[Mlir-commits] [mlir] [mlir][nvgpu] add `nvgpu.rcp` op (PR #100965)

[Mehdi Amini]

================
@@ -20,6 +20,7 @@
 #ifndef NVGPU
 #define NVGPU
 
+include "mlir/Interfaces/InferTypeOpInterface.td"
----------------
joker-eph wrote:

> My point was about inferring the result type. When someone sees an IR like %out = nvgpu.rcp_approx %in : vector<32x16xf32>, the return type is unclear, making the IR unreadable. To

This is how the entirety of the arith dialect is setup: is the return type **must** be the same as the input type this should be the common idiom for formatting dialects upstream IMO.

https://github.com/llvm/llvm-project/pull/100965