Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!know!sdd.hp.com!think.com!mintaka!mit-eddie!uw-beaver!cornell!pat
From: pat@cs.cornell.edu (Pat Stephenson)
Newsgroups: comp.unix.ultrix
Subject: LOCK field of pstat -s ?
Message-ID: <48592@cornell.UUCP>
Date: 19 Nov 90 20:08:48 GMT
Sender: nobody@cornell.UUCP
Reply-To: pat@cs.cornell.edu (Pat Stephenson)
Organization: Cornell University Computer Science Department
Lines: 28

Every so often, on my 12Meg DEcstation 3100, my epoch process (epoch is
a variation of emacs) will hang up for several minutes.  It's not
swapped out; if I run "pstat -s", the hung process has a non-zero
value for the LOCK field.  I have never seen a non-zero value for this
field under any other circumstances.  The man page for pstat doesn't
mention the field, however perusal of the source indicates that this
is the head of a chain of locks that seem to be associated with
symmetric multiprocessing (the p_hlock field defined in <sys/proc.h>).
Last I heard, the 3100 is not a multiprocessor (I wish!)

Can anybody give me any hints about what is going on?  Is this a known
bug?  Is it fixed in the latest version of ultrix?

Possibly relevent information:

epoch is invariably the largest process on the system when this
happens.  I am running:

ULTRIX V4.0 (Rev. 174) System #1: Thu Sep 27 17:03:50 EDT 1990
UWS V4.0 (Rev. 164)

This afternoon, I seem to be able to replicate this bug at will by
starting a compilation subprocess from within epoch.  I'm not sure if
this is always true.

I'd be really grateful for any hints about this; it is quite annoying.

Pat Stephenson