Bug 513751 - Freeze on suspend on Amilo Li3710 because of rtl8187
Summary: Freeze on suspend on Amilo Li3710 because of rtl8187
Keywords:
Status: CLOSED DUPLICATE of bug 469461
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: 11
Hardware: All
OS: Linux
low
medium
Target Milestone: ---
Assignee: Kernel Maintainer List
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2009-07-25 10:41 UTC by Mads Kiilerich
Modified: 2009-08-19 22:13 UTC (History)
7 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2009-08-17 18:44:52 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)
dmesg (71.95 KB, text/plain)
2009-07-26 17:15 UTC, Mads Kiilerich
no flags Details

Description Mads Kiilerich 2009-07-25 10:41:54 UTC
Description of problem:

A friends FUJITSU SIEMENS AMILO Li3710 10601011427 couldn't suspend; it just hang and had to be reset hard. It had been like this since early F10. IIRC there was nothing interesting in /var/log/messages after the hangs.

http://www.smolts.org/client/show/pub_5800899e-b719-47c7-a4ac-5e622fd9f2de

Trial and error showed that it worked when the rtl8187 module was removed before suspending. So adding rtl8187 to SUSPEND_MODULES in /etc/pm/config.d/ was a workaround.

I assume that suspend should work out of the box, but I don't know exactly where it should be addressed. 

Should it be added to hal-info as a quirk? 

Or should the workaround be applied manually until something gets fixed in the kernel so the module can be suspended on this kind of machine?


Version-Release number of selected component (if applicable):

hal-info-20090414-1.fc11.noarch
pm-utils-1.2.5-2.fc11.i586
kernel-PAE-2.6.29.6-213.fc11.i686


How reproducible:

Gnome suspend fails everytime - but probably only on this specific kind of machine

Comment 1 Till Maas 2009-07-26 11:30:03 UTC
This is probably a bug in the kernel, therefore I reassigned it to the kernel component.

Comment 2 Mads Kiilerich 2009-07-26 17:15:56 UTC
Created attachment 355207 [details]
dmesg

Ok. So no hal-info quirks for suspend modules?

I assume dmesg is needed for kernel developers. Some other debug info will probably also be needed. What?

Comment 3 Hin-Tak Leung 2009-08-17 03:23:45 UTC
Yes, it is a kernel problem and a known issue. I just haven't got round to do anything about it. probably should resolve this as duplicate of bug 469461 though. More info there.

Comment 4 Mads Kiilerich 2009-08-17 08:08:23 UTC
Hin-Tak Leung: You didn't mark it is a duplicate - because you are not sure, or because you don't have the rights to do it? Should I do it or should I wait for another expert?

If it is such an old known issue which haven't been fixed properly: It would be very easy to include an ugly workaround for Fedora. Perhaps by just putting the module(s) in SUSPEND_MODULES by default. I propose doing that for F12 if it haven't been resolved properly.

Comment 5 Hin-Tak Leung 2009-08-17 17:09:48 UTC
(In reply to comment #4)
> Hin-Tak Leung: You didn't mark it is a duplicate - because you are not sure, or
> because you don't have the rights to do it? Should I do it or should I wait for
> another expert?

I don't have the redhat bugzilla right to do it. There is no other expert :-(...
 
> If it is such an old known issue which haven't been fixed properly: It would be
> very easy to include an ugly workaround for Fedora. Perhaps by just putting the
> module(s) in SUSPEND_MODULES by default. I propose doing that for F12 if it
> haven't been resolved properly.

It is a known issue - but not too _old_ - almost *none* of the wireless drivers have any suspend/resume code! It is just more noticeable with rtl8187 because it is used as internal usb. Internal PCI/PCI express are unloaded/reset by the PCI core(?) and most people physically detach external USB device on suspend.

I am one of the current maintainers of the rtl8187 kernel code, so *I* (or the other two or John) am supposed to fix this... and indeed there is an incomplete patch bouncing around.

Comment 6 Till Maas 2009-08-17 18:44:52 UTC
closed as duplicate as requested in comment:5.

*** This bug has been marked as a duplicate of bug 469461 ***

Comment 7 Mads Kiilerich 2009-08-19 22:13:28 UTC
Fine. I hope The Perfect Fix will be there soon. Thanks for working on it.

But in the meantime, please consider introducing the SUSPEND_MODULES workaround in Fedora while the modules are _known_ to not be ready yet.


Note You need to log in before you can comment on or make changes to this bug.