[openbeos] Re: Status of the Kits

En réponse à Ingo Weinhold <bonefish@xxxxxxxxxxxxxxx>:

> "Axel Dörfler"  <axeld@xxxxxxxxxxxxxxxx> wrote:
> > "Ingo Weinhold" <bonefish@xxxxxxxxxxxxxxx> wrote:
> > > I still don't know, how _kstart_watching_vnode_() and 
> > > _kstop_watching_vnode_() look like. watch_node() and 
> > > stop_watching() 
> > > are mapped to them, so I have a rough idea, but my x86 assembler 
> > > skills 
> > > are rather rudimentary, so I had to go with try and error, which I
> 
> > > weren't really motivated to do. Node monitoring is a bit more of an
> 
> > > issue, since we can't use the libbe version of watch_node()/
> > > stop_watching(), when we also link against libopenbeos. Currently 
> > > it 
> > > is 
> > > only needed in one place in the registrar (I crudely worked around
> 
> > > by 
> > > running a separate application), which is going to go away soon, 
> > > anyway. Don't know about the app server, though.
> > 
> > I wouldn't mind implementing those right now - if possible, and 
> > testing 
> > won't be rendered impossible due to that.
> 
> You mean _k{start,stop}_watching_vnode_()? I'm afraid that would help 
> until our complete libopenbeos is ready to run on top of of our libroot
> 
> -- which won't be possible, until our libroot is virtually complete and
> 
> libopenbeos doesn't depend on libbe anymore. Kind of an awesome 
> milestone BTW, but still some days in the future. :-)
> 

Should be able to determine that with some bdb-ing...
objdump shows that _kstart_watching_vnode_() has 6 params,
4 of which get -1, -1, -1, 0 in a case 
(I suspect when the BHandler is NULL).

François.

Other related posts: