Re: autoconf AC_FUNC_MKTIME breaks with gcc 4.3
Adrian Bunk <bunk@stusta.de> writes:
> I just tried compiling the emacs22 package from unstable with gcc 4.3,
> and ran into an issue similar to what Martin described back in
> May 2007 for another package in #425544, with the following differences:
> - it's not an infinite loop but a one minute hang
> (see the alarm() in conftest.c)
> - the build therefore proceeds, but emacs wrongly assumes there's no
> working system mktime()
> - Martin said it's "a bug in older versions of autoconf", but as far
> as I can see these older versions include the latest upstream
> version and also the Debian package in unstable
The Debian package for autoconf in unstable is the piece of this
puzzle that is within my power to fix, and so I have uploaded
autoconf 2.61-7 with this patch.
When this kind of thing comes up, please feel free to file a bug
against the autoconf package. I do my best to fix important
problems as soon as I can. It's easy in a case like this where
an upstream fix has already been committed.
--
Ben Pfaff
http://benpfaff.org
Reply to: