fix: correct NVML P2P WRITE status warning to say INDEX_WRITE#2248
Open
EylonKrause wants to merge 1 commit into
Open
fix: correct NVML P2P WRITE status warning to say INDEX_WRITE#2248EylonKrause wants to merge 1 commit into
EylonKrause wants to merge 1 commit into
Conversation
The failure WARN for the NVML_P2P_CAPS_INDEX_WRITE query printed NVML_P2P_CAPS_INDEX_READ, misdirecting anyone debugging a P2P-write capability failure. Match the message to the query. Signed-off-by: EylonKrause <eylon1909@gmail.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
In
ncclNvmlEnsureInitialized(src/misc/nvmlwrap.cc), the P2P-status probe queriesNVML_P2P_CAPS_INDEX_READand thenNVML_P2P_CAPS_INDEX_WRITE. The failureWARNfor the WRITE query incorrectly printsNVML_P2P_CAPS_INDEX_READ— a copy-paste from the READ branch just above it.The effect is purely diagnostic but misleading: an operator debugging a P2P write-capability failure is told the read query failed, pointing them at the wrong thing.
Related Issues
None.
Changes & Impact
src/misc/nvmlwrap.cc: one diagnostic string literal corrected (INDEX_READ→INDEX_WRITE) in the WRITE branch'sWARN. No change to the NVML call, control flow, stored status, or return value. Non-breaking.Performance Impact
None — diagnostic message text only.
Testing
make src.build(no new warnings).WARNcorrectly says READ; the WRITE branch'sWARNpreviously also said READ, and now says WRITE to match its query.