musa: fix failures in test-backend-ops for mul_mat_id op (#15236)

* musa: fix failures in test-backend-ops for mul_mat_id op

Signed-off-by: Xiaodong Ye <redacted>
* Address review comments

Signed-off-by: Xiaodong Ye <redacted>
---------

Signed-off-by: Xiaodong Ye <redacted>

diff --git a/ggml/src/ggml-cuda/common.cuh b/ggml/src/ggml-cuda/common.cuh
index 2e5d48797fa49350b9fc731a37746304c56cc093..a23da57e3a1dcb9c12d44e6c3cb5b27dbc921ed0 100644 (file)
--- a/ggml/src/ggml-cuda/common.cuh
+++ b/ggml/src/ggml-cuda/common.cuh
@@ -312,11 +312,11 @@ static bool turing_mma_available(const int cc) {
 }
 
 static bool ampere_mma_available(const int cc) {
-    return cc < GGML_CUDA_CC_OFFSET_AMD && ggml_cuda_highest_compiled_arch(cc) >= GGML_CUDA_CC_AMPERE;
+    return GGML_CUDA_CC_IS_NVIDIA(cc) && ggml_cuda_highest_compiled_arch(cc) >= GGML_CUDA_CC_AMPERE;
 }
 
 static bool cp_async_available(const int cc) {
-    return cc < GGML_CUDA_CC_OFFSET_AMD && ggml_cuda_highest_compiled_arch(cc) >= GGML_CUDA_CC_AMPERE;
+    return GGML_CUDA_CC_IS_NVIDIA(cc) && ggml_cuda_highest_compiled_arch(cc) >= GGML_CUDA_CC_AMPERE;
 }
 
 static constexpr __device__ int ggml_cuda_get_physical_warp_size() {