kern/41264: ptyfs ttys unidle on write(1)

[dholland%eecs.harvard.edu@localhost]

>Number:         41264
>Category:       kern
>Synopsis:       ptyfs ttys unidle on write(1)
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Apr 22 08:00:00 +0000 2009
>Originator:     David A. Holland
>Release:        NetBSD 5.99.7 (20090209)
>Organization:
>Environment:
System: NetBSD tanaqui 5.99.7 NetBSD 5.99.7 (TANAQUI) #24: Mon Feb 9 11:19:51 
EST 2009 root@tanaqui:/usr/src/sys/arch/i386/compile/TANAQUI i386
Architecture: i386
Machine: i386
>Description:

        write(1) causes all(?) ptyfs ptys to unidle. This is quite
        annoying if on a system where people actually use it.

        Maybe write is doing something foolish, but it's much more
        likely ptyfs is handling atimes or mtimes wrong.

        There are some related older PRs, which I'll look up sometime
        when I have direct access to the PR database, which I don't
        right now.

>How-To-Repeat:

tanaqui% w
 3:43AM  up 56 days, 23:14, 5 users, load averages: 0.00, 0.15, 0.27
USER     TTY     FROM              LOGIN@  IDLE WHAT
dholland pts/17  [...]            24Feb09     1 -csh 
dholland pts/3   [...]            24Feb09     0 w 
dholland pts/35  [...]            Mon02PM     2 -tcsh 
[...]
tanaqui% write dholland
write: dholland is logged in more than once; writing to pts/17
test
tanaqui% w
 3:44AM  up 56 days, 23:14, 5 users, load averages: 0.00, 0.14, 0.26
USER     TTY     FROM              LOGIN@  IDLE WHAT
dholland pts/17  [...]            24Feb09     0 -csh 
dholland pts/3   [...]            24Feb09     0 w 
dholland pts/35  [...]            Mon02PM     0 -tcsh 
[...]
tanaqui% 
                                            ^^^

Note that this also caused write to pick the wrong tty.

>Fix:
dunno.