*BSD News Article 16405


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!munnari.oz.au!spool.mu.edu!darwin.sura.net!news-feed-1.peachnet.edu!emory!dragon!rokkaku!kml
Newsgroups: comp.os.386bsd.bugs
Subject: FIFOs cause lockup
Message-ID: <C7DtDE.4A@rokkaku.atl.ga.us>
From: kml@rokkaku.atl.ga.us (Kevin Lahey)
Date: Fri, 21 May 1993 14:53:37 GMT
Organization: Geeks-R-Us
Lines: 31

I'm running 386BSD 0.1 with a mishmash of patches from the patchkit up 
to 0.2.2, along with some patches from the net and a fix or two of my own.
I'm not a little curious, then, to see if others are experiencing the
same problems I am.

I rebuilt the kernel to support FIFOs using the 'options FIFO' line in the
config file.  It built fine, and the mkfifo command now works.  The following
sequence of commands will lock up the system:

% mkfifo /tmp/goop
% cat > /tmp/goop < /etc/passwd &
% cat /tmp/goop &
% cat > /tmp/goop < /etc/passwd &
% cat > /tmp/goop < /etc/passwd &

At this point, any command I enter fails to complete, and I can no longer
get the system to do anything.  On the console, this means that I can
continue to type, and I see the characters as I type them, but tcsh doesn't
respond.  When X is running, it will continue to respond to mouse movements,
but I can't get any shell I bring up to respond to my input.  Since TWM
forks a shell to deal with commands, that locks up as well.

Does anybody have any ideas about where I should look to fix this problem?
Does it show up for anybody else, or has my haphazard approach to applying
patches finally caught up to me?

Thanks,
Kevin

kml@rokkaku.atl.ga.us
emory!rokkaku!kml