[PATCH v13 0/9] Per-user clock constraints

[Stephen Boyd]

On 09/26/14 16:20, Mike Turquette wrote:
> Quoting Tomeu Vizoso (2014-09-26 01:09:20)
>> On 09/26/2014 03:29 AM, Stephen Boyd wrote:
>>
>>> We already have the consumer/provider split in the struct clk_hw and
>>> struct clk separation. Why don't we just use struct clk_hw throughout
>>> the provider APIs? The only op that isn't doing this is determine_rate()
>>> which might be able to accept a flag day. Otherwise we rename it to
>>> something else and migrate everyone over to a different named function
>>> that doesn't take a struct clk **. Then we introduce new APIs for the
>>> providers to use that are struct clk_hw focused instead of struct clk
>>> focused and migrate them too. The benefit being that we get proper
>>> review of this stuff because the patches are small. We can let
>>> coccinelle do it too.
> I have been opposed to mucking with clk_hw before, but that was because
> the goal was never clear. Now that we know that we're trying to split
> the API then it might be reasonable to use it.
>
> Stephen, does your above proposal still allow for unique struct clk
> cookies for each user of a clock?
>
>

Yes. The clkdev code would know that it's getting clk_hw pointers back
from the provider instead of struct clk pointers because it uses a
different callback:

diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c
index bacc06ff939b..76c356b779d1 100644
--- a/drivers/clk/clk.c
+++ b/drivers/clk/clk.c
@@ -2337,6 +2337,7 @@ struct of_clk_provider {
 
        struct device_node *node;
        struct clk *(*get)(struct of_phandle_args *clkspec, void *data);
+       struct clk_hw *(*get_hw)(struct of_phandle_args *clkspec, void *data);
        void *data;
 };
 

 

It would test for the presence of the get_hw callback and then fall back
to the get callback. Similarly we would have a clk_hw pointer in the
clkdev lookup structure. Once clkdev has the hw pointer it can ask the
ccf to generate a struct clk cookie. This would happen either in
__of_clk_get_from_provider() or clk_get_sys() depending on if we're
using DT or not.

The clk_hw structure will need to be updated to have a pointer to
clk_core. During the transition we would have struct clk and struct
clk_core next to each other in struct clk_hw. Eventually once we convert
all drivers we can remove struct clk from clk_hw. This would require
getting rid of the DEFINE_CLK macro in clk-private.h though.

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
hosted by The Linux Foundation