Skip to content

omamobile-telephony-tegu: Phone and Messages on the Pixel 9a through omamobile-mmd (stacked on #367 and #363) - #371

Merged
derekross merged 9 commits into
derekross:mainfrom
W4HO-ham:tegu-mmd-attach
Nov 9, 2026
Merged

derekross merged 9 commits into
derekross:mainfrom
W4HO-ham:tegu-mmd-attach

Conversation

@W4HO-ham

@W4HO-ham W4HO-ham commented Nov 3, 2026

Copy link
Copy Markdown
Contributor

Stacked on #367 (and so #364, #356) and #363; mine is the top commit. Draft until the stack below merges.

TELEPHONY-APPS-PLAN (#370) step one: the 9a's sitd wiring packaged (it runs against an already active tegu modem; cheetah's and raven's start-up unchanged), and an opt-in omamobile-telephony-tegu package with the bridge, the omatel agent and an attended helper that starts only omamobile-mmd and the user agent, with readiness checks and a terminal confirmation. Installing it enables nothing.

On the palmtop (2026-11-03, by hand, the same steps as the helper): with the modem and sitd running, omamobile-mmd exported the modem on ModemManager's D-Bus API and handed the inbox's four texts over; with the user's omatel-agent started, Messages showed them, sent a text that arrived on the owner's phone, and Phone placed a call to the owner's phone with speakerphone audio. Two findings on the way: the agent is D-Bus activated but the app didn't activate it ("Messages service isn't running" until it was started), and sitd read the 9a's radio state 10 as "unknown", so the bridge showed the modem disabled and the bar an airplane glyph (fixed in #367's branch as 0.3.12).

All seven affected package suites (21 attachment tests, offline systemd verification), the new package's build and tools/pr-check pass; one reference-XML comparison is skipped.

@W4HO-ham

W4HO-ham commented Nov 3, 2026

Copy link
Copy Markdown
Contributor Author

Ran the packaged path on the palmtop (2026-11-03): omamobile-device-tegu 0.2.9, omamobile-modem 0.1.11 (installed normally, no -dd), omamobile-modem-sit 0.3.13 and omamobile-mmd 0.1.1 from this branch, the old /etc sitd copy moved aside as the README says. The packaged sitd unit required the running tegu starter, checked CP ONLINE and came up to "Calls and SMS service ready"; mmd exported the modem as state 11 (connected), SIM ready, radio on; the bar's cell watcher read connected, LTE. Added on top: 0.3.13, sitd reading radio state 10 as on (it was unknown, so the bar showed airplane). Not installed: omamobile-telephony-tegu itself, which asks for omatel-agent >= 0.2.4 and the palmtop's repository has 0.2.1; the agent from the repository worked with Messages and Phone. Signal strength reads -1 (the bar shows the searching glyph): the bridge doesn't report it yet.

@derekross

Copy link
Copy Markdown
Owner

Thanks. Splitting sitd's modem dependency into device drop-ins, with a virtual that forces them along, keeps the 7 Pro's unit exactly as it is, and I appreciate the care in the attach helper. While the stack below lands: the 9a plan's header still says no phone contact happened, while your comment reports a full run. Could you record that run with uname -r and the package versions, and add a tools/phone-smoke report? With #372 starting the bridge at boot, which way of starting it stays: the attended helper, the boot path, or both for different cases? #381 is now on main with radio state 10 and the signal decode, so those parts can drop in the rebase, and I'll pick the omamobile-mmd version at merge.

@W4HO-ham

W4HO-ham commented Nov 3, 2026

Copy link
Copy Markdown
Contributor Author

Rebased onto main and #367. The radio-state and signal-decode commits are dropped, since #381 has both. The plan (your notes on #370) now opens with what ran on 2026-11-02: kernel, versions, steps 1 to 6, and that opening Messages didn't start the agent. It cites #363 instead of the branch, and the task-report lines are gone. On which start stays: both, for different cases. #372's boot unit is for daily use (modem, sitd and mmd once the CP is ONLINE), and the attach helper is for a modem and sitd started by hand during bring-up and tests, where nothing should start the modem for you; the plan says so. The tools/phone-smoke report needs the palmtop, which is down until the morning; I'll add it then. mmd's version file says 0.1.2; set it as you like.

