[llvm] [AArch64] Keep floating-point conversion in SIMD (PR #147707)

[Paul Walker via llvm-commits]

================
@@ -24026,6 +24026,66 @@ static SDValue combineBoolVectorAndTruncateStore(SelectionDAG &DAG,
                       Store->getMemOperand());
 }
 
+// Combine store (fp_to_int X) with to use vector semantics around the
+// conversion when NEON is available.
----------------
paulwalker-arm wrote:

Perhaps "Combine store (fp_to_int X) to use vector semantics around the conversion when NEON is available. This allows us to store the in-vector result directly without needing a GPR transfer."?

https://github.com/llvm/llvm-project/pull/147707