Bug 901767

Summary: Need to wait 2 minutes before wireless connects
Product: [openSUSE] openSUSE Distribution Reporter: Dominique Leuenberger <dimstar>
Component: KernelAssignee: E-mail List <kernel-maintainers>
Status: RESOLVED FIXED QA Contact: E-mail List <qa-bugs>
Severity: Major    
Priority: P5 - None CC: bwiedemann, coolo, dimstar, forgotten_0FuaAO3939, forgotten_5x9cKIYF8b, jslaby, lurodriguez, os.gnome.maintainers, rmilasan, systemd-maintainers, tiwai, wicked-maintainers
Version: 13.2 RC 1   
Target Milestone: ---   
Hardware: x86   
OS: openSUSE 13.2   
Whiteboard:
Found By: --- Services Priority:
Business Priority: Blocker: ---
Marketing QA Status: --- IT Deployment: ---
Bug Depends on: 891645    
Bug Blocks:    

Description Dominique Leuenberger 2014-10-17 14:47:42 UTC 
+++ This bug was initially created as a clone of Bug #891645 +++

User-Agent:       Mozilla/5.0 (X11; Linux i686; rv:31.0) Gecko/20100101 Firefox/31.0

openSUSE 13.2 will NOT be shipping the 3.17 kernel, which has been checked in to Factory with the fix of bug 891645

This means that 13.2 USERS of Centrino-6x00 cards will be affected by this issue now in a final product.

We will need a fix in the Kernel 3.16.x stable branch of openSUSE 13.2
Comment 1 Stephan Kulow 2014-10-17 14:57:21 UTC 
According to Jiri this is about impossible to backport ;(
Comment 3 Jiri Slaby 2014-10-17 15:17:34 UTC 
(In reply to Stephan Kulow from comment #1)
> According to Jiri this is about impossible to backport ;(

This is maybe too bold statement. Luis, I think, has a better insight?
Comment 4 Luis Chamberlain 2014-10-17 15:33:44 UTC 
(In reply to Jiri Slaby from comment #3)
> (In reply to Stephan Kulow from comment #1)
> > According to Jiri this is about impossible to backport ;(
> 
> This is maybe too bold statement. Luis, I think, has a better insight?
Comment 5 Luis Chamberlain 2014-10-17 15:35:09 UTC 
(In reply to Luis Rodriguez from comment #4)
> (In reply to Jiri Slaby from comment #3)
> > (In reply to Stephan Kulow from comment #1)
> > > According to Jiri this is about impossible to backport ;(
> > 
> > This is maybe too bold statement. Luis, I think, has a better insight?

Takashi backported the requested firmware request changes to let us get rid of the udev firmware loading for SLE12 so I do believe we can get this on OpenSUSE as well, and probably should. In the meantime drivers that don't want to use udev helper explicitly can use request_firmware_direct() on older kernels.
Comment 6 Dominique Leuenberger 2014-10-17 17:03:19 UTC 
bug 900710 relates to this one (same issue, different module, nouveau)
Comment 7 Takashi Iwai 2014-10-20 08:33:47 UTC 
If we don't need to freeze kABI at this point, it should be easy to backport.  It's just a backport of commit 5a1379e8748a5cfa3eb068f812d61bde849ef76c, and *disable* a new Kconfig CONFIG_FW_LOADER_USER_HELPER_FALLBACK.

Let me know if it's OK to do the above at this late stage.
Comment 8 Stephan Kulow 2014-10-20 08:56:07 UTC 
I had hopes we could give the 3rd party repos a green light for stable kernel ABI, but if the bug is so severe I'm fine with waiting ;(
Comment 9 Takashi Iwai 2014-10-20 09:00:07 UTC 
The kABI change would be minimal (request_firmware_direct() becomes inlined or unlinined), and on x86 default and desktop kernels, there won't be even any change, I guess.
Comment 10 Stephan Kulow 2014-10-20 10:38:29 UTC 
Even better
Comment 11 Takashi Iwai 2014-10-20 12:06:31 UTC 
OK, pushed now to openSUSE-13.2 git branch.
Comment 12 Forgotten User 5x9cKIYF8b 2014-10-31 16:03:23 UTC 
When is this fix expected to land in 13.2?