diff options
| author | Chao Chen <chao.chen@intel.com> | 2025-05-12 09:16:21 -0500 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-05-12 09:16:21 -0500 |
| commit | db42345dc660329e34fd119fc8edab74521f7c06 (patch) | |
| tree | 5d9bded65816408b64fa6801d7a25592bbd45d97 /llvm/lib/IR/DIBuilder.cpp | |
| parent | d39ca81fdda0ad4700e6a16671ce435b9c10973e (diff) | |
| download | llvm-db42345dc660329e34fd119fc8edab74521f7c06.zip llvm-db42345dc660329e34fd119fc8edab74521f7c06.tar.gz llvm-db42345dc660329e34fd119fc8edab74521f7c06.tar.bz2 | |
[MLIR][XeGPU] Add unroll patterns for XeGPU (1/N) (#137010)
Similar to vector ops, XeGPU ops need to be unrolled into smaller shapes
such that they can be dispatched into a hardware instruction. This PR
marks the initial phase of a series dedicated to incorporating unroll
patterns for XeGPU operations. In this installment, we introduce
patterns for the following operations:
1. createNd
2. updateNd
3. prefetchNd
4. loadNd
5. storeNd
6. dpas
Diffstat (limited to 'llvm/lib/IR/DIBuilder.cpp')
0 files changed, 0 insertions, 0 deletions