[Automated-testing] power-control standard (was RE: Linaro Connect Report)

[Dave Pigott]

I’ve added an alternative on pages 3 and 4 <https://docs.google.com/document/d/1-f2VNVlOnaJUSKUUWeYko3wXh7_ertbFD55y_0dTZvI/edit>. This simplifies, and extends to allow control of any switchable device.

Dave

> On 11 Apr 2019, at 19:50, <Tim.Bird at sony.com> <Tim.Bird at sony.com> wrote:
> 
>> -----Original Message-----
>> From: Tim.Bird at sony.com
>> 
>> So: "power-control minnowboard1 cycle" would be the command for turning
>> the power off and on again, to the board named minnowboard1.
> 
> OK - that's a terrible example, and is the type of layer-conflating that I think
> many of us have been doing.  The power control layer of the board management
> stack should have pdus and ports as its objects, so a better example would be:
> "power-control pdu=sony-debug-board port=1 command=cycle"
> for the minnowboard in my lab.
> 
> The board management layer (LAVA's device layer or ttc's 'target' configuration layer)
> should be responsible for maintaining the relationship between the board
> and the pdu and port where it is connected.
> 
> Also, is it worthwhile to create a pdu port 'resource locator' syntax, like this:
> <pdu>:<port>.  So the above becomes "power-control sony-debug-board:1 cycle"?
> 
> That's just syntactic sugar for specifying the power connection to the board,
> but it's handy to eliminate multiple fields in the device, board, or power layer.
> 
> -- Tim
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.yoctoproject.org/pipermail/automated-testing/attachments/20190412/f335ce7f/attachment.html>