This is in 3.7.5, plus one ath9k patch to over-ride eeprom regdomain.
Code that causes this is here...I seem to have a good test case for
this currently, so if there is any other debugging I could put here
that would be useful, let me know.
if (WARN_ON(!tid_tx)) {
ht_dbg(sdata, "addBA was not requested!\n");
goto unlock;
}
[ 104.499464] ------------[ cut here ]------------
[ 104.509486] WARNING: at /home/greearb/git/linux-3.7.dev.y/net/mac80211/agg-tx.c:625 ieee80211_start_tx_ba_cb+0xcb/0xd0 [mac80211]()
[ 104.534083] Hardware name: To Be Filled By O.E.M.
[ 104.542979] Modules linked in: xt_CT[ 104.543121] Pid: 44, comm: kworker/u:1 Tainted: G C 3.7.5+ #35
[ 104.543124] Call Trace:
[ 104.543132] [<c04440dd>] warn_slowpath_common+0x6d/0xa0
[ 104.543148] [<f8baeaab>] ? ieee80211_start_tx_ba_cb+0xcb/0xd0 [mac80211]
[ 104.543179] [<f8baeaab>] ? ieee80211_start_tx_ba_cb+0xcb/0xd0 [mac80211]
[ 104.543205] [<c044412d>] warn_slowpath_null+0x1d/0x20
[ 104.543216] [<f8baeaab>] ieee80211_start_tx_ba_cb+0xcb/0xd0 [mac80211]
[ 104.543243] [<f8bb221f>] ieee80211_iface_work+0x1ff/0x300 [mac80211]
[ 104.543270] [<c0459327>] process_one_work+0x127/0x2f0
[ 104.543280] [<f8bb2020>] ? ieee80211_netdev_select_queue+0x10/0x10 [mac80211]
[ 104.543306] [<c045bea2>] worker_thread+0x102/0x380
[ 104.543314] [<c045fedc>] kthread+0x9c/0xa0
[ 104.543321] [<c045bda0>] ? manage_workers+0x2b0/0x2b0
[ 104.543328] [<c0460000>] ? kthread_stop+0x10/0x70
[ 104.543336] [<c0917177>] ret_from_kernel_thread+0x1b/0x28
[ 104.543345] [<c045fe40>] ? kthread_freezable_should_stop+0x50/0x50
[ 104.543357] ---[ end trace 84f17e8f8b9c9dda ]---
Thanks,
Ben
--
Ben Greear <[email protected]>
Candela Technologies Inc http://www.candelatech.com
On 01/29/2013 01:46 PM, Johannes Berg wrote:
> On Tue, 2013-01-29 at 13:44 -0800, Ben Greear wrote:
>> This is in 3.7.5, plus one ath9k patch to over-ride eeprom regdomain.
>>
>> Code that causes this is here...I seem to have a good test case for
>> this currently, so if there is any other debugging I could put here
>> that would be useful, let me know.
>>
>> if (WARN_ON(!tid_tx)) {
>> ht_dbg(sdata, "addBA was not requested!\n");
>> goto unlock;
>> }
>
> Heh, I think Kalle just had the same. Do you have the ht_dbg()
> statements (i.e. HT debugging) enabled?
No, it seems I do not.
Want me to enable it?
Thanks,
Ben
--
Ben Greear <[email protected]>
Candela Technologies Inc http://www.candelatech.com
On Tue, 2013-01-29 at 13:44 -0800, Ben Greear wrote:
> This is in 3.7.5, plus one ath9k patch to over-ride eeprom regdomain.
>
> Code that causes this is here...I seem to have a good test case for
> this currently, so if there is any other debugging I could put here
> that would be useful, let me know.
>
> if (WARN_ON(!tid_tx)) {
> ht_dbg(sdata, "addBA was not requested!\n");
> goto unlock;
> }
Heh, I think Kalle just had the same. Do you have the ht_dbg()
statements (i.e. HT debugging) enabled?
johannes
On Tue, 2013-01-29 at 13:50 -0800, Ben Greear wrote:
> On 01/29/2013 01:46 PM, Johannes Berg wrote:
> > On Tue, 2013-01-29 at 13:44 -0800, Ben Greear wrote:
> >> This is in 3.7.5, plus one ath9k patch to over-ride eeprom regdomain.
> >>
> >> Code that causes this is here...I seem to have a good test case for
> >> this currently, so if there is any other debugging I could put here
> >> that would be useful, let me know.
> >>
> >> if (WARN_ON(!tid_tx)) {
> >> ht_dbg(sdata, "addBA was not requested!\n");
> >> goto unlock;
> >> }
> >
> > Heh, I think Kalle just had the same. Do you have the ht_dbg()
> > statements (i.e. HT debugging) enabled?
>
> No, it seems I do not.
>
> Want me to enable it?
Please. Actually, if you can (and it applies) this patch would be
helpful too:
http://git.kernel.org/?p=linux/kernel/git/jberg/mac80211-next.git;a=commit;h=0a214d3f7e5041685ebff054628a4354bedd78c3
johannes
On Tue, 2013-01-29 at 16:55 -0800, Ben Greear wrote:
> On 01/29/2013 01:55 PM, Johannes Berg wrote:
> > On Tue, 2013-01-29 at 13:50 -0800, Ben Greear wrote:
> >> On 01/29/2013 01:46 PM, Johannes Berg wrote:
> >>> On Tue, 2013-01-29 at 13:44 -0800, Ben Greear wrote:
> >>>> This is in 3.7.5, plus one ath9k patch to over-ride eeprom regdomain.
> >>>>
> >>>> Code that causes this is here...I seem to have a good test case for
> >>>> this currently, so if there is any other debugging I could put here
> >>>> that would be useful, let me know.
> >>>>
> >>>> if (WARN_ON(!tid_tx)) {
> >>>> ht_dbg(sdata, "addBA was not requested!\n");
> >>>> goto unlock;
> >>>> }
> >>>
> >>> Heh, I think Kalle just had the same. Do you have the ht_dbg()
> >>> statements (i.e. HT debugging) enabled?
> >>
> >> No, it seems I do not.
> >>
> >> Want me to enable it?
> >
> > Please. Actually, if you can (and it applies) this patch would be
> > helpful too:
> > http://git.kernel.org/?p=linux/kernel/git/jberg/mac80211-next.git;a=commit;h=0a214d3f7e5041685ebff054628a4354bedd78c3
>
> I added that, and the patch that allows two VAPs to register for
> beacons so I can run both VAPs I have configured on this system...
>
> I can grab more history if it helps...
>
>
> [ 142.852997] vap0: delba from 00:0e:8e:32:12:cf (initiator) tid 0 reason code 37
> [ 142.853110] vap0: switched off addBA timer for 00:0e:8e:32:12:cf tid 0
> [ 142.853122] vap0: got addBA resp for 00:0e:8e:32:12:cf tid 0 but we already gave up
> [ 142.853168] vap0: switched off addBA timer for 00:11:22:99:88:00 tid 0
> [ 142.853192] vap0: got addBA resp for 00:11:22:99:88:00 tid 0 but we already gave up
> [ 142.853219] vap0: Stopping Tx BA session for 00:0e:8e:32:12:cf tid 0
> [ 142.853513] vap0: Stopping Tx BA session for 00:11:22:99:88:00 tid 0
> [ 142.853769] vap0: Open BA session requested for 00:0e:8e:32:12:cf tid 0
> [ 142.853961] vap0: Open BA session requested for 00:0e:8e:32:12:cf tid 0
> [ 142.853976] vap0: BA request denied - session is not idle on 00:0e:8e:32:12:cf tid 0
> [ 142.854166] vap0: Rx A-MPDU request on 00:0e:8e:32:12:cf tid 0 result 0
> [ 142.854235] vap0: Open BA session requested for 00:0e:8e:32:12:cf tid 0
> [ 142.854251] vap0: BA request denied - session is not idle on 00:0e:8e:32:12:cf tid 0
> [ 142.854344] vap0: delba from 00:11:22:99:88:00 (initiator) tid 0 reason code 37
> [ 142.854436] vap0: Rx A-MPDU request on 00:11:22:99:88:00 tid 0 result 0
> [ 142.854556] ------------[ cut here ]------------
> [ 142.863992] vap0: Open BA session requested for 00:11:22:99:88:00 tid 0
> [ 142.864166] vap0: Open BA session requested for 00:11:22:99:88:00 tid 0
> [ 142.864181] vap0: BA request denied - session is not idle on 00:11:22:99:88:00 tid 0
> [ 142.864218] vap0: Open BA session requested for 00:11:22:99:88:00 tid 0
> [ 142.864232] vap0: BA request denied - session is not idle on 00:11:22:99:88:00 tid 0
> [ 142.864276] vap0: Open BA session requested for 00:11:22:99:88:00 tid 0
> [ 142.864297] vap0: BA request denied - session is not idle on 00:11:22:99:88:00 tid 0
> [ 142.864332] vap0: Open BA session requested for 00:11:22:99:88:00 tid 0
> [ 142.864346] vap0: BA request denied - session is not idle on 00:11:22:99:88:00 tid 0
> [ 142.866319] WARNING: at /home/greearb/git/linux-3.7.dev.y/net/mac80211/agg-tx.c:631 ieee80211_start_tx_ba_cb+0x1d2/0x220 [mac80211]()
> [ 142.891075] Hardware name: To Be Filled By O.E.M.
