Skip to content

omamobile-shell 0.2.35: the Pixel 9a's display scale - #117

Closed
W4HO-ham wants to merge 11 commits into
derekross:mainfrom
W4HO-ham:shell-tegu-scale
Closed

W4HO-ham wants to merge 11 commits into
derekross:mainfrom
W4HO-ham:shell-tegu-scale

Conversation

@W4HO-ham

Copy link
Copy Markdown
Contributor

Stacked on #115 (and #99 under it); only the last commit is new.

omamobile-setup-user seeded every phone's Hyprland scale at 3, which suits the Pixel 7 Pro (1440x3120 at 3 is 480x1040 logical). On the Pixel 9a's 1080x2424 panel, 3 is only 360 pixels wide. On 2026-10-25 the 9a's owner picked 8/3 (405x909, "about right" next to his iPhone; device/tegu notes). So the 9a now gets 8 / 3: monitors.lua is Lua, so the expression is its value, and Hyprland takes 2.6667 (1080 / 405). Other phones keep 3. It's still seeded once, so a scale a user set stays.

GDK_SCALE: that step still runs only when the monitor scale is exactly 3, so on the 9a X11 GTK apps keep Omarchy's 2. That's the 9a's state today; whether 3 suits it better is a choice for the owner, not something I'd guess.

Tests: the seeded line stays valid for both values (3 and 8 / 3), and Lua evaluates 8 / 3 to 2.6667, which gives 405 logical pixels on 1080. test/all and tools/pr-check pass.

A poll loop is the NFC radio searching for cards. On the Pixel 9a it costs
about 60 mA with the screen off (252 mA against 315 mA, awake, 10-27),
and stock Android polls only while the screen is on. omamobile-nfc-open
now stops neard's poll loop once every display connector has been DPMS
off for 4 s, and starts it again when the screen lights, unless the NFC
switch is off. Tested on the 9a: no controller interrupts with the screen
off, polling again within 2 s of it lighting.
@derekross

Copy link
Copy Markdown
Owner

Thanks, @W4HO-ham. A 360-pixel-wide desktop on the 9a is too small, and 8/3 gives a whole number of logical pixels both ways (405x909), so it's a good value. Nothing changes on the 7 Pro: it still gets 3, and the line comes out the same.

Two things before it merges:

  1. The note it cites isn't there. The comment and the PR cite "device/tegu NOTES, 2026-10-25" for the owner's choice of 8/3, but tegu's NOTES has no such entry, and its summary table still says "1080x2424, scale 2". Please add a short dated entry (what was compared, on which phone) and fix that table row. There are two 9a owners now, so it's worth a line saying this sets the default for every new 9a install.
  2. The package test fails on the 9a. omamobile-setup-user reads the real /proc/device-tree/compatible, and test/setup-user.sh expects omarchy_monitor_scale = 3. Building the package on the 9a (makepkg runs test/all) fails on exactly the phone this is for, and no test ever runs the 9a branch. Please read the compatible through a path the test can override (for example ${OMAMOBILE_DT:-/proc/device-tree}/compatible) and run setup-user once more with a fake tegu compatible, checking 8 / 3 and that the GDK_SCALE step is skipped.

…ays dark

The stop was a single shot on the lit-to-dark transition, so anything that started polling afterwards kept it on until the next screen cycle: the unblock path's start_polling retries, the 9a's sleep hook after a wake, or neard restarted by systemd. Now every 2 s check stops any poll loop once the screen has been dark for 4 s, and start_polling powers the adapter but leaves the loop off while it's dark. The one-shot stop after an unblock is gone. test/nfc-screen.py checks screen_lit() against temporary sysfs trees.
The OWL host answered every device and system query as a Pixel 7 Pro (gs201, 1440x3120, codename cheetah), so a widget on the Pixel 9a or any other port showed the wrong phone. The model and the compatible list now come from the device tree; known boards add the panel's size, rate and density, and an unknown one reports its model and SoC with no display.
…atomically

The watchdog's jump test changes a control from a background job while the watchdog reads the fake mixer's state file, and CI once caught the read halfway through the write (JSONDecodeError, the test failed). Both fakes now write a temporary file and rename it over the state.
The flake had a second writer: JsonCtl, the json: control backend the tests run the speaker code against, rewrites the fake mixer's state on every heartbeat read, in place. The fake mixer reading at that moment saw an empty file. JsonCtl now writes a temporary file and renames it, as the fakes do. Nothing changes on the palmtop: JsonCtl is used only with OMAMOBILE_CTL_BACKEND=json:.
…s first reading

