Repository navigation
Conversation
The board code already reads the model, maker, codename and SoC from the device tree, so husky shows as "Pixel 8 Pro" on zuma. Only the panel's size, density and top refresh rate come from the BOARDS table, which had cheetah and tegu but not husky, so android.system.display was null there. Add husky: 1344x2992, 120 Hz, 489 ppi (the DT's 70 x 155 mm gives 488 to 490). Test: boardFrom on husky's model and compatible. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
|
Thanks, Lef. That's a tidy fix. I checked the numbers: 1344x2992 at 120 Hz is right for the HK3, and the 70 x 155 mm in the panel driver comes out at about 490 ppi, so 489 is fine. The only problem is the version. On testing: a unit test is enough for a table row like this. Once a published shell with it reaches the development channel, a screenshot of the home widget on your 8 Pro would be a nice confirmation, but I won't hold the merge for it. |
0.2.64 went to the Pixel 8 Pro's panel row (#251). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…erekross#251) The Pixel 8 Pro's HK3 panel (1344x2992, 120 Hz, about 490 ppi) in the shell's board table, keyed on google,zuma-husky, with a test (Somnius). Renumbered to 0.2.64 at merge; 0.2.63 went to the voice release. Closes derekross#251 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
What and why
My 8 Pro's home widget says "Pixel 7 Pro" and gs201. The package on my phone is omamobile-shell 0.2.45, the published development channel still has nothing newer, and 0.2.50 (W4HO's #115, which Derek put on main) already reads the model, maker, codename and SoC from the device tree, so on husky it gives Pixel 8 Pro, Google, husky and zuma. The one thing it can't know is the panel, because BOARDS only has cheetah and tegu, so android.system.display is null on husky. This adds husky's row: 1344x2992, 120 Hz, 489 ppi (the DT's 70 x 155 mm gives 488 to 490 ppi, 120 Hz is the panel's top rate, the same thing the other rows mean). One test line for boardFrom on husky's model and compatible.
Device and build
uname -r): 7.3.0-rc6-omamobile-push1-gc7a0b2ef2f66-dirty (Brett's c6 plus the commits of kernel omamobile-modem 0.1.10: the modem profile starts once per boot #163)Checks (no phone needed)
tools/pr-check origin/mainpassestests/allpassestests/packages: omamobile-shell's own tests pass except the call-watch one, which importsdbus, and omamobile-mmd and omamobile-sensors fail the same way on my desktop (no pythondbusandgihere), none of them touched by this changeversionbumped, and the new version is in the commit subjectCo-Authored-By:orAssisted-by:trailer, never aSigned-off-by:On the phone
userdata, weakens the storage guard, or reads registers from userspace (AGENTS.md §1)tools/phone-smoke: not pasted, it would be a report of the battery kernel on my phone and says nothing about this change.