79ff70fda2
sparse_tensor ops cannot be bufferized with One-Shot Bufferize. (They can only be analyzed.) The sparse compiler does the actual lowering to memref. Produce a proper error message instead of crashing. This fixes #61311. Differential Revision: https://reviews.llvm.org/D158728
13 lines
527 B
MLIR
13 lines
527 B
MLIR
// RUN: mlir-opt %s -one-shot-bufferize -verify-diagnostics
|
|
|
|
#SparseVector = #sparse_tensor.encoding<{
|
|
lvlTypes = ["compressed"]
|
|
}>
|
|
|
|
func.func @sparse_tensor_op(%arg0: tensor<64xf32, #SparseVector>) -> tensor<64xf32, #SparseVector> {
|
|
// expected-error @below{{sparse_tensor ops must be bufferized with the sparse compiler}}
|
|
// expected-error @below{{failed to bufferize op}}
|
|
%0 = sparse_tensor.convert %arg0 : tensor<64xf32, #SparseVector> to tensor<64xf32, #SparseVector>
|
|
return %0 : tensor<64xf32, #SparseVector>
|
|
}
|