Below is what I saw today(compat-wireless 2009-05-21), I have both an ath9k and an ath5k card. ath9k crashes (see below), ath5k will just reboot the board after "ifconfig mesh up", any idea?
thanks,
X Xiao
BusyBox v1.13.4 (2009-06-01 21:18:32 CDT) built-in shell (ash)
Enter 'help' for a list of built-in commands.
_______ ________ __
| |.-----.-----.-----.| | | |.----.| |_
| - || _ | -__| || | | || _|| _|
|_______|| __|_____|__|__||________||__| |____|
|__| W I R E L E S S F R E E D O M
KAMIKAZE (bleeding edge, r16290) -------------------
* 10 oz Vodka Shake well with ice and strain
* 10 oz Triple sec mixture into 10 shot glasses.
* 10 oz lime juice Salute!
---------------------------------------------------
root@OpenWrt:/# iw dev wlan0 interface add mesh type mp mesh_id mesh
root@OpenWrt:/# ifconfig -a | grep mesh
mesh Link encap:Ethernet HWaddr 00:0B:6B:57:09:37
root@OpenWrt:/# ifconfig mesh up
PCI error 1 at PCI addr 0x100109dc
Data bus error, epc == c024f12c, ra == c024f128
Oops[#1]:
Cpu 0
$ 0 : 00000000 00000000 deadc0de 000ffc00
$ 4 : b00109dc b0010000 b0010000 0000000a
$ 8 : 80170000 83aeaed0 00080000 00000000
$12 : 00000000 839f04a8 802a0818 00000000
$16 : 000009dc 8016b710 83aea000 000003ff
$20 : 83aeaea8 8016b6bc 0000001c 00000007
$24 : 00000000 80070000
$28 : 82c5c000 82c5dc90 7fb0d068 c024f128
Hi : 000000d8
Lo : 00000144
epc : c024f12c 0xc024f12c
Not tainted
ra : c024f128 0xc024f128
Status: 1000f403 KERNEL EXL IE
Cause : 1080001c
PrId : 00019374 (MIPS 24Kc)
Modules linked in: nf_nat_tftp nf_conntrack_tftp nf_nat_irc nf_conntrack_irc nf_nat_ftp nf_conntrack_ftp i
Process ifconfig (pid: 896, threadinfo=82c5c000, task=839f0000, tls=00000000)
Stack : 83aea000 83a9d680 00000000 00000018 83aea000 83b109a0 83aea000 83aea000
00000300 83b101e0 82cddeb8 83b109a8 c025977c c0259700 ffffffff 00000000
00000002 00000000 00000003 00000000 00080000 00000000 00000000 00000000
00000000 00000000 00000000 00000000 00000001 83b109a0 82cddec0 c02599e0
00000000 c025a164 00000001 8009525c 82cddb80 00000300 83b101e0 82cdd800
...
Call Trace:[<c025977c>] 0xc025977c
[<c0259700>] 0xc0259700
[<c02599e0>] 0xc02599e0
[<c025a164>] 0xc025a164
[<8009525c>] 0x8009525c
[<c01dcb64>] 0xc01dcb64
[<c01ff3b8>] 0xc01ff3b8
[<800dc19c>] 0x800dc19c
[<c01eb4b0>] 0xc01eb4b0
[<800b4040>] 0x800b4040
[<801b85e8>] 0x801b85e8
[<801b4454>] 0x801b4454
[<800bdc94>] 0x800bdc94
[<800bdc94>] 0x800bdc94
[<801b632c>] 0x801b632c
[<801b6b80>] 0x801b6b80
[<801ffbf8>] 0x801ffbf8
[<801ff9c8>] 0x801ff9c8
[<801b75f8>] 0x801b75f8
[<801b75ec>] 0x801b75ec
[<801a75b8>] 0x801a75b8
[<800e0890>] 0x800e0890
[<800e0db0>] 0x800e0db0
[<801a7e30>] 0x801a7e30
[<800d5344>] 0x800d5344
[<800d51c0>] 0x800d51c0
[<801a7ebc>] 0x801a7ebc
[<800e0e64>] 0x800e0e64
[<80062504>] 0x80062504
[<80062504>] 0x80062504
Code: 02a0f809 02042021 8e450008 <34440800> 0220f809 02052821 8e440008 26c21100 3053fffc
Segmentation fault
On Wed, Jun 3, 2009 at 12:59 PM, Nick Kossifidis <[email protected]> wrote:
> 2009/6/3 Andrey Yurovsky <[email protected]>:
>> On Wed, Jun 3, 2009 at 11:21 AM, Nick Kossifidis <[email protected]> wrote:
>>> 2009/6/2 Andrey Yurovsky <[email protected]>:
>>>> With wireless-testing from a couple of days ago, an X86 system, and this patch:
>>>> http://osdir.com/ml/linux-wireless/2009-05/msg01241.html
>>>> ath9k can bring up a mesh point interface and doesn't crash for me,
>>>> but there are no mesh beacons, so mesh isn't working. ?On ath5k, there
>>>> are also no beacons and mesh doesn't work, but there's no crash.
