Bug 104874

Summary: module "lp" unloaded during startup of YaST printer config
Product: [openSUSE] SUSE LINUX 10.0 Reporter: Johannes Meixner <jsmeix>
Component: PrintingAssignee: Steffen Winterfeldt <snwint>
Status: RESOLVED FIXED QA Contact: Johannes Meixner <jsmeix>
Severity: Major    
Priority: P5 - None CC: gholmer, jdd, jsmeix, peter
Version: Beta 1   
Target Milestone: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Found By: Development Services Priority:
Business Priority: Blocker: ---
Marketing QA Status: --- IT Deployment: ---

Description Johannes Meixner 2005-08-16 07:42:10 UTC 
I installed beta1 on x86_64 (on caps.suse.de).
I connected a HP DeskJet 990C to the parallel port.

The "lp" module becomes unloaded during startup of "yast2 printer"
so that even manual setup of a parallel port printer doesn't work
(in particular printing of testpage cannot work)
until I do manually "modprobe lp".

Again this is caused by hwinfo
(see the old bug 64412, in particular comment number 57):
-----------------------------------------------------------------
caps:~ # lsmod | egrep 'parport|lp'
lp                     30912  0 
parport_pc             62192  1 
parport                61324  2 lp,parport_pc
usblp                  31488  0 
usbcore               157996  3 usblp,ohci_hcd

caps:~ # hwinfo --printer &>/dev/null

caps:~ # lsmod | egrep 'parport|lp'
parport_pc             62192  0 
parport                61324  1 parport_pc
usblp                  31488  0 
usbcore               157996  3 usblp,ohci_hcd
-----------------------------------------------------------------

See also bug #102932 which may be the same problem.
Comment 1 Johannes Meixner 2005-08-16 15:03:50 UTC 
*** Bug 102932 has been marked as a duplicate of this bug. ***
Comment 2 Johannes Meixner 2005-08-16 15:05:15 UTC 
Happens als on i686 architecture, see bug #102932.
=> Assume it happens on any architecture (which has a parallel port).
Comment 3 Michal Zugec 2005-08-17 11:14:28 UTC 
lp module unloaded when execute SCR::Read (.probe.printer);
this is problem of hwinfo (hwinfo --printer makes the same reason)
Comment 4 Steffen Winterfeldt 2005-08-18 09:47:21 UTC 
Sure it unloads lp. AFAIK it's still necessary to reload parport_pc to 
trigger parport device probing (correct me if I'm wrong). 
 
So what should it do? Load lp again? 
 
BTW, Johannes, why didn't you drop me a note in bug 64412??? I surely can't 
fix bugs I don't know about.
Comment 5 Johannes Meixner 2005-08-18 10:36:24 UTC 
As far as I know it is still true that parport device detection happens
only when the parport/parport_pc are loaded.
But I am no kernel module developer.

Therefore we may
either rely on what was detected when those modules are loaded during booting
or we may like to force re-detection by unloading lp, parport_pc, parport
and re-loading parport, parport_pc, lp.

Steffen,
I remembered bug #52548 which is about 9.1 where it seemed
that hwinfo is not responsible to re-load the lp module
because it was loaded when there was a device node access.
But since 9.1 things are changing so much that I don't know
which component is at the moment responsible for what.
Comment 6 Steffen Winterfeldt 2005-08-18 10:40:31 UTC 
Now that we have that dynamic device thing, autoloading no longer 
works. I'll change hwinfo to load lp, too, then. Let's see what that gives.
Comment 7 Steffen Winterfeldt 2005-08-19 14:10:06 UTC 
done
Comment 8 Johannes Meixner 2005-08-22 07:24:59 UTC 
*** Bug 105920 has been marked as a duplicate of this bug. ***
Comment 9 Michal Zugec 2005-08-23 13:35:24 UTC 
*** Bug 105379 has been marked as a duplicate of this bug. ***
Comment 10 Michal Zugec 2005-09-02 11:35:14 UTC 
*** Bug 113558 has been marked as a duplicate of this bug. ***