> Date: Mon, 17 Jul 2006 21:45:05 +0200 > From: Lennart Borgman <[EMAIL PROTECTED]> > CC: [EMAIL PROTECTED], [EMAIL PROTECTED], [email protected] > > Eli Zaretskii wrote: > > > >> Could something has changed in the arithmetics on w32? > >> > > > > Only if the two versions were compiled by different versions of the C > > compiler, and even then it's extremely unlikely. > > > The simplest way to check for this is perhaps to recompile 21.3 with the > compilers we now use for 22 on w32.
FWIW, I just built Emacs 21.4a with the same development environment I use to build Emacs 22 (MinGW port of GCC 3.4.2, MinGW port of Binutils version 2.15.91 20040904, and MinGW runtime version 3.7), and tested file-attributes on several files. The results are identical (as expected, since the code didn't change at all). So the question is not why 21.3 succeeded whereas 22.0 fails, but rather how come the results returned by Emacs are one second off. I will try to look into this, but anyone is invited to beat me to it. _______________________________________________ emacs-pretest-bug mailing list [email protected] http://lists.gnu.org/mailman/listinfo/emacs-pretest-bug
