On Mon, Jun 10, 2002 at 12:51:44PM -0700, Andrew Morton wrote:
> Tom Rini wrote:
> >
> > ...
> > This reminds me of another slightly annoying issue. At least for
> > toolchains, Documentation/Changes works poorly for !i386. How about we
> > try and take care of things like this in <linux/compiler.h> ?
> > Eg:
> >
> > #if defined(CONFIG_SPARC) || defined(CONFIG_SPARC64)
> > ... egcs 1.1.2 check ...
> >
> > #define __func__ __FUNCTION__
> > #endif
> >
> > #if defined(CONFIG_X86) || ...
> > ... gcc-2.95.3 check ...
> > #endif
> >
>
> That won't work very well - if SPARC wants 2.91.66 then
> we need to support that compiler on x86 as well. So that
> people won't use later-supported compiler features. And
> because many compiler bugs are platform-independent, so
> they will be detected (and worked around) on x86.
Well, didn't someone just find a bug where egcs-1.1.2 falls down on x86
in an important area?
> wrt the __func__ thing: is it possible to do:
>
> #if (compiler version test)
> #define __FUNCTION__ __func__
> #endif
>
> to kill the 3.x warning?
Well, the warning (at least from what I've seen) is when you do:
"In " __FUNCTION__ " something bad happened\n", which __func__ just
won't do. Doing:
"In %s something bad happened\n", __FUNCTION__
Is OK[1].
-- Tom Rini (TR1265) http://gate.crashing.org/~trini/[1]: gcc version 3.1.1 20020606 (Debian prerelease), just a simple program with -Wall. - 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 : Sat Jun 15 2002 - 22:00:19 EST