| Message ID | 20160920205833.19638-4-alexandre.belloni@free-electrons.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 09/20, Alexandre Belloni wrote: > Assume that if the oscillator is enabled (OSC32EN bit is present), the > delay has already elapsed as the bootloader probably waited for the > oscillator to settle. This could waste up to 1.2s. > > Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com> > --- Applied to clk-next
diff --git a/drivers/clk/at91/sckc.c b/drivers/clk/at91/sckc.c index 60d8225715a2..b1265f2cdae7 100644 --- a/drivers/clk/at91/sckc.c +++ b/drivers/clk/at91/sckc.c @@ -69,7 +69,7 @@ static int clk_slow_osc_prepare(struct clk_hw *hw) void __iomem *sckcr = osc->sckcr; u32 tmp = readl(sckcr); - if (tmp & AT91_SCKC_OSC32BYP) + if (tmp & (AT91_SCKC_OSC32BYP | AT91_SCKC_OSC32EN)) return 0; writel(tmp | AT91_SCKC_OSC32EN, sckcr);
Assume that if the oscillator is enabled (OSC32EN bit is present), the delay has already elapsed as the bootloader probably waited for the oscillator to settle. This could waste up to 1.2s. Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com> --- drivers/clk/at91/sckc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)