musa: override warp_size of musa device to 32 (#12445)

Signed-off-by: Xiaodong Ye <redacted>

diff --git a/ggml/src/ggml-cuda/ggml-cuda.cu b/ggml/src/ggml-cuda/ggml-cuda.cu
index 8fb063822cfb74d5b4492615b99c3100c9d240b8..5cb56df9a81ae82e97d00b21952ac2438826540d 100644 (file)
--- a/ggml/src/ggml-cuda/ggml-cuda.cu
+++ b/ggml/src/ggml-cuda/ggml-cuda.cu
@@ -262,6 +262,8 @@ static ggml_cuda_device_info ggml_cuda_init() {
                       id, prop.name, prop.gcnArchName, info.devices[id].cc & 0xffff,
                       device_vmm ? "yes" : "no", prop.warpSize);
 #elif defined(GGML_USE_MUSA)
+        // FIXME: Ensure compatibility with varying warp sizes across different MUSA archs.
+        info.devices[id].warp_size = 32;
         // TODO: refine the .cc to reflect MUSA's actual CC capabilities
         info.devices[id].smpbo = prop.sharedMemPerBlockOptin;
         info.devices[id].cc = 100*prop.major + 10*prop.minor;