Re: ioctl's handling with a trivfs translator

[Marcus Brinkmann]

On Thu, Mar 14, 2002 at 05:41:04PM +0100, Ludovic Courtès wrote:
> > One of the reason to not have a ioctl call is that not all traditional UNIX
> > ioctls can be dealt with by simple rpcs, it needs support in the user task. 
> > Another is cleanliness:  ioctl is a horrible hack to avoid designing proper
> > interfaces, or maybe to save syscall numbers.  In the Hurd, we have a proper
> > IPC system, so we don't really need ioctls.  But we need them anyway for
> > compatibility, so we have them, but then a ioctl id corresponds to an RPC,
> > or to a special case in the C library.
> 
> So, if I understand well, libstore is the user interface that makes it 
> possible
> for a user program to communicate with a storeio server, right?

No, libstore is the library that provides a store abstraction to, for
example, servers.  storeio is one server using libstore.  But I am not sure
what this has to do with ioctl's :)  A store is basically one or multiple
runs of blocks.  The blocks can be collected from different sub-stores, like
devices, and manipulated upfront, like gunzip'ed.

Play with the storeinfo program a bit to get the feel of its power.

> I've been looking for the source code of the mouse and kbd translators in the
> hurd source tree but I haven't been able to find it. Any pointers? Is it 
> related
> to streamio?

mouse and kbd are only in the Debian package.  We want them to be part of a
more general scheme, the channel abstraction, which is to character devices
what a store is to block devices (see the analogy?).  Nobody has written
libchannel yet, though.

Thanks,
Marcus

-- 
`Rhubarb is no Egyptian god.' Debian http://www.debian.org brinkmd@debian.org
Marcus Brinkmann              GNU    http://www.gnu.org    marcus@gnu.org
Marcus.Brinkmann@ruhr-uni-bochum.de
http://www.marcus-brinkmann.de