https://bugzilla.novell.com/show_bug.cgi?id=751712
https://bugzilla.novell.com/show_bug.cgi?id=751712#c45
Johannes Meixner changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |hare@suse.com
Summary|USB resets break printing |since kernel 3.1 USB resets
| |break printing for HP
| |all-in-one devices with
| |built-in cardreader
--- Comment #45 from Johannes Meixner 2012-03-23 08:20:47 UTC ---
I asked Hannes Reinecke and we agree that the whole idea
to have long lists of individual models in a generic udev rule
like 60-persistent-storage.rules it broken by design.
This means attachment #482601
is broken by design.
Reasons:
1)
by design generic udev rules should be generic - surprise! surprise!
(i.e. lists of individual models therein contradict the design)
2)
udev upstream which provides the generic udev rules
cannot maintain lists of individual models
(i.e. sooner or later the list will be outdated)
Solution in compliance with the udev design:
Have the list of individual models totally outside of udev
(i.e. even not as a udev rules file in /etc/)
and have a generic udev rule that uses IMPORT (see "man udev")
to import the list of individual models as a set of variables
into the udev event environment.
The generic udev rule that does the IMPORT
must run before 60-persistent-storage.rules
so that the imported variables are available
in 60-persistent-storage.rules.
Then add a generic test in 60-persistent-storage.rules
which enables or skips the media presence polling
depending on the values of the imported variables.
This way the list of individual models outside of udev
can be provided and maintained by HPLIP upstream.
For each HPLIP version HPLIP upstream can keep this
list up-to-date and HPLIP upstream does no longer
need to care about the internal details of udev.
In the end this means the whole HPLIP related
udev rules stuff needs a re-design.
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
