diff options
author | Howard Su <howard0su@gmail.com> | 2023-06-12 20:44:16 +0800 |
---|---|---|
committer | GitHub <noreply@github.com> | 2023-06-12 14:44:16 +0200 |
commit | 58970a4c39124a647ac2a640d9e178ea6c961e65 (patch) | |
tree | 28c041035a12587390b4263667532cb416ce16dc /ggml-opencl.h | |
parent | 8c0a10e64dbf60fd9946c0cd5e6f59690800b123 (diff) |
Leverage mmap for offloading tensors to GPU (#1597)
* Rebase to latest
* Show progress
* Add assert to make sure we only allocate temp buffer for non-CPU backend tensor
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
---------
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
Diffstat (limited to 'ggml-opencl.h')
-rw-r--r-- | ggml-opencl.h | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/ggml-opencl.h b/ggml-opencl.h index bf95e5c..a92b445 100644 --- a/ggml-opencl.h +++ b/ggml-opencl.h @@ -18,8 +18,7 @@ void ggml_cl_host_free(void * ptr); void ggml_cl_free_data(const struct ggml_tensor* tensor); -void ggml_cl_transform_tensor(struct ggml_tensor * tensor); -void ggml_cl_load_data(const char * fname, struct ggml_tensor * tensor, size_t offset); +void ggml_cl_transform_tensor(void * data, struct ggml_tensor * tensor); #ifdef __cplusplus } |