> [ 142.901277] Modules linked in: iptable_raw xt_CT nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack bridge veth ip_gre gre 8021q garp stp llc
> macvlan pktgen fuse nfsv3 nfs_acl nfsv4 auth_rpcgss nfs fscache lockd sunrpc binfmt_misc uinput arc4 ath9k coretemp snd_hda_codec_realtek iTCO_wdt mac80211
> iTCO_vendor_support gpio_ich snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device ath9k_common ath9k_hw snd_pcm ath lpc_ich microcode pcspkr i2c_i801
> serio_raw cfg80211 snd_timer snd r8169 soundcore snd_page_alloc mii rfkill i915 drm_kms_helper drm i2c_algo_bit i2c_core video [last unloaded: iptable_nat]
> [ 142.938975] Pid: 6, comm: kworker/u:0 Tainted: G C 3.7.5+ #36
> [ 142.938980] Call Trace:
> [ 142.939062] [<c04451bd>] warn_slowpath_common+0x6d/0xa0
> [ 142.939100] [<f8e933b2>] ? ieee80211_start_tx_ba_cb+0x1d2/0x220 [mac80211]
> [ 142.939160] [<f8e933b2>] ? ieee80211_start_tx_ba_cb+0x1d2/0x220 [mac80211]
> [ 142.939202] [<c044520d>] warn_slowpath_null+0x1d/0x20
> [ 142.939232] [<f8e933b2>] ieee80211_start_tx_ba_cb+0x1d2/0x220 [mac80211]
> [ 142.939273] [<c049d64b>] ? trace_hardirqs_on+0xb/0x10
> [ 142.939287] [<c09746df>] ? _raw_spin_unlock_irqrestore+0x3f/0x80
> [ 142.939321] [<f8e977bf>] ieee80211_iface_work+0x20f/0x310 [mac80211]
> [ 142.939364] [<c045de76>] process_one_work+0x1a6/0x3d0
> [ 142.939377] [<c045de09>] ? process_one_work+0x139/0x3d0
> [ 142.939389] [<c0460e58>] ? worker_thread+0x198/0x380
> [ 142.939417] [<f8e975b0>] ? ieee80211_netdev_select_queue+0x10/0x10 [mac80211]
> [ 142.939494] [<c0460dc2>] worker_thread+0x102/0x380
> [ 142.939506] [<c0973362>] ? preempt_schedule+0x32/0x50
> [ 142.939520] [<c046596c>] kthread+0xac/0xb0
> [ 142.939532] [<c0460cc0>] ? manage_workers+0x2b0/0x2b0
> [ 142.939543] [<c0974742>] ? _raw_spin_unlock_irq+0x22/0x50
> [ 142.939559] [<c097a0b7>] ret_from_kernel_thread+0x1b/0x28
> [ 142.939571] [<c04658c0>] ? __init_kthread_worker+0x60/0x60
> [ 142.939587] ---[ end trace 0f04f0a807663584 ]---
> [ 142.939594] vap0: addBA was not requested!
Darn, it looks like I forgot to add the MAC address/TID to this debug
statement. Could you do that?
johannes
On 01/29/2013 01:55 PM, Johannes Berg wrote:
> On Tue, 2013-01-29 at 13:50 -0800, Ben Greear wrote:
>> On 01/29/2013 01:46 PM, Johannes Berg wrote:
>>> On Tue, 2013-01-29 at 13:44 -0800, Ben Greear wrote:
>>>> This is in 3.7.5, plus one ath9k patch to over-ride eeprom regdomain.
>>>>
>>>> Code that causes this is here...I seem to have a good test case for
>>>> this currently, so if there is any other debugging I could put here
>>>> that would be useful, let me know.
>>>>
>>>> if (WARN_ON(!tid_tx)) {
>>>> ht_dbg(sdata, "addBA was not requested!\n");
>>>> goto unlock;
>>>> }
>>>
>>> Heh, I think Kalle just had the same. Do you have the ht_dbg()
>>> statements (i.e. HT debugging) enabled?
>>
>> No, it seems I do not.
>>
>> Want me to enable it?
>
> Please. Actually, if you can (and it applies) this patch would be
> helpful too:
> http://git.kernel.org/?p=linux/kernel/git/jberg/mac80211-next.git;a=commit;h=0a214d3f7e5041685ebff054628a4354bedd78c3
I added that, and the patch that allows two VAPs to register for
beacons so I can run both VAPs I have configured on this system...
I can grab more history if it helps...
[ 142.852997] vap0: delba from 00:0e:8e:32:12:cf (initiator) tid 0 reason code 37
[ 142.853110] vap0: switched off addBA timer for 00:0e:8e:32:12:cf tid 0
[ 142.853122] vap0: got addBA resp for 00:0e:8e:32:12:cf tid 0 but we already gave up
[ 142.853168] vap0: switched off addBA timer for 00:11:22:99:88:00 tid 0
[ 142.853192] vap0: got addBA resp for 00:11:22:99:88:00 tid 0 but we already gave up
[ 142.853219] vap0: Stopping Tx BA session for 00:0e:8e:32:12:cf tid 0
[ 142.853513] vap0: Stopping Tx BA session for 00:11:22:99:88:00 tid 0
[ 142.853769] vap0: Open BA session requested for 00:0e:8e:32:12:cf tid 0
[ 142.853961] vap0: Open BA session requested for 00:0e:8e:32:12:cf tid 0
[ 142.853976] vap0: BA request denied - session is not idle on 00:0e:8e:32:12:cf tid 0
[ 142.854166] vap0: Rx A-MPDU request on 00:0e:8e:32:12:cf tid 0 result 0
[ 142.854235] vap0: Open BA session requested for 00:0e:8e:32:12:cf tid 0
[ 142.854251] vap0: BA request denied - session is not idle on 00:0e:8e:32:12:cf tid 0
[ 142.854344] vap0: delba from 00:11:22:99:88:00 (initiator) tid 0 reason code 37
[ 142.854436] vap0: Rx A-MPDU request on 00:11:22:99:88:00 tid 0 result 0
[ 142.854556] ------------[ cut here ]------------
[ 142.863992] vap0: Open BA session requested for 00:11:22:99:88:00 tid 0
[ 142.864166] vap0: Open BA session requested for 00:11:22:99:88:00 tid 0
[ 142.864181] vap0: BA request denied - session is not idle on 00:11:22:99:88:00 tid 0
[ 142.864218] vap0: Open BA session requested for 00:11:22:99:88:00 tid 0
[ 142.864232] vap0: BA request denied - session is not idle on 00:11:22:99:88:00 tid 0
[ 142.864276] vap0: Open BA session requested for 00:11:22:99:88:00 tid 0
[ 142.864297] vap0: BA request denied - session is not idle on 00:11:22:99:88:00 tid 0
[ 142.864332] vap0: Open BA session requested for 00:11:22:99:88:00 tid 0
[ 142.864346] vap0: BA request denied - session is not idle on 00:11:22:99:88:00 tid 0
[ 142.866319] WARNING: at /home/greearb/git/linux-3.7.dev.y/net/mac80211/agg-tx.c:631 ieee80211_start_tx_ba_cb+0x1d2/0x220 [mac80211]()
[ 142.891075] Hardware name: To Be Filled By O.E.M.