@derekross

Copy link
Copy Markdown
Owner

Thanks for recording the run and for the clear answer on the two ways to start the bridge. The split of sitd's modem dependency into device drop-ins checks out: on the 7 Pro the base unit plus cheetah's drop-in gives exactly today's Requires= and After=, and #398's boot target is untouched. The attach helper's preflight is careful in the right places: a typed confirmation, a re-check after it, and nothing that starts or stops the modem.

Before it comes out of draft:

  1. The tools/phone-smoke report is still to come (the plan says so too). Now that the palmtop is back, one run with this branch's packages installed would close it.
  2. Versions: this branch and reboot-to, omamobile-modem: a deliberate restart doesn't cost the next boot its modem #403 both claim device-cheetah 0.1.44 and device-raven 0.1.31, and with omamobile-device-tegu 0.2.9, device/tegu: the radio partitions tools/nv-backup reads #407 device-tegu 0.2.9. Git merges them without a conflict, so I'll set the next free numbers at merge (a call-audio fix of mine also bumps device-cheetah and modem-sit first) and give them here. No push needed for that.
  3. On my side, the order matters: modem-sit's new dependency on omamobile-sitd-policy is only satisfied by the new device-cheetah, and reboot-to, omamobile-modem: a deliberate restart doesn't cost the next boot its modem #403 makes me restage cheetah's boot files before that package can publish again. I'll land this after the restage, so no 7 Pro gets an update it can't resolve.

Two small ones: the plan's new line dates the removal of mmd's Conflicts= to 0.1.1, while this PR is what removes it (and the bullet above still lists it as built). And has #372's boot path run with this package installed? Its mmd drop-in's --service-check wants a registered radio, which may not be there yet when the boot unit queues mmd.

@W4HO-ham

W4HO-ham commented Nov 3, 2026

Copy link
Copy Markdown
Contributor Author

Thanks. The plan and the package README now put the Conflicts= removal in mmd 0.1.3, this PR. The bullet describing the unit no longer lists it as built, and says it was there up to 0.1.2 (87c356d). #372 to #380 are replayed on top.

Your question: no, #372's boot path hasn't run with omamobile-telephony-tegu installed. The palmtop has mmd 0.1.4 (from #380) without this package, so its drop-in and --service-check weren't in the path. In the boot-unit boot of 10:44 today, plain mmd started 3 s after the boot unit queued it, and the modem was already exported registered ("state 11, failed reason none, SIM ready, radio on"). So the radio was up in time there, but that doesn't cover the check's timing. After the palmtop's sleep series (about 18:00 MST), I'll install this branch's packages, reboot with the boot unit on, and post what --service-check does at boot, along with the tools/phone-smoke report.

@W4HO-ham

W4HO-ham commented Nov 4, 2026

Copy link
Copy Markdown
Contributor Author

Your question found a real bug, thanks. I ran #372's boot path with this package installed, on the palmtop (2026-11-03, 7.3.0-rc6-omamobile-tegu-s2mpg15-m1-g991042e2eff3-dirty, the package built on the palmtop from this branch). The modem came up ONLINE, and then the bridge's --service-check refused every start: Attach refused: bridge or sitd has an unexpected start dependency. mmd also restarted every 5 s.

Two causes:

Neither pulls the RIL or the CP into a start, so the check now allows both, and only those (657e7a7). It still refuses sitd wanting anything else and the bridge wanting anything at all. A --service-check refusal now exits 3, and the drop-in has RestartPreventExitStatus=3, so a refused bridge stays down instead of retrying. Two tests cover this (23 pass).

Re-run on the palmtop with the fixed helper and drop-in: systemctl reboot with the boot unit on. The check passed ({"sitd_pid": …, "modem_pid": …}), mmd took org.freedesktop.ModemManager1 and exported the modem (state 11, failed reason none, SIM ready, radio on), and NameHasOwner reads true.

The tools/phone-smoke report from that boot is below. The dmesg row is CPIF dropping packets for a channel no service has opened (in #372's §2.7 record too), and the backlight unit fails on every boot. The plan's line about mmd's Conflicts= was fixed earlier (87c356d). #372 to #380 are replayed on top.

