This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: patch to enable using C alloca for CC=gcc
- To: Fred Fish <fnf at ninemoons dot com>
- Subject: Re: patch to enable using C alloca for CC=gcc
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Mon, 15 Sep 1997 01:34:55 -0600
- cc: egcs-bugs at cygnus dot com
- Reply-To: law at cygnus dot com
In message <199709060620.XAA00880@ninemoons.com>you write:
> I don't believe this change will cause any ill effects for other
> configurations that define USE_C_ALLOCA in the manner suggested in
> gcc.texi:
>
> #ifndef __GNUC__
> #define USE_C_ALLOCA
> #else
> #define alloca __builtin_alloca
> #endif
Well, some ports just define it regardless of the value of __GNUC__;
the idea being that just because you're compiling with GCC doesn't
mean you want the __builtin_alloca. For example you might be compiling
with gcc -fno-builtin -- which some ports have done by default in the
past.
And, IMHO, compiling alloca and including it on the link line isn't
going to hurt anything. In fact, I do this locally because I often
end up with mixed cc and gcc compiled code -- the "cc" code doesn't
expand alloca as a builtin.
Always compiling alloca and including it on the link line would also
simplify the makefiles.
That's the direction I'd like to see things go; I don't know if Jim
has other opinions.
If we get to that stage, then you should be able to get by with just
using -fno-builtin.
jeff