flash-attention/csrc/flash_attn_ck/flash_api.cpp

123 lines
8.1 KiB
C++

/******************************************************************************
* Copyright (c) 2024, Tri Dao.
******************************************************************************/
#include "flash_common.hpp"
std::vector<at::Tensor>
mha_fwd(at::Tensor &q, // batch_size x seqlen_q x num_heads x round_multiple(head_size, 8)
const at::Tensor &k, // batch_size x seqlen_k x num_heads_k x round_multiple(head_size, 8)
const at::Tensor &v, // batch_size x seqlen_k x num_heads_k x round_multiple(head_size, 8)
c10::optional<at::Tensor> &out_, // batch_size x seqlen_q x num_heads x round_multiple(head_size, 8)
c10::optional<at::Tensor> &alibi_slopes_, // num_heads or batch_size x num_heads
const float p_dropout,
const float softmax_scale,
bool is_causal,
int window_size_left,
int window_size_right,
const float softcap,
const bool return_softmax,
c10::optional<at::Generator> gen_);
std::vector<at::Tensor>
mha_varlen_fwd(at::Tensor &q, // total_q x num_heads x head_size, total_q := \sum_{i=0}^{b} s_i
const at::Tensor &k, // total_k x num_heads_k x head_size, total_k := \sum_{i=0}^{b} s_i or num_blocks x page_block_size x num_heads_k x head_size if there's a block_table.
const at::Tensor &v, // total_k x num_heads_k x head_size, total_k := \sum_{i=0}^{b} s_i or num_blocks x page_block_size x num_heads_k x head_size if there's a block_table.
c10::optional<at::Tensor> &out_, // total_q x num_heads x head_size, total_k := \sum_{i=0}^{b} s_i
const at::Tensor &cu_seqlens_q, // b+1
const at::Tensor &cu_seqlens_k, // b+1
c10::optional<at::Tensor> &seqused_k, // b. If given, only this many elements of each batch element's keys are used.
c10::optional<const at::Tensor> &leftpad_k_, // batch_size
c10::optional<at::Tensor> &block_table_, // batch_size x max_num_blocks_per_seq
c10::optional<at::Tensor> &alibi_slopes_, // num_heads or b x num_heads
int max_seqlen_q,
const int max_seqlen_k,
const float p_dropout,
const float softmax_scale,
const bool zero_tensors,
bool is_causal,
int window_size_left,
int window_size_right,
const float softcap,
const bool return_softmax,
c10::optional<at::Generator> gen_);
std::vector<at::Tensor>
mha_bwd(const at::Tensor &dout, // batch_size x seqlen_q x num_heads, x multiple_of(head_size_og, 8)
const at::Tensor &q, // batch_size x seqlen_q x num_heads x head_size
const at::Tensor &k, // batch_size x seqlen_k x num_heads_k x head_size
const at::Tensor &v, // batch_size x seqlen_k x num_heads_k x head_size
const at::Tensor &out, // batch_size x seqlen_q x num_heads x head_size
const at::Tensor &softmax_lse, // b x h x seqlen_q
c10::optional<at::Tensor> &dq_, // batch_size x seqlen_q x num_heads x head_size
c10::optional<at::Tensor> &dk_, // batch_size x seqlen_k x num_heads_k x head_size
c10::optional<at::Tensor> &dv_, // batch_size x seqlen_k x num_heads_k x head_size
c10::optional<at::Tensor> &alibi_slopes_, // num_heads or batch_size x num_heads
const float p_dropout, // probability to drop
const float softmax_scale,
const bool is_causal,
int window_size_left,
int window_size_right,
const float softcap,
const bool deterministic,
c10::optional<at::Generator> gen_,
c10::optional<at::Tensor> &rng_state);
std::vector<at::Tensor>
mha_varlen_bwd(const at::Tensor &dout, // total_q x num_heads x head_size
const at::Tensor &q, // total_q x num_heads x head_size, total_q := \sum_{i=0}^{b} s_i
const at::Tensor &k, // total_k x num_heads_k x head_size, total_k := \sum_{i=0}^{b} s_i
const at::Tensor &v, // total_k x num_heads_k x head_size, total_k := \sum_{i=0}^{b} s_i
const at::Tensor &out, // total_q x num_heads x head_size
const at::Tensor &softmax_lse, // b x h x s softmax logsumexp
c10::optional<at::Tensor> &dq_, // total_q x num_heads x head_size, total_q := \sum_{i=0}^{b} s_i
c10::optional<at::Tensor> &dk_, // total_k x num_heads_k x head_size, total_k := \sum_{i=0}^{b} s_i
c10::optional<at::Tensor> &dv_, // total_k x num_heads_k x head_size, total_k := \sum_{i=0}^{b} s_i
const at::Tensor &cu_seqlens_q, // b+1
const at::Tensor &cu_seqlens_k, // b+1
c10::optional<at::Tensor> &alibi_slopes_, // num_heads or b x num_heads
const int max_seqlen_q,
const int max_seqlen_k, // max sequence length to choose the kernel
const float p_dropout, // probability to drop
const float softmax_scale,
const bool zero_tensors,
const bool is_causal,
int window_size_left,
int window_size_right,
const float softcap,
const bool deterministic,
c10::optional<at::Generator> gen_,
c10::optional<at::Tensor> &rng_state);
std::vector<at::Tensor>
mha_fwd_kvcache(at::Tensor &q, // batch_size x seqlen_q x num_heads x head_size
const at::Tensor &kcache, // batch_size_c x seqlen_k x num_heads_k x head_size or num_blocks x page_block_size x num_heads_k x head_size if there's a block_table.
const at::Tensor &vcache, // batch_size_c x seqlen_k x num_heads_k x head_size or num_blocks x page_block_size x num_heads_k x head_size if there's a block_table.
c10::optional<const at::Tensor> &k_, // batch_size x seqlen_knew x num_heads_k x head_size
c10::optional<const at::Tensor> &v_, // batch_size x seqlen_knew x num_heads_k x head_size
c10::optional<const at::Tensor> &seqlens_k_, // batch_size
c10::optional<const at::Tensor> &rotary_cos_, // seqlen_ro x (rotary_dim / 2)
c10::optional<const at::Tensor> &rotary_sin_, // seqlen_ro x (rotary_dim / 2)
c10::optional<const at::Tensor> &cache_batch_idx_, // indices to index into the KV cache
c10::optional<const at::Tensor> &leftpad_k_, // batch_size
c10::optional<at::Tensor> &block_table_, // batch_size x max_num_blocks_per_seq
c10::optional<at::Tensor> &alibi_slopes_, // num_heads or batch_size x num_heads
c10::optional<at::Tensor> &out_, // batch_size x seqlen_q x num_heads x head_size
const float softmax_scale,
bool is_causal,
int window_size_left,
int window_size_right,
const float softcap,
bool is_rotary_interleaved, // if true, rotary combines indices 0 & 1, else indices 0 & rotary_dim / 2
int num_splits);
PYBIND11_MODULE(TORCH_EXTENSION_NAME, m)
{
m.doc() = "FlashAttention";
m.def("fwd", &mha_fwd, "Forward pass");
m.def("varlen_fwd", &mha_varlen_fwd, "Forward pass (variable length)");
m.def("bwd", &mha_bwd, "Backward pass");
m.def("varlen_bwd", &mha_varlen_bwd, "Backward pass (variable length)");
m.def("fwd_kvcache", &mha_fwd_kvcache, "Forward pass, with KV-cache");
}