happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

ggml-cuda: disable gdn for musa (#20278)

This commit is contained in:
Aman Gupta 2026-03-09 16:15:36 +08:00 committed by GitHub
parent b518195101
commit e8bbc736cb
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 7 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
ggml/src/ggml-cuda/ggml-cuda.cu
View File

@ -4992,9 +4992,15 @@ static bool ggml_backend_cuda_device_supports_op(ggml_backend_dev_t dev, const g
case GGML_OP_LEAKY_RELU:
case GGML_OP_RWKV_WKV6:
case GGML_OP_GATED_LINEAR_ATTN:
case GGML_OP_GATED_DELTA_NET:
case GGML_OP_RWKV_WKV7:
return true;
case GGML_OP_GATED_DELTA_NET:
//TODO: enable once MUSA compiler is solved https://github.com/ggml-org/llama.cpp/pull/19504#issuecomment-4018634327
#ifdef GGML_USE_MUSA
return false;
#else
return true;
#endif // GGML_USE_MUSA
case GGML_OP_FLASH_ATTN_EXT:
return ggml_cuda_flash_attn_ext_supported(dev_ctx->device, op);
case GGML_OP_CROSS_ENTROPY_LOSS: