Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!uunet!visenix!beattie
From: beattie@visenix.UUCP (Brian Beattie)
Newsgroups: comp.os.minix
Subject: Re: (none)
Message-ID: <841@visenix.UUCP>
Date: 31 May 90 02:34:04 GMT
References: <20534@nigel.udel.EDU> <1990May30.215212.6245@sun.soe.clarkson.edu>
Reply-To: beattie@visenix.UUCP (Brian Beattie)
Organization: Department of Redundancy Department
Lines: 33

In article <1990May30.215212.6245@sun.soe.clarkson.edu> mikey@clutx.clarkson.edu writes:
>From article <20534@nigel.udel.EDU>, by uucp@hgatenl.hobby.nl (0000-uucp(0000)):
>
>v7 on an 11/34?  The 11/34 I used to work on was running v6 because it
>did not have a large enough address space for v7 (or so they said).  Also
>the 11/34 WAS NOT a 64K machine!  It had a seventeen bit address bus

actually it was eighteen bits if I am not mistaken (0-17).

>(thanx to the MMU) addressing 16-bit words yielding 256KB.  That is four
>times the 64K figure being bounced around.  If v7 would not run on the
>11/34 with 256KB core, paging (which impies swapping) etc, how can it
>run on an 8-bit machine with no MMU and a 64KB address spce (and less
>register => more instructions)?
>
.
:
>
>Mike
v7 in the standard configuration required "split I&D" for the kernel.
Further with 256KB of memory it would consistantly hang when more
than two or three users tried to use the system at the same time.
I sure both of these problems could have been over-come with some
effort but it was not considered worth it in most cases.  v6 on
the other hand would run fine on an LSI 11/23+ (needed the eighteen bit 
address bus)


-- 
Performance is not a   | Brian Beattie          (703)471-7552
security requirement.  | 11525 Hickory Cluster, Reston, VA. 22090 
            A nameless | beattie@visenix.UU.NET
Security Architect     | ...uunet!visenix!beattie