Skip to content

Fix CUDA minmax collector copies from volatile shared memory #127

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

Open
xsjk wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Fix CUDA minmax collector copies from volatile shared memory #127

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
309d82a
Fix CUDA minmax reduction for volatile collectors
xsjk May 7, 2026
5adb32d
Support Python 3.14 AST constants
xsjk May 11, 2026
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
16 changes: 16 additions & 0 deletions compyle/array.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -97,6 +97,22 @@ def get_backend(backend=None):

minmax_operator_tpl = """

__device__ ${dtype}()
{
}

__device__ ${dtype}(${dtype} const volatile &src)
{
% for prop in prop_names:
% if not only_max:
this->cur_min_${prop} = src.cur_min_${prop};
% endif
% if not only_min:
this->cur_max_${prop} = src.cur_max_${prop};
% endif
% endfor
}
Comment on lines +100 to +114
Copy link
Copy Markdown
Author

@xsjk xsjk May 8, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@copilot apply changes based on this feedback


__device__ ${dtype} volatile &operator=(
${dtype} const &src) volatile
{
Expand Down
Loading