[clang] [CIR] add vsqrt and vsqrtq support (PR #192282)

[Andrzej Warzyński via cfe-commits]

================
@@ -2615,17 +2568,16 @@ CIRGenFunction::emitAArch64BuiltinExpr(unsigned builtinID, const CallExpr *expr,
   case NEON::BI__builtin_neon_vmul_laneq_v:
   case NEON::BI__builtin_neon_vpmaxnm_v:
   case NEON::BI__builtin_neon_vpmaxnmq_v:
+  case NEON::BI__builtin_neon_vpminnm_v:
+  case NEON::BI__builtin_neon_vpminnmq_v:
+  case NEON::BI__builtin_neon_vsqrth_f16:
     cgm.errorNYI(expr->getSourceRange(),
                  std::string("unimplemented AArch64 builtin call: ") +
                      getContext().BuiltinInfo.getName(builtinID));
     return mlir::Value{};
-  case NEON::BI__builtin_neon_vpminnm_v:
-  case NEON::BI__builtin_neon_vpminnmq_v:
-    intrName = "aarch64.neon.fminnmp";
-    return emitNeonCall(cgm, builder, {ty, ty}, ops, intrName, ty, loc);
-  case NEON::BI__builtin_neon_vsqrth_f16:
   case NEON::BI__builtin_neon_vsqrt_v:
   case NEON::BI__builtin_neon_vsqrtq_v:
+    return emitNeonCall(cgm, builder,{ty}, ops,"sqrt",ty, loc);
----------------
banach-space wrote:

```suggestion
    assert(!cir::MissingFeatures::emitConstrainedFPCall());
    return emitNeonCall(cgm, builder,{ty}, ops,"sqrt",ty, loc);
```

https://github.com/llvm/llvm-project/pull/192282