[clang] [CIR][AArch64] Upstream vmull_*/vmull_high_* Neon builtins (PR #188371)

[Andrzej Warzyński via cfe-commits]

================
@@ -982,3 +982,220 @@ int64_t test_vshld_u64(int64_t a,int64_t b) {
   return (int64_t)vshld_u64(a, b);
 }
 
+//===------------------------------------------------------===//
+// 2.1.1.2.8. Widening Multiplication
----------------
banach-space wrote:

I see that you have also included https://arm-software.github.io/acle/neon_intrinsics/advsimd.html#polynomial-multiply, great! Could you move those to a dedicated section? Also, looks like `vmul_p8` and `vmul_high_p8` are missing.

https://github.com/llvm/llvm-project/pull/188371