On 02/03/2015 11:53 AM, Zubair Lutfullah Kakakhel wrote:
On 03/02/15 10:32, Lars-Peter Clausen wrote:
On 02/03/2015 11:17 AM, Zubair Lutfullah Kakakhel wrote:
[...]
V4 Changes
Removed clock binding because of pending work in clock tree. Will add
binding later. Rather than introduce a bad binding now and change later.
But this patch is introducing a bad binding. The part needs the clock to work. The binding you are specifying right now says that it works fine without any clocks.
Facing a chicken and egg problem with these patches here..
When the new clock driver gets in we can add the correct clock binding and replace
devm_clk_get(&pdev->dev, "uhc"); with devm_clk_get(&pdev->dev, NULL);
Specifying the current binding got NAKed by Arnd who mentioned that clock names shouldn't be needed as required properties.
And even if needed, then it shouldn't be "uhc" and more close to what other ohci drivers have.
It's not a chicken and egg problem. The order is very clear, first convert
the clock driver to the CCF, then update the driver to pass NULL instead of
"uhc", then add the devicetree bindings.
- Lars
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
