On Mon, May 20, 2013 at 6:28 AM, Guenter Roeck <linux@xxxxxxxxxxxx> wrote:
Agreed.On Mon, May 20, 2013 at 02:10:15PM +0200, Jean Delvare wrote:
> Hi Robert,
>
> Adding the list back, there's really no reason to keep this discussion
> private.
>
> On Fri, 17 May 2013 07:51:13 -0700, Robert Coulson wrote:
> > As you know, the ds1631 device cannot be uniquely identified,
> > but it is displayed if a ds1621 device is detected; my thought is to
> > remove the DS1631 and replace it DS1625. This change would
> > provide more accurate device information for the user.
> >
> > If you agree, then I can submit a small patch for it, so please let me know.
>
> I'm afraid I don't quite understand what the problem is nor how you
> intend to solve it. A patch showing your point might be the best
> explanation.
>
> All these DS16xx devices are hard if not impossible to detect reliably,
> because they lack device and vendor identification registers. If
> anything I'd rather drop detection of these from sensors-detect
> altogether. This script is really only meant to detect sensor chips on
> PC systems, and I don't think any DS16xx was ever used on a consumer PC
> system.
>
> Detection should probably be dropped from the ds1621 driver too, the
> proper way to instantiate these devices is explicitly.
>
Robert, can you send a patch ?
Thanks,
Guenter
Yes, I will send one shortly.
thanks again,
Rob.
_______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
