Skip to content

Update from scarthgap to walnascar #298

@MagneFire

Description

@MagneFire

Yocto Styhead has been released in October, we should consider updating to this version.

See:

Currently tracking changes via:

PRs are available at:

Current state:

  • Community packages have not been built and/or tested yet.

Platforms:

Name Built Tested Notes
anthias Don't own watch
bass Don't own watch
beluga GPS needs to be tested
catfish
dory 🟡 Tilt-to-wake doesn't work (existing issue)
firefish Don't own watch
harmony Don't own watch
hoki
inharmony Don't own watch
koi Don't own watch
lenok
minnow 🟡 Stuck on boot logo, USB doesn't work (existing issue)
mooneye Don't own watch
narwhal
nemo Don't own watch
pike
qemux86
ray
rinato
sawfish
skipjack Don't own watch
smelt
sparrow 🟡 Tilt-to-wake doesn't work (existing issue)
sparrow-mainline 🟡 Doesn't boot (existing issue)
sprat Don't own watch
sturgeon
swift
tetra Launcher crashes Fixed
triggerfish Don't own watch
wren Don't own watch

Legend:

  • ✅: Tested, works as intended
  • ⚪: Has not been tested
  • 🟡: Tested, but has existing issues
  • ❌: Tested, but has new issues

Notes

dory: tilt-to-wake doesn't work

Log from walnascar

Mar 06 14:40:06 dory sensorfwd[405]: "incorrect event version (version=-1254940152, expected=104"
Mar 06 14:40:06 dory kernel: lis3dsh_acc 4-001e: DEVICE IN TILT
Mar 06 14:40:10 dory sensorfwd[405]: "incorrect event version (version=-1254928864, expected=104"
Mar 06 14:40:10 dory kernel: lis3dsh_acc 4-001e: DEVICE IN TILT
Mar 06 14:40:15 dory sensorfwd[405]: "incorrect event version (version=-1254914204, expected=104"
Mar 06 14:40:15 dory kernel: lis3dsh_acc 4-001e: DEVICE IN TILT

scarthgap has the same issue:

Mar 06 14:52:32 dory kernel: lis3dsh_acc 4-001e: DEVICE IN TILT
Mar 06 14:52:32 dory sensorfwd[406]: "incorrect event version (version=13, expected=104"
Mar 06 14:52:37 dory sensorfwd[406]: "incorrect event version (version=1695154228, expected=104"
Mar 06 14:52:37 dory kernel: lis3dsh_acc 4-001e: DEVICE IN TILT
Mar 06 14:52:40 dory kernel: lis3dsh_acc 4-001e: DEVICE IN TILT
Mar 06 14:52:40 dory sensorfwd[406]: "incorrect event version (version=1173, expected=104"
Mar 06 14:52:45 dory sensorfwd[406]: "incorrect event version (version=52, expected=104"
Mar 06 14:52:45 dory kernel: lis3dsh_acc 4-001e: DEVICE IN TILT

The most recent sensorfw update was with: AsteroidOS/meta-asteroid@1adfde8, further research needed, but outside of the scope of this work.

sparrow: tilt-to-wake doesn't work

Nothing is logged, but also occurs on current scarthgap...
Further testing, it seems that tilt-to-wake works on a nightly release from January.

sparrow-mainline: Doesn't boot

For some reason can't get the kernel to boot always getting:

Sending 'boot.img' (12901 KB)                      OKAY [  0.408s]
Booting                                            FAILED (remote: 'dtb not found')
fastboot: error: Command failed

This also occurs with the current scarthgap.

Looking at the recipe (https://github.com/AsteroidOS/meta-smartwatch/blob/9ce9e605cb9ffe27a280624c295393d3a29a3d96/meta-sparrow/recipes-kernel/linux/linux-sparrow-mainline_6.0.bb#L24) a file ending with the name *-dtb should be created. However, this file does not seem to exist.

tetra: Launcher crashes

For some reason asteroid-launcher crashes. This is reported in the logs:

Mar 06 15:25:08 tetra asteroid-launcher[746]: Unable to load hwcomposer module
Mar 06 15:25:08 tetra asteroid-launcher[746]: QPA-HWC: backend in HwComposerContext returned 0

The Qt hwcomposer backend seems to try and load hwcomposer.tetra.so or hwcomposer.default.so. But this does not exist on tetra instead hwcomposer.bcm_java.so exists.
Providing a symlink solves this issue. But the launcher crashes later on, nothing is logged but gdb reports:

Thread 12 "QSGRenderThread" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0xa8b782e0 (LWP 808)]
0x00000000 in ?? ()
(gdb) bt full

#0  0x00000000 in ?? ()
No symbol table info available.
#1  0xb41bd8e0 in ?? ()
No symbol table info available.
Backtrace stopped: previous frame identical to this frame (corrupt stack?)
(gdb) 
#0  0x00000000 in ?? ()
No symbol table info available.
#1  0xb41bd8e0 in ?? ()
No symbol table info available.
Backtrace stopped: previous frame identical to this frame (corrupt stack?)

It's worth noting that this does not happen on our current nightly builds.
I've already tried if downgrading libhybris works, but it results in the same issues.

Solved via AsteroidOS/meta-smartwatch@8ced113

minnow: Stuck on boot logo, USB doesn't work

This is not a new issue, minnow has been broken for a while now. While the kernel seems to boot fine it doesn't seem to successfully load asteroid-launcher for some reason.
This one is particularly hard to debug as USB doesn't work either anymore.
MCE seems to run as the backlight seems to change when the power button is pressed.

The only USB activity observed is as follows:

[ 1811.519376] usb 1-1: new high-speed USB device number 9 using xhci_hcd
[ 1811.642505] usb 1-1: no configurations
[ 1811.642512] usb 1-1: can't read configurations, error -22
[ 1811.754331] usb 1-1: new high-speed USB device number 10 using xhci_hcd
[ 1812.141455] usb 1-1: no configurations
[ 1812.141465] usb 1-1: can't read configurations, error -22
[ 1812.141594] usb usb1-port1: attempt power cycle
[ 1813.146372] usb 1-1: new high-speed USB device number 11 using xhci_hcd
[ 1813.159159] usb 1-1: New USB device found, idVendor=18d1, idProduct=d001, bcdDevice=ff.ff
[ 1813.159168] usb 1-1: New USB device strings: Mfr=2, Product=3, SerialNumber=4
[ 1813.159173] usb 1-1: Product: Moto 360
[ 1813.159176] usb 1-1: Manufacturer: Motorola
[ 1813.159179] usb 1-1: SerialNumber: P027XY2LNL
[ 1813.485150] usb 1-1: USB disconnect, device number 11

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions