Matthias Springer 6040044f2f [mlir][vector] VectorToSCF: Omit redundant out-of-bounds check ...

There was a bug in `TransferWriteNonPermutationLowering`, a pattern that extends the permutation map of a TransferWriteOp with leading transfer dimensions of size ones. These newly added transfer dimensions are always in-bounds, because the starting point of any dimension is in-bounds. VectorToSCF inserts out-of-bounds checks based on the "in_bounds" attribute and dims that are marked as out-of-bounds but that are actually always in-bounds lead to unnecessary "scf.if" ops.

Differential Revision: https://reviews.llvm.org/D155196

2023-07-14 09:50:37 +02:00

..

bufferize-invalid.mlir

…

bufferize.mlir

…

canonicalize.mlir

[mlir][Vector] Disable 'vector.extract' folding for unsupported 0-D vectors

2023-06-01 22:22:15 +00:00

constant-fold.mlir

…

fold-arith-extf-into-vector-contract.mlir

[mlir][Vector] Adds a pattern to fold arith.extf into vector.contract

2023-06-05 23:22:20 +00:00

invalid.mlir

[mlir][vector] Refine diagnostic messages

2023-07-12 07:42:46 +01:00

lower-vector-mask.mlir

…

one-shot-bufferize.mlir

[mlir][vector][bufferize] Better analysis for vector.transfer_write

2023-06-14 09:38:51 +02:00

ops.mlir

[mlir][Vector] Add support for 0-D vectors to vector.insert/extract

2023-07-11 19:28:16 +00:00

scalar-vector-transfer-to-memref.mlir

[mlir][Vector] Prevent vector-to-scalar xfer patterns from triggering on sub-vectors

2023-06-01 22:22:16 +00:00

sink-vector-broadcast.mlir

[mlir][Vector] Add pattern to reorder elementwise and broadcast ops

2023-06-15 10:13:41 +01:00

test-create-broadcast.mlir

…

transform-vector.mlir

[mlir][transform] Improve transform.get_closest_isolated_parent

2023-07-04 16:23:08 +02:00

vector-break-down-bitcast.mlir

[mlir][vector] Add pattern to break down vector.bitcast

2023-04-25 20:18:02 -04:00

vector-broadcast-lowering-transforms.mlir

[mlir][transform] Use separate ops instead of PatternRegistry

2023-06-06 11:53:03 +02:00

vector-contract-matmul-transforms.mlir

[mlir][vector] Fix integer promotion type mismatch

2023-04-17 11:29:23 -07:00

vector-contract-matvec-transforms.mlir

[mlir][transform] SequenceOp: Top-level operations can be used as matchers

2023-06-19 09:06:18 +02:00

vector-contract-to-dot-transforms.mlir

[mlir][transform] Use separate ops instead of PatternRegistry

2023-06-06 11:53:03 +02:00

vector-contract-to-matrix-intrinsics-transforms.mlir

[mlir][transform] Use separate ops instead of PatternRegistry

2023-06-06 11:53:03 +02:00

vector-contract-to-outerproduct-transforms.mlir

[mlir][transform] Use separate ops instead of PatternRegistry

2023-06-06 11:53:03 +02:00

vector-contract-to-parallel-arith-transforms.mlir

[mlir][transform] Use separate ops instead of PatternRegistry

2023-06-06 11:53:03 +02:00

vector-dropleadunitdim-transforms.mlir

[vector] When trimming leading insertion dimensions, base the final result on the ranks

2023-04-18 18:49:29 +02:00

vector-emulate-narrow-type.mlir

[mlir] Narrow bitwidth emulation for vector.load

2023-07-11 13:38:15 -07:00

vector-extract-strided-slice-lowering.mlir

…

vector-gather-lowering.mlir

…

vector-mask-lowering-transforms.mlir

[mlir][transform] Use separate ops instead of PatternRegistry

2023-06-06 11:53:03 +02:00

vector-mem-transforms.mlir

…

vector-multi-reduction-lowering.mlir

[mlir][transform] SequenceOp: Top-level operations can be used as matchers

2023-06-19 09:06:18 +02:00

vector-multi-reduction-outer-lowering.mlir

[mlir][transform] SequenceOp: Top-level operations can be used as matchers

2023-06-19 09:06:18 +02:00

vector-outerproduct-lowering-transforms.mlir

[mlir][transform] Use separate ops instead of PatternRegistry

2023-06-06 11:53:03 +02:00

vector-reduce-to-contract.mlir

…

vector-scalable-outerproduct.mlir

[mlir][VectorType] Allow arbitrary dimensions to be scalable

2023-06-27 19:21:59 +01:00

vector-scan-transforms.mlir

…

vector-shape-cast-lowering-transforms.mlir

[mlir][transform] Use separate ops instead of PatternRegistry

2023-06-06 11:53:03 +02:00

vector-transfer-collapse-inner-most-dims.mlir

…

vector-transfer-drop-unit-dims-patterns.mlir

[mlir][transform] SequenceOp: Top-level operations can be used as matchers

2023-06-19 09:06:18 +02:00

vector-transfer-flatten.mlir

…

vector-transfer-full-partial-split-copy-transform.mlir

[mlir][transform] SequenceOp: Top-level operations can be used as matchers

2023-06-19 09:06:18 +02:00

vector-transfer-full-partial-split.mlir

[mlir][vector] Handle memory space conflicts in VectorTransferSplit patterns

2023-07-11 22:58:23 -04:00

vector-transfer-permutation-lowering.mlir

[mlir][vector] Fix bug in TransferWriteNonPermutationLowering

2023-07-10 17:21:03 +02:00

vector-transfer-to-vector-load-store.mlir

[mlir][vector] VectorToSCF: Omit redundant out-of-bounds check

2023-07-14 09:50:37 +02:00

vector-transfer-unroll.mlir

[mlir][vector] Add unroll pattern for vector.gather

2023-04-24 14:02:59 -04:00

vector-transferop-opt.mlir

…

vector-transforms.mlir

[mlir][Vector] Remove lhs and rhs masks from vector.contract

2023-03-29 19:53:29 +00:00

vector-transpose-lowering.mlir

[mlir][transform] SequenceOp: Top-level operations can be used as matchers

2023-06-19 09:06:18 +02:00

vector-unroll-options.mlir

…

vector-warp-distribute.mlir

[mlir][Vector] Fix a propagation bug with broadcast

2023-06-06 16:40:15 +02:00