On Mon, Feb 16, 2009 at 2:20 AM, Johannes Berg <johannes@xxxxxxxxxxxxxxxx> wrote: > On Mon, 2009-02-16 at 02:11 -0800, Luis R. Rodriguez wrote: > >> >> * Driver hints which never have the wiphy NULL >> >> * __regulatory_hint: called from 11d hints, and from the workqueue >> >> that processes the pending requests (which can come from userspace or >> >> from drivers right now). >> > >> > Ok, but in the first case you can't have NULL, and in the second you >> > could just inline it. It's not an operation that requires an inline here >> > for it. >> >> Sure, its just called twice > > Yeah but the first call is entirely pointless since wiphy cannot be NULL > there. I don't want to encourage random error-checking accessors where > writing wiphy->wiphy_idx is perfectly fine. That's the thing wiphy does not have a wiphy_idx, the cfg80211_registered_device does. >> and I think it makes the code more >> readable to use a routine, what if I just move it to core.c or reg.c. > > If you really think it makes the code that much more readable then > please just use it once in the case where a NULL argument can happen > (the workqueue), and move it to that code. OK Luis -- 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
