mpi: Patch unnecessary halo updates inserted with functions on subdomains - #2986
Closed
EdCaunt wants to merge 2 commits into
Closed
mpi: Patch unnecessary halo updates inserted with functions on subdomains#2986EdCaunt wants to merge 2 commits into
EdCaunt wants to merge 2 commits into
Conversation
…operators using functions on subdomains
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #2986 +/- ##
==========================================
- Coverage 83.57% 83.53% -0.05%
==========================================
Files 257 257
Lines 53783 53917 +134
Branches 4608 4614 +6
==========================================
+ Hits 44951 45038 +87
- Misses 8035 8078 +43
- Partials 797 801 +4
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
When constructing an operator with a trivial equation in which a field defined on a subdomain is updated without using any surrounding points (i.e.
Eq(u.forward, u)), halo exchanges get inserted as the symbolic subdomain offsets do not get evaluated until runtime, and thus the compiler cannot currently determine that it will not need data from within the halo. These halo exchanges are erroneous as they read and write from the same spatial position and thickness values are populated to avoid indexing into the halo.Note: currently has a lot of comments as there are aspects here I'm not entirely sure about and which need sanity checking by someone with more experience with MPI data dependence