Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!think.com!snorkelwacker.mit.edu!bloom-beacon!bu.edu!transfer!lectroid!jjmhome!acestes!paradis
From: paradis@acestes.UUCP (Jim Paradis)
Newsgroups: comp.os.minix
Subject: Re: emacs not executable
Message-ID: <161@acestes.UUCP>
Date: 11 Mar 91 20:55:36 GMT
References: <1107@qusunb.queensu.CA> <1108@qusunb.queensu.CA>
Reply-To: paradis@acestes.UUCP (Jim Paradis)
Distribution: na
Organization: Alberichs Rings & Things
Lines: 24

In article <1108@qusunb.queensu.CA> cox@qucis.queensu.CA (Donald Cox) writes:

>The problem, I have discovered after some experimentation is that if the
>executable text + stack + malloc > 400000 bytes the systems refuses to 
>recognize it as executable....

Hmmm... I think the problem may simply be that you're short on memory...
especially if you've got a login or shell running on each of your virtual
windows.  Think about it: MINIX is 100-odd Kbytes, the buffer cache is
300-odd kbytes (with the default NR_BUFS for INTEL_32BITS), each login
is about 70Kbytes, update is another 65Kbytes, your shell is prob'ly
64K or so...

A kilobyte here and a kilobyte there, and soon you're talking real memory 8-)

Under MINIX-386, the only limit on process size is the amount of free
contiguous physical memory available in the machine when the process
is started up.  I have 4Mb on my machine, and I can run 2Mb processes
no problem (I had to chmem "pathalias" to 2Mb in order to get it to
successfully crunch the full UUCP map).

-- 
Jim Paradis                  UUCP:  harvard!m2c!jjmhome!acestes!paradis
9 Carlstad St.               AT&T:  (508) 792-3810
Worcester, MA 01607-1569     ICBM:  42deg 13' 52",  71deg 47' 51"