[Mlir-commits] [mlir] [mlir][NVVM] Add InferTypeOpInterface to sync and ldmatrix ops (PR #188238)

[llvmlistbot at llvm.org]

github-actions[bot] wrote:

<!--LLVM CODE FORMAT COMMENT: {darker}-->


:warning: Python code formatter, darker found issues in your code. :warning:

<details>
<summary>
You can test this locally with the following command:
</summary>

``````````bash
darker --check --diff -r origin/main...HEAD mlir/test/python/dialects/nvvm.py
``````````

:warning:
The reproduction instructions above might return results for more than one PR
in a stack if you are using a stacked PR workflow. You can limit the results by
changing `origin/main` to the base branch/commit you want to compare against.
:warning:

</details>

<details>
<summary>
View the diff from darker here.
</summary>

``````````diff
--- nvvm.py	2026-03-24 12:21:10.000000 +0000
+++ nvvm.py	2026-03-24 12:52:09.490250 +0000
@@ -228,19 +228,19 @@
     i32 = T.i32()
     f32 = T.f32()
 
     @func.FuncOp.from_py_func(i32, i32, f32, i32, i32)
     def shfl_sync_ops(mask, i32val, f32val, offset, clamp):
-        i32_result = nvvm.shfl_sync(
-            mask, i32val, offset, clamp, nvvm.ShflKind.bfly
-        )
-        f32_result = nvvm.shfl_sync(
-            mask, f32val, offset, clamp, nvvm.ShflKind.bfly
-        )
+        i32_result = nvvm.shfl_sync(mask, i32val, offset, clamp, nvvm.ShflKind.bfly)
+        f32_result = nvvm.shfl_sync(mask, f32val, offset, clamp, nvvm.ShflKind.bfly)
         struct_result = nvvm.shfl_sync(
-            mask, i32val, offset, clamp, nvvm.ShflKind.bfly,
-            return_value_and_is_valid=True
+            mask,
+            i32val,
+            offset,
+            clamp,
+            nvvm.ShflKind.bfly,
+            return_value_and_is_valid=True,
         )
         return i32_result
 
 
 # CHECK-LABEL:   func.func @shfl_sync_ops(
@@ -260,16 +260,20 @@
     elt_b16 = Attribute.parse("#nvvm.ld_st_matrix_elt_type<b16>")
 
     @func.FuncOp.from_py_func(ptr_shared)
     def ldmatrix_ops(ptr):
         r1 = nvvm.ldmatrix(
-            ptr, num=1, layout=nvvm.MMALayout.row,
+            ptr,
+            num=1,
+            layout=nvvm.MMALayout.row,
             shape=shape_8x8,
             elt_type=elt_b16,
         )
         r4 = nvvm.ldmatrix(
-            ptr, num=4, layout=nvvm.MMALayout.row,
+            ptr,
+            num=4,
+            layout=nvvm.MMALayout.row,
             shape=shape_8x8,
             elt_type=elt_b16,
         )
         return r1
 

``````````

</details>


https://github.com/llvm/llvm-project/pull/188238