bug-autoconf
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: autoconf-2.69 pkgsrc patch


From: Eric Blake
Subject: Re: autoconf-2.69 pkgsrc patch
Date: Fri, 14 Sep 2012 08:13:02 -0600
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:15.0) Gecko/20120828 Thunderbird/15.0

On 09/14/2012 07:24 AM, Thomas Klausner wrote:
> Hi!
> 
> Jörg Sonnenberger added the attached patch to the autoconf package in
> pkgsrc in 2010; it seems the patch still isn't integrated, so I
> thought I'd send it upstream.
> 
> The commit message was: AC_FUNC_ALLOCA should never define a prototype
> on the BSDs. (The patch comment is "Do not fall through to the
> implicit prototype as it conflicts with stdlib.h.")
> 
> Please include it in the next release.

Thanks for reporting this.  However, it needs tweaking before it can be
applied...

> @@ -369,6 +369,8 @@ AC_CACHE_CHECK([for alloca], ac_cv_func_
>  # ifdef _MSC_VER
>  #  include <malloc.h>
>  #  define alloca _alloca
> +# elif defined(__NetBSD__) || defined(__FreeBSD__) || defined(__DragonFly__) 
> || defined(__OpenBSD__)
> +#   include <stdlib.h>

The autoconf philosophy is to avoid system-specific defines, and instead
test features.  I would rather do one of two things - identify the
specific feature (that is, write a test that fails if <stdlib.h> is not
included, and define something like FUNC_ALLOCA_REQUIRES_STDLIB), or
make the use of <stdlib.h> unconditional (does it hurt any other
platform, other than the possibility of namespace pollution?).

Also, I'm worried that there is some matching documentation that will
need touching up.

Can you please investigate these two points, and resubmit the patch
accordingly?

-- 
Eric Blake   address@hidden    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

Attachment: signature.asc
Description: OpenPGP digital signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]