Re: [PATCH #upstream-fixes 2/4] libata: beef up iterators

[Elias Oltmanns <eo@nebensachen.de>]

Tejun Heo <tj@kernel.org> wrote:
> Elias Oltmanns wrote:
>> Tejun Heo <tj@kernel.org> wrote:
>
>>> Hello,
>>>
>>> Elias Oltmanns wrote:
>>>>> -		ata_port_for_each_link(link, ap) {
>>>>> -			ata_link_for_each_dev(dev, link)
>>>>> +		ata_for_each_link(link, ap, EDGE) {
>>>>> +			ata_for_each_dev(dev, link, ALL)
>>>> Where did these short forms (EDGE, ALL) spring from? Does this code even
>>>> compile?
>>> +#define ata_for_each_link(link, ap, mode) \
>>> +	for ((link) = ata_link_next(NULL, (ap), ATA_LITER_##mode); (link); \
>>> +	     (link) = ata_link_next((link), (ap), ATA_LITER_##mode))
>>> +
>>> +#define ata_for_each_dev(dev, link, mode) \
>>> +	for ((dev) = ata_dev_next(NULL, (link), ATA_DITER_##mode); (dev); \
>>> +	     (dev) = ata_dev_next((dev), (link), ATA_DITER_##mode))
>> 
>> Sorry, I should have been more explicit. I was referring to EDGE and ALL
>> as opposed to ATA_LITER_EDGE and ATA_DITER_ALL. Unless I've missed
>> something, the former aren't defined anywhere in your patch.
>
> I seriously can't be more explicit.  Please take a shower and read my
> reply again, especially, the "ATA_[LD]ITER_##mode" part.  :-)

Oh dear, I really need to drop that idea that a macro does what I think
it does. Sorry for the noise.

Regards,

Elias