sunjizzle/diskless-workstation
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

docs/TODO #18: kiosk render bench-verified on VM + 820 (i915 firmware confirmed)

Browse Source 
f59eb79 now fully bench-verified: the fleet SPA renders on both the VM
(virtio-gpu) and the 820 (i915), so the i915 DMC/GuC firmware loads on real
hardware and the firmware keep-list is validated end-to-end. The two later
commits (nss-mdns drop, omit_drivers + keep-list tightening) don't touch i915,
so render should be safe, but still need their own build+boot.
This commit is contained in:
Audiobuzz
2026-07-03 06:58:52 +00:00
parent bf0c61a025
commit 8e105c8bf1
1 changed files with 10 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/TODO.md
+10 -10
View File
@@ -1499,16 +1499,16 @@ DDP insurance for the `ice`/`mlx5` server-NIC build variant + small Intel blobs)
~600 MB). Wi-Fi (`iwl*-firmware`, ~83 MB) removed as packages (wired-only fleet). ~600 MB). Wi-Fi (`iwl*-firmware`, ~83 MB) removed as packages (wired-only fleet).
Strip-in-place (`find … -exec rm -rf`, not `rpm -e` — can't drop the package without Strip-in-place (`find … -exec rm -rf`, not `rpm -e` — can't drop the package without
losing i915); a `du -sh /usr/lib/firmware` line reports the result in the build log. losing i915); a `du -sh /usr/lib/firmware` line reports the result in the build log.
**BOOT-VERIFIED 2026-07-03 (commit `f59eb79`): built AND booted successfully** — langpacks **BENCH-VERIFIED 2026-07-03 (commit `f59eb79`): built, booted, AND the kiosk SPA RENDERS
en-only + the pre-dracut firmware strip (keep i915/intel/mellanox/intel-ucode) do NOT on both the VM (virtio-gpu) and the 820 (i915).** This clears the firmware strip's one
break the boot/root path (firmware 678→147 MB, initramfs bundles only the keep-list). real risk — the i915 DMC/GuC blobs load on real hardware, so the keep-list
Still to confirm on that build: (a) the kiosk actually **RENDERS** (i915 DMC/GuC load) — (i915/intel/mellanox/intel-ucode, pre-dracut) is validated end-to-end (firmware 678→147 MB,
"booted" ≠ "display up"; only exercised on real i915 hardware (the 820), not a virtio VM; initramfs bundles only the keep-list). Remaining caveat: if a future host has a non-Intel
(b) if any host has a non-Intel GPU or an `ice`/`mlx5` NIC, widen the keep-list before GPU or an `ice`/`mlx5` NIC, widen the keep-list before trusting fleet-wide. **NOT yet in
trusting fleet-wide. **NOT yet in `f59eb79`** (bench-verify separately): the `nss-mdns`/ `f59eb79`** (bench-verify separately — none touch i915, so render should be safe): the
avahi drop (`5cc72cc`) and the `omit_drivers` GPU-kmod trim + keep-list tightening — `nss-mdns`/avahi drop (`5cc72cc`) and the `omit_drivers` GPU-kmod trim + keep-list
dropped `mellanox/`, narrowed `intel/` to `ice` only (`853ae20`). Audited models so far: tightening — dropped `mellanox/`, narrowed `intel/` to `ice` only (`853ae20`). Audited
HP-820-class (i915 + e1000e + iwlwifi-dropped) and virtio VMs. models: HP-820-class (i915 + e1000e + iwlwifi-dropped) and virtio VMs.
**Build-verified 2026-07-03:** firmware **678 MB → 147 MB**; the initramfs diagnostic **Build-verified 2026-07-03:** firmware **678 MB → 147 MB**; the initramfs diagnostic
confirms only the keep-list is bundled (i915 all-gens + intel/ice + intel/ish + mellanox) confirms only the keep-list is bundled (i915 all-gens + intel/ice + intel/ish + mellanox)