search

c3sr / comm_scope

NUMA-aware multi-CPU multi-GPU data transfer benchmarks
https://github.com/c3sr/scope
Apache License 2.0
21 stars 3 forks source link

Use aligned_alloc in numa-memcpy/pinned_to_gpu #8

Closed cwpearson closed 6 years ago

cwpearson commented 6 years ago

https://github.com/c3sr/comm_scope/blob/bfaf31a8cb9ea8ae997ff0c93c0427fbf31b92be/src/numa-memcpy/pinned_to_gpu.cpp#L43