[PATCH v9 4/4] clk: meson: add sub MMC clock controller driver

[Liang Yang]

On 2022/1/15 7:01, Stephen Boyd wrote:
> [ EXTERNAL EMAIL ]
> 
> Quoting Liang Yang (2022-01-13 21:14:46)
>> On 2022/1/14 5:35, Stephen Boyd wrote:
>>> Quoting Liang Yang (2022-01-13 03:57:45)
>>>> diff --git a/drivers/clk/meson/mmc-clkc.c b/drivers/clk/meson/mmc-clkc.c
>>>> new file mode 100644
>>>> index 000000000000..f53977f61390
>>>> --- /dev/null
>>>> +++ b/drivers/clk/meson/mmc-clkc.c
>>>> @@ -0,0 +1,300 @@
> [..]
>>>> +
>>>> +static int mmc_clkc_probe(struct platform_device *pdev)
>>>> +{
>>>> +       struct clk_hw_onecell_data *onecell_data;
>>>> +       struct device *dev = &pdev->dev;
>>>> +       struct mmc_clkc_data *data;
>>>> +       struct regmap *map;
>>>> +       struct clk_regmap *clk, *core;
>>>> +       struct meson_sclk_div_data *div_data;
>>>> +
>>>> +       /*cast to drop the const in match->data*/
>>>
>>> Space after *, also why do we need to cast away const? The user of this
>>> pointer passes it all the way down to mmc_clkc_register_clk() which
>>> could take the data as const void pointer and decide to cast away const
>>> there.
>>
>> if use 'const' here, it  will report a warning:
>> drivers/clk/meson/mmc-clkc.c:224:7: error: assignment discards ‘const’
>> qualifier from pointer targe
>> t type [-Werror=discarded-qualifiers]
>>
>>     data = (const struct mmc_clkc_data *)of_device_get_match_data(dev);
> 
> Of course. The type declaration up above needs const added to it.The parm of mmc_clkc_register_clk_with_parent(...., void *data) does not 
have 'const', so make the type declaration cause a further 'const' cast 
warning. Could i copy these infos just like below:

struct mmc_clkc_data *data_clkc;
......
data = of_device_get_match_data(dev);
data_clkc = devm_kmalloc(dev, sizeof(*data_clkc), GFP_KERNEL);
memcpy(data_clkc, data, sizeof(*data_clkc));
......
clk = mmc_clkc_register_clk_with_parent(...., &data_clkc->rx);

> 
> .