Please use GitHub pull requests for new patches. Avoid migrating existing patches. Phabricator shutdown timeline

Differential D147010 Diff 508837 mlir/test/Dialect/SparseTensor/codegen_sparse_dealloc.mlir

Changeset View

Standalone View

mlir/test/Dialect/SparseTensor/codegen_sparse_dealloc.mlir

This file was added.

				// RUN: mlir-opt %s --post-sparsification-rewrite="enable-runtime-library=false" \
				// RUN: --sparse-tensor-codegen=create-sparse-deallocs=false \
				// RUN: --canonicalize --cse \| FileCheck %s -check-prefix=CHECK-NO-DEALLOC

				// RUN: mlir-opt %s --post-sparsification-rewrite="enable-runtime-library=false" \
				// RUN: --sparse-tensor-codegen=create-sparse-deallocs=true \
				// RUN: --canonicalize --cse \| FileCheck %s -check-prefix=CHECK-DEALLOC

				#CSR = #sparse_tensor.encoding<{ dimLevelType = ["dense", "compressed"]}>
				#CSC = #sparse_tensor.encoding<{
				dimLevelType = ["dense", "compressed"],
				dimOrdering = affine_map<(i,j) -> (j,i)>
				}>

				//
				// No memref.dealloc is user-requested so
				// CHECK-NO-DEALLOC-LABEL: @sparse_convert_permuted
				aartbikUnsubmitted Not Done Reply Inline Actions user-requested aartbik: user-requested
				// CHECK-NO-DEALLOC-NOT: memref.dealloc
				aartbikUnsubmitted Not Done Reply Inline Actions can you add a CHECK-NO-DEALLOC-LABEL: @sparse_convert_permuted and CHECK-DEALLOC-LABEL: @sparse_convert_permuted It clearly works without, but it feels a bit cleaner to avoid matching output elsewhere if this every changes aartbik: can you add a CHECK-NO-DEALLOC-LABEL: @sparse_convert_permuted and CHECK-DEALLOC-LABEL…
				//
				// Otherwise memref.dealloc is created to free temporary sparse buffers.
				// CHECK-DEALLOC-LABEL: @sparse_convert_permuted
				// CHECK-DEALLOC: memref.dealloc
				//
				func.func @sparse_convert_permuted(%arg0: tensor<?x?xf32, #CSR>) -> tensor<?x?xf32, #CSC> {
				%0 = sparse_tensor.convert %arg0 : tensor<?x?xf32, #CSR> to tensor<?x?xf32, #CSC>
				return %0 : tensor<?x?xf32, #CSC>
				}