Krzysztof Drewniak 81950f6de4

[mlir][GPU] Bump static bound on cluster IDs (#187106) ...

Hardware (like AMD's gfx1250) allows 16 workgroups per cluster, but the
static bound of 8 from many years ago hasn't been updated. This commit
adds such an update and adds a test for that bound.

2026-03-18 11:33:20 -07:00

..

IR

[mlir][GPU] Bump static bound on cluster IDs (#187106)

2026-03-18 11:33:20 -07:00

Pipelines

[MLIR][XeGPU] Handle index element type in the layout propagation (#184322)

2026-03-11 15:25:44 +00:00

TransformOps

[mlir][gpu] Harden verification constraints in transform.gpu.map_nested_forall_to_threads (#170282)

2026-02-18 08:41:45 +09:00

Transforms

[mlir][GPU] Set nsw/nuw when expanding out subgroup ID (#187099)

2026-03-17 19:32:19 +00:00

Utils

[mlir] Use llvm accumulate wrappers. NFCI. (#162957)

2025-10-11 11:33:18 -04:00

CMakeLists.txt

[MLIR][GPU] Add xevm-attach-target transform pass. (#147372)

2025-07-10 15:44:26 -05:00