Matlab crashes 2010a: Aborted (core dumped)
조회 수: 3 (최근 30일)
이전 댓글 표시
Hi we were trying to run GPUmat 0.270 on matlab in order to utilize the NVIDIA CUDA graphic card. GPUmat: http://gp-you.org/index.php?option=com_content&view=frontpage&Itemid=53
All CUDA SDK ran through C code, but GPUmat crashes matlab without any error message.
GPUstart.m and GPUmatSystemCheck.m both abort, I would appreciate any info on how I may troubleshoot this further, thank you!:
>> GPUmatSystemCheck * GPUmat system diagnostics * Running on -> "glnxa64" * Matlab ver. -> "7.10.0.499 (R2010a)" * GPUmat version -> 0.270 * GPUmat build -> 13-Dec-2010 * GPUmat architecture -> "glnxa64"
* ARCHITECTURE TEST * GPUmat architecture test -> passed.
* CUDA TEST * CUDA CUBLAS -> installed. * CUDA CUFFT -> installed. * CUDA CUDART -> installed.
* GPUmat device check Aborted (core dumped)
Here is the matlab (7.10) crash dump:
[ 0] 0x00007f0b82987789 /data/apps/matlab/2010a/bin/glnxa64/../../bin/glnxa64/libmwfl.so+00522121 fl::sysdep::linux::unwind_stack(void const**, unsigned long, unsigned long)+000009 [ 1] 0x00007f0b8293194b /data/apps/matlab/2010a/bin/glnxa64/../../bin/glnxa64/libmwfl.so+00170315 fl::diag::stacktrace_base::capture(unsigned long)+000027 [ 2] 0x00007f0b81a08aa2 /data/apps/matlab/2010a/bin/glnxa64/../../bin/glnxa64/libmwmcr.so+00330402 [ 3] 0x00007f0b81a0904d /data/apps/matlab/2010a/bin/glnxa64/../../bin/glnxa64/libmwmcr.so+00331853 [ 4] 0x00007f0b80caa866 /data/apps/matlab/2010a/bin/glnxa64/../../sys/os/glnxa64/libstdc++.so.6+00780390 [ 5] 0x00007f0b80caa893 /data/apps/matlab/2010a/bin/glnxa64/../../sys/os/glnxa64/libstdc++.so.6+00780435 [ 6] 0x00007f0b80caa97a /data/apps/matlab/2010a/bin/glnxa64/../../sys/os/glnxa64/libstdc++.so.6+00780666 [ 7] 0x00007f0b03b80667 /usr/local/cuda/lib64/libcudart.so.3+00067175 [ 8] 0x00007f0b03dba680 /usr/local/cuda/lib64/libcudart.so.3+02401920
If this problem is reproducible, please submit a Service Request via: http://www.mathworks.com/support/contac ... est_1.html
A technical support engineer might contact you with further information.
Thank you for your help.
[ 0] 0x00007f0b82987789 /data/apps/matlab/2010a/bin/glnxa64/../../bin/glnxa64/libmwfl.so+00522121 fl::sysdep::linux::unwind_stack(void const**, unsigned long, unsigned long)+000009 [ 1] 0x00007f0b8293194b /data/apps/matlab/2010a/bin/glnxa64/../../bin/glnxa64/libmwfl.so+00170315 fl::diag::stacktrace_base::capture(unsigned long)+000027 [ 2] 0x00007f0b81a08aa2 /data/apps/matlab/2010a/bin/glnxa64/../../bin/glnxa64/libmwmcr.so+00330402 [ 3] 0x00007f0b81a0908e /data/apps/matlab/2010a/bin/glnxa64/../../bin/glnxa64/libmwmcr.so+00331918 [ 4] 0x00007f0b8054e4a0 /lib64/libpthread.so.0+00062624 [ 5] 0x00007f0b801f38f5 /lib64/libc.so.6+00207093 gsignal+000053 [ 6] 0x00007f0b801f50d5 /lib64/libc.so.6+00213205 abort+000373 [ 7] 0x00007f0b80caa86b /data/apps/matlab/2010a/bin/glnxa64/../../sys/os/glnxa64/libstdc++.so.6+00780395 [ 8] 0x00007f0b80caa893 /data/apps/matlab/2010a/bin/glnxa64/../../sys/os/glnxa64/libstdc++.so.6+00780435 [ 9] 0x00007f0b80caa97a /data/apps/matlab/2010a/bin/glnxa64/../../sys/os/glnxa64/libstdc++.so.6+00780666 [ 10] 0x00007f0b03b80667 /usr/local/cuda/lib64/libcudart.so.3+00067175 [ 11] 0x00007f0b03dba680 /usr/local/cuda/lib64/libcudart.so.3+02401920
If this problem is reproducible, please submit a Service Request via: http://www.mathworks.com/support/contac ... est_1.html
A technical support engineer might contact you with further information.
Thank you for your help.
Here is my card info:
./deviceQuery Starting...
CUDA Device Query (Runtime API) version (CUDART static linking)
There is 1 device supporting CUDA
Device 0: "GeForce GTX 570" CUDA Driver Version: 3.20 CUDA Runtime Version: 3.20 CUDA Capability Major/Minor version number: 2.0 Total amount of global memory: 1341325312 bytes Multiprocessors x Cores/MP = Cores: 15 (MP) x 32 (Cores/MP) = 480 (Cores) Total amount of constant memory: 65536 bytes Total amount of shared memory per block: 49152 bytes Total number of registers available per block: 32768 Warp size: 32 Maximum number of threads per block: 1024 Maximum sizes of each dimension of a block: 1024 x 1024 x 64 Maximum sizes of each dimension of a grid: 65535 x 65535 x 1 Maximum memory pitch: 2147483647 bytes Texture alignment: 512 bytes Clock rate: 1.46 GHz Concurrent copy and execution: Yes Run time limit on kernels: Yes Integrated: No Support host page-locked memory mapping: Yes Compute mode: Default (multiple host threads can use this device simultaneously) Concurrent kernel execution: Yes Device has ECC support enabled: No Device is using TCC driver mode: No
deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 3.20, CUDA Runtime Version = 3.20, NumDevs = 1, Device = GeForce GTX 570
PASSED
댓글 수: 0
답변 (2개)
Jill Reese
2011년 3월 4일
The MATLAB R2010b release of the Parallel Computing Toolbox contains support for NVIDIA GPU devices. If you are already licensed for the Parallel Computing Toolbox, then you may want to upgrade to R2010b.
That aside, if your R2010a installation of MATLAB worked properly before you integrated with GPUmat, then I would suggest contacting the GPUmat developers directly to correct the issue.
If your R2010a MATLAB installation crashes reproducibly even when you aren't using GPUmat, then (like Walter suggested) I think your best bet is to submit a service request using the link that appears below your crash dump.
댓글 수: 0
Walter Roberson
2011년 3월 4일
Did it ever work for you? If not then I would treat it as an "installation problem" and contact Mathworks support; there is free support for installation issues.
댓글 수: 2
Walter Roberson
2011년 3월 4일
Sorry, my response was based upon thinking that GPUmat was a Mathworks product.
참고 항목
카테고리
Help Center 및 File Exchange에서 GPU Computing에 대해 자세히 알아보기
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!