[ 142.901277] Modules linked in: iptable_raw xt_CT nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack bridge veth ip_gre gre 8021q garp stp llc
macvlan pktgen fuse nfsv3 nfs_acl nfsv4 auth_rpcgss nfs fscache lockd sunrpc binfmt_misc uinput arc4 ath9k coretemp snd_hda_codec_realtek iTCO_wdt mac80211
iTCO_vendor_support gpio_ich snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device ath9k_common ath9k_hw snd_pcm ath lpc_ich microcode pcspkr i2c_i801
serio_raw cfg80211 snd_timer snd r8169 soundcore snd_page_alloc mii rfkill i915 drm_kms_helper drm i2c_algo_bit i2c_core video [last unloaded: iptable_nat]
[ 142.938975] Pid: 6, comm: kworker/u:0 Tainted: G C 3.7.5+ #36
[ 142.938980] Call Trace:
[ 142.939062] [<c04451bd>] warn_slowpath_common+0x6d/0xa0
[ 142.939100] [<f8e933b2>] ? ieee80211_start_tx_ba_cb+0x1d2/0x220 [mac80211]
[ 142.939160] [<f8e933b2>] ? ieee80211_start_tx_ba_cb+0x1d2/0x220 [mac80211]
[ 142.939202] [<c044520d>] warn_slowpath_null+0x1d/0x20
[ 142.939232] [<f8e933b2>] ieee80211_start_tx_ba_cb+0x1d2/0x220 [mac80211]
[ 142.939273] [<c049d64b>] ? trace_hardirqs_on+0xb/0x10
[ 142.939287] [<c09746df>] ? _raw_spin_unlock_irqrestore+0x3f/0x80
[ 142.939321] [<f8e977bf>] ieee80211_iface_work+0x20f/0x310 [mac80211]
[ 142.939364] [<c045de76>] process_one_work+0x1a6/0x3d0
[ 142.939377] [<c045de09>] ? process_one_work+0x139/0x3d0
[ 142.939389] [<c0460e58>] ? worker_thread+0x198/0x380
[ 142.939417] [<f8e975b0>] ? ieee80211_netdev_select_queue+0x10/0x10 [mac80211]
[ 142.939494] [<c0460dc2>] worker_thread+0x102/0x380
[ 142.939506] [<c0973362>] ? preempt_schedule+0x32/0x50
[ 142.939520] [<c046596c>] kthread+0xac/0xb0
[ 142.939532] [<c0460cc0>] ? manage_workers+0x2b0/0x2b0
[ 142.939543] [<c0974742>] ? _raw_spin_unlock_irq+0x22/0x50
[ 142.939559] [<c097a0b7>] ret_from_kernel_thread+0x1b/0x28
[ 142.939571] [<c04658c0>] ? __init_kthread_worker+0x60/0x60
[ 142.939587] ---[ end trace 0f04f0a807663584 ]---
[ 142.939594] vap0: addBA was not requested!
[ 142.939634] vap0: Stopping Tx BA session for 00:11:22:99:88:00 tid 0
[ 142.939648] vap0: unexpected callback to A-MPDU stop for 00:11:22:99:88:00 tid 0
[ 142.939671] vap0: delba from 00:11:22:99:88:00 (initiator) tid 0 reason code 37
[ 142.939682] vap0: Rx BA session stop requested for 00:11:22:99:88:00 tid 0 inititator reason: 0
[ 142.939772] vap0: Rx A-MPDU request on 00:11:22:99:88:00 tid 0 result 0
[ 142.939850] vap0: delba from 00:11:22:99:88:00 (initiator) tid 0 reason code 37
[ 142.939860] vap0: Rx BA session stop requested for 00:11:22:99:88:00 tid 0 inititator reason: 0
[ 142.939895] vap0: unexpected AddBA Req from 00:0e:8e:32:12:cf on tid 0
[ 142.939903] vap0: Rx BA session stop requested for 00:0e:8e:32:12:cf tid 0 recipient reason: 32
[ 142.940130] vap0: Rx A-MPDU request on 00:0e:8e:32:12:cf tid 0 result 0
[ 142.940240] vap0: Rx A-MPDU request on 00:11:22:99:88:00 tid 0 result 0
[ 142.940290] vap0: delba from 00:0e:8e:32:12:cf (initiator) tid 0 reason code 37
[ 142.940300] vap0: Rx BA session stop requested for 00:0e:8e:32:12:cf tid 0 inititator reason: 0
[ 142.940444] vap0: Rx A-MPDU request on 00:0e:8e:32:12:cf tid 0 result 0
>
> johannes
>
--
Ben Greear <[email protected]>
Candela Technologies Inc http://www.candelatech.com