GNU bug report logs - #27736
OSX 10.6.8: Building from master branch fails.

Previous Next

Full log

Message #132 received at 27736 <at> debbugs.gnu.org (full text, mbox):

From: charles <at> aurox.ch (Charles A. Roelli)
To: Paul Eggert <eggert <at> cs.ucla.edu>
Cc: 27736 <at> debbugs.gnu.org, esq <at> lawlist.com, bug-gnulib <at> gnu.org,
 npostavs <at> users.sourceforge.net
Subject: Re: bug#27736: OSX 10.6.8: Building from master branch fails.
Date: Sat, 31 Mar 2018 17:02:07 +0200

> From: Paul Eggert <eggert <at> cs.ucla.edu>
> Organization: UCLA Computer Science Department
> Date: Wed, 28 Mar 2018 14:29:38 -0700
> 
> On 03/28/2018 11:54 AM, Charles A. Roelli wrote:
> >
> > -            time_t t = -67768038400665600;
> > +            time_t t = -67768038400666600;
> 
> Thanks, I did that in Gnulib and propagated the result into Emacs master 
> (see attached patch).
> 
> > And then I got the following in the output of ./configure:
> >
> >    checking whether localtime loops forever near extrema... ./configure: line 30561: 35760 Alarm clock             TZ=QQQ0 ./conftest$EXEEXT
> 
> Thanks for mentioning that; I attempted to suppress the "Alarm clock" 
> stuff in the attached patch too.

Thank you.  Strangely, it looks like it still prints the invocation,
however:

  checking whether localtime loops forever near extrema... ./configure: line 30562: 55052 Alarm clock             ( TZ=QQQ0 ./conftest$EXEEXT ) > /dev/null 2>&1
  yes

Not sure why that is still so.

> > I will try to find the exact boundary of the hang again.  And the hang
> > is timezone-dependent, so that complicates things too.
> 
> I think we should be OK here, as the test case uses UTC so that it is 
> reproducible. The use of UTC could explain why the abovementioned 
> numeric change was needed.

Thanks, it should be okay to leave it at that.  (I hadn't realized the
test case was using UTC.)

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.