Hi Mauro,
my bad.
On Thu, Jun 18, 2015 at 02:32:39PM -0300, Mauro Carvalho Chehab wrote:
> Signed-off-by: Mauro Carvalho Chehab <mchehab@xxxxxxxxxxxxxxx>
> drivers/media/pci/mantis/mantis_i2c.c: In function 'mantis_i2c_init':
> drivers/media/pci/mantis/mantis_i2c.c:222:15: warning: variable 'intmask' set but not used [-Wunused-but-set-variable]
> u32 intstat, intmask;
>
> Signed-off-by: Mauro Carvalho Chehab <mchehab@xxxxxxxxxxxxxxx>
>
> diff --git a/drivers/media/pci/mantis/mantis_i2c.c b/drivers/media/pci/mantis/mantis_i2c.c
> index a93823490a43..d72ee47dc6e4 100644
> --- a/drivers/media/pci/mantis/mantis_i2c.c
> +++ b/drivers/media/pci/mantis/mantis_i2c.c
> @@ -219,7 +219,7 @@ static struct i2c_algorithm mantis_algo = {
>
> int mantis_i2c_init(struct mantis_pci *mantis)
> {
> - u32 intstat, intmask;
> + u32 intstat;
> struct i2c_adapter *i2c_adapter = &mantis->adapter;
> struct pci_dev *pdev = mantis->pdev;
>
> @@ -242,7 +242,7 @@ int mantis_i2c_init(struct mantis_pci *mantis)
> dprintk(MANTIS_DEBUG, 1, "Initializing I2C ..");
>
> intstat = mmread(MANTIS_INT_STAT);
> - intmask = mmread(MANTIS_INT_MASK);
> + mmread(MANTIS_INT_MASK);
I'm not sure if the mmread() is still needed. But as I don't have any
docs about the chipset I would keep it because that's how it has been
tested. I could re-test without it but I guess nobody cares about the
extra mmread() in the init path anyway.
Acked-by: Jan Klötzke <jan@xxxxxxxxxxxx>
Regards,
Jan
> mmwrite(intstat, MANTIS_INT_STAT);
> dprintk(MANTIS_DEBUG, 1, "Disabling I2C interrupt");
> mantis_mask_ints(mantis, MANTIS_INT_I2CDONE);
> --
> 2.4.3
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
