Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

perf: use cuda-core implemention for io-bound block-sparse attention #560

Merged
merged 2 commits into from
Oct 26, 2024
Merged

perf: use cuda-core implemention for io-bound block-sparse attention #560

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2d7e85f
upd
yzh119 Oct 26, 2024
068fdf4
format
yzh119 Oct 26, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
156 changes: 109 additions & 47 deletions python/flashinfer/sparse.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,7 @@
from typing import Optional, Union, Tuple
import logging
import torch
from .decode import get_batch_decode_module
from .prefill import _compute_page_qk_indptr, get_batch_prefill_module
from .quantization import segment_packbits
from .utils import (
Expand Down Expand Up @@ -299,31 +300,65 @@ def plan(

kv_indptr_host = indptr.to("cpu", non_blocking=True)

self._cached_module = get_batch_prefill_module(
q_data_type,
kv_data_type,
q_data_type,
indptr.dtype,
head_dim,
PosEncodingMode[pos_encoding_mode].value,
mask_mode,
False, # use_sliding_window
logits_soft_cap > 0, # use_logits_soft_cap
allow_fp16_qk_reduction,
)
# NOTE(Zihao): we haven't supported mask in cuda-core implementations but it should
# be easy to add support for it if needed, leave it as a future work.
# at this moment, when mask is provided, we use the tensor-core implementation
if (
R * (num_qo_heads // num_kv_heads) < 4
and mask_mode == MaskMode.NON_CAUSAL.value
):
# If the operation is not compute-bound, we use the cuda-core implementation
self._use_tensor_cores = False
self._cached_module = get_batch_decode_module(
q_data_type,
kv_data_type,
q_data_type,
indptr.dtype,
head_dim,
PosEncodingMode[pos_encoding_mode].value,
False, # use_sliding_window
logits_soft_cap > 0, # use_logits_soft_cap
)

self._plan_info = self._cached_module.plan(
self._float_workspace_buffer,
self._int_workspace_buffer,
self._pin_memory_int_workspace_buffer,
qo_indptr_host,
kv_indptr_host,
num_blocks_row,
num_qo_heads,
num_kv_heads,
C,
False, # is_cuda_graph_enabled
)
self._plan_info = self._cached_module.plan(
self._float_workspace_buffer,
self._int_workspace_buffer,
self._pin_memory_int_workspace_buffer,
kv_indptr_host,
num_blocks_row,
num_qo_heads,
num_kv_heads,
C,
False, # is_cuda_graph_enabled
)
else:
# if the operation is compute-bound, we use the tensor-core implementation
self._use_tensor_cores = True
self._cached_module = get_batch_prefill_module(
q_data_type,
kv_data_type,
q_data_type,
indptr.dtype,
head_dim,
PosEncodingMode[pos_encoding_mode].value,
mask_mode,
False, # use_sliding_window
logits_soft_cap > 0, # use_logits_soft_cap
allow_fp16_qk_reduction,
)

self._plan_info = self._cached_module.plan(
self._float_workspace_buffer,
self._int_workspace_buffer,
self._pin_memory_int_workspace_buffer,
qo_indptr_host,
kv_indptr_host,
num_blocks_row,
num_qo_heads,
num_kv_heads,
C,
False, # is_cuda_graph_enabled
)

self._pos_encoding_mode = pos_encoding_mode
self._allow_fp16_qk_reduction = allow_fp16_qk_reduction
Expand Down Expand Up @@ -404,30 +439,57 @@ def run(
k = k.reshape(-1, self.C, *k.shape[-2:]).contiguous()
v = v.reshape(-1, self.C, *v.shape[-2:]).contiguous()

out = self._cached_module.paged_run(
self._float_workspace_buffer,
self._int_workspace_buffer,
self._plan_info,
q,
k,
v,
self._packed_mask_buf,
_get_cache_alibi_slopes_buf(q.shape[1], self.device),
self._qo_indptr,
self._paged_kv_indptr_buf,
self._paged_kv_indices_buf,
self._paged_kv_last_page_len,
self._qk_indptr_buf,
TensorLayout[self._kv_layout].value,
-1, # window_left
logits_soft_cap,
sm_scale,
rope_scale,
rope_theta,
return_lse,
)
lse = None
if return_lse:
lse = torch.empty(
(q.size(0), q.size(1)), dtype=torch.float32, device=q.device
)

if self._use_tensor_cores:
out = self._cached_module.paged_run(
self._float_workspace_buffer,
self._int_workspace_buffer,
self._plan_info,
q,
k,
v,
self._packed_mask_buf,
_get_cache_alibi_slopes_buf(q.shape[1], self.device),
self._qo_indptr,
self._paged_kv_indptr_buf,
self._paged_kv_indices_buf,
self._paged_kv_last_page_len,
self._qk_indptr_buf,
TensorLayout[self._kv_layout].value,
-1, # window_left
logits_soft_cap,
sm_scale,
rope_scale,
rope_theta,
lse,
)
else:
out = self._cached_module.run(
self._float_workspace_buffer,
self._int_workspace_buffer,
self._plan_info,
q,
k,
v,
self._paged_kv_indptr_buf,
self._paged_kv_indices_buf,
self._paged_kv_last_page_len,
_get_cache_alibi_slopes_buf(q.shape[1], self.device),
TensorLayout[self._kv_layout].value,
-1, # window_left
logits_soft_cap,
sm_scale,
rope_scale,
rope_theta,
lse,
)

return out if return_lse else out[0]
return (out, lse) if return_lse else out

def end_forward(self) -> None:
r"""Warning: This method is deprecated and has no effect."""
Expand Down