linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Mike Fedyk <mfedyk@matchmail.com>
To: Andrey Borzenkov <arvidjaar@mail.ru>
Cc: Oliver Neukum <oliver@neukum.org>, linux-kernel@vger.kernel.org
Subject: Re: Re[2]: Input layer demand loading
Date: Thu, 17 Jul 2003 11:38:59 -0700	[thread overview]
Message-ID: <20030717183859.GA1003@matchmail.com> (raw)
In-Reply-To: <E19d16P-00044X-00.arvidjaar-mail-ru@f19.mail.ru>

On Thu, Jul 17, 2003 at 09:14:49AM +0400, "Andrey Borzenkov"  wrote:
> 
> 
> -----Original Message-----
> 
> > 
> > On Wed, Jul 16, 2003 at 10:19:17PM +0400, Andrey Borzenkov wrote:
> > > >> True, but then if you try to open the port, you will only get the base
> > > >> joydev.o module loaded, not the gameport driver, which is what you
> > > >> _really_ want to have loaded, right?
> > > >> 
> > > >> So there really isn't much benifit to doing this, sorry.
> > > >
> > > > Why? It could work the way PCMCIA SCSI works.
> > > > Cardmgr loads the LLDD, but sd, sg, etc. are loaded
> > > > on demand.
> > > 
> > > how? SCSI (or USB, PCI, EISA etc) have driver-independent means to identify 
> > > device or at least device class.
> > > 
> > > But how are you going you going to know you need to load specific mouse driver 
> > > (logitech not microsoft) or specific joystick flavour? All that you possibly 
> > > know that you have _some_ mouse or _some_ joystick ...
> > 
> > Isn't that why we have hotplug in userspace?
> > 
> > That way, we know we have a mouse, but it's up to userspace to find out
> > what kind.
> > 
> 
> that is not what hotplug is for. What you describe is called "probing". 
> 
> Hotplug is there to load hardware driver once hardware is known, not
> to probe what hardware is there.
> 
> It can be used this way but it is rather abuse (or misuse) I'd say.

In many cases, you only know you have a class of devices, and you need more
probing after the initial hardware notification (hotplug).

There are many people that want all probing to be in user space.  Let's face
it, it's easier (no need to reboot) to upgrade a userspace probing app than
to install a new kernel.

Unfortunatly, it doesn't look like that will happen for 2.6, but I'd bet it
will in 2.7, and hotplug will probably be a considerable player in that area
(unless something better is used instead).

      parent reply	other threads:[~2003-07-17 18:27 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-07-16 18:19 Input layer demand loading Andrey Borzenkov
2003-07-16 18:29 ` Oliver Neukum
2003-07-16 18:33 ` Mike Fedyk
     [not found]   ` <E19d16P-00044X-00.arvidjaar-mail-ru@f19.mail.ru>
2003-07-17 18:38     ` Mike Fedyk [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20030717183859.GA1003@matchmail.com \
    --to=mfedyk@matchmail.com \
    --cc=arvidjaar@mail.ru \
    --cc=linux-kernel@vger.kernel.org \
    --cc=oliver@neukum.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).