diff options
| author | Henri Vasserman <henv@hot.ee> | 2023-04-29 02:31:56 +0300 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-04-29 01:31:56 +0200 |
| commit | b1ee8f59b4101b46999a0995d9a34506f7285466 (patch) | |
| tree | 8d9fced5966780e5ea3b95741029590cef2c2754 /ggml-cuda.h | |
| parent | 36d19a603b221d1bd7897fcb10e823e2103b052d (diff) | |
cuBLAS: non-contiguous tensor support (#1215)
* Cuda: non-contiguous tensor support
* remove extra stuff
* rename
* fix error
* more fixes, now OpenBLAS and CLBlast build too
* now then?
Diffstat (limited to 'ggml-cuda.h')
| -rw-r--r-- | ggml-cuda.h | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/ggml-cuda.h b/ggml-cuda.h index b105ed0..1fd67eb 100644 --- a/ggml-cuda.h +++ b/ggml-cuda.h @@ -1,5 +1,6 @@ #include <cublas_v2.h> #include <cuda_runtime.h> +#include "ggml.h" #ifdef __cplusplus extern "C" { @@ -38,6 +39,8 @@ void dequantize_row_q5_0_cuda(const void * vx, float * y, int k, cudaStream_t st void dequantize_row_q5_1_cuda(const void * vx, float * y, int k, cudaStream_t stream); void dequantize_row_q8_0_cuda(const void * vx, float * y, int k, cudaStream_t stream); +cudaError_t ggml_cuda_h2d_tensor_2d(void * dst, const struct ggml_tensor * src, uint64_t i3, uint64_t i2, cudaStream_t stream); + #ifdef __cplusplus } #endif |