Bug 134586

Summary:	atheros driver: Does not work after suspend
Product:	[openSUSE] SUSE Linux 10.1	Reporter:	Andreas Jaeger <aj>
Component:	Network	Assignee:	Forgotten User ZhJd0F0L3x <forgotten_ZhJd0F0L3x>
Status:	VERIFIED FIXED	QA Contact:	E-mail List <qa-bugs>
Severity:	Minor
Priority:	P5 - None	CC:	andreas.hanke, forgotten_waLIv0Y828, hpj
Version:	Beta 2
Target Milestone:	Beta 4
Hardware:	Other
OS:	Other
Whiteboard:
Found By:	Other	Services Priority:
Business Priority:		Blocker:	---
Marketing QA Status:	---	IT Deployment:	---

Description Andreas Jaeger 2005-11-20 17:54:52 UTC

After both suspend2disk and suspend2ram, a restart of the network connection does not happen with the atheros driver.  A rcnetwort restart does not help.  Only unloading of all ath* modules helped.

This is with current 2.6.14.2 kernel recompiled with updated wireless tools package, the new wireless and wpa packages.

Comment 1 Andreas Jaeger 2005-12-04 09:28:10 UTC

fails also on Alpha3plus with update to kernel 2.6.15-rc4-20051202095602-default.

Comment 2 Christoph Thiel 2006-02-01 07:53:19 UTC

Still fails with Beta2 (and Beta3 kernel). Doing a "rmmod ath_pci; modprobe ath_pci" fixes the problem.

Comment 3 Christoph Thiel 2006-02-02 13:24:08 UTC

Just a short sidenote: This bug has been reported upstream some time ago at http://madwifi.org/ticket/201

I'v tested the suggested patch that is attached to the upstream bugreport but it unfortunately didn't fix the problem ;(

Comment 4 Christoph Thiel 2006-02-05 22:46:03 UTC

Just in case we can't get this fixed in time for 10.1 -- let's add "ath_pci" to UNLOAD_MODULES_BEFORE_SUSPEND2DISK. (Which works great for me ;))

Comment 5 Joachim Gleissner 2006-02-09 18:39:47 UTC

Yes, may be necessary.

Comment 6 Forgotten User ZhJd0F0L3x 2006-02-10 19:12:13 UTC

*** Bug 147815 has been marked as a duplicate of this bug. ***

Comment 7 Forgotten User ZhJd0F0L3x 2006-02-10 19:14:51 UTC

i have added the ath_pci driver to the unload-before-suspend list in SVN for beta4. Package is not yet submitted, and of course the driver should be fixed somewhen :-)

Comment 8 Joachim Gleissner 2006-05-03 09:40:14 UTC

Indeed. ;-) Resolving as LATER.

Comment 9 Stefan Behlert 2006-09-13 12:16:58 UTC

Can you please look at that for 10.2 again? Thanks.

Comment 10 Cyril Hrubis 2006-10-20 15:52:50 UTC

*** Bug 213931 has been marked as a duplicate of this bug. ***

Comment 11 Forgotten User waLIv0Y828 2006-11-01 22:59:51 UTC

This behaviour is also seen in SLED 10.

Comment 13 Joachim Gleissner 2007-01-18 09:48:51 UTC

SLED10 has ath_pci in its default modules-to-unload list. 10.2 is using pm-utils instead of powersave for suspending, and I don't think pm-utils unload ath_pci by default. We didn't ship madwifi with 10.2 anyway. I'll reassign to seife just in case he wants to add it, although I doubt it.

Comment 14 Forgotten User ZhJd0F0L3x 2007-01-18 11:03:14 UTC

no. I can't, for technical reasons with rpm.

Whoever maintains / ships madwifi just needs to put a custom hook for use by pm-utils into their kmp-package that does the unloading. I can help them with that, but i don't want to ship it with our pm-utils.

For 10.2, this is basically the same as bug 232420. For SLED, this should be fixed.