Adding extra width to the filesystem interface

[Michael Smith <miff@apanix.apana.org.au>]

Greetings people, and apologies if I am resurrecting a dead thread here.
(speaking of which, is anyone archiving this list, and if so, can
I steal a copy of it?)

Some time ago, a friend and I started work on an ethernet interface unit
for the ST, as much as a learning experience as anything else.  Since 
then, it's come along slowly, and developed into a somewhat more diverse
device, however it's roots as a SCSI/ACSI <--> ethernet adapter remain.
(cause the first)

Slightly more recently, the australian public access network association
finally got their collective act together and started a nationwide
SLIP network.
(cause the second)

X likes to run on the berkley socket interface.
(cause the third)

For a while now I've been working on a neat and portable way of adding a
set of Berkley-compatible socket calls to MiNT - possibly drawing heavily
on the net-2 code ( it is a little elaborate in some places, but I can 
certainly vouch for its effectiveness ).  

The obvious way to do this is via the loadable filesystem interface,
unfortunately there is no pleasant way to map the rather large number
of calls that actually need to be handled by the kernel onto that semantic.
( Especially as the kernel likes to preprocess quite a few of them before
the filesystem finds out )

I'm reasonably certain that this issue, or something similar must have been
discussed before - so what conclusions, if any, were reached? 

-- 
# mike smith : miff@apanix.apana.org.au - Silicon grease monkey        #
# "The question 'why are the fundamental laws of nature mathematical'  #
# then invites the trivial response 'because we define as fundamental  #
# those laws which are mathematical'". Paul Davies, _The_Mind_of_God_. #