Re: [PATCH v2 1/6] mmc: omap_hsmmc: use devm_clk_get

[Balaji T K <balajitk@ti.com>]

On Monday 12 May 2014 07:20 PM, Ulf Hansson wrote:
> On 12 May 2014 15:33, Balaji T K <balajitk@ti.com> wrote:
>> On Monday 12 May 2014 02:03 PM, Ulf Hansson wrote:
>>>
>>> On 9 May 2014 18:46, Balaji T K <balajitk@ti.com> wrote:
>>>>
>>>> With devm_clk_get conversion clk_put can be removed in clean up path
>>>>
>>>> Signed-off-by: Balaji T K <balajitk@ti.com>
>>>> ---
>>>>    drivers/mmc/host/omap_hsmmc.c |   15 ++++-----------
>>>>    1 files changed, 4 insertions(+), 11 deletions(-)
>>>>
>>>> diff --git a/drivers/mmc/host/omap_hsmmc.c
>>>> b/drivers/mmc/host/omap_hsmmc.c
>>>> index b4de63b..b8ae7ee 100644
>>>> --- a/drivers/mmc/host/omap_hsmmc.c
>>>> +++ b/drivers/mmc/host/omap_hsmmc.c
>>>> @@ -1922,7 +1922,7 @@ static int omap_hsmmc_probe(struct platform_device
>>>> *pdev)
>>>>
>>>>           spin_lock_init(&host->irq_lock);
>>>>
>>>> -       host->fclk = clk_get(&pdev->dev, "fck");
>>>> +       host->fclk = devm_clk_get(&pdev->dev, "fck");
>>>>           if (IS_ERR(host->fclk)) {
>>>>                   ret = PTR_ERR(host->fclk);
>>>>                   host->fclk = NULL;
>>>> @@ -1941,7 +1941,7 @@ static int omap_hsmmc_probe(struct platform_device
>>>> *pdev)
>>>>
>>>>           omap_hsmmc_context_save(host);
>>>>
>>>> -       host->dbclk = clk_get(&pdev->dev, "mmchsdb_fck");
>>>> +       host->dbclk = devm_clk_get(&pdev->dev, "mmchsdb_fck");
>>>>           /*
>>>>            * MMC can still work without debounce clock.
>>>>            */
>>>> @@ -1949,7 +1949,6 @@ static int omap_hsmmc_probe(struct platform_device
>>>> *pdev)
>>>>                   host->dbclk = NULL;
>>>>           } else if (clk_prepare_enable(host->dbclk) != 0) {
>>>>                   dev_warn(mmc_dev(host->mmc), "Failed to enable debounce
>>>> clk\n");
>>>> -               clk_put(host->dbclk);
>>>>                   host->dbclk = NULL;
>>>
>>>
>>> You should use the IS_ERR macro, no need to reset dbclk to NULL.
>>>
>>
>> Agreed, IS_ERR macro usage deserves patch on its own.
>> will create separate patch on top of this series.
>
> Or you just update this patch, since it would touch there very same
> piece of code. :-)

plus few more code additionally, since these are two different logical changes,
have sent a separate patch. But feel free to squash them if needed.

Thanks and Regards,
Balaji T K