The jump tests changed the estimate 0.4 s after starting the watchdog. On a slow CI runner the watchdog wasn't reading yet, took the raised value as its baseline, saw no jump and didn't mute. The jump now waits for the watchdog's start line and a few intervals, and the watchdog runs 60 iterations.
…top rate

As asked on the pull request: display.hz is what the panel can do, not the rate it runs at now (the 9a's published kernel still runs it at 60 Hz).
omamobile-setup-user seeded every phone's Hyprland scale at 3, the Pixel 7 Pro's (480x1040 logical). On the Pixel 9a's 1080x2424 panel that is 360 pixels wide, smaller than its owner chose on 2026-10-25: 8/3, 405x909. The 9a now gets 8 / 3 (monitors.lua is Lua, so the expression is its value); other phones keep 3. Seeded once, as before, so a user's own scale stays.
… tegu's NOTES

setup-user reads the compatible through OMAMOBILE_DT, so its test no longer depends on the machine it runs on (it failed when the package was built on the 9a itself). The test now runs a 7 Pro and a 9a device tree and checks 8 / 3 and that GDK_SCALE is left alone on the 9a. tegu's NOTES gets the dated entry for the 8/3 choice and its display row says so.
@W4HO-ham

Copy link
Copy Markdown
Contributor Author

Both done, thanks.

  1. NOTES: tegu's NOTES now has the dated entry. It records what was compared (scale 2, 540x1212, against 8/3, 405x909, on the reference 9a's running desktop, next to an iPhone) and says this sets the default for every new 9a install. The display row in the summary table now says scale 8/3 since that date.
  2. The test: omamobile-setup-user reads the compatible through ${OMAMOBILE_DT:-/proc/device-tree}. test/setup-user.sh gives every run a fake 7 Pro device tree by default, so the result no longer depends on where it runs. It also runs once more with a fake 9a tree and checks 8 / 3 and that GDK_SCALE stays 2.

Rebased onto #115's new head.

derekross added a commit that referenced this pull request Oct 28, 2026
…ry work

Open PRs hold device-cheetah 0.1.35 (#102) and omamobile-shell 0.2.33
to 0.2.36 (#99, #115, #117, #124); this branch takes the next free ones.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
derekross added a commit that referenced this pull request Oct 28, 2026
0.2.33 to 0.2.37 are held by #99, #115, #117, #124 and the low-battery
branch.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
derekross added a commit that referenced this pull request Oct 28, 2026
…ry work

Open PRs hold device-cheetah 0.1.35 (#102) and omamobile-shell 0.2.33
to 0.2.36 (#99, #115, #117, #124); this branch takes the next free ones.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@derekross

Copy link
Copy Markdown
Owner

Thanks, @W4HO-ham, both points are done, and done well:

  1. tegu's NOTES has the dated entry: what was compared, on which phone, and that 8/3 is now the default for every new 9a install. The summary row says scale 8/3 since that date.
  2. omamobile-setup-user reads the compatible through OMAMOBILE_DT, and the test gives every run a fake 7 Pro tree, so the result no longer depends on the machine it runs on. The 9a run checks 8 / 3 and that GDK_SCALE stays at 2. To make sure that run really tests something, I swapped the 9a's value back to 3 in the script, and the test failed as it should.

On the 7 Pro nothing changes: it still gets local omarchy_monitor_scale = 3, and the GDK_SCALE step runs as before. The package tests pass here with #115 under this and #124 and my touch batch on top.

A small optional one: the new NOTES entry says "7.3.0-rc6 integration kernel". If you still have the full uname -r from that session, please put it in, as the other entries do. It doesn't hold up the merge.

This merges right after #115. Once #115 is squash-merged, this branch will conflict on the version file only. Put your two commits onto main the same way:

git fetch origin
git checkout -B shell-tegu-scale origin/main
git cherry-pick d7d7220b82 02e8d5ed01
git push --force-with-lease origin shell-tegu-scale

I checked that sequence too: both commits apply cleanly after #99 and #115, and give this branch's tree. As with #115, I can do it myself if you'd rather.

