Hi Juerg,
On Tue, 13 Apr 2010 10:49:15 +0200, Jean Delvare wrote:
> On Fri, 9 Apr 2010 10:20:40 -0700, Juerg Haefliger wrote:
> > I don't understand your proposal. Could you elaborate?
>
> I propose that you don't change the value of force_id in the driver.
> This value can be set by the user and should be considered read-only by
> the driver itself. The resulting code might be slightly larger, but it
> is also more correct:
>
> /* Skip chip detection if module is loaded with force_id parameter */
> switch (force_id) {
> case SCH5127_ID:
> data->type = sch5127;
> break;
> case SCH3112_ID:
> data->type = sch311x;
> break;
> default:
> company = dme1737_read(data, DME1737_REG_COMPANY);
> device = dme1737_read(data, DME1737_REG_DEVICE);
>
> if ((company == DME1737_COMPANY_SMSC) &&
> (device == SCH311X_DEVICE)) {
> data->type = sch5127;
> } else if ((company == DME1737_COMPANY_SMSC) &&
> (device == SCH5127_DEVICE)) {
> data->type = sch311x;
> } else {
> err = -ENODEV;
> goto exit_kfree;
> }
> }
>
> data->name = data->type == sch5127 ? "sch5127" : "sch311x";
Any progress with this patch? Would be nice to finally have support for
the SCH5127 in kernel 2.6.35.
Thanks,
--
Jean Delvare
_______________________________________________
lm-sensors mailing list
lm-sensors@xxxxxxxxxxxxxx
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
