Hi Guenter and Jean, Sorry for the delay. Much stuff at work... Here's (Jean's ;) patch. BR Per On 05/28/2011 04:04 PM, Guenter Roeck wrote: > Hi Jean, > > On Sat, May 28, 2011 at 04:36:31AM -0400, Jean Delvare wrote: >> Hi Guenter, >> >> On Fri, 27 May 2011 21:51:13 -0700, Guenter Roeck wrote: >>> The attached patch (on top of Per's most recent patch) works quite nicely. >>> Per, maybe you can just merge it with your patch and resubmit it. >> >> Looks good, pretty much what I had in mind. But I think you could make >> the code even more compact: >> >> /* sanity check */ >> if (i2c_smbus_read_byte_data(client, 0x04) != 0x4D >> || i2c_smbus_read_byte_data(client, 0x06) != 0x4D >> || i2c_smbus_read_byte_data(client, 0xff) != 0x4D) >> return -ENODEV; >> > Yes, you are right. I didn't do it to avoid a checkpatch warning, but forgot > that I don't use a variable anymore. > > Guenter
Improve the detection of MAX6642 by reading non exciting registers (0x04, 0x06 and 0xff). The value of those registers should be the same as the last valid resister read. Signed-off-by: Per Dalen <per.dalen@xxxxxxxxxxxx> --- diff --git a/drivers/hwmon/max6642.c b/drivers/hwmon/max6642.c index 0f9fc40..4fb0564 100644 --- a/drivers/hwmon/max6642.c +++ b/drivers/hwmon/max6642.c @@ -136,6 +136,12 @@ static int max6642_detect(struct i2c_client *client, if (man_id != 0x4D) return -ENODEV; + /* sanity check */ + if (i2c_smbus_read_byte_data(client, 0x04) != 0x4D + || i2c_smbus_read_byte_data(client, 0x06) != 0x4D + || i2c_smbus_read_byte_data(client, 0xff) != 0x4D) + return -ENODEV; + /* * We read the config and status register, the 4 lower bits in the * config register should be zero and bit 5, 3, 1 and 0 should be
_______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
