clang-p2996

Files

Matthias Springer 77124386fe [mlir][tensor] Add transform to make tensor.pad loop-independent

Add a transform to make `tensor.pad` and `tensor.empty` ops independent of SCF loop IVs. Such ops can then be hoisted.

E.g.:
```
scf.for %iv = %lb to %ub step %step {
  %high = affine.apply affine_map<(d0)[s0] -> (s0 - d0)> (%i)[%ub]
  %p = tensor.pad %t low[5] high[%high] ...
  ...
}
```
Is transformed to:
```
%high_new = affine.apply affine_map<()[s0, s1] -> (-s0 + s1)> ()[%lb, %ub]
%p_hoistable = tensor.pad %t low[5] high[%high_new]
%dim = tensor.dim %t, %c0
%size = affine.apply affine_map<(d0)[s0, s1] -> (-d0 + s0 + s1 + 5)>(%iv)[%ub, %dim]
%slice = tensor.extract_slice %p_hoistable [0] [%size] [1]
```

Differential Revision: https://reviews.llvm.org/D143910

2023-04-28 11:46:32 +09:00

bufferize.mlir

[mlir][tensor|memref] Harden the checks on dim op

2023-02-02 11:34:03 +01:00

canonicalize.mlir

[mlir] Fix folding into tensor.pad op.

2023-03-30 11:30:06 +02:00

extract-slice-from-collapse-shape.mlir

[mlir][scf] Rename ForeachThreadOp->ForallOp, PerformConcurrentlyOp->InParallelOp.

2023-02-17 09:59:39 +01:00

fold-consecutive-insert-extract-slice.mlir

[mlir] Add a pattern to fold single- and zero-iteration scf.forall ops.