Sample program executable hanging Why is it hanging?

When I compile matrxMul by line commands I get an executable that will work. When I compile either singly or batch with the make command, I get a executable that will hang when it is invoked. The files are also of different sizes. The output shown should demonstrate what I am saying.

errol@fermi:~/NVIDIA_GPU_Computing_SDK/C/bin/linux/release$ ls -al matrixMul
-rwxr-xr-x 1 root root 173988 2010-04-26 10:07 matrixMul
errol@fermi:~/NVIDIA_GPU_Computing_SDK/C/bin/linux/release$ ./matrixMul
[ matrixMul ]
Device 0: “GeForce GTX 275” with Compute 1.3 capability

errol@fermi:~/NVIDIA_GPU_Computing_SDK/C/src/matrixMul$ ./matrixMul
[ matrixMul ]
./matrixMul Starting…

Device 0: “GeForce GTX 275” with Compute 1.3 capability

Using Matrix Sizes: A(80 x 160), B(80 x 80), C(80 x 160)

Run Kernels…

matrixMul, Throughput = 14.7586 GFlop/s, Time = 0.00014 s, Size = 2048000 Ops, NumDevsUsed = 1, Workgroup = 256

Check against Host computation…


Press to Quit…

errol@fermi:~/NVIDIA_GPU_Computing_SDK/C/src/matrixMul$ ls -al matrixMul
-rwxr-xr-x 1 errol errol 379803 2010-04-26 10:04 matrixMul
errol@fermi:~/NVIDIA_GPU_Computing_SDK/C/src/matrixMul$ matrixMul

I am unsure as to why it is hanging.


Only you can answer that, I suspect. The big clue is that the build process isn’t the same and the resulting executable isn’t the same, so what is different in the build process? You can’t seriously expect someone to answer that.

How long are you waiting before ctrl-c’ing? I ask because I’ve run into this mysterious problem where in the first execution after compiling the first cuda call takes an extra 12-14 secs and subsequent executions seem to fine after that (even after rebooting). Maybe if you wait long enough it will eventually complete.