[Mlir-commits] [mlir] [mlir][gpu] Add gpu.rotate operation (PR #142796)
Jakub Kuderski
llvmlistbot at llvm.org
Mon Jun 16 08:37:22 PDT 2025
================
@@ -1364,6 +1364,49 @@ def GPU_ShuffleOp : GPU_Op<
];
}
+def GPU_RotateOp : GPU_Op<
+ "rotate", [Pure, AllTypesMatch<["value", "rotateResult"]>]>,
+ Arguments<(ins AnyIntegerOrFloatOr1DVector:$value, I32:$offset, I32:$width)>,
+ Results<(outs AnyIntegerOrFloatOr1DVector:$rotateResult)> {
+ let summary = "Rotate values within a subgroup.";
+ let description = [{
+ The "rotate" op moves values across lanes (a.k.a., invocations, work items)
+ within the same subgroup. The `width` argument specifies the number of lanes
+ that participate in the rotation, and must be uniform across all lanes.
+ Further, the first `width` lanes of the subgroup must be active.
+
+ `width` must be a power of two, and `offset` must be in the range
+ `[0, width)`.
+
+ Return the `rotateResult` of the invocation whose id within the group is
+ calculated as follows:
+
+ Invocation ID = ((LocalId + Delta) & (width - 1)) + (LocalId & ~(width - 1))
+
+ Returns the `rotateResult` if the current lane id is smaller than `width`.
----------------
kuhar wrote:
What if the lane id is `>= width`?
https://github.com/llvm/llvm-project/pull/142796
More information about the Mlir-commits
mailing list