[Mlir-commits] [mlir] [MLIR][Linalg] Scalable Vectorization of Reduction on the Trailing Dimension (PR #97788)
Andrzej WarzyĆski
llvmlistbot at llvm.org
Sun Jul 21 09:09:00 PDT 2024
================
@@ -0,0 +1,134 @@
+// DEFINE: %{compile} = mlir-opt %s \
+// DEFINE: -transform-interpreter -test-transform-dialect-erase-schedule \
+// DEFINE: -one-shot-bufferize="bufferize-function-boundaries" -buffer-deallocation-pipeline -cse -canonicalize -convert-vector-to-scf -arm-sve-legalize-vector-storage \
+// DEFINE: -convert-vector-to-llvm="enable-arm-sve" -test-lower-to-llvm -o %t
+// DEFINE: %{entry_point} = reduce_1d_f32
+// DEFINE: %{run} = %mcr_aarch64_cmd %t -e %{entry_point} -entry-point-result=void --march=aarch64 --mattr="+sve"\
+// DEFINE: -shared-libs=%mlir_native_utils_lib_dir/libmlir_runner_utils%shlibext,%mlir_native_utils_lib_dir/libmlir_c_runner_utils%shlibext
+
+// RUN: %{compile}
+
+// RUN: %{run} | FileCheck %s --check-prefix=REDUCE
+
+// REDEFINE: %{entry_point} = generic_reduce_1d_f32
+// RUN: %{run} | FileCheck %s --check-prefix=GENERIC
+
+func.func @reduce_1d_f32() {
----------------
banach-space wrote:
Would you mind adding a test for `i16` as well? Or any integer value. Just to make sure that we check both FP and integers. Thanks!
https://github.com/llvm/llvm-project/pull/97788
More information about the Mlir-commits
mailing list