Mail Archives: opendos/1997/02/06/20:57:14
On Thu, 6 Feb 1997 05:52:05 +0000 "Ian 'DrDebug' Day"
<Ian AT darkblak DOT demon DOT co DOT uk> writes:
>In article
><Pine DOT GSO DOT 3 DOT 95 DOT 970204222207 DOT 15188C-100000 AT sparkie DOT gnofn DOT org>,
>"Colin W. Glenn" <cwg01 AT gnofn DOT org> writes
>>On Tue, 4 Feb 1997, Benjamin D Chambers wrote:
>>> Well, I'm getting tired of 3mil threads with the same title, so I'm
>>> carrying this one to a new subject...
>>Psst. Good idea.
>>
>>> Now, as long as the driver code is, say, under 1 meg than most
>people
>>
>>A MEG Driver! Are you nuts?
>
>JESUS! If he dares submit a 1meg driver, I'll personally flail him
>alive! ;-)
>
>GUI OS, yes, sure. Everyone seems to do it, but DOS! FOAD.
SHEESH!! I picked a RANDOM number that I was SURE that ALL drivers would
be LESS than and that NOONE would mind missing too much!!! How many
problems can that cause?? I could easily have said 64k or so, but you
never know what kind of fatware people come up with, so a safer bet would
probably be 128k.
Anyways, I'd still like to hear what people think about the original
sugestion: Storing the access routines at the beginning of the partition,
loading them in the first time a partition is accessed, after no accesses
for a while, dumping them from memory. AFAIS, this would solve most
compatability problems - emulating a FAT drive _should_ be child's play
like this (though I don't know for _sure_ - that's why I asked for
comments in the original post.)
...Chambers
- Raw text -