On Mon, Oct 21, 2013 at 10:28:34AM +0200, Jean-Francois Moine wrote:
On Mon, 21 Oct 2013 09:06:57 +0100 Russell King - ARM Linux linux@arm.linux.org.uk wrote:
On Mon, Oct 21, 2013 at 09:46:13AM +0200, Jean-Francois Moine wrote:
This patch checks the return value of the clk_prepare_enable of the external clock and removes the test about a same internal and external clock which would quite never occur and won't work in most cases it would occur.
NAK. It can occur.
In which case? And, what would be the right treatment?
priv->clk = devm_clk_get(&pdev->dev, NULL); priv->extclk = devm_clk_get(&pdev->dev, "extclk");
Supplying the first clock to this driver without a separate "extclk" via clkdev will return it as the second clock.
Again, NAK. Your change is against the principles of the clk API.