Thomas Raoux d77f483640 [mlir][gpu] Relax restriction on mma load/store op ...

Those ops can support more complex layout as long as the most inner
dimension is contiguous.

Differential Revision: https://reviews.llvm.org/D122452

2022-03-25 04:03:40 +00:00

..

all-reduce-max.mlir

[mlir] Split out a new ControlFlow dialect from Standard

2022-02-06 14:51:16 -08:00

all-reduce.mlir

[mlir] Split out a new ControlFlow dialect from Standard

2022-02-06 14:51:16 -08:00

async-region.mlir

[mlir] Support pre-existing tokens in 'gpu-async-region'

2021-06-10 08:43:45 +02:00

canonicalize.mlir

[MLIR] Replace std ops with arith dialect ops

2021-10-13 03:07:03 +00:00

invalid.mlir

[mlir][gpu] Relax restriction on mma load/store op

2022-03-25 04:03:40 +00:00

mapping.mlir

[MLIR] Replace std ops with arith dialect ops

2021-10-13 03:07:03 +00:00

multiple-all-reduce.mlir

[mlir] Replace StrEnumAttr -> EnumAttr in core dialects

2022-01-18 17:15:00 +00:00

ops.mlir

[mlir][gpu] Relax restriction on mma load/store op

2022-03-25 04:03:40 +00:00

outlining.mlir

[mlir][gpu] Split ops sinking from gpu-kernel-outlining pass into separate pass

2022-02-17 10:34:20 +03:00

promotion.mlir

[mlir] Replace StrEnumAttr -> EnumAttr in core dialects

2022-01-18 17:15:00 +00:00

sink-ops.mlir

[mlir][gpu] Split ops sinking from gpu-kernel-outlining pass into separate pass

2022-02-17 10:34:20 +03:00