On Fri, Apr 4, 2014 at 2:39 AM, Counihan, Tom <tom.couni...@intel.com> wrote:
>
>
>> -----Original Message-----
>> From: ivi-boun...@lists.tizen.org [mailto:ivi-boun...@lists.tizen.org] On 
>> Behalf
>> Of Jussi Laako
>> Sent: Friday, April 04, 2014 10:15 AM
>> To: ivi@lists.tizen.org
>> Subject: Re: Understanding AMB dependencies on Murphy
>>
>> On 2.4.2014 23:59, Rees, Kevron wrote:
>> > One possible advantage murphy has over AMB is that murphy rules do not
>> > have to be compiled like AMB plugins.  This is because Murphy uses lua
>> > scripting.  I'm not sure if that feature is necessarily a game-changer
>> > though.  I'm not sure how much system implementers will care whether
>> > or not they have to compile rules.
>>
>> To me, as a realtime system guy, the question is more like if the policy 
>> handling
>> chain is O(1) or not.
>
> That's a really great observation Jussi. It triggers a thought on the time 
> domain.
> What entity is the master clock, and is there an expectation that a) clock is 
> synchronised across the entire system and/or b) are sources expected to time 
> stamp their events or the receiving entity expected to stamp them?

Timestamp is automatically set when the source (AMB source plugin)
passes the raw event to AMB's core.  It relies on the system clock.

> What if I want collate 2 raw events into 1 logical property event, but only 
> within a certain time window frame?
>

This is kinda what the DBus sink plugin does.  Latitude and Longitude
are two raw events that are collated into the "Location" dbus object.

>>
>> I wouldn't like to have ESC or ABS system decision making having garbage
>> collection or similar since it could cause non-deterministic delays in time
>> critical operation.
>
> I would go even further, and apply this use case right up to the UI space. 
> What is the garbage collection behaviour of the WRT? I wouldn't think a 
> driver continually distracted touching the screen and getting no response 
> because of garbage collection is in progress.
>
>>
>> Well, I don't know if AMB is deterministic either...
>
> No idea, but I'm less worried today. My approach here is to discard all 
> concepts of AMB and Murphy, look at the problem space, figure out where the 
> logic is most suitably applied and then return to what we have and figure out 
> who owns the solution.
> That could end up in Murphy, AMB or even one may discover you need another 
> entity behind AMB to handle this... I have not figured this out yet.
> My challenge is I only have 1 tool today for groking this space, that being 
> the colourful 'system architecture' diagram - I leave it to you to decide 
> whether that is adequate or not.
>
>>
>> _______________________________________________
>> IVI mailing list
>> IVI@lists.tizen.org
>> https://lists.tizen.org/listinfo/ivi
> --------------------------------------------------------------
> Intel Shannon Limited
> Registered in Ireland
> Registered Office: Collinstown Industrial Park, Leixlip, County Kildare
> Registered Number: 308263
> Business address: Dromore House, East Park, Shannon, Co. Clare
>
> This e-mail and any attachments may contain confidential material for the 
> sole use of the intended recipient(s). Any review or distribution by others 
> is strictly prohibited. If you are not the intended recipient, please contact 
> the sender and delete all copies.
>
>
> _______________________________________________
> IVI mailing list
> IVI@lists.tizen.org
> https://lists.tizen.org/listinfo/ivi
_______________________________________________
IVI mailing list
IVI@lists.tizen.org
https://lists.tizen.org/listinfo/ivi

Reply via email to