amd64/101248: vi(1) can crash in ncurses(3) on amd64

[Andrew Pantyukhin]

The following reply was made to PR amd64/101248; it has been noted by GNATS.

From: "Andrew Pantyukhin" <infofarmer at FreeBSD.org>
To: "Yar Tikhiy" <yar at comp.chem.msu.su>
Cc: FreeBSD-gnats-submit at freebsd.org
Subject: Re: amd64/101248: vi(1) can crash in ncurses(3) on amd64
Date: Tue, 27 Mar 2007 12:44:41 +0400

 On 8/2/06, Yar Tikhiy <yar at comp.chem.msu.su> wrote:
 >
 > >Number:         101248
 > >Category:       amd64
 > >Synopsis:       vi(1) can crash in ncurses(3) on amd64
 > >Confidential:   no
 > >Severity:       non-critical
 > >Priority:       medium
 > >Responsible:    freebsd-amd64
 > >State:          open
 > >Quarter:
 > >Keywords:
 > >Date-Required:
 > >Class:          sw-bug
 > >Submitter-Id:   current-users
 > >Arrival-Date:   Wed Aug 02 09:00:28 GMT 2006
 > >Closed-Date:
 > >Last-Modified:
 > >Originator:     Yar Tikhiy
 > >Release:        FreeBSD 7.0-CURRENT amd64
 > >Organization:
 > None
 > >Environment:
 > System: FreeBSD sledge.freebsd.org 7.0-CURRENT FreeBSD 7.0-CURRENT #741: Tue Aug 1 14:17:00 UTC 2006 root at sledge.freebsd.org:/h/src/sys/amd64/compile/SLEDGE amd64
 >
 > >Description:
 >         If a +N command line option is given to vi(1) on amd64,
 >         where N is greater than the actual number of lines in the
 >         file to edit, vi(1) will crash on signal 11 reproducably.
 >         The stack trace indicates that the crash happened in a
 >         ncurses(3) function, but it isn't evident yet where the
 >         bug itself lurks (vi may pass bogus data to ncurses.)
 >
 >         The problem won't reproduce on ia64 or i386.
 
 > #0  0x00000008006bc1f8 in tputs () from /lib/libncurses.so.6
 
 I see this crash on latest CURRENT/amd64 when trying to
 go into ex mode (Shift-Q).