On Thu, Aug 09, 2012 at 04:54:27PM +0530, Tushar Behera wrote:
> + linux-kernel@xxxxxxxxxxxxxxx
> - linux-kernel@xxxxxxxxxxxxxxxxxxx
>
> On 08/09/2012 02:45 PM, Thierry Reding wrote:
> > On Thu, Aug 09, 2012 at 02:33:55PM +0530, Tushar Behera wrote:
> >> Signed-off-by: Tushar Behera <tushar.behera@xxxxxxxxxx>
> >> ---
> >> arch/arm/mach-exynos/mach-origen.c | 11 +++++++++++
> >> 1 files changed, 11 insertions(+), 0 deletions(-)
> >>
> >> diff --git a/arch/arm/mach-exynos/mach-origen.c b/arch/arm/mach-exynos/mach-origen.c
> >> index 5ca8030..4c4048d 100644
> >> --- a/arch/arm/mach-exynos/mach-origen.c
> >> +++ b/arch/arm/mach-exynos/mach-origen.c
> > [...]
> >> @@ -613,6 +614,10 @@ static struct platform_device origen_lcd_hv070wsa = {
> >> .dev.platform_data = &origen_lcd_hv070wsa_data,
> >> };
> >>
> >> +static struct pwm_lookup origen_pwm_lookup[] = {
> >> + PWM_LOOKUP("s3c24xx-pwm.0", 0, "pwm-backlight.0", NULL),
> >> +};
> >> +
> >
> > This might conflict with some other patches that Jingoo (Cc'ed) is
> > working on. His patches were going to rework the Samsung PWM driver to
> > register multiple PWM devices per chip. In that case the s3c24xx-pwm
> > device should probably be modified to use .id = -1 and then the .0 can
> > be dropped from the provider name above.
> >
> > Otherwise this patch looks good to me.
> >
> > Thierry
> >
> Thanks for your review.
>
> It was wrong for me to create s3c24xx-pwm platform device as it is
> already created in samsung_bl_set(). Hence I would be dropping that.
>
> However, dropping .0 from provider name results in following. (as the
> device name is s3c24xx-pwm.N).
>
> [ 0.240000] pwm-backlight pwm-backlight.0: unable to request PWM,
> trying legacy API
That's to be expected. Dropping the .0 requires further changes to the
driver and in arch/arm/plat-samsung/devs.c to register only a single
instance of the s3c24xx-pwm device which can handle all 5 PWM devices.
This should be done separately, though.
Thierry
Attachment:
pgpoED_YJ8kkd.pgp
Description: PGP signature
