Skip to content

Fix parquet row number check for page bounds #20217

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
Oct 13, 2025
Merged

Fix parquet row number check for page bounds #20217

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2e2d6fe
fix row number check for page bounds
pmattione-nvidia Oct 9, 2025
908c25a
Merge branch 'branch-25.12' into fix_row_limits
pmattione-nvidia Oct 9, 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
10 changes: 5 additions & 5 deletions cpp/src/io/parquet/page_decode.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1174,23 +1174,23 @@ inline __device__ bool setup_local_page_info(page_state_s* const s,
// NOTE: s->page.num_rows, s->col.chunk_row, s->first_row and s->num_rows will be
// invalid/bogus during first pass of the preprocess step for nested types. this is ok
// because we ignore these values in that stage.
auto const max_row = min_row + num_rows;
auto const end_row = min_row + num_rows;

// if we are totally outside the range of the input, do nothing
if ((page_start_row > max_row) || (page_start_row + s->page.num_rows < min_row)) {
auto const page_end_row = page_start_row + s->page.num_rows;
if ((page_start_row >= end_row) || (page_end_row <= min_row)) {
Copy link
Contributor Author

Choose a reason for hiding this comment

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

The only real change: adding = on the comparisons

s->first_row = 0;
s->num_rows = 0;
}
// otherwise
else {
s->first_row = page_start_row >= min_row ? 0 : min_row - page_start_row;
auto const max_page_rows = s->page.num_rows - s->first_row;
s->num_rows = (page_start_row + s->first_row) + max_page_rows <= max_row
s->num_rows = (page_start_row + s->first_row) + max_page_rows <= end_row
? max_page_rows
: max_row - (page_start_row + s->first_row);
: end_row - (page_start_row + s->first_row);
}
}

__syncthreads();

// zero counts
Expand Down