### phone-smoke: tegu

Kernel `7.3.0-rc6-omamobile-tegu-s2mpg15-m1-g991042e2eff3-dirty`, tools at `330a7f2`, 2026-11-04T02:54Z.

| Check | Result | Detail |
| --- | --- | --- |
| boot (required) | ok | kernel 7.3.0-rc6-omamobile-tegu-s2mpg15-m1-g991042e2eff3-dirty, slot b, up 174s |
| storage guard (required) | ok | 62 partitions; only userdata and slot B's boot images writable |
| root (required) | ok | ext4 on userdata, 16G free |
| kernel health (required) | ok | no oops or BUG; 0 WARN, 6 error lines in dmesg (0 known, 6 new) |
| new dmesg errors | not ok | cpif: rx_demux: PCIE: ERR! umts_router is not opened x1; cpif: rx_demux: PCIE: ERR! N is not opened x1; cpif: pass_skb_to_demux: PCIE: umts_router<N: ERR! umts_router->recv_skb fail (N) x1 |
| systemd units | not ok | failed: systemd-backlight@backlight:19440000.dsi.0.service |
| USB gadget (required) | ok | bound to 11210000.usb |
| display | ok | card1-DSI-1 connected, 1080x2424 |
| GPU | ok | render node on panthor |
| session (required) | ok | Hyprland and the Omarchy shell running |
| lock | ok | PAM file present, the shell can lock (passwordPam true, locked true) |
| touch | ok | Synaptics TouchComm Touchscreen |
| buttons | ok | gpio-keys |
| Wi-Fi | skip | no wireless interface |
| Bluetooth | ok | hci0, powered true |
| battery | ok | 89% Unknown, 30 °C |
| sound cards | skip | none |
| sensors | skip | no sensor proxy |
| cameras | ok | 3 listed by libcamera |
| thermal | ok | hottest zone 30 °C |

Packages: hyprland 0.56.2-5.1, libcamera 0.7.2-4.22, mesa 1:26.2.4-1.4, omacam 0.3.3-1, omamessages 0.2.2-1, omamobile-apps 0.2.2-1, omamobile-audio 0.5.13-1, omamobile-device-tegu 0.2.12-1, omamobile-fingerprint 0.1.8-1, omamobile-firstboot 0.1.0-1, omamobile-gnss 0.2.7-1, omamobile-keyring 20261013-1, omamobile-mirror 0.1.1-1, omamobile-mmd 0.1.4-1, omamobile-modem 0.1.11-1, omamobile-modem-sit 0.3.36-1, omamobile-omarchy 0.1.31-1, omamobile-owl-widgets 0.1.2-1, omamobile-sensors 0.2.1-1, omamobile-shell 0.2.100-1, omamobile-telephony-tegu 0.1.0-1, omaphone 0.2.3-1, omarchy 4.0.6-1, omatel-agent 0.2.4-2

@derekross

Copy link
Copy Markdown
Owner

Thanks for running the boot path with this package installed. It caught a real problem: allowing exactly the bridge's own WantedBy and tmp.mount, and nothing more, is the right fix, and the re-run shows the check passing and mmd exporting the modem. Thanks for the smoke report too.

One thing in the new commit doesn't do what it says. systemd applies RestartPreventExitStatus= only to the main process. An ExecStartPre= that exits 3 is still restarted under Restart=on-failure, and with RestartSec=5 the start limit is never reached. I checked with two throwaway units on systemd 262. One had ExecStartPre=/bin/sh -c 'exit 3' and RestartPreventExitStatus=3, and it restarted five times, the same as without the line. The other had ExecCondition=/bin/sh -c 'exit 3', which skipped the start cleanly with no restart. Could you move the check to ExecCondition=/usr/bin/omamobile-tegu-mmd-attach --service-check and drop RestartPreventExitStatus=? Any exit from 1 to 254 then means "don't start, don't retry", without marking the bridge failed. Your passing re-run stays valid; only the refusal path changes, and a throwaway systemd-run --user -p Restart=on-failure -p ExecCondition="/bin/sh -c 'exit 3'" /bin/sleep 30 shows the behaviour in a minute, with no phone needed.

