search

hpcgarage / spatter

Benchmark for measuring the performance of sparse and irregular memory access.
Other
75 stars 15 forks source link

Added fix to use nvhpc instead of cuda module #169

Closed jyoung3131 closed 10 months ago

jyoung3131 commented 10 months ago

Updated cuda runner to use the latest nvhpc, which is available more consistently across the CRNCH testbed.

jyoung3131 commented 10 months ago

Hi @plavin

It seems we get a linker error with the hilbert3d section of the code. Do you have a preference on how we fix it?

1) We can add -no-pie or -fPIE to the C++ compilation flags 2) We can deprecate the Hilbert tests if we don't think we need it with upcoming rewrites.

Linking CUDA executable spatter
/usr/bin/ld: CMakeFiles/spatter.dir/src/hilbert3d.c.o: relocation R_X86_64_32 against `.rodata' can not be used when making a PIE object; recompile with -fPIE
/usr/bin/ld: failed to set dynamic section sizes: bad value
collect2: error: ld returned 1 exit status