Messages in this thread | | | Date | Sat, 07 Oct 2006 17:39:18 -0700 | From | Jeremy Fitzhardinge <> | Subject | Re: [PATCH] Fix WARN_ON / WARN_ON_ONCE regression |
| |
Steven Rostedt wrote: > http://marc.theaimsgroup.com/?l=linux-kernel&m=114935833125957&w=2 > > That was different, since we were putting a likely condition in an > unlikely(). But I still don't see why we would ever want to test > __warn_once before the condition, since it doesn't save on anything and > just adds extra work. I don't see the savings. >
Also, in that thread you cite (__warn_once && (condition)) is flat-out wrong, because condition may have a side-effect. There are plenty of places in the code which use BUG_ON or WARN_ON as a general error checking mechanism which expect the condition to be always evaluated once; WARN_ON_ONCE should be the same.
Personally I think it is poor style, but there you are.
J - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |