On 08/04/2019 12:22:49+0200, Thierry Reding wrote: > On Sun, Apr 07, 2019 at 11:16:44PM +0200, Alexandre Belloni wrote: > > The Tegra 20 RTC is a 32bit seconds counter (with an unused millisecond > > counter). > > > > Signed-off-by: Alexandre Belloni <alexandre.belloni@xxxxxxxxxxx> > > --- > > drivers/rtc/rtc-tegra.c | 24 ++++++++++++++---------- > > 1 file changed, 14 insertions(+), 10 deletions(-) > > > > diff --git a/drivers/rtc/rtc-tegra.c b/drivers/rtc/rtc-tegra.c > > index c6b0a99aa3a9..3f93a1f7abb5 100644 > > --- a/drivers/rtc/rtc-tegra.c > > +++ b/drivers/rtc/rtc-tegra.c > > @@ -306,6 +306,13 @@ static int __init tegra_rtc_probe(struct platform_device *pdev) > > > > info->tegra_rtc_irq = ret; > > > > + info->rtc_dev = devm_rtc_allocate_device(&pdev->dev); > > + if (IS_ERR(info->rtc_dev)) > > + return PTR_ERR(info->rtc_dev); > > + > > + info->rtc_dev->ops = &tegra_rtc_ops; > > + info->rtc_dev->range_max = U32_MAX; > > The placement of this new block seems somewhat arbitrary, but either > way: > Well, I needed it to be before the clk_prepare_enable call so it can just return and I thought it would be better to leave devm_clk_get close to that clk_prepare_enable call. I don't mind, I can respin with it placed somewhere else. > Acked-by: Thierry Reding <treding@xxxxxxxxxx> -- Alexandre Belloni, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
