2014-07-31 13:34 GMT-07:00 Rafał Miłecki <zajec5@xxxxxxxxx>:
> On 31 July 2014 22:25, Florian Fainelli <f.fainelli@xxxxxxxxx> wrote:
>>> +static inline void b43_write16f(struct b43_wldev *dev, u16 offset, u16 value)
>>> +{
>>> + b43_write16(dev, offset, value);
>>> +#if defined(CONFIG_BCM47XX) || defined(CONFIG_BCM63XX)
>>> + b43_read16(dev, offset);
>>> +#endif
>>> +}
>>
>> How about you leave the calls to b43_write16() to avoid missing call
>> sites, but you re-define it for BCM47XX and BCM63XX? Or, you could
>> introduce a __b43_write16() that defaults to dev->dev->write16 in the
>> normal case, and gets called by b43_write16() such that we you'd get:
>
> I didn't mean (and I didn't) enable flushing for all writes. Only
> these touching specific registers. I don't think we need to flush
> every single write.
Ok, then the current approach looks good to me, thanks!
--
Florian
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
