Re: [PATCH v3] w1: Add Maxim/Dallas DS2780 Stand-Alone Fuel Gauge IC support.

[Simon INIZAN <simon.inizan@goobie.fr>]

On 31/05/2011 14:42, Barnes, Clifton A. wrote:
> On Monday, May 30, 2011, Simon INIZAN wrote:
>
>> However from time to time I a getting stuck in the driver.
> Getting stuck meaning a deadlock?  Is it during a read on a
> sysfs file?  Are you using the driver compiled in or as a
> module?
I have tested the driver with the fix you provided me, I am not getting 
a deadlock anymore.

>> FYI, I am using the GPIO 1-wire driver, and from time to time the gauge is not responding to the w1 queries.
> Is this the same condition you're talking about above?  Does this mean that sometimes
> the sysfs file returns and sometimes it hangs?
>
It seems I am getting some flipped bits on the 1-wire bus (using the 
w1-gpio master),
this is not directly related to your driver but it makes the results not 
very reliable
on my platform (sometimes the rsgain is wrong, so are the currents).
I am getting a 1-wire bus analyzer to investigate the issue.



In the get_current function, you are returning:

sense_res = 1000 / sense_res_raw;
*current_uA = current_raw * (DS2780_CURRENT_UNITS / sense_res);


Which means that we are getting current (in µA) = current_raw (in 
µV/Ohms) * sense_res_raw (in Ohms).
I have replaced it with

sense_res = sense_res_raw;
*current_uA = current_raw * (DS2780_CURRENT_UNITS / sense_res);


Which gives me more consistent results with my measures.

Simon