ubuntu-phone team mailing list archive

[Ricardo Salveti de Araujo <ricardo.salveti@xxxxxxxxxxxxx>]

On Thu, Feb 27, 2014 at 4:46 AM, Didier Roche <didrocks@xxxxxxxxxx> wrote:
> Le 27/02/2014 06:04, Ricardo Salveti de Araujo a écrit :
>> On Thu, Feb 27, 2014 at 1:44 AM, Ricardo Salveti de Araujo
>> <ricardo.salveti@xxxxxxxxxxxxx> wrote:
>>> On Wed, Feb 26, 2014 at 7:22 PM, Ricardo Salveti de Araujo
>>> <ricardo.salveti@xxxxxxxxxxxxx> wrote:
>> Alright, flashed manta again with 202 and already got the crash during
>> first boot.
>> Program terminated with signal SIGSEGV, Segmentation fault.
>> #0  0x4160d8f8 in ?? ()
>>
>> current build number: 202
>> device name: manta
>> channel: trusty-proposed
>> last update: 2014-02-27 04:50:01
>> version version: 202
>> version ubuntu: 20140222.1
>> version device: 20140115.1
>>
>> So it is indeed not related with 4.4.2.
>
> Hey Ricardo!
>
> Nice investigation. Ok, so the old crash that we didn't get a fix for
> (because it was 5.0 and 5.2 was around the corner) is finally hitting us as
> I was fearing when we, as a team, didn't investigate it. However, something
> changed on the platform to make it way more often now? I guess all the dbus
> timeout (as the app/service crashes) and other issues are due to that. So
> instance of having one crash on the whole testsuite (550 tests), we now have
> it 10/15 times on the same test suite. Any idea of what can make this crash
> happening more now (scheduler change, so more races?)

Well, this issue was always there on Manta, as I could easily
reproduce it with the older Android stack. I believe this is probably
a race, and it's just happening more with Mako now as the 4.4.2 stack
is better and faster.

Cheers,
-- 
Ricardo Salveti de Araujo