Le 30/06/2011 15:13, Uwe Kleine-König :
> Hello Nicolas,
>
> On Thu, Jun 30, 2011 at 03:49:41PM +0200, Nicolas Ferre wrote:
>> Take care of slots while going to suspend state.
>>
>> Signed-off-by: Nicolas Ferre <[email protected]>
>> Signed-off-by: Uwe Kleine-König <[email protected]>
>> ---
>> V2: move to pm_ops
>>
>> drivers/mmc/host/atmel-mci.c | 47
>> ++++++++++++++++++++++++++++++++++++++++++
>> 1 files changed, 47 insertions(+), 0 deletions(-)
>>
>> diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
>> index aa8039f..058f1842 100644
>> --- a/drivers/mmc/host/atmel-mci.c
>> +++ b/drivers/mmc/host/atmel-mci.c
>> @@ -1878,10 +1878,57 @@ static int __exit atmci_remove(struct
>> platform_device *pdev)
>> return 0;
>> }
>>
>> +#ifdef CONFIG_PM
>> +static int atmci_suspend(struct device *dev)
>> +{
>> + struct atmel_mci *host = dev_get_drvdata(dev);
>> + struct atmel_mci_slot *slot;
>> + int i, ret;
> slot and ret can have a more local scope.
ok.
>> +
>> + for (i = 0; i < ATMEL_MCI_MAX_NR_SLOTS; i++) {
>> + slot = host->slot[i];
>> + if (!slot)
>> + continue;
>> + ret = mmc_suspend_host(slot->mmc);
>> + if (ret < 0) {
>> + while (--i >= 0) {
>> + slot = host->slot[i];
>> + if (slot)
>> + mmc_resume_host(host->slot[i]->mmc);
> hmm, mmc_resume_host could fail. But probably you cannot handle that in
> a sane way, do you?
Well, actually for the current implementation the only return code is... 0.
And here I try to have a king of best effort approach ;-)
>> + }
>> + return ret;
>> + }
>> + }
>> +
>> + return 0;
>> +}
>> +
>> +static int atmci_resume(struct device *dev)
>> +{
>> + struct atmel_mci *host = dev_get_drvdata(dev);
>> + struct atmel_mci_slot *slot;
>> + int i, ret;
>> +
>> + for (i = 0; i < ATMEL_MCI_MAX_NR_SLOTS; i++) {
>> + slot = host->slot[i];
>> + if (!slot)
>> + continue;
>> + ret = mmc_resume_host(slot->mmc);
>> + if (ret < 0)
>> + return ret;
> Maybe you should try to resume host 5 even if resuming host 4 failed?
In fact all other drivers that are dealing with multiple slots are doing
the same... So I have difficulties to know the truth.
The question is: is it better to return an error to the "resume"
function so that we are called later again or do we have to do our best
to thaw everything out?
>> + }
>> +
>> + return 0;
>> +}
>> +#endif
>> +static SIMPLE_DEV_PM_OPS(atmci_pm, atmci_suspend, atmci_resume);
>> +
>> +
> For my taste a single empty line is enough.
Ok.
>> static struct platform_driver atmci_driver = {
>> .remove = __exit_p(atmci_remove),
>> .driver = {
>> .name = "atmel_mci",
>> + .pm = &atmci_pm,
>> },
>> };
>
Thanks, best regards,
--
Nicolas Ferre
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
