864a0b67a6
* CUDA: use mma PTX instructions for FlashAttention * __shfl_sync workaround for movmatrix * add __shfl_sync to HIP Co-authored-by: Diego Devesa <slarengh@gmail.com>
3 lines
115 B
Text
3 lines
115 B
Text
#include "common.cuh"
|
|
|
|
void ggml_cuda_flash_attn_ext_wmma_f16(ggml_backend_cuda_context & ctx, ggml_tensor * dst);
|