786cbb09ed
This reverts commit d0650d1089.
Original commit message:
Subviews are supposed to be expanded before we hit the lowering
code.
The expansion is done with the pass called
expand-strided-metadata.
Add a test that demonstrate how these passes can be linked up to achieve
the desired lowering.
This patch is NFC in spirit but not in practice because `subview` gets
lowered into `reinterpret_cast(extract_strided_metadata, <some math>)`
which lowers in two memref descriptors (one for `reinterpert_cast` and
one for `extract_strided_metadata`), which creates some noise of the
form: `extractvalue(unrealized_cast(extractvalue[0]))[0]` that is
currently not simplified within MLIR but that is really just noop in
that case.
Differential Revision: https://reviews.llvm.org/D136377
25 lines
1.1 KiB
MLIR
25 lines
1.1 KiB
MLIR
// RUN: mlir-opt %s -linalg-bufferize \
|
|
// RUN: -arith-bufferize -tensor-bufferize -func-bufferize \
|
|
// RUN: -finalizing-bufferize -buffer-deallocation \
|
|
// RUN: -convert-linalg-to-loops -convert-scf-to-cf -convert-linalg-to-llvm -expand-strided-metadata -lower-affine -convert-arith-to-llvm --convert-memref-to-llvm -convert-func-to-llvm -reconcile-unrealized-casts | \
|
|
// RUN: mlir-cpu-runner -e main -entry-point-result=void \
|
|
// RUN: -shared-libs=%mlir_lib_dir/libmlir_runner_utils%shlibext \
|
|
// RUN: | FileCheck %s
|
|
|
|
func.func @main() {
|
|
%const = arith.constant dense<10.0> : tensor<2xf32>
|
|
%insert_val = arith.constant dense<20.0> : tensor<1xf32>
|
|
%inserted = tensor.insert_slice %insert_val into %const[0][1][1] : tensor<1xf32> into tensor<2xf32>
|
|
|
|
%unranked = tensor.cast %inserted : tensor<2xf32> to tensor<*xf32>
|
|
call @printMemrefF32(%unranked) : (tensor<*xf32>) -> ()
|
|
|
|
// CHECK: Unranked Memref base@ = {{0x[-9a-f]*}}
|
|
// CHECK-SAME: rank = 1 offset = 0 sizes = [2] strides = [1] data =
|
|
// CHECK-NEXT: [20, 10]
|
|
|
|
return
|
|
}
|
|
|
|
func.func private @printMemrefF32(%ptr : tensor<*xf32>)
|