Re: [PATCH 2/3] Driver for user access to internal clocks

[Davide Rizzo]

> you've borken the const of the clk_get() call. I think that
> this is either something that the caller needs to deal with
> as all the drivers will be passing a 'dev' structure in to
> help identify common clocks.
>
As I modified it the clk_get() function it should be compatible with
the past, simply now it recognizes ALSO the format clkname.X to allow
a generic driver to have access to all clocks.
The other way to achieve this is to simulate a specific id in a fake
dev parameter, but it's awful.
For the driver to be generic, all low-level drivers should recognize
the clkname.X format. But this is the standard syntax for devices
name, so I think this should be correct.
Also clk_for_each() and clk_name() should be added to clock
infrastructure, so they should be implemented in all drivers.
Maybe this should be proposed and discussed with the clock
infrastructure maintainer, but I didn't find him in the MAINTAINER
list. Do you know who is ?
Where do you think the generic user access driver should be placed ?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/