On Mon, 2014-02-03 at 19:32 -0500, Peter Gavin wrote: > On Mon, Feb 3, 2014 at 5:54 PM, Brian Drummond > <[email protected]> wrote: > so here, the library was compiled after its source timestamp, > and before > the testcase. If the package build machine was in a different > timezone... > > > You're correct, the timezone was actually the problem. I found a log > of the build on launchpad, which lists the timestamps of the packaged > files. I'm in EST, and the timestamps are consistent with the server > using GMT (5 hours later). And in any case, the problem appears to > have gone away (it's 5 hours after I ran the build), and the package > is now working. :)
Joris - are you watching this thread? If the OS package is built in one timezone and installed in another (further West) the same day, bad things can (temporarily) happen thanks to timestamps embedded in obj.cf files. Possible answers : force the std libraries to be built during package installation (if that's an option) build, and wait 12 hours before pushing! any others? - Brian _______________________________________________ Ghdl-discuss mailing list [email protected] https://mail.gna.org/listinfo/ghdl-discuss
