Skip to content

issue with the implementation of column_sum_reduce #804

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
RezaYazdaniAminabadi merged 2 commits into from
Feb 28, 2021
Merged

issue with the implementation of column_sum_reduce #804

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
afa147b
issue with the implementation of column_sum_reduce
zmxdream Feb 28, 2021
f33e662
Merge branch 'master' into zmx-patch-1
RezaYazdaniAminabadi Feb 28, 2021
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
2 changes: 1 addition & 1 deletion csrc/transformer/general_kernels.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,7 +43,7 @@ __global__ void column_sum_reduce(const T* __restrict__ inp,

if (threadIdx.x == 0) {
int pos = blockIdx.x * TILE_DIM + threadIdx.y;
if (pos < (rows * width)) out[pos] = sum;
if (pos < width) out[pos] = sum;
Copy link
Contributor

@RezaYazdaniAminabadi RezaYazdaniAminabadi Feb 28, 2021 •
edited
Loading

Choose a reason for hiding this comment

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

Thanks for fixing this! I would say it still was working when the hidden dimension was dividable by 32, however, it would have caused a memory leak for when the hidden is not dividable by 32!

Copy link
Contributor Author

@zmxdream zmxdream Mar 1, 2021

Choose a reason for hiding this comment

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

yes! thanks for your approval!!

}
}

Expand Down