8000 Implements a Loop Fusion Transformation by kaushikcfd · Pull Request #493 · inducer/loopy · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Implements a Loop Fusion Transformation #493

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

Jump to bottom
Open
wants to merge 3 commits into
base: main
Choose a base branch
from
Open

Implements a Loop Fusion Transformation #493

wants to merge 3 commits into from

Conversation

kaushikcfd
Copy link
Collaborator
@kaushikcfd kaushikcfd commented Oct 4, 2021 •
edited
Loading

Loopy-flavored loop-fusion transformation corresponding to https://doi.org/10.1007/3-540-57659-2_18.

@kaushikcfd kaushikcfd mentioned this pull request Oct 10, 2021
Closed
@kaushikcfd kaushikcfd force-pushed the loop_fusion branch 2 times, most recently from b390c5a to 47d7fa8 Compare October 31, 2021 23:25
@kaushikcfd kaushikcfd force-pushed the loop_fusion branch 2 times, most recently from 5bef1aa to 214e6b9 Compare November 11, 2021 17:14
@kaushikcfd kaushikcfd mentioned this pull request Nov 13, 2021
Merged
@kaushikcfd kaushikcfd force-pushed the loop_fusion branch 2 times, most recently from d844444 to 776ee64 Compare November 17, 2021 17:09
@kaushikcfd kaushikcfd force-pushed the loop_fusion branch 4 times, most recently from acbf9a8 to 921ea56 Compare December 10, 2021 05:33
@kaushikcfd kaushikcfd force-pushed the loop_fusion branch 2 times, most recently from a8dfaf0 to 07194cc Compare December 13, 2021 11:34
@kaushikcfd kaushikcfd force-pushed the loop_fusion branch 3 times, most recently from 4c89630 to a6171e2 Compare March 10, 2022 16:43
@kaushikcfd kaushikcfd force-pushed the loop_fusion branch 2 times, most recently from 371632d to d46bfd1 Compare May 6, 2022 17:41
@kaushikcfd kaushikcfd mentioned this pull request May 6, 2022
Closed
@kaushikcfd kaushikcfd force-pushed the loop_fusion branch 3 times, most recently from f47d987 to 4e97120 Compare May 18, 2022 18:57
kaushikcfd
kaushikcfd commented Jun 21, 2022
View reviewed changes
@kaushikcfd kaushikcfd force-pushed the loop_fusion branch 7 times, most recently from 78ad97a to dbbd55c Compare March 8, 2025 01:48
@kaushikcfd kaushikcfd marked this pull request as ready for review March 8, 2025 02:04
@kaushikcfd kaushikcfd force-pushed the loop_fusion branch 3 times, most recently from 020afc7 to 0d15c6a Compare March 8, 2025 02:27
@majosm
Copy link
Contributor
majosm commented Mar 11, 2025

FYI @kaushikcfd, while I was browsing through this code the other day trying to understand a warning that was being emitted (which turned into inducer/meshmode#453), I spotted a few opportunities to avoid recomputation and speed things up a fair amount in get_kennedy_unweighted_fusion_candidate 8000 s. Specifically, the calls I noticed that were being repeated were _get_partial_loop_nest_tree_for_fusion, _get_ldg_nodes_from_loopy_insn, and (I think, need to revisit and confirm) get_insn_access_map. If I can find some time this week I'll finish my changes and push them for you take a look at.

@kaushikcfd
Copy link
Collaborator Author

@majosm: Thanks for the potential bottlenecks. I memoized those routines.

inducer
inducer reviewed Mar 14, 2025
View reviewed changes
@kaushikcfd kaushikcfd force-pushed the loop_fusion branch 2 times, most recently from 60a1d38 to 78174b8 Compare March 19, 2025 15:58
@inducer inducer force-pushed the loop_fusion branch 6 times, most recently from 06598b8 to 3127ecb Compare May 20, 2025 20:36
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@inducer inducer inducer left review comments

@majosm majosm 5AE0 majosm left review comments

@matthiasdiener matthiasdiener Awaiting requested review from matthiasdiener

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@kaushikcfd @inducer @majosm @matthiasdiener
0