Paul Mackerras wrote:
>
> On Sun, 27 Feb 2000, Jeff Garzik wrote:
>
> > PPC arch compiles with -fno-builtin. Why is this? It seems sub-optimal
> > and non-uniform, leading to little platform discrepancies like this...
>
> Have you tried it without -fno-builtin? If you do, gcc grumbles about
> various procedures not having the prototypes it expects, in particular
> memset, memcpy, and memcmp. It's not obvious from the warning messages
> you get exactly what prototype gcc is expecting.
>
> In general there's no particular reason why a function named `memcpy' in
> the kernel should have the same prototype (or even the same behaviour) as
> a function named `memcpy' in a userland program, which is why we use
> -fno-builtin.
So... you are choosing to lose extremely valuable optimizations because
of a few prototype warnings? This logic seems bogus to me... it also
seems like a slap fix instead of fixing the real problem...
Jakub commented to me not a couple weeks ago about how -fno-builtin was
a bad idea. Maybe you could dig this message up in the linux-kernel
archives?
-- Jeff Garzik | "Are you the police?" Building 1024 | MandrakeSoft, Inc. | "No ma'am. We're musicians."- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Feb 29 2000 - 21:00:17 EST