Oleksandr "Alex" Zinenko 9d11accf95

[mlir] move if-condition propagation to a standalone pass (#150278) ...

This offers a significant speedup over running this as a
canonicalizaiton pattern, up to 10x improvement when running on large
(>100k operations) inputs coming from Polygeist.

It is also not clear whether this transformation is a reasonable
canonicalization as it performs non-local rewrites.

2025-07-23 21:02:40 +02:00

..

buffer-deallocation.mlir

…

bufferize.mlir

[mlir][bufferization][NFC] Rename to_memref to to_buffer (#137180)

2025-05-14 11:17:09 +02:00

canonicalize.mlir

[mlir] move if-condition propagation to a standalone pass (#150278)

2025-07-23 21:02:40 +02:00

control-flow-sink.mlir

…

for-loop-canonicalization.mlir

…

for-loop-peeling-front.mlir

…

for-loop-peeling.mlir

…

for-loop-specialization.mlir

…

for-loop-to-while-loop.mlir

…

forall-to-for.mlir

…

forall-to-parallel.mlir

…

foreach-thread-canonicalization.mlir

…

if-cond-prop.mlir

[mlir] move if-condition propagation to a standalone pass (#150278)

2025-07-23 21:02:40 +02:00

invalid.mlir

[mlir][SCF][GPU] Add DeviceMaskingAttrInterface (#146943)

2025-07-07 18:06:41 +02:00

loop-pipelining.mlir

[MLIR][SCF] fix loop pipelining pass use of uninitialized value (#146991)

2025-07-05 12:21:28 -07:00

loop-range.mlir

[mlir][SCF] Multiply lower bound in loop range folding (#111875)

2024-10-14 20:15:12 +02:00

loop-unroll.mlir

[mlir][affine]make affine-loop-unroll a FunctionOpInterface pass. (#126475)

2025-02-13 15:32:50 +08:00

one-shot-bufferize-allow-return-allocs-no-deallocs.mlir

…

one-shot-bufferize-analysis.mlir

…

one-shot-bufferize-encodings.mlir

[mlir][bufferization][NFC] Rename to_memref to to_buffer (#137180)

2025-05-14 11:17:09 +02:00

one-shot-bufferize-invalid.mlir

…

one-shot-bufferize-tensor-copy-insertion.mlir

…

one-shot-bufferize.mlir

[mlir][Bufferization] Do not have read semantics for destination of tensor.parallel_insert_slice. (#134169)

2025-04-03 09:47:36 -07:00

ops.mlir

…

parallel-loop-fusion.mlir

…

parallel-loop-invalid.mlir

…

parallel-loop-specialization.mlir

…

parallel-loop-tiling-inbound-check.mlir

…

parallel-loop-tiling.mlir

…

transform-loop-fuse-sibling.mlir

…

transform-op-coalesce.mlir

[mlir][SCF] Use Affine ops for indexing math. (#108450)

2024-09-27 18:25:41 -07:00

transform-op-forall-to-for.mlir

…

transform-op-forall-to-parallel.mlir

…

transform-op-take-assumed-branch.mlir

…

transform-ops-invalid.mlir

…

transform-ops.mlir

…

uplift-while.mlir

[mlir][scf] Allow different forwarding ordering in uplift

2025-03-27 18:09:07 -04:00

value-bounds-op-interface-impl.mlir

[mlir][scf]: Add value bound for the computed upper bound of for loop (#126426)

2025-02-10 11:35:02 +02:00

while-op-builder.mlir

…

wrap-while-loop-in-zero-trip-check.mlir

[MLIR][SCF] Define -scf-rotate-while pass (#99850)

2024-07-30 10:06:01 +02:00