>>>>
>>>
>>>
>>> Hmm, try adding NL80211_IFTYPE_MESH_POINT here ->
>>>
>>> http://git.kernel.org/?p=linux/kernel/git/linville/wireless-testing.git;a=blob;f=drivers/net/wireless/ath/ath5k/pcu.c;h=ec35503f6a40e6b6a1cc4323287b31e7cdd3a428;hb=HEAD#l733
>>
>> It's already there, in ath5k_hw_set_opmode. ?We'll need to trace
>> what's going on and see why we don't get beacons. ?Last night I had an
>> ath5k MP suddenly start beaconing but I haven't been able to reproduce
>> that yet.
>
> 736 ? ? ? ? /* When in AP mode zero timer0 to start TSF */
> 737 ? ? ? ? if (ah->ah_op_mode == NL80211_IFTYPE_AP)
> 738 ? ? ? ? ? ? ? ? ath5k_hw_reg_write(ah, 0, AR5K_TIMER0);
>
> We need to do this for NL80211_IFTYPE_MESH_POINT or DBA won't fire up
> and beacon queue wont start...
Thanks Nik, that gets beaconing going but only after a issuing a scan
(we have the same problem on rt2x00).
> --
> GPG ID: 0xD21DB2DB
> As you read this post global entropy rises. Have Fun ;-)
> Nick
>
I was using compat-wireless-0521. the reboot of ath9k/80211s could be a soft-watchdog reset caused by the lockup you mentioned. thanks. going to try a new wireless release.
ath5k does not look good to me, going back to madwifi for a typical b/g AP setting instead.
80211s looks neat to me, though it's still not well tested in the field. tried olsrd with ath9k it seems working fine.
xxiao
--- On Tue, 6/2/09, Andrey Yurovsky <[email protected]> wrote:
> From: Andrey Yurovsky <[email protected]>
> Subject: Re: ath9k/ath5k + 80211s crash
> To: [email protected], [email protected]
> Cc: [email protected], [email protected]
> Date: Tuesday, June 2, 2009, 11:58 AM
> With wireless-testing from a couple
> of days ago, an X86 system, and this patch:
> http://osdir.com/ml/linux-wireless/2009-05/msg01241.html
> ath9k can bring up a mesh point interface and doesn't crash
> for me,
> but there are no mesh beacons, so mesh isn't working.?
> On ath5k, there
> are also no beacons and mesh doesn't work, but there's no
> crash.
>
> On Mon, Jun 1, 2009 at 8:52 PM,? <[email protected]>
> wrote:
> >
> > Sorry, it should be "ath9k reboots, ath5k crashes, the
> dump log is for ath5k, while ath9k displayed nothing".
> >
> > --- On Mon, 6/1/09, [email protected]
> <[email protected]>
> wrote:
> >
> >> From: [email protected]
> <[email protected]>
> >> Subject: ath9k/ath5k + 80211s crash
> >> To: [email protected],
> [email protected],
> [email protected]
> >> Date: Monday, June 1, 2009, 10:43 PM
> >>
> >> Below is what I saw today(compat-wireless
> 2009-05-21), I
> >> have both an ath9k and an ath5k card. ath9k
> crashes (see
> >> below), ath5k will just reboot the board after
> "ifconfig
> >> mesh up", any idea?
> >> thanks,
> >> X Xiao
> >>
> >>
> >> BusyBox v1.13.4 (2009-06-01 21:18:32 CDT) built-in
> shell
> >> (ash)
> >> Enter 'help' for a list of built-in commands.
> >>
> >> ? _______
> >> ? ? ? ???________
> >> ? ? __
> >> ?|
> >> ???|.-----.-----.-----.|? |
> >> |? |.----.|? |_
> >> ?|???-???||? _
> >> |? -__|? ???||? |
> >> |? ||???_||???_|
> >>
> >> |_______||???__|_____|__|__||________||__|
> >> |____|
> >> ? ? ? ? ? |__| W I R E L E S
> >> S???F R E E D O M
> >> ?KAMIKAZE (bleeding edge, r16290)
> -------------------
> >> ? * 10 oz Vodka? ? ???Shake
> >> well with ice and strain
> >> ? * 10 oz Triple sec? mixture into 10 shot
> >> glasses.
> >> ? * 10 oz lime juice? Salute!
