[openbeos] Re: Network Update
- From: "David Reid" <mail@xxxxxxxxxxxxxx>
- To: <openbeos@xxxxxxxxxxxxx>
- Date: Sun, 24 Mar 2002 11:56:48 -0000
> > Well, in an exciting week the code now runs in kernel space, with all
> > modules being loaded at run time. The drivers are still the original
> > be ones
> > being used via the same interface as the userland stack, which does
> > mean
> > that our code can be built to run as either a userland application or
> > as the
> > kernel drivers!
>
> Excuse my ignorance on the matter, but a few queries....
No problem.
> How does integration into kernelspace affect the way users have control
> over
> the networking modules?
Virtually none. The devil is in the details, but we hope to allow for
restarts to be "total".
> What I mean to say is, if there is a problem, is it still possible to
> simply 'restart'
> the net_server?
again, at the present time there are some issues, but hopefully we can
resolve them.
>And how about stability... is it wise to move things
> into kernel
> space when they're still untested... even when they are tested?
Well, a huge piece of the work to be done is to test the stack to
destruction and beyond. As has been pointed out the actual stack is being
based heavily on the BSD sockets code, used by BSD's with some modifications
to keep up with the latest security fixes and portability issues for
protocols. This gives us a nice set of code that is proven and so the
challenge is getting it working under beos with all the challenges that
poses.
As for stability, the stack sat and served ping replies for over 2 hours
being sent to both the network cards it was configured to use. while not a
real worls test it does show that the lower levels are working nicely...
Now we have UDP I'm hoping to add a UDP ping/reply and run that to see what
sort of throughput and stress level we can achieve!
> As much as we might like to think otherwise, there are plenty of
> situations
> when things break, and killing the kernel team is probably not a
> NiceThing[tm]
Well, we'd never reccomend doing that! As I said, we have some small issues
that we're working on, so we'll see what happens :)
One thing I can say - we won't push for an unstable stack to EVER be
included in OpenBEOS, and of course Michael wouldn't allow us to anyways!
david
Other related posts:
