Skip to Content.
Sympa Menu

sm-commit - Re: [SM-Commit] PERFORCE change 79107 by Juuso Alasuutari for review

sm-commit AT lists.ibiblio.org

Subject: Source Mage code commit list

List archive

Chronological Thread  
  • From: Juuso Alasuutari <iuso AT sourcemage.org>
  • To: Andrew Stitt <a AT t.armory.com>, Arjan Bouter <abouter AT sourcemage.org>, SM-Commit Daemon <sm-commit AT lists.ibiblio.org>, duane_malcolm <d.malcolm AT auckland.ac.nz>, Ethan Grammatikidis <eekee AT eekee.is-a-geek.org>, Gareth Clay <gareth AT caffeinefuelled.co.uk>, Maurizio Boriani <baux AT member.fsf.org>, Pieter Lenaerts <e-type AT sourcemage.org>, Robin Cook <rcook AT wyrms.net>, Seth Woolley <seth AT swoolley.homeip.net>, Unet <unet AT sourcemage.org>, vladimir_marek <vlmarek AT volny.cz>
  • Subject: Re: [SM-Commit] PERFORCE change 79107 by Juuso Alasuutari for review
  • Date: Thu, 11 May 2006 01:01:44 +0300

On Thursday 11 May 2006 00:07, Jeremy Blosser wrote:
> On May 10, Perforce Review Daemon [p4review AT smee.org] wrote:
> > Change 79107 by juuso_alasuutari@juuso_alasuutari-kamayuq on 2006/05/10
> > 21:18:26
> >
> > Fixed usb printer node naming, added new rules, added new
> > keywords
>
> There's been a lot of this tweaking going on lately, and 50-default.rules
> has 224 lines by now, and the spell is looking quite complex overall. Can
> I ask for a sanity check... are we still installing only upstream + the
> minimal config to make things work and leaving the rest for the local admin
> to take care of? Or at least making all the helpful autoconfiguration
> optional?
>
> Just checking...

I'd like this to receive more testing, yes. But I guess that's why we have
test, right?

I don't know what exactly you mean by sanity check. Can you clarify?

I admit the udev spell isn't as bare and stripped as possible. But udev is
also unlike most other programs. If we would leave it precisely in
its "upstream state", there wouldn't be any rules at all.

The udev source file includes rulesets for a few distros such as gentoo,
debian, frugalware, etc. There are no generic rules, only distribution
specific ones.

The 50-default.rules that existed before I laid my fingers on it was mainly
thiefed from the Gentoo rules file. I've simply added node naming rules from
other distros' rule files for various devices.

I admit the file is large, and there probably are differing opinions about
what it should include. But as I see it, device node rules don't directly
impose anything on the user: If you have a device, you'll find that a node
exists when you plug it in. If you haven't that device, you aren't losing
anything. The node names I've used are the "standard" names found in almost
any distro.

If there is a general wish for 50-default.rules to be split into optional and
non-optional rules files, then we also need to separate the wheat from the
chaff. That might turn out to be a difficult task. For instance, in the old
50-default.rules we only had one rule for printers:
KERNEL="lp*", NAME="%k", MODE="0660"
Does that mean that the rule for usb printers in the new file is optional?
And while a rule for the Lego usb tower isn't really a must-have (yes, I
admit
I put that in the new file), is a rule for jaz devices also something we can
do without? Are jaz drives "lesser" devices than e.g. v4l devices (for which
there are rules in the old file)?

Well, I'm tired and probably missing good points as well as spewing out bad
ones. :) Be glad to continue this discussion tomorrow.

--
Juuso Alasuutari
[[ Source Mage GNU/Linux ]]

Attachment: pgpzcY9CX1QGH.pgp
Description: PGP signature




Archive powered by MHonArc 2.6.24.

Top of Page