Skip to content

fix pin memory device #755

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

Merged
merged 2 commits into from
Jan 27, 2025
Merged

fix pin memory device #755

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
04c3d6c
fix pin memory
youkaichao Jan 27, 2025
7ffad07
fix various device
youkaichao Jan 27, 2025
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
8 changes: 6 additions & 2 deletions flashinfer/decode.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -651,7 +651,8 @@ def __init__(
(8 * 1024 * 1024,), dtype=torch.uint8, device=self.device
)
self._pin_memory_int_workspace_buffer = torch.empty(
(8 * 1024 * 1024,), dtype=torch.uint8, pin_memory=True
(8 * 1024 * 1024,), dtype=torch.uint8,
pin_memory=True, device="cpu",
)

if use_cuda_graph:
Expand Down Expand Up @@ -718,6 +719,7 @@ def reset_workspace_buffer(
self._pin_memory_int_workspace_buffer = torch.empty(
self._int_workspace_buffer.shape,
dtype=self._int_workspace_buffer.dtype,
device="cpu",
pin_memory=True,
)

Expand Down Expand Up @@ -1277,7 +1279,8 @@ def __init__(
(8 * 1024 * 1024,), dtype=torch.uint8, device=self.device
)
self._pin_memory_int_workspace_buffer = torch.empty(
(8 * 1024 * 1024,), dtype=torch.uint8, pin_memory=True
(8 * 1024 * 1024,), dtype=torch.uint8,
pin_memory=True, device="cpu",
)

if use_cuda_graph:
Expand Down Expand Up @@ -1330,6 +1333,7 @@ def reset_workspace_buffer(
self._pin_memory_int_workspace_buffer = torch.empty(
self._int_workspace_buffer.shape,
dtype=self._int_workspace_buffer.dtype,
device="cpu",
pin_memory=True,
)

Expand Down
8 changes: 6 additions & 2 deletions flashinfer/prefill.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1099,6 +1099,7 @@ def __init__(
self._pin_memory_int_workspace_buffer = torch.empty(
self._int_workspace_buffer.shape,
dtype=self._int_workspace_buffer.dtype,
device="cpu",
pin_memory=True,
)
self._use_cuda_graph = use_cuda_graph
Expand Down Expand Up @@ -1165,6 +1166,7 @@ def reset_workspace_buffer(
self._pin_memory_int_workspace_buffer = torch.empty(
self._int_workspace_buffer.shape,
dtype=self._int_workspace_buffer.dtype,
device="cpu",
pin_memory=True,
)

Expand Down Expand Up @@ -1415,7 +1417,7 @@ def plan(
if page_size != 1:
vector_sparse_indptr_host = torch.cat(
[
torch.tensor([0], dtype=torch.int32),
torch.tensor([0], dtype=torch.int32, device=kv_lens_arr_host.device),
torch.cumsum(kv_lens_arr_host, dim=0, dtype=torch.int32),
],
dim=0,
Expand Down Expand Up @@ -1858,7 +1860,8 @@ def __init__(
(8 * 1024 * 1024,), dtype=torch.uint8, device=self.device
)
self._pin_memory_int_workspace_buffer = torch.empty(
self._int_workspace_buffer.shape, dtype=torch.uint8, pin_memory=True
self._int_workspace_buffer.shape, dtype=torch.uint8,
pin_memory=True, device="cpu",
)
self._use_cuda_graph = use_cuda_graph
if use_cuda_graph:
Expand Down Expand Up @@ -1911,6 +1914,7 @@ def reset_workspace_buffer(
self._pin_memory_int_workspace_buffer = torch.empty(
self._int_workspace_buffer.shape,
dtype=self._int_workspace_buffer.dtype,
device="cpu",
pin_memory=True,
)

Expand Down