On Tue, Jun 04 2002, Martin Dalecki wrote:
> >WRT the unlikely(), if you have the hints available, why not pass them
> >on?
>
> Well it's kind like the answer to the question: why don't do it all in hand
> optimized assembler? Or in other words - let's give the GCC guys good
[snip]
If I didn't know better Martin, I would say that you are trolling.
Surely you have better ways to spend your time than to fly fuck [1]
every single patch posted on lkml? :-)
I would agree with your entire email if you were talking about inlining.
Yes we should not inline excessively without a good reason. However, the
likely/unlikely hints provide both clues to the code reader (as someone
else already established) about which is the likely code path or not,
and of course to the compiler. I would agree that profiling would in
some cases be needed before slapping on an unlikely() sticker. These are
the cases where you do not know what the call path is typically like.
When I put unlikely() in there, it's because I know there's a 1:50 ratio
or more. Or maybe a 1:inf ratio, meaning it should only trigger because
of a bug.
[1] To "fuck a fly" is a danish expression, meaning excessive attention
to detail.
-- Jens Axboe- 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/
This archive was generated by hypermail 2b29 : Fri Jun 07 2002 - 22:00:23 EST