On Wed, Dec 14, 2011 at 04:05:12PM +0100, Lars-Peter Clausen wrote: > On 12/14/2011 03:31 PM, Dan Carpenter wrote: > > On Wed, Dec 14, 2011 at 11:15:49AM +0100, Lars-Peter Clausen wrote: > >> Marking the function as __must_check doesn't make much sense here. Since it > >> will either return an error or the buffer length. So you'll always use the > >> returned result one way or the other. > > > > Isn't that the point of a __must_check? > > My understanding is that you should use __must_check if it is potentially > dangerous to ignore the return value. Which is not the case here, if you > don't look at the return value it's kind of pointless to call the function > in the first, but it is not dangerous. > I only responded to the previous email because you described exactly the situation that __must_check is designed for, as a reason to not use it. It struck me as humourous. ERR_PTR() is likewise not dangerous. It's just a cast, but it doesn't make sense to not check it, so that's why it has a __must_check tag. If a function is part of the infrastructure and gets called a lot then a __must_check is appropriate. regards, dan carpenter
Attachment:
signature.asc
Description: Digital signature
_______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
