[BUG] nil value field getGpuTopology
Closed this issue · 3 comments
Atrisan commented
getGpuTopology()
is a null value, in the likwid.lua, when executing the triadCU test with the GPUMarkerAPI via likwid-perfctr.
Installation:
Sources
Installation from Github branch: write_marker_file
Commit Hash: 3a54b98
Config
Install command
make && make install
Execution:
likwid-perfctr -G 0 -W FLOPS_DP -m -V 3 test/triadCU
Output:
Environment
Operating system: Rocky 8.9
CPU Setup:
===== Processor composition =====
Processor name : Intel(R) Xeon(R) Platinum 8160
Packages(sockets) : 2
Cores : 48
Processors(CPUs) : 48
Cores per package : 24
Threads per core : 1
GPU Setup:
2x Nvidia Tesla V100-SXM2 with CUDA 12.0 Driver
Compiler
CPU: GCC 11.3.0
GPU: CUDA 11.8
Application
triad.cu from the likwid testcases.
TomTheBear commented
Similar reason than the non-working markers. I renamed getGpuTopology
to getCudaTopology
but unfortunately not everywhere.
TomTheBear commented
This should now be fixed in the master
branch. It would be great if you could test it and comment on/close the issue.
Atrisan commented
Thanks, everything works now.