autoconf-patches
[Top][All Lists]
Advanced

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

Re: AC_CHECK_SIZEOF([int *]) is error in autoconf-2.66


From: Eric Blake
Subject: Re: AC_CHECK_SIZEOF([int *]) is error in autoconf-2.66
Date: Sat, 10 Jul 2010 08:49:17 -0600
User-agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.10) Gecko/20100621 Fedora/3.0.5-1.fc13 Lightning/1.0b2pre Mnenhy/0.8.3 Thunderbird/3.0.5

On 07/10/2010 06:47 AM, Ralf Wildenhues wrote:
> * Paolo Bonzini wrote on Sat, Jul 10, 2010 at 02:39:16PM CEST:
>> On 07/10/2010 02:20 PM, Ralf Wildenhues wrote:
>>> * Paolo Bonzini wrote on Sat, Jul 10, 2010 at 02:15:41PM CEST:
> 
>>>> This breaks GCC.  Is this worth a blitz 2.67 release?

That's been our plan, getting 2.67 out the door as soon as possible.

>>>
>>> What breaks GCC?  The commit I just pushed this morning or the `+'
>>> thingy that I excluded from the commit?  Please do not answer yes or no.
>>
>> Pointer type sizeof.
>>
>> There is "AC_CHECK_SIZEOF(void *)" in configure.ac.  Of course, GCC
>> forces you to use one particular Autoconf version, so strictly
>> speaking it is not "broken", but many other packages do not and they
>> will fail.
> 
> Agreed.  Eric, if you agree I can roll a 2.67 tomorrow (I used to have
> upload privs at one time, so I guess I still do).

It would still be nice to address Bruno's report about AC_INIT and valid
version strings first.

In thinking about that, would it be worth adding a new macro
AS_LITERAL_HEREDOC_IF, which is comparable to AS_LITERAL_IF and
AS_LITERAL_WORD_IF, but the difference is that it permits any strings
that pass through unchanged in both quoted and unquoted heredocs?  Then,
instead of AS_LITERAL_IF, we make the version string validation of
AC_INIT use AS_LITERAL_HEREDOC_IF use the new AS_LITERAL_HEREDOC_IF and
that should resolve Bruno's report.

A rough implementation is that it only needs to filter out @S|@ (the
quadrigraph for $) and $,`,\.  I've got a few minutes, let me see if I
can propose a more formal patch for review...

> 
> I'm not sure how available Eric is by email, though, so Cc:ing Paul
> just in case.

I'm likewise unsure of how much email time I'll have today.  The idea of
getting 2.67 out the door now, instead of waiting another week for me to
get home to my normal development environment, does have merit though,
so I am not opposed to pursuing the idea if we can come to consensus on
whether the AC_INIT regression should be fixed now or postponed till
post-release.

-- 
Eric Blake   address@hidden    +1-801-349-2682
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]