[lm-sensors] [PATCH 1/3] hwmon: (it87) Fix VID reading on IT8718F/IT8720F

[Jean Delvare]

On Thu, 22 Oct 2009 15:19:23 +0200, Jean Delvare wrote:
> Comparing apples to bananas doesn't seem right. Consistently use the
> chips enum for chip type comparisons, to avoid such bugs in the
> future.
> 
> The bug has been there since support for the IT8718F was added, so
> VID never worked for this chip nor for the similar IT8720F.
> 
> Signed-off-by: Jean Delvare <khali at linux-fr.org>
> ---
> Stable team, this fix applies to 2.6.27 and 2.6.31, and everything in
> between if they are going to see one more stable release (I guess not.)

My bad, it won't apply to 2.6.27 because IT8720F support was added in
2.6.29. I'll send a separate patch for 2.6.27.

> 
>  drivers/hwmon/it87.c |    5 ++---
>  1 file changed, 2 insertions(+), 3 deletions(-)
> 
> --- linux-2.6.32-rc5.orig/drivers/hwmon/it87.c	2009-10-21 16:06:43.000000000 +0200
> +++ linux-2.6.32-rc5/drivers/hwmon/it87.c	2009-10-21 16:44:16.000000000 +0200
> @@ -1028,12 +1028,11 @@ static int __init it87_find(unsigned sho
>  		chip_type, *address, sio_data->revision);
>  
>  	/* Read GPIO config and VID value from LDN 7 (GPIO) */
> -	if (chip_type != IT8705F_DEVID) {
> +	if (sio_data->type != it87) {
>  		int reg;
>  
>  		superio_select(GPIO);
> -		if ((chip_type == it8718) ||
> -		    (chip_type == it8720))
> +		if (sio_data->type == it8718 || sio_data->type == it8720)
>  			sio_data->vid_value = superio_inb(IT87_SIO_VID_REG);
>  
>  		reg = superio_inb(IT87_SIO_PINX2_REG);
> 
> 


-- 
Jean Delvare