Fix maximum number of threads per SM

lattice / quda

QUDA is a library for performing calculations in lattice QCD on GPUs.

https://lattice.github.io/quda

Other

287 stars 94 forks source link

Fix maximum number of threads per SM #1405

Closed hummingtree closed 11 months ago

hummingtree commented 11 months ago

The Mobius local operator uses launch bounds. However if blockDim * minBlock > maxThreadsPerSM, the STRICT build fails. This PR fixes this by not launching nor instantiating the failing cases.

Fixes #1403.