Path: utzoo!attcan!uunet!lll-winken!elroy.jpl.nasa.gov!usc!zaphod.mps.ohio-state.edu!swrinde!ucsd!ucbvax!pasteur!canuck.Berkeley.EDU!luciano
From: luciano@canuck.Berkeley.EDU (Luciano Lavagno)
Newsgroups: comp.unix.ultrix
Subject: cc and ar have a funny idea about time...
Message-ID: <28146@pasteur.Berkeley.EDU>
Date: 25 Sep 90 01:50:04 GMT
Sender: news@pasteur.Berkeley.EDU
Reply-To: luciano@canuck.Berkeley.EDU (Luciano Lavagno)
Organization: UC Berkeley IC CAD Group
Lines: 34

I am posting this for a friend without access to the net.
The following thing happens on a Decstation 3100, running Ultrix Worksystem
V2.1 (Rev. 14). The machine is installed in Italy. My DEC3100 here in the USA
has the same version of Ultrix Worksystem and it does not exhibit the strange
behavior.

% date
Mon Sep 17 08:06:48 MET DST 1990
% touch foo.c
% ls -l foo.c
-rw-rw-r--  1 luciano       16050 Sep 17 08:07 foo.c
% make foo_bar.a
cc -c foo.c
ar ruv foo_bar.a foo.o
r - foo.o
% ls -l foo.o foo_bar.a
-rw-r-----  1 luciano       62452 Sep 17 08:39 foo.o
-rw-r-----  1 luciano     1329756 Sep 17 08:40 foo_bar.a
% date
Mon Sep 17 08:09:10 MET DST 1990
% touch foo.c
% ls -l foo.c
-rw-rw-r--  1 luciano       16050 Sep 17 08:09 foo.c
% make foo_bar.a
`foo_bar.a' is up to date.
%

Notice that foo.o and foo_bar.a have a "future" date, so even if I touch
foo.c, it isn't recompiled. Moreover the time is wrong by about 30 minutes, so
it does not seem to be a problem with DST.
Any clue ?

Luciano Lavagno
luciano@ic.berkeley.edu