[llvm] Add `llvm.vector.partial.reduce.fadd` intrinsic (PR #159776)

[Matt Arsenault via llvm-commits]

================
@@ -20614,6 +20614,48 @@ performance, and an out-of-loop phase to calculate the final scalar result.
 By avoiding the introduction of new ordering constraints, these intrinsics
 enhance the ability to leverage a target's accumulation instructions.
 
+'``llvm.vector.partial.reduce.fadd.*``' Intrinsic
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+This is an overloaded intrinsic.
+
+::
+
+      declare <4 x f32> @llvm.vector.partial.reduce.fadd.v4f32.v8f32(<4 x f32> %a, <8 x f32> %b)
+      declare <vscale x 4 x f32> @llvm.vector.partial.reduce.fadd.nxv4f32.nxv8f32(<vscale x 4 x f32> %a, <vscale x 8 x f32> %b)
+
+Overview:
+"""""""""
+
+The '``llvm.vector.partial.reduce.fadd.*``' intrinsics reduce the
+concatenation of the two vector arguments down to the number of elements of the
+result vector type.
+
+Arguments:
+""""""""""
+
+The first argument is a floating-point vector with the same type as the result.
+
+The second argument is a vector with a length that is a known integer multiple
+of the result's type, while maintaining the same element type.
+
+Semantics:
+""""""""""
+
+Other than the reduction operator (e.g. fadd) the way in which the concatenated
+arguments is reduced is entirely unspecified. By their nature these intrinsics
+are not expected to be useful in isolation but instead implement the first phase
+of an overall reduction operation.
+
+The typical use case is loop vectorization where reductions are split into an
+in-loop phase, where maintaining an unordered vector result is important for
+performance, and an out-of-loop phase to calculate the final scalar result.
+
+By avoiding the introduction of new ordering constraints, these intrinsics
+enhance the ability to leverage a target's accumulation instructions.
----------------
arsenm wrote:

This is a recurring design mistake. e.g. originally the sqrt intrinsic was undefined for negative values. The net result is now the optimizer can't do anything with it, in the name of freedom for the backend to select it.

The only way to resolve this problem is to expose the full range of behaviors, and let optimization hints guide the backend to the preferred lowering. 

https://github.com/llvm/llvm-project/pull/159776