[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: autoconf bug
From: |
Paul Eggert |
Subject: |
Re: autoconf bug |
Date: |
14 Jul 2003 13:27:23 -0700 |
User-agent: |
Gnus/5.09 (Gnus v5.9.0) Emacs/21.3 |
Joerg Schilling <address@hidden> writes:
> Why would you call this a GCC bug?
>
> GCC just detects that 'dummy' is not (except for an address) used and does
> not reserve stack space for this object.
The C Standard requires that different objects must have different
addresses. It doesn't matter whther an object's contents are never
accessed; if the object exists, it must have a unique address.
The standard does allow a few objects to be coelesced and to share
addresses. These include string literals and (in C99) compound
literals with const-qualified types. But objects must be distinct
unless otherwise specified.
> >But I'm puzzled as to why this matters, if you're using GCC.
> >GCC has builtin alloca, so shouldn't this issue be moot for GCC?
>
> ????
With AC_FUNC_ALLOCA, if 'configure' detects that the compiler-supplied
alloca works, then it shouldn't even test for any of this stuff. It
should say something like this:
checking for working alloca.h... yes
checking for alloca... yes
and then go on to the next subject.
> I was just working on a portable software signal system that needs to
> know the direction in which the stack grows... so I was looking for a
> usable test mechanism.
OK, but that part of the Autoconf code isn't general; it is intended
to be used only for configuring the LIBOBJ replacement for alloca.