Re: 答复: low can service

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi, sorry forn the delay,

Could you send me the log messages of the binding please ? You could find the log using the journalctl command:

journalctl -u afm-service-agl-service-can-low-level--6.99--main@0

Also, I saw this line in the Dashboard.qml:

value: engineSpeed / 65535

So, the engineSpeed would be received but divided but 65535, so there isn't so much left after that. And always be near to 0.

BR

On 1/25/19 10:19 AM, Li, Xiaoming wrote:
Hi Romain,

I am testing AGL's DASHBOARD with vcan,  VehicleAvgSpeed work well, but PT_EngineSpeed(RPM) shows no change.

I have already enable vcan,edit the mapping file,restart vcan service.
Signals.json: EngineSpeed id  0x3D9
Could you give some hints? Thanks.

Hardware: m3ulcb&kingfish m06 / RPI 3B
AGL:GG RC4
Cmd:cansend vcan0 3D9#00003A
      

BR
Li

-----邮件原件-----
发件人: automotive-discussions-bounces@xxxxxxxxxxxxxxxxxxxxxxxxx
[mailto:automotive-discussions-bounces@xxxxxxxxxxxxxxxxxxxxxxxxx] 代表
Romain Forlot
发送时间: 2018年10月16日 16:44
收件人: automotive-discussions@xxxxxxxxxxxxxxxxxxxxxxxxx
主题: Re:  low can service

Ho, yes, correct.


I forgot the dashboard app.


The low-can try to open the socket if the /etc/dev-mapping.conf exists and
correctly set but if your CAN interface doesn't exists then you'll see error
messages:


DEBUG: [API low-can] BCM socket ifr_name is : can0
ERROR: [API low-can] ioctl failed. Error was : No such device


This will not fail the binding which continues to operate. As we might have
several CAN buses, like in a real car, the binding should be able to process the
operational buses even if one is failing. Here, it doesn't exists at all and the
binding lacks some recovery capabilities which, for example, would retry to
connect to a socket if this one failed.


I think that the binding should not fails completely if a bus comes down. I'm
open to discussion about that, binder have also some recovery capabilities
which retries a connection to a failed/disconnected binding, we could imagine
to relaunch the binding until is fine but then the CAN service would be more
degraded than running with errors on one of its bus. Also, we maybe needs to
define priorities or level of importance on certain buses and not the other which
would define which ones are considerate as mandatory to be up and running,
and others with lower importance would be ignored even if failing.


The binding have some lacks here but will be improved, thanks.


Best regards.


On 16/10/2018 04:55, Jason Reich wrote:
Hi Romain and Giuseppe,

The sample app "dashboard" app also uses the low can service, and will
show the vehicle speed on the UI if the low-can service detects the
correct message.  If you setup a vcan interface with low-can, and then
run "cansend vcan0 3e9#22" the speed should change on the UI.  Make
sure the vcan interface is up before starting the low-can service and
dashboard app.


Hope that helps,

Jason.



----------------------------------------------------------------------
--

On 10/10/18 12:29 PM, "Romain Forlot [IoT.bzh]"
<romain.forlot@xxxxxxx> wrote:

 Humm, I don't think that there is an existing apps that use the
low-can  API except the signal-composer for now which is also
installed by  default but the low-can configuration isn't activated by default.
Anyway, signal-composer is also a service without UI. I suggest you
to  read the usage guide of the low-can service here :

http://docs.automotivelinux.org/docs/apis_services/en/dev/reference/s
ignaling/low-can-usage-guide.html

          
If you haven't already read it. :)


_______________________________________________
automotive-discussions mailing list
automotive-discussions@xxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linuxfoundation.org/mailman/listinfo/automotive-discussi
ons
--
Romain Forlot - Embedded Engineer - IoT.bzh romain.forlot@xxxxxxx -
www.iot.bzh - +33675142438

_______________________________________________
automotive-discussions mailing list
automotive-discussions@xxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linuxfoundation.org/mailman/listinfo/automotive-discussions



-- 
Romain Forlot - Embedded Engineer - IoT.bzh
romain.forlot@xxxxxxx - www.iot.bzh - +33675142438 
_______________________________________________
automotive-discussions mailing list
automotive-discussions@xxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linuxfoundation.org/mailman/listinfo/automotive-discussions

[Index of Archives]     [LARTC]     [Bugtraq]     [Yosemite Forum]     [Photo]

  Powered by Linux