[cherry-pick][release/2.13] NCCL ldconfig symlink fix + atomic symlink#4403
Closed
apbose wants to merge 3 commits into
Closed
[cherry-pick][release/2.13] NCCL ldconfig symlink fix + atomic symlink#4403apbose wants to merge 3 commits into
apbose wants to merge 3 commits into
Conversation
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Collaborator
Author
|
This was incorrect. Closing in favor of #4408 |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Cherry-pick of the last two commits from
abose/release_2.12_mdtrt_ld_library_fixontorelease/2.13.Both applied cleanly (release/2.13's
_nccl_utils.pywas at the same base as release/2.12); the resulting_nccl_utils.pyis byte-identical to the source branch's final version.Commits:
c4fce46b3setup_nccl_for_torch_tensorrt change — replaces the oldLD_LIBRARY_PATH+ctypespreload approach with creating alibnccl.sosymlink in a system libdir thatld.soalready searches (Debian/Ubuntu multiarch, RHELlib64, or/usr/libfallback via new helper_sys_libdir_on_ldso_path()), thenldconfig. Removescheck_nccl_library_path()and its two tests, and trims the warning incheck_nccl_engine_requirements().55b3ea7d8atomic symlink — makes the symlink swap race-safe for multi-process/distributed runs: symlink to a unique per-pid temp name thenos.replaceonto the target, with temp cleanup and "another rank already produced the correct symlink" handling.Files:
py/torch_tensorrt/distributed/_nccl_utils.py,tests/py/dynamo/distributed/test_native_nccl.py🤖 Generated with Claude Code