ggml : use __builtin_amdgcn_sudot4 in __dp4a for gfx11 (llama/4787)

ggml-cuda.cu

@@ -183,7 +183,7 @@ static __device__ __forceinline__ int __vsubss4(const int a, const int b) {
 static __device__ __forceinline__ int __dp4a(const int a, const int b, int c) {
 #if defined(__gfx906__) || defined(__gfx908__) || defined(__gfx90a__) || defined(__gfx1030__)
     c = __builtin_amdgcn_sdot4(a, b, c, false);
-#elif defined(__gfx1100__)
+#elif defined(RDNA3)
     c = __builtin_amdgcn_sudot4( true, a, true, b, c, false);
 #elif defined(__gfx1010__) || defined(__gfx900__)
     int tmp1;