On Oct 8,  3:40pm, Peter J. Braam wrote:
> Subject: Re: Coda Win95 port
> None of the clients you mention use a user level program to service kernel
> calls [which effectively sit below the mutex that is taken].
>
> Coda could in principle be implemented in the kernel, like they are.  From
> practical perspectives that would be a disaster.

Which perspective(s) is/are making it the worst?  Is it simply too much and
too complex to port?  Are there some modules (LWP, rx, cache manager I/O, etc.)
more difficult (to get into the kernel) than the rest?  Would it help to not
to do disk caching (only memory caching) at all?  I guess this would defeat
one main purpose of Coda - disconnected mode operation, but would it be simpler
at all that way?

> There will be a release of the 95 code for Coda that actually works in
> about 6 weeks from now.  A visiting student to our group Marc Schnieder
> has made good progress fixing a few bugs and implementing some missing
> components.

Nice to hear this!  Would this DOS application solution be an interim solution
or it will be "the" solution for Window 95?  Would this solution be very
vulnerable to the announced plan of discountinuing DOS support beyond
Windows98?

Shyh-Wei Luan


--