6 lines
131 B
Plaintext
Vendored
6 lines
131 B
Plaintext
Vendored
#include "common.cuh"
|
|
|
|
#define CUDA_PAD_BLOCK_SIZE 256
|
|
|
|
void ggml_cuda_op_pad(ggml_backend_cuda_context & ctx, ggml_tensor * dst);
|