[Mlir-commits] [mlir] [mlir][gpu] Support `minimumnumf` and `maximumnumf` in GPU dialect (PR #226903)

[Lewis Crawford]

================
@@ -0,0 +1,21 @@
+// RUN: mlir-opt --convert-gpu-to-nvvm %s | FileCheck %s
+
+gpu.module @kernels {
+  // CHECK-LABEL: llvm.func @minimum_number
+  // CHECK: llvm.intr.minimumnum
+  // CHECK: llvm.return
+  gpu.func @minimum_number(%arg0 : f32, %out : memref<f32>) kernel {
+    %result = gpu.all_reduce minimumnumf %arg0 uniform {} : (f32) -> f32
----------------
LewisCrawford wrote:

Would it be worth adding tests for vector variants? Do they map to something like `llvm.vector.reduce.fmaximumnum`

https://github.com/llvm/llvm-project/pull/226903