Skip to content

Conversation

csarofeen
Copy link
Owner

The way we unroll pointwise ops can cause far too many registers to be used and predicates to be generated. This will help cover perf gap until we make more sophisticated unrolling/predicate generation/schedules.

Copy link
Collaborator

@jjsjann123 jjsjann123 left a comment

Choose a reason for hiding this comment

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

LGTM

@csarofeen csarofeen merged commit c68fba8 into 20_8_18_devel Aug 31, 2020
@csarofeen csarofeen deleted the schedule_pwise branch June 9, 2021 13:46
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants