ci: move Megatron-Bridge to d352aced - #3824
Merged
Merged
Conversation
|
Auto-sync is disabled for ready for review pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
tdene
added
the
CI:Lfast
Contributor
Author
|
/ok to test e07ff41 |
Contributor
|
Superseded by a newer MBridge bump PR. |
Contributor
|
Superseded by a newer MBridge bump PR. |
tdene
changed the title
Signed-off-by: Teodor-Dumitru Ene <teodord.ene@gmail.com>
See NVIDIA-NeMo/Megatron-Bridge#5262 Signed-off-by: Teodor-Dumitru Ene <teodord.ene@gmail.com>
See NVIDIA/Megatron-LM#6476 Signed-off-by: Teodor-Dumitru Ene <teodord.ene@gmail.com>
Contributor
Author
|
/ok to test 9cab81e |
tdene
force-pushed
the
tde/bump_mbridge_20260819
branch
from
August 26, 2026 15:53
291596c to
9cab81e
Compare
tdene
added
CI:L1
terrykong
approved these changes
Aug 27, 2026
asolergi-nv
added a commit
that referenced
this pull request
Aug 27, 2026
14 upstream commits; four of them produced seven conflicts. Each resolution below. #3612 feat(sglang): megatron backend weight refit for sglang rollouts - weight_sync/factory.py: it rewrote the train_cluster/inference_cluster/ refit_buffer_size_gb docstrings (SGLang owns its own process group, so it needs neither cluster handle). Took its wording and kept our refit_timeout_s entry, which it never saw. refit_timeout_s still reaches NcclReshardWeightSynchronizer and CollectiveWeightSynchronizer; the new SGLang synchronizer does not take it, which is correct -- our watchdog bounds a JOINT communicator and SGLang does not build one. - base_policy_worker.py: it added _refit_transport_state and connect_sglang_rollout_engines at the same insertion point as our stand_down_refit_watchdog. Disjoint additions; kept both. - pyrefly.toml: it swapped http_weight_synchronizer for sglang_weight_synchronizer. Corroborated by the merge deleting http_weight_synchronizer.py outright. #3773 feat(sc): support PPO in single controller - single_controller_utils/setup.py: the SC path is no longer GRPO-only, so it renamed grpo_config to algo_cfg. Kept our nccl_reshard precondition guard and applied the rename to the val-period line inside it; grpo_config no longer appears anywhere. - L1_Functional_Tests_SingleController.sh: it added a ppo_async run_test and padded every non-fast entry to align with "run_test fast". Kept our annotation -- it says which of skip-vs-pass a green lane actually means, which its one-line version does not -- and adopted the alignment, including on our seven recovery entries, so the file does not end up half-converted. - pyrefly.toml: it re-sorted the list, moving vllm_remote_sparse_weight_synchronizer to its correct alphabetical slot. Our side had added membership.py AND held that entry in the old position, so taking our block verbatim would have duplicated it. Kept membership.py only; verified the result is sorted and has no duplicates. #3545 fix(vllm): support native BF16 FlashInfer TRTLLM refit - vllm_backend.py: its _nrl_layerwise_reload_* class attributes landed where our model_update_group declaration is. Disjoint; kept both. - tests/unit/models/generation/test_vllm_backend.py: its layerwise-reload suite against our init_collective release tests plus the _RecordingGroup fixture. Disjoint; kept both. 53 tests collect. #3768 feat: add MOPD to single-controller text path - Touched setup.py alongside #3773; no separate resolution needed. Submodule: the merge advances Megatron-Bridge to d352aced (#3824). Verified the STAGED pointer is upstream's and not our stale 8c46dc42 -- staging the local one is what breaks the fast-forward check and `uv lock --check` together. Gym is untouched by the merge. Verified after resolving: no conflict markers remain, all four lint hooks clean (the one pyrefly error is the pre-existing unrelated transfer_queue import), and 841 unit tests pass across single_controller, refit_watchdog, worker_refit_signatures and weight_sync -- up from 725, because #3773 brings a large new SC suite that passes alongside ours. Signed-off-by: asolergibert <asolergibert@nvidia.com>
asolergi-nv
added a commit
that referenced
this pull request
Aug 27, 2026
…o PR4
Brings PR3's sync with upstream/main down the stack. Three conflicts, all between PR4's
own additions and changes that arrived from main:
pyrefly.toml
#3773 re-sorted the list and moved generation/fleet_health.py to its correct
alphabetical slot. PR4's block added engine_supervisor.py AND held fleet_health.py in
the old position, so taking it verbatim would have duplicated the entry -- the same
trap this file set on PR3 with vllm_remote_sparse_weight_synchronizer. Kept
engine_supervisor.py only, placed where it sorts (after dynamo/, before fleet_health).
Verified the whole nemo_rl list is sorted and duplicate-free.
nemo_rl/algorithms/single_controller.py
PR4's EngineSupervisor wiring against #3768's MOPD TQTeacherLogprobCoordinator, both
landing in the same __init__ region. Disjoint, so both kept, with the coordinator
first because it installs a post-write enricher on the buffer.
Also removed a duplicate this merge would otherwise have introduced: main relocated
the `_rollout_manager._tq_buffer = self._buffer` rebind next to the assignment it
guards, so PR4's copy at the old site became redundant. Confirmed main carries it
exactly once before deleting the second.
tests/functional/L1_Functional_Tests_SingleController.sh
PR4's RESTART_DEAD_SHARDS entry against the column alignment #3773 introduced. Kept
the entry and applied the alignment to it and to every other unpadded line, so the
file is not left half-converted.
Submodules: both pointers match upstream/main exactly (Megatron-Bridge d352aced from
#3824, Gym c3bac963), so the fast-forward check and `uv lock --check` both see a clean
state. The locally dirty submodule working trees were deliberately not staged.
Verified: no markers remain, all four lint hooks clean (the single pyrefly error is the
pre-existing unrelated transfer_queue import), 849 unit tests pass.
Signed-off-by: asolergibert <asolergibert@nvidia.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
3 participants
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.
What does this PR do ?
Add a one line overview of what this PR aims to accomplish.
Issues
List issues that this PR closes (syntax):
Usage
# Add a code snippet demonstrating how to use thisBefore your PR is "Ready for review"
Pre checks:
Additional Information