[Mlir-commits] [mlir] [MLIR][Linalg] Remove unary named ops (PR #220905)
Renato Golin
llvmlistbot at llvm.org
Fri Sep 4 08:04:46 PDT 2026
================
@@ -842,15 +842,15 @@ module attributes { transform.with_named_sequence } {
// expected-remark @below {{op result}}
// expected-note @below {{value handle points to an op result #0}}
// expected-remark @below {{single user}}
- linalg.negf ins(%2 : tensor<42x42xf32>) outs(%0 : tensor<42x42xf32>) -> tensor<42x42xf32>
+ linalg.elementwise kind=#linalg.elementwise_kind<negf> ins(%2 : tensor<42x42xf32>) outs(%0 : tensor<42x42xf32>) -> tensor<42x42xf32>
// expected-remark @below {{matched result value}}
// expected-remark @below {{op result}}
// expected-note @below {{value handle points to an op result #0}}
- linalg.exp ins(%3 : tensor<42x42xf32>) outs(%0 : tensor<42x42xf32>) -> tensor<42x42xf32>
+ linalg.elementwise kind=#linalg.elementwise_kind<abs> ins(%3 : tensor<42x42xf32>) outs(%0 : tensor<42x42xf32>) -> tensor<42x42xf32>
----------------
rengolin wrote:
It doesn't matter. This is an artifact of me trying to change from `linalg.exp` to `linalg.abs` when I was removing just the `exp` and later moving all to `linalg.elementwise`. I consider moving them, but none of the tests the actual op is load bearing, and they're there just to validate matchers, so I decided against changing it.
https://github.com/llvm/llvm-project/pull/220905
More information about the Mlir-commits
mailing list