Fix missing numpy dependency in pyproject.toml - #524
Merged
Merged
Conversation
timethink
pushed a commit
to timethink/sglang
that referenced
this pull request
Mar 9, 2025
vschandramourya
pushed a commit
to vschandramourya/sglang
that referenced
this pull request
Feb 3, 2026
* update * update * update * Add null-to-string conversion for string-type parameters in KimiK2Detector When a parameter schema expects a string type but receives null, convert it to the string "null" instead. This matches the behavior in Qwen3CoderDetector and ensures proper type handling for string parameters. * update --------- Co-authored-by: Qingyang Wu <qywu@research-common-h200-01.cloud.together.ai>
efschu
added a commit
to efschu/htsglang
that referenced
this pull request
Aug 16, 2026
…th collective sgl-project#517 named five BAR1 broadcasts on the HOST path per NEXTN decode round. Verified against the current tree, that inventory is now THREE, not five: the three verify-result syncs were fused into one by pack_accept_payload (#616c). The remaining two are the draft-pick sync in _draft_extend_for_decode, which capture_safe_tp_broadcast still issues one collective PER tensor from its loop. This fuses that pair into one buffer. The picks have mixed dtypes (topk_index integer, topk_p float32 at the SAME shape), so unlike the int32 accept payload the fusion is byte-level. Contiguity is required rather than coerced: reshape(-1) on a non-contiguous tensor returns a copy, so an in-place unpack through it would silently drop the broadcast. fuse is opt-in per call site and defaults OFF. _broadcast_draft_picks is also called from inside the draft CUDA-graph capture (draft_forward), and enabling fusion globally would change what a captured region allocates -- a capture change. Only the site that is both per-decode-round and outside any captured region opts in; the prefill, catchup and in-loop sites stay byte-identical. Not claimed here: a ms/round figure. sgl-project#476's 6.64 pp Seam A cost was measured with the pre-sgl-project#517 guard, which synchronised once per collective; sgl-project#517 removed that sync, so the per-collective price must be re-measured on metal rather than carried over. The A/B is window-gated. Tests: test_draft_pick_fusion_524.py, 7 passed (red first: the three symbols did not exist). Regression over test/registered/unit/debug -k "616 or spec_util or draft_pick or eagle or broadcast": 351 passed, 4 failed. Those 4 are pre-existing and unrelated -- verified by reverting both source files and re-running: test_barlink_bar1_abort_poll_616f.py fails identically on the clean tree (SimpleNamespace missing _round_dev, a sgl-project#624-class stub drift). ruff check clean on both changed files (the one F401 in eagle_worker_v2.py is pre-existing on HEAD); ruff format applied.
efschu
added a commit
to efschu/htsglang
that referenced
this pull request
Aug 16, 2026
… dry run, and the vouch The notes were written at 3c984ad and had gone stale in three ways. Chain: 18 commits at e21e87f -> 22 at 5939d0e. Added sgl-project#697 (8fb86ef), sgl-project#441 (5e0fa1e), sgl-project#524 (5939d0e) and the revision-1 docs commit. Runtime-touching is now 8 of 22, and sgl-project#441 is recorded as NOT runtime despite its ticket -- it adds a handover script and one test guard. Target: integration/r2 has moved a73a0d8 -> a157bf1, so the dry run was re-run rather than carried over. Still zero conflicts. Scope shrank with the target's advance: 115 -> 109 commits, 129 -> 124 files, +22731 -> +22196. Merged-tree tests: managers 2093 passed / 0 failed; planner plus the three new test files 2594 passed / 2 failed. Those 2 are PpWithSpecEvidenceTest in test_rejected_evidence_pins.py and were re-verified as pre-existing by aborting the merge and running that file on clean a157bf1 -- same 2 failed, 2 passed. The chain introduces no new failure. Because the target tip moved, this was re-checked instead of quoting revision 1's finding. Holds: sgl-project#690 (e21e87f) and sgl-project#697 (8fb86ef) are in F4-r4's bundle, so revision 1's objection to landing sgl-project#690 ahead of the sgl-project#694 soak is spent; both lift to group 1 on his boot commit. sgl-project#524 (5939d0e) joins the hold list -- desk-verified only, never on metal, A/B window-gated. sgl-project#685 (5301b94) stays held on the unmade R' decision. Vouch (new section 5): the four F4-r4 cherry-picks are byte-identical to the originals by git patch-id --stable -- f630947/c41645c8c9, 7c58aba/ce6035884d, f1f31d2/658ea3ac11, de92bb6/84b0171fa6 -- so preferring the originals is free and the only real risk is double attribution. 1073702 is excluded: patch-id 8595e66885ac matches nothing here, it is diagnostic scaffolding and not mine to vouch for. Section 6 records that 5af1531 (sgl-project#696) and c738ef5 (sgl-project#689) are F4-r4's line, not this branch, and are listed only because they lift holds; sgl-project#696 is what was DoSing the lanes. Docs only. No merge performed; the dry-run worktree was removed.
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
2 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.
numpy is currently listed as an optional dependency. With the current version on pypi and on the master branch, sglang will fail to import without numpy installed. See the log output below:
This commit proposes a simple fix by requiring numpy in the
pyproject.toml.