On Dec 11, 2007 8:46 PM, bruno randolf <[EMAIL PROTECTED]> wrote:
> On Tuesday 11 December 2007 21:38:25 Jiri Slaby wrote:
> > > + if (unlikely(tx_tries0 == 0)) {
> > > + ATH5K_ERR(ah->ah_sc, "zero retries\n");
> > > + WARN_ON(1);
> > > return -EINVAL;
> > > + }
> >
> > I would just do
> > if (WARN_ON(tx_tries0 == 0))
> > return -EINVAL;
> >
> > Or maybe plus the ath error. Comments?
>
> i had it like this first. then i thought it would be nice to get a descriptive
> error too, using ATH5K_ERR, but if we put it inside the if it will be printed
> after the warn info, which is completely counter-intuitive.
>
> of course having a error message is not really that important, so i'm ok with
> your more minimalistic version too.
I think your approach is more readable really.
> it would be nice to have a WARN_ON(condition, message) macro.
That would be nice.
BTW -- If rate 0 can be used to make noise, I can find good uses for
it. Perhaps we should just warn on it? The retries==0 should not be
allowed, however, because IIRC the hardware doesn't like it, I
remember playing with it only vaguely.
Luis
_______________________________________________
ath5k-devel mailing list
ath5k-devel@lists.ath5k.org
https://lists.ath5k.org/mailman/listinfo/ath5k-devel
