claude-noether
31da35a549
build and publish packages / distcc-avahi-aarch64 (push) Successful in 1m20s
build and publish packages / lmcp-any (push) Successful in 13s
build and publish packages / lmcp-debian (push) Successful in 7s
build and publish packages / claude-his-any (push) Successful in 13s
build and publish packages / ffmpeg-v4l2-request-aarch64 (push) Successful in 12m39s
build and publish packages / claude-his-debian (push) Successful in 8s
build and publish packages / libva-v4l2-request-fourier-aarch64 (push) Successful in 27s
build and publish packages / mpv-fourier-aarch64 (push) Has been cancelled
build and publish packages / libva-v4l2-request-fourier-debian (push) Has been cancelled
build and publish packages / mpv-fourier-debian (push) Has been cancelled
build and publish packages / daedalus-v4l2-debian (push) Has been cancelled
build and publish packages / daedalus-v4l2-dkms-debian (push) Has been cancelled
build and publish packages / ffmpeg-v4l2-request-debian (push) Has been cancelled
Bumps pkgrel from 5 to 6. Cumulative regenerated from the redone per-series reconstruction in kernel-agent (replacing the broken PR #33 attempt with a proper rebase onto v7.0-danctnix1 baseline). cumulative.patch: 162 716 -> 279 554 bytes b2sum: 50397711a6a3... -> eb179c03f35a... manifest.lock: 32 resolved patches (was 4 — c5x interim collapsed the bes2600 driver scope into one cumulative blob; per-series properly tracks each fix) bes2600 series-dirs in kernel-agent: now 25 individual series-dirs (one per cleanups commit + Patch H), each with the proper in-tree drivers/staging/bes2600/* paths Built + installed on ohm 2026-05-19 23:39. Functional verification: uname: 7.0.0-danctnix1-6-pinetab2-danctnix-besser srcversion: 1A919EED0E6DC2478559B17 (differs from pkgrel=5's BEB625FA7443171EA8D55F7 — not byte-equivalent; functional equivalence verified: wlan0 associates, bes2600_btuart loads, Pattern A 0, no WARN/BUG) Per-fix revertability now real: removing an include from fleet/ohm.yaml drops that fix from the cumulative. Bisecting on kernel-agent side becomes practical.
linux-pinetab2-danctnix-besser
DanctNIX PineTab2 kernel + the BESser bes2600 driver patchset. kernel-agent managed as of pkgrel=4 (2026-05-18).
TL;DR
- What: drop-in replacement for
linux-pinetab2with the BESser bes2600 patchset applied. - For: PineTab2 (RK3566) with the on-board BES2600 SDIO WiFi/BT chip.
- Install:
pacman -U linux-pinetab2-danctnix-besser-*.pkg.tar.zst linux-pinetab2-danctnix-besser-headers-*.pkg.tar.zst. Itprovidesandconflictswithlinux-pinetab2, so it slots in cleanly. - Source of truth for patches:
git.reauktion.de/marfrit/kernel-agent
fleet/ohm.yaml.
How the patch cumulative is built
kernel-agent/fleet/ohm.yamllists the patch series-dir includes for this kernel.kernel-agent/bin/ka-promote ohmwalks those includes, concatenates them into a singlecumulative.patch, and emits amanifest.lockrecording per-patch sha256s and the cumulative b2sum.- The cumulative is dropped into this directory as
0001-bes2600-besser-kernel-agent-cumulative.patch, the b2sum inPKGBUILDis updated to match, andpkgrelis bumped. manifest.lockis committed alongside the cumulative for an audit trail (which patches went in, in what order).
pkgrel history
| pkgrel | Date | Notes |
|---|---|---|
| 1–3 | 2026-05-08…05-18 | hand-managed in marfrit/besser/danctnix-besser-pkgbuild/ (DEPRECATED) |
| 4 | 2026-05-18 | first kernel-agent-managed release; same source contents as the hand-managed pkgrel=3 (Patch I + SCS workaround folded into single cumulative); fixes pkgrel=3's duplicated 0003-arm64-...patch source-array bug |
Known caveats
CONFIG_SHADOW_CALL_STACK=nin the config: GCC 15.2.1 fails to buildarch/arm64/lib/xor-neon.cwitharm_neon.h#pragma GCC target()blocks and-fsanitize=shadow-call-stack. Seereference_arm64_scs_arm_neon_gcc15memory and the Makefile workaround patch included in the cumulative. Track upstream GCC resolution before re-enabling.- The
cumulative-c5x-danctnixseries-dir in kernel-agent is currently a single-file interim (the working hand-curated cumulative) because the per-series mirrors created inkernel-agent#17use DKMS-style root paths instead of in-tree staging paths and do not apply cleanly. Per-series reconstruction is a separate kernel-agent follow-up. The b2sum integrity check inPKGBUILDstill pins the exact bytes shipped.