[clang] [clang] Restrict use of scalar types in vector builtins (PR #119423)

[Eli Friedman via cfe-commits]

================
@@ -68,15 +66,18 @@ void test_builtin_elementwise_add_sat(float f1, float f2, double d1, double d2,
                                       long long int i2, si8 vi1, si8 vi2,
                                       unsigned u1, unsigned u2, u4 vu1, u4 vu2,
                                       _BitInt(31) bi1, _BitInt(31) bi2,
-                                      unsigned _BitInt(55) bu1, unsigned _BitInt(55) bu2) {
+                                      unsigned _BitInt(55) bu1, unsigned _BitInt(55) bu2,
+                                      char c1, char c2, unsigned char uc1,
+                                      unsigned char uc2, short s1, short s2,
+                                      unsigned short us1, unsigned short us2) {
   // CHECK:      [[I1:%.+]] = load i64, ptr %i1.addr, align 8
   // CHECK-NEXT: [[I2:%.+]] = load i64, ptr %i2.addr, align 8
   // CHECK-NEXT: call i64 @llvm.sadd.sat.i64(i64 [[I1]], i64 [[I2]])
   i1 = __builtin_elementwise_add_sat(i1, i2);
 
   // CHECK:      [[I1:%.+]] = load i64, ptr %i1.addr, align 8
   // CHECK-NEXT: call i64 @llvm.sadd.sat.i64(i64 [[I1]], i64 10)
-  i1 = __builtin_elementwise_add_sat(i1, 10);
+  i1 = __builtin_elementwise_add_sat(i1, (long long int)10);
----------------
efriedma-quic wrote:

It's a little awkward, but integer constants in C are awkward in general; not sure there's any better way to handle it.  You can write this more concisely as `10ll`.

https://github.com/llvm/llvm-project/pull/119423