On my side, as before: #416 shipped device-cheetah 0.1.44 and modem-sit 0.3.30, which this branch's files now equal. At merge I'll give device-cheetah, device-raven, device-tegu, modem-sit and mmd the next free numbers after #403 and #367 (#407 has taken device-tegu 0.2.9 since). I'll also bring the new package's depends up to those versions, since modem-sit>=0.3.12 and mmd>=0.1.1 predate what it relies on. It lands after #403 and my restage of the 7 Pro's boot files.

A question from the smoke report: that boot shows no sound card and no sensor proxy, while the §2.7 boot in #372 had both. Was the AoC down that time?

@W4HO-ham

W4HO-ham commented Nov 6, 2026

Copy link
Copy Markdown
Contributor Author

I reproduced the refusal loop on systemd 262 and prepared the ExecCondition change with RestartPreventExitStatus removed. The existing package and unit tests pass (23 tests). I haven't installed that change on the palmtop yet; it is in a suspend test, so I'll verify the boot path when that finishes.

For the earlier smoke report: it shows neither a sound card nor a sensor proxy, and also no Wi-Fi interface. The saved boot excerpts don't establish whether AoC was down or why. I'll check the journal from that boot if it is still available, rather than guess.

@W4HO-ham

W4HO-ham commented Nov 6, 2026

Copy link
Copy Markdown
Contributor Author

I checked the palmtop's journal for the November 3, 19:55 MST smoke run. That boot is no longer retained; the oldest available journal begins at 20:37 MST. I can't tell from the remaining evidence whether AoC was down on that run.

@W4HO-ham

W4HO-ham commented Nov 6, 2026

Copy link
Copy Markdown
Contributor Author

The check is now ExecCondition=/usr/bin/omamobile-tegu-mmd-attach --service-check, and RestartPreventExitStatus= is gone (26751a9). The test asserts the condition and the absence of both old lines; 23 tests pass. On systemd 262, an ExecStartPre exiting 3 with RestartPreventExitStatus=3 restarted five times, while the ExecCondition skipped the start once, as you found.

It hasn't run on the palmtop yet, because the palmtop is in a suspend series. I'll do the boot path with it when that's done.

@W4HO-ham

W4HO-ham commented Nov 6, 2026

Copy link
Copy Markdown
Contributor Author

I installed the package from 26751a9 on the palmtop. The installed files match that head, the package integrity check is clean, and the modem, sitd and bridge stayed active. The real --service-check passes. In a disposable systemd unit, I gave the installed helper an invalid preflight root: it refused, systemd skipped the start, and there was one start with no retry. This covers the refusal path; the boot-path test is still to come.

The PR currently conflicts with main in packages/omamobile-device-raven/version: main has 0.1.32 and this branch has 0.1.31.

@W4HO-ham

W4HO-ham commented Nov 6, 2026

Copy link
Copy Markdown
Contributor Author

I ran the boot-path check with 26751a9 installed. During an orderly reboot the modem reported CP OFFLINE before shutdown. On the next boot the modem came ONLINE, sitd reached LTE and IMS voice/SMS registration, and the bridge started and exported the ready SIM. Its service check passed, with zero bridge restarts. The protected-partition backup check matched before and after.

The required smoke checks passed, but the full smoke was not clean: the saved kernel journal has the same Wi-Fi invalid-chanspec warning seen on the preceding boot and 46 AoC mailbox queue-full messages (45 on the preceding boot). The Wi-Fi warning is covered by kernel #191. The boot-reason field says watchdog; the matching archived shutdown log shows systemd's requested reboot and the kernel reaching Restarting system, so that field alone doesn't establish a spontaneous watchdog reset.

@derekross

Copy link
Copy Markdown
Owner

Thank you, @W4HO-ham. ExecCondition is exactly what I meant, and it's good to see it run on the palmtop both ways: the refusal path skipped the start once with no retry, and the boot path passed with zero bridge restarts and a matching NV check. I re-ran the package's tests on this head (23 pass), and nothing in the docs still mentions the old lines. Thank you also for looking up the 11-03 journal. It's fine that it's gone.

