From owner-freebsd-current  Fri Dec 29 03:21:12 1995
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id DAA14854
          for current-outgoing; Fri, 29 Dec 1995 03:21:12 -0800 (PST)
Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id DAA14849
          Fri, 29 Dec 1995 03:21:05 -0800 (PST)
Received: from sax.sax.de by irz301.inf.tu-dresden.de (8.6.12/8.6.12-s1) with ESMTP id MAA25797; Fri, 29 Dec 1995 12:21:02 +0100
Received: by sax.sax.de (8.6.11/8.6.12-s1) with UUCP
	id MAA22997; Fri, 29 Dec 1995 12:20:57 +0100
Received: (from j@localhost) by uriah.heep.sax.de (8.7.3/8.6.9) id KAA10015; Fri, 29 Dec 1995 10:59:04 +0100 (MET)
From: J Wunsch <j@uriah.heep.sax.de>
Message-Id: <199512290959.KAA10015@uriah.heep.sax.de>
Subject: Re: cvs commit: src/lib/libc/gen getpass.c
To: freebsd-current@FreeBSD.org (FreeBSD-current users)
Date: Fri, 29 Dec 1995 10:59:04 +0100 (MET)
Cc: wollman@lcs.mit.edu, pst@FreeBSD.org
Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch)
In-Reply-To: <199512282218.OAA12618@precipice.shockwave.com> from "Paul Traina" at Dec 28, 95 02:18:15 pm
X-Phone: +49-351-2012 669
X-Mailer: ELM [version 2.4 PL23]
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
Sender: owner-current@FreeBSD.org
Precedence: bulk

As Paul Traina wrote:
> 
> Two points:
> (a) you two just changed the semantics of this routine, you need to check
>     all of the callers to make sure THEY block SIGINT and SIGTSTP.
> (b) This is a bad idea if you use ^C or ^Z as part of your password, which
>     was legal.

>   From: "Garrett A. Wollman" <wollman@lcs.mit.edu>

>   >   Don't block SIGINT in getpass(3); this doesn't make sense.
>     
>   It shouldn't block SIGTSTP, either.  (The calling program should if
>   it's important.)

Paul:

* Where have you been on Dec 10, when i've been asking in -current
  (Subject: getpass() and SIGINT) whether somebody objects?

* No, this hasn't changed the behaviour of getpass() in accepting any
  of the signalling characters; they had to be quoted with ^V (or
  whatever your lnext character is set to) anyway, as well as for
  example ^\.  The only change is that the ^C interpretation is now
  immediate, while you first had to press enter in order to see the
  effect of you intr previously.

Garrett:

* No, _blocking_ the signals isn't the right thing to do in any case.
  It has only been done to avoid the tty hassles involved (turning
  echo on/off).  Well, i just noticed that we should install a SIGINT
  handler anyway (i didn't notice this on tcsh since it's restoring
  the old tty settings itself, but it's very obvious if you interrupt
  a `su' attempt right now, say on plain ol' csh).  So it's perhaps a
  good idea to also install handlers for SIGTSTP and SIGCONT as well,
  in which case we could also avoid to block SIGTSTP.

-- 
cheers, J"org

joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE
Never trust an operating system you don't have sources for. ;-)