> >>
> ?---------------------------------------------------
> >> root@OpenWrt:/# iw dev wlan0 interface add mesh
> type mp
> >> mesh_id mesh
> >> root@OpenWrt:/# ifconfig -a | grep mesh
> >> mesh? ? ? Link encap:Ethernet? HWaddr
> >> 00:0B:6B:57:09:37
> >> root@OpenWrt:/# ifconfig mesh up
> >> PCI error 1 at PCI addr 0x100109dc
> >> Data bus error, epc == c024f12c, ra == c024f128
> >> Oops[#1]:
> >> Cpu 0
> >> $ 0???: 00000000 00000000 deadc0de 000ffc00
> >> $ 4???: b00109dc b0010000 b0010000 0000000a
> >> $ 8???: 80170000 83aeaed0 00080000 00000000
> >> $12???: 00000000 839f04a8 802a0818 00000000
> >> $16???: 000009dc 8016b710 83aea000 000003ff
> >> $20???: 83aeaea8 8016b6bc 0000001c 00000007
> >> $24???: 00000000 80070000
> >>
> >> $28???: 82c5c000 82c5dc90 7fb0d068 c024f128
> >> Hi? ? : 000000d8
> >> Lo? ? : 00000144
> >> epc???: c024f12c 0xc024f12c
> >> ? ? Not tainted
> >> ra? ? : c024f128 0xc024f128
> >> Status: 1000f403? ? KERNEL EXL IE
> >> Cause : 1080001c
> >> PrId? : 00019374 (MIPS 24Kc)
> >> Modules linked in: nf_nat_tftp nf_conntrack_tftp
> nf_nat_irc
> >> nf_conntrack_irc nf_nat_ftp nf_conntrack_ftp i
> >> Process ifconfig (pid: 896, threadinfo=82c5c000,
> >> task=839f0000, tls=00000000)
> >> Stack : 83aea000 83a9d680 00000000 00000018
> 83aea000
> >> 83b109a0 83aea000 83aea000
> >> ? ? ? ? 00000300 83b101e0 82cddeb8
> >> 83b109a8 c025977c c0259700 ffffffff 00000000
> >> ? ? ? ? 00000002 00000000 00000003
> >> 00000000 00080000 00000000 00000000 00000000
> >> ? ? ? ? 00000000 00000000 00000000
> >> 00000000 00000001 83b109a0 82cddec0 c02599e0
> >> ? ? ? ? 00000000 c025a164 00000001
> >> 8009525c 82cddb80 00000300 83b101e0 82cdd800
> >> ? ? ? ? ...
> >> Call Trace:[<c025977c>] 0xc025977c
> >> [<c0259700>] 0xc0259700
> >> [<c02599e0>] 0xc02599e0
> >> [<c025a164>] 0xc025a164
> >> [<8009525c>] 0x8009525c
> >> [<c01dcb64>] 0xc01dcb64
> >> [<c01ff3b8>] 0xc01ff3b8
> >> [<800dc19c>] 0x800dc19c
> >> [<c01eb4b0>] 0xc01eb4b0
> >> [<800b4040>] 0x800b4040
> >> [<801b85e8>] 0x801b85e8
> >> [<801b4454>] 0x801b4454
> >> [<800bdc94>] 0x800bdc94
> >> [<800bdc94>] 0x800bdc94
> >> [<801b632c>] 0x801b632c
> >> [<801b6b80>] 0x801b6b80
> >> [<801ffbf8>] 0x801ffbf8
> >> [<801ff9c8>] 0x801ff9c8
> >> [<801b75f8>] 0x801b75f8
> >> [<801b75ec>] 0x801b75ec
> >> [<801a75b8>] 0x801a75b8
> >> [<800e0890>] 0x800e0890
> >> [<800e0db0>] 0x800e0db0
> >> [<801a7e30>] 0x801a7e30
> >> [<800d5344>] 0x800d5344
> >> [<800d51c0>] 0x800d51c0
> >> [<801a7ebc>] 0x801a7ebc
> >> [<800e0e64>] 0x800e0e64
> >> [<80062504>] 0x80062504
> >> [<80062504>] 0x80062504
> >>
> >>
> >> Code: 02a0f809? 02042021? 8e450008
> >> <34440800> 0220f809? 02052821
> >> 8e440008? 26c21100? 3053fffc
> >> Segmentation fault
> >>
> >> _______________________________________________
> >> Devel mailing list
> >> [email protected]
> >> http://open80211s.com/mailman/listinfo/devel
> >>
> > _______________________________________________
> > Devel mailing list
> > [email protected]
> > http://open80211s.com/mailman/listinfo/devel
> >
>
2009/6/3 Andrey Yurovsky <[email protected]>:
> On Wed, Jun 3, 2009 at 11:21 AM, Nick Kossifidis <[email protected]> wrote:
>> 2009/6/2 Andrey Yurovsky <[email protected]>:
>>> With wireless-testing from a couple of days ago, an X86 system, and this patch:
>>> http://osdir.com/ml/linux-wireless/2009-05/msg01241.html
>>> ath9k can bring up a mesh point interface and doesn't crash for me,
>>> but there are no mesh beacons, so mesh isn't working. On ath5k, there
>>> are also no beacons and mesh doesn't work, but there's no crash.
