mirror of
https://github.com/intel/llvm.git
synced 2026-01-23 07:58:23 +08:00
70c2e0618a
This work adds `nvgpu.tma.async.load` Op that requests tma load asyncronusly using mbarrier object. It also creates nvgpu.tma.descriptor type. The type is supposed be created by `cuTensorMapEncodeTiled` cuda drivers api. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D155453