Re: [PATCH V3 2/4] soc/tegra: pmc: Correct type of variable for tegra_pmc_readl()

From: Jon Hunter
Date: Thu May 05 2016 - 08:48:49 EST



On 05/05/16 13:35, Laxman Dewangan wrote:
>
> On Thursday 05 May 2016 06:13 PM, Jon Hunter wrote:
>> On 05/05/16 10:52, Laxman Dewangan wrote:
>>> On Thursday 05 May 2016 03:19 PM, Jon Hunter wrote:
>>>> On 04/05/16 12:39, Laxman Dewangan wrote:
>>>>> The function tegra_pmc_readl() returns the u32 type data and hence
>>>>> change the data type of variable where this data is stored to u32
>>>>> type.
>>>>>
>>>>> Signed-off-by: Laxman Dewangan <ldewangan@xxxxxxxxxx>
>>>>>
>>>>> ---
>>>>> Changes from V1:
>>>>> -This is new in series as per discussion on V1 series to use u32 for
>>>>> tegra_pmc_readl.
>>>>>
>>>>> Changes from V2:
>>>>> - Make unsigned long to u32 for some missed variable from V1.
>>>>> ---
>>>>> drivers/soc/tegra/pmc.c | 24 ++++++++++++++----------
>>>>> 1 file changed, 14 insertions(+), 10 deletions(-)
>>>>>
>>>>> diff --git a/drivers/soc/tegra/pmc.c b/drivers/soc/tegra/pmc.c
>>>>> index 2c3f1f9..eff9425 100644
>>>>> --- a/drivers/soc/tegra/pmc.c
>>>>> +++ b/drivers/soc/tegra/pmc.c
>>>>> @@ -844,7 +844,8 @@ static void tegra_powergate_init(struct tegra_pmc
>>>>> *pmc)
>>>>> static int tegra_io_rail_prepare(unsigned int id, unsigned long
>>>>> *request,
>>>>> unsigned long *status, unsigned int *bit)
>>>>> {
>>>>> - unsigned long rate, value;
>>>>> + unsigned long rate;
>>>>> + u32 value;
>>>>> *bit = id % 32;
>>>>> @@ -868,17 +869,18 @@ static int tegra_io_rail_prepare(unsigned int
>>>>> id, unsigned long *request,
>>>>> tegra_pmc_writel(DPD_SAMPLE_ENABLE, DPD_SAMPLE);
>>>>> /* must be at least 200 ns, in APB (PCLK) clock cycles */
>>>>> - value = DIV_ROUND_UP(1000000000, rate);
>>>>> - value = DIV_ROUND_UP(200, value);
>>>>> + rate = DIV_ROUND_UP(1000000000, rate);
>>>>> + rate = DIV_ROUND_UP(200, rate);
>>>>> + value = (u32)rate;
>>>> Although it is unlikely, I think that we should check it is less
>>>> than U32_MAX, return an error if it is not.
>>> rate = DIV_ROUNC_UP(200, rate) means
>>>
>>> rate = (200 + rate -1)/rate
>>>
>>> and can not be more than 200 in any case (if rate =1).
>>> So no need of the error check.
>> OK, yes you are right. In that case there is no need to cast and so I
>> would leave this code as-is and not change the type.
>>
>
> You mean keep value as unsigned long for value?

Yes.

> I think we can still say value as u32 and simply write
> value = rate
>
> Just remove the casting.

I would not change this at all. I don't see any benefit.

Jon