On Mon, 2005-11-14 at 23:28 -0700, Joel Aelwyn wrote: > tags 323425 -patch > done > > dann frazier wrote: > > Package: cronolog > > Version: 1.6.2-5 > > Severity: important > > Tags: patch > > > > Our automated buildd log filter[1] detected a problem[2] that will cause > > your package to segfault on architectures where the size of a pointer is > > greater than the size of an integer, such as ia64. > > > > _XOPEN_SOURCE must be defined when using the strptime() function. See > > strptime(3) for details. There might be a more autoconfy patch, but > > this one works. > > > > [1]http://people.debian.org/~dannf/check-implicit-pointer-functions > > [2]Function `strptime' implicitly converted to pointer at cronoutils.c:673 > > I have applied a fix to the package, and stand duly chastized for having > failed to follow up on it in a more useful fashion. Mea culpa. However, the > patch provided is not the most appropriate way to fix the problem; since it > involves adding a -D flag to CFLAGS, this is perfectly capable of happening > in > the debian/rules file, rather than requiring a source patch (I guess you > could > call that 'more autoconfy'; certainly, it seems cleaner). > > However, since I have changed the behavior, it would be helpful if I could > arrange to validate that the -6 package will, in fact, correctly solve this > problem on a 64-bit architecture. Since I don't particularly trust that I can > verify this sufficiently, myself, would it be possible to get a test build > verified in some fashion?
Of course - you could either post a source package for me, or log into merulo and perform a build and run the log through the above script. -- dann frazier <[EMAIL PROTECTED]> -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
