chore: sync fork-main with upstream through #159 - #2
Merged
Conversation
The backlog was 16 while the server holds up to 128 connections, so a burst of connects overflowed the listen queue before the accept loop drained it. macOS 27 answers that overflow with RST, which made connectionsBeyondTheCapAreClosed fail with ECONNRESET on 17 of 20 runs there (drumih#151); macOS 26 only stalls the burst, so CI stayed green. Sixteen predates the cap and was never a decision - NIO's own default is 128. The new test reads the configured backlog from the server channel and pins it to at least the cap, so the mismatch fails on every OS rather than only where the kernel resets.
* Speed up prefill on pre-Apple10 Macs * Apply the prefill diagnostics review fixes Brings this branch up to the reviewed state of the private tree. Four findings, all of which this branch previously carried unfixed. Reject a clipping window on the TensorOps path, not only on the fallback selection. The full-attention kernels start their key loop at zero and ignore `slidingWindow`, so a `.full`-labelled request carrying a window that actually clips produced wrong output. Production passes `slidingWindow == kvValidCount` for full layers and never clips, so selection there is unchanged; the guard now also requires `layerKind == .full`, which this tree was missing. Restrict the AGX interactivity mitigation to macOS at compile time. The `AGX_RELAX_CDM_CTXSTORE_TIMEOUT` variable is read by the AGX userspace driver and does not exist elsewhere. The operator override is unchanged: the overwrite argument stays 0, so exporting the variable as 0 still restores stock behaviour. Make the system-device source audit fail closed. It swallowed enumerator and file-read errors with `try?` and exempted by bare filename, so it could report green while auditing nothing. It now requires the enumerator, records walk and read failures, asserts them empty, and exempts by exact relative path. Route all three VisionRuntime command-buffer waits through the status-aware diagnostic helper. They stringified `buffer.error`, so a buffer that failed with `status == .error` and no error object was consumed as success. The fourth wait is inside a `defer` on the throw path and stays deliberately unchecked, which is now stated at the site. Each fix lands with a regression: a clipping-window fixture at the production 512/16/2 shape that reaches the real TensorOps pipeline, a compile-selected non-macOS negative test, positive assertions on the audit's failure arrays, and a call-site check on the VisionRuntime waits. * Remove internal working documents committed by mistake The previous commit staged the whole docs/ directory and swept in six untracked working files that were never meant to be published, plus a stray directory left by a failed sync whose name contained a newline. Removed: the four ISSUE_84 review and planning documents, the terminal 3D engine plan, and issue84_simulate_payloads.py. These are internal working notes; they contain local filesystem paths and unreleased planning material. They remain reachable in this branch's history at 04b36ce. Rewriting that history would discard the main merge made on top of it, so it is left for a deliberate decision rather than done here.
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.
Fast-forward sync of
fork-maintoupstream/main.Brings in two upstream commits:
ec8ad21Speed up prefill on pre-Apple10 Macs (Speed up prefill on pre-Apple10 Macs drumih/turbo-fieldfare#159)1206ff2Size the server listen backlog to the connection cap (Size the server listen backlog to the connection cap drumih/turbo-fieldfare#153)Routed through a PR because
fork-mainis protected (required PR rule blocks the direct fast-forward push and GitHub's Sync button alike).No local changes — this is a pure fast-forward of upstream content. Needed as the base for the Qwen 3.6 architecture work, since
ec8ad21touches the prefill path that the Qwen port re-dispatches.