gmap / npb-gpu Goto Github PK

When I run your code in linux with cuda and gcc, I meet with the problem which describes "/usr/bin/ld: cannot open output file ../bin/ep.B: No such file or directory".
I dn not know the reason of it, so I need your help to assit me to deal with the problem.

I built and ran the FT program for the class D problem size, but the verification fails. The verification routine is failing because an overflow error is occurring with class D (int type is not large enough for this class). Would you please give me some help?

The definition of atomic operations for floating-pointing numbers of double precision must be included for old GPUs (compute capability < 6.0).

Hi.

It seems that instead of default mode, benchmarks also get an input file. For example, BT can get inputbt.data. I was wondering if there is any place where I can the input files for all benchmarks ?

Thanks

The read-me should include information about the GPU compute capability. This is important for the compilation of the benchmarks.

Hi, many thanks for this porting.

I tried it with CUDA 12.3 but all failed in the verification, could you please give me a hint on why this happened, please?
Thanks.

Also, what is the CUDA version that you have particularly used, please?
Thanks.

Hi

I built and ran the IS program for the class D problem size, but the verification fails. It is quite fast to run the programs with smaller problem sizes on a p100 gpu. Thank you for your updates.