When you're happy with it, please take this out of draft. Everything else is on my side at merge. I'll resolve the device-raven version conflict you spotted and give every package here the next free number. I'll also raise depends to match, since omamobile-device-tegu>=0.2.9 is already met by the published 0.2.9 from #407, which has none of this. And I'll write the squash message myself, because the branch still carries #367's two commits underneath. Since #403 is on hold for now, I'll land this one first if my 7 Pro boot files are current when I merge.

@W4HO-ham
W4HO-ham marked this pull request as ready for review November 7, 2026 14:00
@W4HO-ham

W4HO-ham commented Nov 7, 2026

Copy link
Copy Markdown
Contributor Author

Thanks. Both paths have run on the palmtop and the tests pass, so I've taken it out of draft. Versions, depends and the squash message are yours at merge, as you said.

@derekross

Copy link
Copy Markdown
Owner

Thank you, @W4HO-ham, and thanks for taking it out of draft. I have nothing
more to ask of you here. When I merge I'll take main's docs/DEVICES.md: that
conflict comes from #367's commits underneath, and main already has the 9a row
they added. I'll also set the versions to the next free numbers (mmd too,
since #461 also uses 0.1.3), raise depends to match, and write the squash
message without the "stacked on" part, since #367 and #363 have both landed.

Trace the existing app, session agent, ModemManager-compatible bridge and sitd path. Record the tegu installation and unit prerequisites still missing on main, and an attended bridge-only test that preserves the running modem stack.

Document the passing offline bridge and apps package suites, activation and audio limits, and the eventual board-specific image integration.
Keep sitd's canonical name while moving the modem dependency into device-owned policy drop-ins. Tegu checks an already running modem and CP ONLINE; cheetah and raven retain their existing starter relationship. Require the device policy capability during RIL upgrades.

Add an opt-in telephony package with a bridge policy and attended helper. It checks live sitd readiness, refuses legacy unit shadows and competing managers, and starts only mmd and the graphical user's agent. The service recovery check permits live calls and registration loss; no helper enables or recovers the modem.

Remove mmd's service-level conflict so a bridge start cannot stop another modem manager. Package exclusion and D-Bus name ownership remain. Retain the previously developed SIM, safe-release and audio changes in this stack.

Validate the helper, dependency policies and composed units offline; run all seven affected/dependency package suites and build the new architecture-independent package without installing dependencies.
…start the bridge, and a plan that reads as one

Record the palmtop run of Phone and Messages (kernel, versions, what passed, and that opening an app didn't start the agent), say which start fits which case, cite derekross#363 instead of a branch, and drop the lines that read like a task report.
…in 0.1.3, with this pull request

The plan and the README dated the removal of the unit's Conflicts=ModemManager.service to mmd 0.1.1, and the bullet above still listed it as built; this pull request is what removes it, in 0.1.3.
…WantedBy and tmp.mount, and a refusal isn't retried

With derekross#372's boot unit on and this package installed, the bridge's ExecStartPre refused every start: sitd wanted omamobile-mmd.service (the bridge's own WantedBy, once enabled) and both units wanted tmp.mount (systemd adds it). Neither pulls the RIL or the CP into a start, so both are allowed now. A refusal of --service-check exits 3, which the drop-in names in RestartPreventExitStatus=, so a refused bridge stays down instead of retrying every five seconds.
RestartPreventExitStatus= applies only to the main process, so an ExecStartPre= that refused with exit 3 was still restarted every RestartSec under Restart=on-failure. As an ExecCondition=, a refusal (any exit from 1 to 254) skips the start without a retry and without marking the bridge failed. Checked on systemd 262: ExecStartPre exiting 3 with RestartPreventExitStatus=3 restarted five times, ExecCondition exiting 3 skipped the start once.
…: versions past main after the rebase

The rebase onto main dropped the two commits main already had and left these three packages changed without a version above main's.
@derekross

Copy link
Copy Markdown
Owner

Thank you, @W4HO-ham, for rebasing this onto main and fixing the versions
the rebase had left behind. I compared it with the head I reviewed last
time: your commits are the same apart from device-raven's version line, and
the DEVICES conflict is gone because main already has those lines. The
package tests pass on today's main, with #461 on top too.