@W4HO-ham

Copy link
Copy Markdown
Contributor Author

Added the full release to the NOTES entry (df271d6, a third small commit). Once #99 and #115 are in, I'll redo this branch on main the way you described, with the three commits cherry-picked.

@derekross

Copy link
Copy Markdown
Owner

Thanks, @W4HO-ham, the NOTES entry now has the full release, so that's everything for this PR.

Since there are three commits now, the list for after #99 and #115 are in is:

git fetch origin
git checkout -B shell-tegu-scale origin/main
git cherry-pick d7d7220b82 02e8d5ed01 df271d605b
git push --force-with-lease origin shell-tegu-scale

I tried that sequence on main with #99 (this head) squashed in and #115's two commits picked on top. All three apply cleanly, the package tests pass, and the omamobile-shell tree comes out the same as this branch's, apart from #99's latest omamobile-nfc-open change, which this branch was stacked below. If you'd rather I do the cherry-picks myself, just say so.

# Conflicts:
#	packages/omamobile-shell/plugins/omamobile.home/owl/phone.js
#	packages/omamobile-shell/version
@W4HO-ham

Copy link
Copy Markdown
Contributor Author

Merged tonight's main, version to 0.2.50 (it sits on #115, now 0.2.49). The phone.js export-list conflict is the same one as #115 and resolved the same way, keeping your CLOCK and frame entries alongside the board ones. Nothing dropped. pr-check passes and all 239 package checks pass.

@W4HO-ham

Copy link
Copy Markdown
Contributor Author

One thing I noticed while running the suite repeatedly on this stack, not caused by it: test/owl-idle's assertion that the widget tick slows while the home screen is paused is timing-sensitive. It failed twice with different measured values while I had two suites running at once, then passed four times in a row on this branch and three times on plain main. Nothing in these branches touches that code, so it's only a heads-up in case it ever fails in CI under load.

@derekross derekross closed this in def7efe Oct 31, 2026
@derekross

Copy link
Copy Markdown
Owner

Same story as #115: I put your change onto main as 0.2.51, authored by you; setup-user and the package tests pass. Heads-up: #204 adds a per-device scale seed too, so that one will adapt to this.

derekross added a commit that referenced this pull request Oct 31, 2026
… rounds

L-348 to L-361: shared drivers judged against every int-line user (#150),
board pins against every stock user (#151), ownership flags under the
consumer's lock (#152), /chosen children and Google's panel constants
(#154), same-day notes PRs on one device (#208/#209), merge simulations for
DEVICES rows, ours included (#165), stacked PRs that merged main (#115/#117),
titles behind renumbered versions, a contributor merge that dropped the PR
(#99), smoke package lists as provenance and "owner-authorized" writes
(#200), promised fixes and named tags checked on the remote (#153/#157/#158),
and file swaps into old packages (#193). C-23 (usbc-deadman's device scope,
kernel #107) and C-24 (BNEP in a lynx-only fragment, #98).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
robertzas pushed a commit to robertzas/omarchy-mobile that referenced this pull request Nov 14, 2026
omamobile-setup-user seeded every phone's Hyprland scale at 3, which
leaves the 9a's 1080x2424 panel only 360 logical pixels wide. The 9a now
gets 8 / 3 (405x909, its owner's choice in the tegu notes); monitors.lua
is Lua, so the expression is the value. Other phones keep 3, and it is
still seeded once, so a scale the user set stays.

Rebuilt on main by the maintainer from the pull request's own change;
renumbered to 0.2.51.

Closes derekross#117

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
robertzas pushed a commit to robertzas/omarchy-mobile that referenced this pull request Nov 14, 2026
… rounds

L-348 to L-361: shared drivers judged against every int-line user (derekross#150),
board pins against every stock user (derekross#151), ownership flags under the
consumer's lock (derekross#152), /chosen children and Google's panel constants
(derekross#154), same-day notes PRs on one device (derekross#208/derekross#209), merge simulations for
DEVICES rows, ours included (derekross#165), stacked PRs that merged main (derekross#115/derekross#117),
titles behind renumbered versions, a contributor merge that dropped the PR
(derekross#99), smoke package lists as provenance and "owner-authorized" writes
(derekross#200), promised fixes and named tags checked on the remote (derekross#153/derekross#157/derekross#158),
and file swaps into old packages (derekross#193). C-23 (usbc-deadman's device scope,
kernel derekross#107) and C-24 (BNEP in a lynx-only fragment, derekross#98).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.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