Announce: TILT tracking Androidization trees

Andy Green at
Mon Oct 10 04:49:29 UTC 2011

On 10/07/2011 05:59 AM, Somebody in the thread at some point said:

Hi -

>>   - Suspend borked
>>   - WLAN borked

> It passes:
> Boots 0xBench Busybox YouTube Ethernet ADB GLMark2 Wi-Fi Bluetooth

Actually WLAN is broken on that build.  It will start up, associate and 
transfer data, but it will not recover from runtime_pm taking it idle if 
it sits there for a bit.  You just get a bunch of ERP timeout errors.

Not sure if that's worth adding to your WLAN test arrangements or not 
but that is the case.

We have worked around that in current tilt-android-tracking.

> It fails:
> Resumes

Several bugs contributed to that but I think I cleared them all now. 
Our vanilla tracking branch has mem suspend working fine.  On our 
current Android tracking HEAD it suspends and resumes fine, but on 
resume immediately returns to suspend, I presume either due to what the 
rootfs is telling it to do, or possibly still some bug on our side.  So 
I updated the status on the launchpad bug.

> It also only boots on the 4430 not the 4460.

That should also be solved thanks to Seb Jan.

> The following bugs are related to this effort so far:
> Android tracking-panda build 9 fails to boot on 4460

The problem we had was not to do with U-Boot but on our side fwiw

> Android tracking-panda build 9 crashes during suspend
>   __linux__ is not getting defined in android-build

I took Dave Martin's patchset in (not just drm one) so that should also 
be solved.

Only live issue I am expecting at the moment is figure out why we go 
straight back to suspend after what is a full resume.  That only happens 
on Android, not our vanilla tracking tree fwiw.


Andy Green | TI Landing Team Leader │ Open source software for ARM SoCs | Follow Linaro  -!/linaroorg -

More information about the linaro-dev mailing list