Krzysztof Drewniak 051612c018

[mlir][ValueBounds] memref.dim and tensor.dim are always positive (#122804) ...

Add the constraint that the length of a memref or tensor dimension is
always non-negative (at least 0) even if we don't know which dimension
we're querying the length of.

2025-01-13 17:48:03 -06:00

..

bubble-reshapes.mlir

…

bufferize.mlir

…

canonicalize.mlir

[mlir][tensor] Introduce FoldTensorCastUnPackOp (#121393)

2025-01-03 09:09:23 +00:00

decompose-concat.mlir

…

drop-redundant-insert-slice-rank-expansion.mlir

…

extract-slice-from-collapse-shape.mlir

…

fold-consecutive-insert-extract-slice.mlir

…

fold-constant-extract-slice.mlir

…

fold-empty-op.mlir

…

fold-into-pack-and-unpack.mlir

…

fold-reassociative-reshapes.mlir

…

fold-tensor-subset-ops-into-vector-transfers.mlir

…

fold-tensor-subset-ops.mlir

…

invalid.mlir

[mlir][tensor] Introduce FoldTensorCastUnPackOp (#121393)

2025-01-03 09:09:23 +00:00

mesh-spmdization.mlir

…

one-shot-bufferize-encodings.mlir

…

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

…

one-shot-bufferize.mlir

…

ops.mlir

…

resolve-shaped-type-result-dims.mlir

…

rewrite-as-constant.mlir

…

simplify-pack-unpack.mlir

…

tiling.mlir

…

tracking-listener.mlir

…

transform-op-casting.mlir

…

transform-op-make-loop-independent.mlir

…

value-bounds-op-interface-impl.mlir

[mlir][ValueBounds] memref.dim and tensor.dim are always positive (#122804)

2025-01-13 17:48:03 -06:00