CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-11 10:36:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

ggml: Re-enable CUDA graphs in presence of CONT and DUP nodes (#12970)

Browse Source
This commit is contained in:
Alan Gray
2025-04-17 14:19:42 +01:00
committed by GitHub
parent 7a395f67a7
commit 207c22ec2d
3 changed files with 7 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/src/ggml-cuda/cpy.cuh
View File

@@ -2,7 +2,7 @@
#define CUDA_CPY_BLOCK_SIZE 64
void ggml_cuda_cpy(ggml_backend_cuda_context & ctx, const ggml_tensor * src0, ggml_tensor * src1);
void ggml_cuda_cpy(ggml_backend_cuda_context & ctx, const ggml_tensor * src0, ggml_tensor * src1, bool disable_indirection = false);
void ggml_cuda_dup(ggml_backend_cuda_context & ctx, ggml_tensor * dst);