l4-hurd
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Task destruction


From: Marcus Brinkmann
Subject: Re: Task destruction
Date: Wed, 7 Aug 2002 00:56:06 +0200
User-agent: Mutt/1.4i

On Wed, Aug 07, 2002 at 12:45:48AM +0200, Marcus Brinkmann wrote:
> On Tue, Aug 06, 2002 at 05:18:19PM +0200, Niels Möller wrote:
> > I'm aware of that. I think I'm willing to assume that all
> > personalities using shared servers like the task server use the same
> > libport conventions, though, for things like transfer of port
> > rights/handles.
> 
> An interesting point.  I am tending to think that this assumption is way too
> strong, and we should split up the discussion into a generic task server and
> a Hurdish proxy task server.

BTW, for those people who think that this opens another can of worms:
What I mean here is that I feel that there is a lot of room between the
things that must be in a task server that sits directly on top of L4, and
the proc server in the Hurd system.

The discussion always says the task server, but I actually think there must
be two of them, one which only does simple operations: Cretae a task,
destroy a task, create a thread, destroy a thread, etc.  Those operations
that must be in this server because of L4 semantics.  This server could have
some light attributes associated with tasks, like task IDs, and maybe some
sort of parent-child or grouping relationship (not a full blown handle, just
an integer, for example).

And then there should be a Hurdish task server, that knows about the Hurd
IPC semantics (like object handles etc), can create task death
notifications, and things like that.  I would likely be less concerned about
how much features end up in this server, as long as it is reasonable to be
in there instead in proc.  It's also interesting to think about if there
should be one such server for all Hurd and Hurd-alike systems running, or
one for each, or a mixture of both.

As for real world development, we can very well start out with a Hurdish
task server, as long as we keep in mind that at some time there might be
another task server put between that and L4.

Thanks,
Marcus

-- 
`Rhubarb is no Egyptian god.' GNU      http://www.gnu.org    address@hidden
Marcus Brinkmann              The Hurd http://www.gnu.org/software/hurd/
address@hidden
http://www.marcus-brinkmann.de/



reply via email to

[Prev in Thread] Current Thread [Next in Thread]