fix(pd): shape send list by swap count#5850
Open
njzjz-bot wants to merge 1 commit into
Open
Conversation
Pack exactly one send-list pointer address per communication swap and shape the Paddle tensor from that packed array instead of the total atom send count. Add a regression with two swaps and seven sent atoms to keep pointer-array length independent from per-swap send counts. Coding-Agent: Codex Codex-Version: codex-cli 0.144.4 Model: gpt-5.6-sol Reasoning-Effort: xhigh
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #5850 +/- ##
==========================================
- Coverage 78.58% 78.32% -0.26%
==========================================
Files 1050 1050
Lines 120637 120643 +6
Branches 4356 4356
==========================================
- Hits 94801 94497 -304
- Misses 24278 24579 +301
- Partials 1558 1567 +9 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Member
|
Possible reviewers based on changed lines, exact file history, and exact-file review history:
No review request was made automatically. Coding agent: Codex |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #5659.
Summary
send_listfrom the packed pointer-array length (nswap) instead ofsum(sendnum)send_numas the independent per-swap atom count consumed by the border communication opnswap=2andsendnum={3,4}so pointer-array length cannot be confused with the seven total sent atomsWhy existing tests missed this
The universal Paddle fixture uses an SE(A) model without message passing, so
DeepPotPDnever constructs communication tensors. The Paddle LAMMPS MPI case is skipped and uses the same non-message-passing model. Existing send-list tests covered NULL-atom remapping and send/receive counts, while the Paddle input specification accepts a dynamicsend_listlength; none asserted that the pointer-address tensor contains one entry per swap.Validation
ruff format .ruff check .clang-format --dry-run --Werroron all changed C++ filesrunUnitTests_ccTestRemapCommSendlist.*: 5 passed, including thenswap != sum(sendnum)regressionThe local environment does not include the Paddle inference SDK, so the Paddle-enabled compile/integration job is left to CI.
Coding agent: Codex
Codex version: codex-cli 0.144.4
Model: gpt-5.6-sol
Reasoning effort: xhigh