The thing I was waiting for on my side, the 7 Pro's boot files on the build
machine, is sorted, and nothing here touches them. So this is ready. When I
merge I'll:

  • set the versions to the next free numbers (device-cheetah, device-tegu,
    modem-sit and mmd each collide with another open PR);
  • raise telephony-tegu's omamobile-device-tegu dependency to match;
  • publish the device packages together with modem-sit;
  • write the squash message without the "stacked on" part.

Thanks for your patience with this one.

@derekross
derekross merged commit 9e9ae70 into derekross:main Nov 9, 2026
2 checks passed
W4HO-ham added a commit to W4HO-ham/omarchy-mobile that referenced this pull request Nov 9, 2026
robertzas pushed a commit to robertzas/omarchy-mobile that referenced this pull request Nov 14, 2026
…rough omamobile-mmd (derekross#371)

* tegu: plan Phone and Messages integration through mmd

Trace the existing app, session agent, ModemManager-compatible bridge and sitd path. Record the tegu installation and unit prerequisites still missing on main, and an attended bridge-only test that preserves the running modem stack.

Document the passing offline bridge and apps package suites, activation and audio limits, and the eventual board-specific image integration.

* tegu: attach Phone and Messages to an attended running sitd

Keep sitd's canonical name while moving the modem dependency into device-owned policy drop-ins. Tegu checks an already running modem and CP ONLINE; cheetah and raven retain their existing starter relationship. Require the device policy capability during RIL upgrades.

Add an opt-in telephony package with a bridge policy and attended helper. It checks live sitd readiness, refuses legacy unit shadows and competing managers, and starts only mmd and the graphical user's agent. The service recovery check permits live calls and registration loss; no helper enables or recovers the modem.

Remove mmd's service-level conflict so a bridge start cannot stop another modem manager. Package exclusion and D-Bus name ownership remain. Retain the previously developed SIM, safe-release and audio changes in this stack.

Validate the helper, dependency policies and composed units offline; run all seven affected/dependency package suites and build the new architecture-independent package without installing dependencies.

* device/tegu TELEPHONY-APPS-PLAN: the 2026-11-02 run, the two ways to start the bridge, and a plan that reads as one

Record the palmtop run of Phone and Messages (kernel, versions, what passed, and that opening an app didn't start the agent), say which start fits which case, cite derekross#363 instead of a branch, and drop the lines that read like a task report.

* TELEPHONY-APPS-PLAN, omamobile-telephony-tegu: mmd's Conflicts= goes in 0.1.3, with this pull request

The plan and the README dated the removal of the unit's Conflicts=ModemManager.service to mmd 0.1.1, and the bullet above still listed it as built; this pull request is what removes it, in 0.1.3.

* omamobile-telephony-tegu: the bridge's service check accepts its own WantedBy and tmp.mount, and a refusal isn't retried

With derekross#372's boot unit on and this package installed, the bridge's ExecStartPre refused every start: sitd wanted omamobile-mmd.service (the bridge's own WantedBy, once enabled) and both units wanted tmp.mount (systemd adds it). Neither pulls the RIL or the CP into a start, so both are allowed now. A refusal of --service-check exits 3, which the drop-in names in RestartPreventExitStatus=, so a refused bridge stays down instead of retrying every five seconds.

* omamobile-telephony-tegu: the bridge's service check is an ExecCondition

RestartPreventExitStatus= applies only to the main process, so an ExecStartPre= that refused with exit 3 was still restarted every RestartSec under Restart=on-failure. As an ExecCondition=, a refusal (any exit from 1 to 254) skips the start without a retry and without marking the bridge failed. Checked on systemd 262: ExecStartPre exiting 3 with RestartPreventExitStatus=3 restarted five times, ExecCondition exiting 3 skipped the start once.

* omamobile-device-cheetah 0.1.48, device-tegu 0.2.10, modem-sit 0.3.32: versions past main after the rebase

The rebase onto main dropped the two commits main already had and left these three packages changed without a version above main's.

* omamobile-telephony-tegu: depend on the device-tegu that lands with it (0.2.10)

* omamobile-telephony-tegu: the dependency test reads device-tegu>=0.2.10 too

---------

Co-authored-by: Derek Ross <derekross@gmail.com>
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