[openbeosnetteam] Re: mbuf.h

> > Solution to fix that issue need more consideration.
> > 
> > I don't like the idea of a public variable exported by kernel 
> > module api.
> > Beside that issue, that's global variables aiming to be used *per* 
> > module.
> > max_xxxx variables won't have same value for each network modules 
> > using 
> > mbuffers...
> 
> Why do we need this information per module? 

What I tried to say is that this maximum value should be expanded to 
maximum values asked by 
each network modules.
As, currently, they're all loaded and inited at stack start time, these 
modules could perfectly 
call some set_max_link_header_size(), raising the max_linkhdr mbuf 
internal value if necessary.
Same for the max_protohdr value.

> Could not all modules just set max_linkhdr and max_protohdr globally 
> (using get/set_max_xxx functions exported by > the core module)? IMHO, 
this would at least work until we get a netstack replacement.

Yep, that's exactly what I had in mind when I wrote my reply, but fail 
to explain :-(

> There are many more bugs in the stack. How much is working at the 
> moment? 
> Will TCP work after these mbuf bugs are solved (or does it work 
> already)?

It (should) works already, except for any remaining bugs still to be 
found. And fixed, so.

- Philippe

--
Fortune Cookie Says:

Ducharme's Precept:
        Opportunity always knocks at the least opportune moment.

Other related posts: