forked from OSchip/llvm-project
5535696c38
The ops are very similar to the std variants, but support async GPU execution. gpu.alloc does not currently support an alignment attribute, and the new ops do not have canonicalizers/folders like their std siblings do. Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D91698 |
||
---|---|---|
.. | ||
lower-alloc-to-gpu-runtime-calls.mlir | ||
lower-launch-func-to-gpu-runtime-calls.mlir | ||
lower-wait-to-gpu-runtime-calls.mlir | ||
memory-attrbution.mlir |