[kotevcode]

What this changes

Admit and Ledger move unchanged from zig/src/core/engine_api.zig to a new zig/src/core/engine_admission.zig, and engine_api re-exports them under the same names, so no caller changes. engine_api.zig goes from 600 to 593 lines, which leaves room for the execution witness's Info hook. Behaviour is unchanged.

• Issue: Refs Opt-in coherent execution witness for exact-response cache qualification #2741. This is the behaviour-free extraction agreed there to go ahead of the feature.
• Stacked on: main
• atc overlap --json: run on this head against main; complete: false (otherPullRequestsLimit, pullFilesUnavailable, timeLimit). 7 pull requests report a textual conflict: dsv41 metal: image prompts #1722, dsv41 metal: the vision tower's kernels under the kernel contract #1721, dsv41 metal: the lane target, the prompt cache and the native server over Metal #1720, Vulkan: qmm_q4_vec4 reads the 4-bit projection 16 bytes at a time, 34.8% to 43.0% less run time on an M1 with the same logits (stacked on #1544) #1990, Implement GLP control vectors with a fused Metal kernel for Qwen3.8-27B #1835, app: control vector picker and strength in Engine advanced #1951 and GLM-5.3-Flash on CUDA keeps conversation states on disk, every rank's part (--learn-conversations) #1411. Each has one conflicting file. Six share only CHANGELOG.md; Implement GLP control vectors with a fused Metal kernel for Qwen3.8-27B #1835 also changes engine_api.zig, which merges cleanly, so its conflict is CHANGELOG.md as well. Native Vulkan backend for Qwen3.5-2B (opt-in): engine and kernel-contract registration (stacked on #1526) #1544 is clean; the rest report unknown.

Checks on this head

• Head commit: 6cb4dbc
• Lean checker, python3 tools/zig/lean_check.py, on the base and on this head: 345 and 345, 0 new, after rebasing onto main eab305d (only CHANGELOG.md conflicted). python3 tools/pr_check.py --base upstream/main --pr 2896: 0 errors, 0 warnings.
• Tests run, with the command and pass, fail and skip counts:
  • zig build test-cuda-host (every Admit and Ledger caller): 379 of 379 steps succeeded, 0 failed.
  • zig build test-server-cpu: all steps succeeded, 0 failed.
  • zig fmt --check build.zig zig and git diff --check pass.
• CI: not yet run on this head.
• Not run, and why: Metal and golden steps; my machine has no Xcode Metal toolchain. Nothing here reaches a kernel.

Receipt

n/a: two type declarations move between files with identical fields and names, so nothing that reaches inference changes.

Checklist

• The title and description match the head commit, including the receipt and the stack.
• New tests fail before the change, pass after it, and skip cleanly without their dependency. (No new tests: nothing changes.)
• Comments and docstrings are one line. No measurements or history in the source.
• No personal data, machine names, internal hosts or local paths in the diff or its commits. No AI attribution lines.

The inference checklist is skipped because the Receipt is n/a.

[ashhart]

Thanks @kotevcode, this is the behaviour-free extraction we agreed on #2741: both types move byte for byte and keep their names under engine_api, so every caller is unchanged. Ready from our side. Main is frozen for the 1.2.7 release this afternoon, so it lands just after, once CI is green.

• Synthena (Ash's Agent)

[ashhart]

@kotevcode, our landing train couldn't finish this one's rebase (a tooling problem on our side, now fixed), so we rebased it onto today's main and force-pushed (15a9a32a53). The only conflict was the changelog, where your line stays at the top of Unreleased. The moved code is byte-identical to main's, and the lane API, KV pool, fault-admission and GLM admission steps pass (670/670). It's back in the landing queue.

• Synthena (Ash's Agent)

[ashhart]

Thank you @kotevcode. This is merged into main with your commits under your own name, so you're listed as a TensorFold contributor. It ships in the next release, credited to you in the changelog and the release notes.

[ashhart]

Thanks @kotevcode! This is in TensorFold 1.2.8, out now: https://github.com/ashhart/TensorFold/releases/tag/v1.2.8

• Synthena (Ash's Agent)