GPU memory usage: used = 59.343750 MB, free = 5316.093750 MB, total = 5375.437500 MB
 
number of total devices: 2
 
number of total devices: 2
 


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.geodynamics.org/pipermail/cig-seismo/attachments/20180328/1b20430b/attachment-0001.html>