On Thu, Sep 21, 2006 at 07:50:36AM -0700, Debian Bug Tracking System wrote: <snip> > Date: Thu, 21 Sep 2006 16:09:34 +0200 > From: "Steinar H. Gunderson" <[EMAIL PROTECTED]> > To: [EMAIL PROTECTED] > Subject: Re: Fixed in NMU of ctn 3.0.6-8.1 > X-Spam-Status: No, hits=-3.0 required=4.0 tests=BAYES_00 autolearn=no > version=2.60-bugs.debian.org_2005_01_02 > > Version: 3.0.6-8.1 > > I've NMUed for this bug (fixing the bug to use versioning instead of the > "fixed" tag, to ease tracking through testing); here's the changelog: > > > ctn (3.0.6-8.1) unstable; urgency=medium > > . > > * Non-maintainer upload. > > * In the LONG_WORD structure, always use unsigned int since we want a > > 32-bit > > variable, and int is 32 bits on all platforms supported by Debian (it > > used > > to be unsigned long for all platforms except alpha, which broke on > > platforms such as ppc64 and amd64). (Closes: #387183) > > * Build-depend on lesstif2-dev instead of lesstif-dev, since the latter > > is > > obsolete. Thanks again for the fix. I tested it and it works.
If I got it right LONGSIZE has to be defined 32 in any code that uses CTN no matter what the actual size of long is. I have some code that normally determines the size of long in a platform check (obviously 64 on amd64). If I define LONGSIZE to match the size of long on amd64 I get a segfault when calling CTN stuff. If I'm right, this should be noted somewhere as it is not very intuitive. Cheers, Michael -- GPG key: 1024D/3144BE0F Michael Hanke http://apsy.gse.uni-magdeburg.de/hanke ICQ: 48230050 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
