To: opendos AT mail DOT tacoma DOT net
Subject: Re: [opendos] Filesystems
Message-ID: <19970207.210324.8119.1.chambersb@juno.com>
References: <Pine DOT GSO DOT 3 DOT 95 DOT 970207211655 DOT 21429D-100000 AT sparkie DOT gnofn DOT org>
From: chambersb AT juno DOT com (Benjamin D Chambers)
Date: Sat, 08 Feb 1997 00:01:41 EST
Sender: owner-opendos AT mail DOT tacoma DOT net
Precedence: bulk


On Fri, 7 Feb 1997 21:22:57 -0600 (CST) "Colin W. Glenn"
<cwg01 AT gnofn DOT org> writes:
>On Fri, 7 Feb 1997, Benjamin D Chambers wrote:
>> On Thu, 6 Feb 1997 22:09:29 -0600 (CST) "Colin W. Glenn"
>
>> have, say 10 drives mounted with a different fs on each, you could
>> encounter a lag - but how many people run a program that accesses 10
>
>Also true, I just hope no-one comes up with a driver bigger than 120k, 
If it happens like this, I expect we'll start out with a few big drivers
(say 60-80k) for the most common filesystems, then suddenly everyone will
start pouring out their own FS's with their own (optimized) drivers (say,
10-20k, larger if the FS automagically implements compression (gasp! Did
I just write what I think I did???)).  Because the basic API would be the
same, we'd end up with hundreds of compatable FS's - BEAT THAT, M$!!! :)

>the
>smaller the better.  Better yet, instead of dumping the driver, what 
>about
>caching the driver?  That way you don't have the driver reinitialize
>itself, it's ready to do!  (This assumes the system uses the driver in
>Protected Mode with a smart enough swapmanager.)

Can you run a PM driver while the OS remains in RM?  Nifty >:-)

Problem, though:  I haven't coded anything REMOTELY similar to this.  How
feasible is it, really?  Anyone (I mean who has (pseudo)experience with
things of this nature) know whether or not this might work?

...Chambers