[Mlir-commits] [mlir] [mlir][sparse] fold explicit value during sparsification (PR #90530)

Yinying Li llvmlistbot at llvm.org
Mon Apr 29 15:43:18 PDT 2024


================
@@ -0,0 +1,75 @@
+// RUN: mlir-opt %s --linalg-generalize-named-ops \
+// RUN:             --sparsification-and-bufferization | FileCheck %s
+
+#CSR_ones_complex = #sparse_tensor.encoding<{
+  map = (d0, d1) -> (d0 : dense, d1 : compressed)
+// explicitVal = (1.0, 0.0) : complex<f32>,
+// implicitVal = (1.0, 0.0) : complex<f32>
----------------
yinying-lisa-li wrote:

(0.0, 0.0) for implicit value? I'm planing to add verification to fail anything else for complex implicit value.

https://github.com/llvm/llvm-project/pull/90530


More information about the Mlir-commits mailing list