Re: [PATCH] tmio_mmc: Optionally support using platform clock

[Magnus Damm <magnus.damm@gmail.com>]

On Thu, Jul 30, 2009 at 12:20 AM, pHilipp Zabel<philipp.zabel@gmail.com> wrote:
> On Tue, Jul 28, 2009 at 9:51 AM, Guennadi
> Liakhovetski<g.liakhovetski@gmx.de> wrote:
>> If the platform device has a clock associated with the tmio-mmc device,
>> use it.
>>
>> Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
>> Cc: Magnus Damm <damm@opensource.se>
>> ---
>>
>> Depends (at least logically) on
>> http://marc.info/?l=linux-kernel&m=124782904228865&w=2
>>
>> Sorry if I forgot anyone who volunteered to help out with MMC patches.
>>
>>  drivers/mmc/host/tmio_mmc.c |   21 +++++++++++++++++++++
>>  drivers/mmc/host/tmio_mmc.h |    3 +++
>>  include/linux/mfd/tmio.h    |    1 +
>>  3 files changed, 25 insertions(+), 0 deletions(-)
>>
>> diff --git a/drivers/mmc/host/tmio_mmc.c b/drivers/mmc/host/tmio_mmc.c
>> index c246191..2a01572 100644
>> --- a/drivers/mmc/host/tmio_mmc.c
>> +++ b/drivers/mmc/host/tmio_mmc.c
>> @@ -567,6 +576,11 @@ static int __devinit tmio_mmc_probe(struct platform_device *dev)
>>                        goto unmap_cnf;
>>        }
>>
>> +       host->clk = clk_get(&dev->dev, pdata->clk_name);
>
> Instead of pdata->clk_name, this should be either NULL or better
> "HCLK" (to differentiate from CLK32, not sure if we want to be able to
> toggle that, too, in the future).
> Passing the clock consumer ID via pdata is just wrong and shouldn't be
> needed with clkdev.

Really? I think that depends on the clock framework implementation.
Remember that this needs to work on other architectures than just ARM.

Cheers,

/ magnus