Skip to content

device/lynx: clarify historical Wi-Fi source and safety evidence - #822

Open
latentoperator wants to merge 2 commits into
derekross:mainfrom
latentoperator:docs/lynx-wifi-followups-20261115
Open

latentoperator wants to merge 2 commits into
derekross:mainfrom
latentoperator:docs/lynx-wifi-followups-20261115

Conversation

@latentoperator

Copy link
Copy Markdown
Contributor

The historical Wi-Fi report now links the complete tested MSI source snapshot and records the later post-reboot connection, including the DHCP-readiness failure before the successful bounded activation. It keeps both old-stack successes separate from qualification of the revised signed integration.

This also answers the follow-ups on #751: names alfikpl for the ordering lead, limits the Trusty-storage statement to the recorded lab stages and process checks, and makes clear that the S3C2410 hardware watchdog itself was disabled with WDIOS_DISABLECARD after successful checks. It does not claim a continuous whole-boot storage audit or an unchanged RPMB write counter.

Historical release: 7.3.0-rc6-omamobile-lynx-msi-20261113+. Archival source: latentoperator/linux-omarchy-mobile@3fd82de (provenance only, not proposed for merge). The preserved earlier WMI/TX/USB failure and later DHCP-readiness failure remain part of the evidence. No new hardware operation or runtime pass is claimed by this documentation change.

Validation: tools/pr-check, git diff --check and tests/all pass. tests/packages finishes with only omamobile-modem failing: its pre-existing NV-overlay Broken pipe / “failure at object” test failure also reproduces from pristine upstream ca82fe7. Modem code is unchanged; all other package suites pass. This is disclosed rather than presented as a green full-package gate.

Tie the two historical connection tests to an immutable archival source snapshot, preserve the DHCP-readiness failure and distinguish signed-source preparation from runtime qualification. Clarify the recorded Trusty-storage scope and hardware watchdog disarm, and credit alfikpl for the ordering lead.

Assisted-by: LLM
@derekross

Copy link
Copy Markdown
Owner

Thank you, Chris. This answers all four things I asked on #751: the tested tree is on your fork under a branch (I checked that 3fd82de7aefe exists and that archive/lynx-msi-tested-20261113 holds it), the Trusty line says what the stages did and what it doesn't prove, the watchdog sentence says it was the hardware watchdog your stage had opened, and alfikpl is named. Keeping the DHCP race beside the second pass is the kind of record I want.

Three small things and it can leave draft:

  1. Please add what confined the Wi-Fi chip's DMA in these runs. The record now has two boots with association, HTTPS and autoconnect, and as far as I can see nothing sat between the WCN6855 and RAM (no restricted pool on the node, no IOMMU in front of pcie_1). One sentence in section 1, and the pool in section 3's list of what's left. I've asked for the pool on linux-omarchy-mobile#255.
  2. "Published as draft kernel PRs 251-255" and the head id are true for the 15th; two of them conflict with the base today and every head will move. Please write "as of 2026-11-15" and name the repository (linux-omarchy-mobile#251 to #255), because a bare number here links to a different PR.
  3. "A separately reviewed activation" and "a newly reviewed attended runner": reviewed by whom? If that's your own tooling, say so, or say "revised".

Nothing here needs a phone. Mark it ready when those are in and I'll merge it.

Record the endpoint and SysMMU configuration in the tested DT, date the draft kernel series and clarify the scope of the revised activation.

Assisted-by: LLM
@latentoperator
latentoperator marked this pull request as ready for review November 19, 2026 02:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants