2.6.36-rc4/iwl: warning triggered in net/mac80211/scan.c:266

2.6.36-rc4/iwl: warning triggered in net/mac80211/scan.c:266

[Michael S. Tsirkin]

I am running 2.6.36-rc4 on T60 with an iwlagn wireless card:
03:00.0 Network controller: Intel Corporation PRO/Wireless 5100 AGN [Shiloh] Network Connection

I was compiling a large file which stressed the box heavily, and the
following warning in net/mac80211/scan.c triggered:

        /*
         * It's ok to abort a not-yet-running scan (that
         * we have one at all will be verified by checking
         * local->scan_req next), but not to complete it
         * successfully.
         */
        if (WARN_ON(!local->scanning && !aborted))
                aborted = true;

dmesg below:

[84379.361297] ------------[ cut here ]------------
[84379.361307] WARNING: at net/mac80211/scan.c:266 ieee80211_scan_completed+0x43/0x17f()
[84379.361311] Hardware name: 2241B48
[84379.361313] Modules linked in: tun iwlagn kvm_intel kvm iwlcore e1000e thinkpad_acpi [last unloaded: scsi_wait_scan]
[84379.361325] Pid: 8254, comm: kworker/u:0 Not tainted 2.6.36-rc4-mst #96
[84379.361328] Call Trace:
[84379.361334]  [<ffffffff815d1399>] ? ieee80211_scan_completed+0x43/0x17f
[84379.361340]  [<ffffffff81042046>] warn_slowpath_common+0x80/0x99
[84379.361354]  [<ffffffffa005d8e7>] ? iwl_generic_cmd_callback+0x0/0xe4 [iwlcore]
[84379.361359]  [<ffffffff81042074>] warn_slowpath_null+0x15/0x17
[84379.361363]  [<ffffffff815d1399>] ieee80211_scan_completed+0x43/0x17f
[84379.361374]  [<ffffffffa0065212>] iwl_bg_scan_completed+0x1ce/0x1d7 [iwlcore]
[84379.361380]  [<ffffffff81057153>] process_one_work+0x1e2/0x301
[84379.361391]  [<ffffffffa0065044>] ? iwl_bg_scan_completed+0x0/0x1d7 [iwlcore]
[84379.361395]  [<ffffffff8105752c>] worker_thread+0x130/0x243
[84379.361400]  [<ffffffff810573fc>] ? worker_thread+0x0/0x243
[84379.361404]  [<ffffffff8105bc1b>] kthread+0x81/0x89
[84379.361409]  [<ffffffff81003914>] kernel_thread_helper+0x4/0x10
[84379.361413]  [<ffffffff8105bb9a>] ? kthread+0x0/0x89
[84379.361417]  [<ffffffff81003910>] ? kernel_thread_helper+0x0/0x10
[84379.361420] ---[ end trace c5ebbd902daa4ea8 ]---
[84379.361422] ------------[ cut here ]------------
[84379.361426] WARNING: at net/mac80211/scan.c:269 ieee80211_scan_completed+0x62/0x17f()
[84379.361429] Hardware name: 2241B48
[84379.361431] Modules linked in: tun iwlagn kvm_intel kvm iwlcore e1000e thinkpad_acpi [last unloaded: scsi_wait_scan]
[84379.361441] Pid: 8254, comm: kworker/u:0 Tainted: G        W   2.6.36-rc4-mst #96
[84379.361461] Call Trace:
[84379.361464]  [<ffffffff815d13b8>] ? ieee80211_scan_completed+0x62/0x17f
[84379.361469]  [<ffffffff81042046>] warn_slowpath_common+0x80/0x99
[84379.361479]  [<ffffffffa005d8e7>] ? iwl_generic_cmd_callback+0x0/0xe4 [iwlcore]
[84379.361484]  [<ffffffff81042074>] warn_slowpath_null+0x15/0x17
[84379.361488]  [<ffffffff815d13b8>] ieee80211_scan_completed+0x62/0x17f
[84379.361503]  [<ffffffffa0065212>] iwl_bg_scan_completed+0x1ce/0x1d7 [iwlcore]
[84379.361506]  [<ffffffff81057153>] process_one_work+0x1e2/0x301
[84379.361512]  [<ffffffffa0065044>] ? iwl_bg_scan_completed+0x0/0x1d7 [iwlcore]
[84379.361515]  [<ffffffff8105752c>] worker_thread+0x130/0x243
[84379.361517]  [<ffffffff810573fc>] ? worker_thread+0x0/0x243
[84379.361519]  [<ffffffff8105bc1b>] kthread+0x81/0x89
[84379.361522]  [<ffffffff81003914>] kernel_thread_helper+0x4/0x10
[84379.361524]  [<ffffffff8105bb9a>] ? kthread+0x0/0x89
[84379.361526]  [<ffffffff81003910>] ? kernel_thread_helper+0x0/0x10
[84379.361528] ---[ end trace c5ebbd902daa4ea9 ]---

Should this happen? Does anyone care?
I am not sure how reproducible this is.
Thanks,

-- 
MST

Re: 2.6.36-rc4/iwl: warning triggered in net/mac80211/scan.c:266

[Johannes Berg]

On Tue, 2010-09-14 at 16:28 +0200, Michael S. Tsirkin wrote:
> I am running 2.6.36-rc4 on T60 with an iwlagn wireless card:
> 03:00.0 Network controller: Intel Corporation PRO/Wireless 5100 AGN [Shiloh] Network Connection
> 
> I was compiling a large file which stressed the box heavily, and the
> following warning in net/mac80211/scan.c triggered:
> 
>         /*
>          * It's ok to abort a not-yet-running scan (that
>          * we have one at all will be verified by checking
>          * local->scan_req next), but not to complete it
>          * successfully.
>          */
>         if (WARN_ON(!local->scanning && !aborted))
>                 aborted = true;
> 
> dmesg below:
> 
> [84379.361297] ------------[ cut here ]------------
> [84379.361307] WARNING: at net/mac80211/scan.c:266 ieee80211_scan_completed+0x43/0x17f()

> Should this happen? Does anyone care?
> I am not sure how reproducible this is.

No, it shouldn't happen. But there are various races in iwlwifi that can
cause this. Stanislaw has fixed them recently, but it's a fairly large
patchset. I'm not sure, but I don't think this is really new in .36.

johannes

Re: 2.6.36-rc4/iwl: warning triggered in net/mac80211/scan.c:266

[Michael S. Tsirkin]

On Tue, Sep 14, 2010 at 04:36:28PM +0200, Johannes Berg wrote:
> On Tue, 2010-09-14 at 16:28 +0200, Michael S. Tsirkin wrote:
> > I am running 2.6.36-rc4 on T60 with an iwlagn wireless card:
> > 03:00.0 Network controller: Intel Corporation PRO/Wireless 5100 AGN [Shiloh] Network Connection
> > 
> > I was compiling a large file which stressed the box heavily, and the
> > following warning in net/mac80211/scan.c triggered:
> > 
> >         /*
> >          * It's ok to abort a not-yet-running scan (that
> >          * we have one at all will be verified by checking
> >          * local->scan_req next), but not to complete it
> >          * successfully.
> >          */
> >         if (WARN_ON(!local->scanning && !aborted))
> >                 aborted = true;
> > 
> > dmesg below:
> > 
> > [84379.361297] ------------[ cut here ]------------
> > [84379.361307] WARNING: at net/mac80211/scan.c:266 ieee80211_scan_completed+0x43/0x17f()
> 
> > Should this happen? Does anyone care?
> > I am not sure how reproducible this is.
> 
> No, it shouldn't happen. But there are various races in iwlwifi that can
> cause this. Stanislaw has fixed them recently, but it's a fairly large
> patchset. I'm not sure, but I don't think this is really new in .36.
> 
> johannes

So ... get back to you after 2.6.37?

-- 
MST

Re: 2.6.36-rc4/iwl: warning triggered in net/mac80211/scan.c:266

[Johannes Berg]

On Tue, 2010-09-14 at 17:14 +0200, Michael S. Tsirkin wrote:

> > No, it shouldn't happen. But there are various races in iwlwifi that can
> > cause this. Stanislaw has fixed them recently, but it's a fairly large
> > patchset. I'm not sure, but I don't think this is really new in .36.

> So ... get back to you after 2.6.37?

Err, yes, I guess so, though I suppose it doesn't really solve the
problem. This particular instance should be harmless though, I believe,
other than the warning.

johannes

Re: 2.6.36-rc4/iwl: warning triggered in net/mac80211/scan.c:266

[Maciej Rutecki]

On wtorek, 14 września 2010 o 16:28:15 Michael S. Tsirkin wrote:
> I am running 2.6.36-rc4 on T60 with an iwlagn wireless card:
> 03:00.0 Network controller: Intel Corporation PRO/Wireless 5100 AGN
> [Shiloh] Network Connection
> 
> I was compiling a large file which stressed the box heavily, and the
> following warning in net/mac80211/scan.c triggered:

I created a Bugzilla entry at 
https://bugzilla.kernel.org/show_bug.cgi?id=18692
for your bug report, please add your address to the CC list in there, thanks!

-- 
Maciej Rutecki
http://www.maciek.unixy.pl