Xref: utzoo news.software.nntp:458 news.software.b:3815
Path: utzoo!utstat!helios.physics.utoronto.ca!jarvis.csri.toronto.edu!cs.utexas.edu!usc!brutus.cs.uiuc.edu!ux1.cso.uiuc.edu!uxf.cso.uiuc.edu!chris
From: chris@uxf.cso.uiuc.edu (Christopher Martin)
Newsgroups: news.software.nntp,news.software.b
Subject: Re: rrn dies when brought back into foreground
Summary: I think it's a bug...
Message-ID: <1989Dec12.042641.19976@ux1.cso.uiuc.edu>
Date: 12 Dec 89 04:26:41 GMT
References: <1989Dec10.153749.23565@phri.nyu.edu>
Sender: news@ux1.cso.uiuc.edu (News)
Organization: University of Illinois at Urbana-Champaign
Lines: 43

In article <1989Dec10.153749.23565@phri.nyu.edu> roy@phri.nyu.edu (Roy Smith) writes:
>
>	A few weeks ago I install rrn on our Sun-3/50s running SunOS-3.5.2.
>Every once in a while, if you suspend rrn, when you bring it back into the
>foreground, you get "read error: not owner" and rrn exits.  Anybody have
>any idea what's going on?  We're running rrn 4.3.1.4, patchlevel 39.
>

I just hunted this down this afternoon.  It seems to me that there may be
a bug in the code.  It seems that errno shouldn't be looked at unless read_tty
returns an error.  Here's a diff off rrn 4.3.2.5:


*** term.c.DIST	Mon Dec 11 14:17:41 1989
--- term.c	Mon Dec 11 14:44:16 1989
***************
*** 733,744 ****
      for (;;) {
  	int_count = 0;
  	errno = 0;
! 	if (read_tty(whatbuf,1) < 0 && !errno)
! 	    errno = EINTR;
! 	if (errno && errno != EINTR) {
! 	    perror(readerr);
! 	    sig_catcher(0);
! 	}
  #ifdef PUSHBACK
  	if (*whatbuf & 0200 || no_macros) {
  	    *whatbuf &= 0177;
--- 733,745 ----
      for (;;) {
  	int_count = 0;
  	errno = 0;
! 	if (read_tty(whatbuf,1) < 0)
! 	    if (!errno)
! 		errno = EINTR;
! 	    else {
! 		perror(readerr);
! 		sig_catcher(0);
! 	    }
  #ifdef PUSHBACK
  	if (*whatbuf & 0200 || no_macros) {
  	    *whatbuf &= 0177;