>>>
>>
>>
>> Hmm, try adding NL80211_IFTYPE_MESH_POINT here ->
>>
>> http://git.kernel.org/?p=linux/kernel/git/linville/wireless-testing.git;a=blob;f=drivers/net/wireless/ath/ath5k/pcu.c;h=ec35503f6a40e6b6a1cc4323287b31e7cdd3a428;hb=HEAD#l733
>
> It's already there, in ath5k_hw_set_opmode. We'll need to trace
> what's going on and see why we don't get beacons. Last night I had an
> ath5k MP suddenly start beaconing but I haven't been able to reproduce
> that yet.
736 /* When in AP mode zero timer0 to start TSF */
737 if (ah->ah_op_mode == NL80211_IFTYPE_AP)
738 ath5k_hw_reg_write(ah, 0, AR5K_TIMER0);
We need to do this for NL80211_IFTYPE_MESH_POINT or DBA won't fire up
and beacon queue wont start...
--
GPG ID: 0xD21DB2DB
As you read this post global entropy rises. Have Fun ;-)
Nick
2009/6/4 Andrey Yurovsky <[email protected]>:
> On Wed, Jun 3, 2009 at 12:59 PM, Nick Kossifidis <[email protected]> wrote:
>> 2009/6/3 Andrey Yurovsky <[email protected]>:
>>> On Wed, Jun 3, 2009 at 11:21 AM, Nick Kossifidis <[email protected]> wrote:
>>>> 2009/6/2 Andrey Yurovsky <[email protected]>:
>>>>> With wireless-testing from a couple of days ago, an X86 system, and this patch:
>>>>> http://osdir.com/ml/linux-wireless/2009-05/msg01241.html
>>>>> ath9k can bring up a mesh point interface and doesn't crash for me,
>>>>> but there are no mesh beacons, so mesh isn't working. On ath5k, there
>>>>> are also no beacons and mesh doesn't work, but there's no crash.
>>>>>
>>>>
>>>>
>>>> Hmm, try adding NL80211_IFTYPE_MESH_POINT here ->
>>>>
>>>> http://git.kernel.org/?p=linux/kernel/git/linville/wireless-testing.git;a=blob;f=drivers/net/wireless/ath/ath5k/pcu.c;h=ec35503f6a40e6b6a1cc4323287b31e7cdd3a428;hb=HEAD#l733
>>>
>>> It's already there, in ath5k_hw_set_opmode. We'll need to trace
>>> what's going on and see why we don't get beacons. Last night I had an
>>> ath5k MP suddenly start beaconing but I haven't been able to reproduce
>>> that yet.
>>
>> 736 /* When in AP mode zero timer0 to start TSF */
>> 737 if (ah->ah_op_mode == NL80211_IFTYPE_AP)
>> 738 ath5k_hw_reg_write(ah, 0, AR5K_TIMER0);
>>
>> We need to do this for NL80211_IFTYPE_MESH_POINT or DBA won't fire up
>> and beacon queue wont start...
>
> Thanks Nik, that gets beaconing going but only after a issuing a scan
> (we have the same problem on rt2x00).
>
Cool, i'll send a patch for this asap ;-)
Thanks for testing !
--
GPG ID: 0xD21DB2DB
As you read this post global entropy rises. Have Fun ;-)
Nick
ath9k + 80211s has no crash anymore. however after assigned IP to the mesh node, I could not ping at all, it's not working.
ath5k crashes on ar71xx boards, just learned it might have something to do with silicon bug from atheros instead of ath5k driver bug.
xxiao
--- On Thu, 6/4/09, Nick Kossifidis <[email protected]> wrote:
> From: Nick Kossifidis <[email protected]>
> Subject: Re: ath9k/ath5k + 80211s crash
> To: "Andrey Yurovsky" <[email protected]>
> Cc: [email protected], [email protected], [email protected], [email protected]
> Date: Thursday, June 4, 2009, 10:39 PM
> 2009/6/4 Andrey Yurovsky <[email protected]>:
> > On Wed, Jun 3, 2009 at 12:59 PM, Nick Kossifidis
> <[email protected]>
> wrote:
> >> 2009/6/3 Andrey Yurovsky <[email protected]>:
> >>> On Wed, Jun 3, 2009 at 11:21 AM, Nick
> Kossifidis <[email protected]>
> wrote:
> >>>> 2009/6/2 Andrey Yurovsky <[email protected]>:
> >>>>> With wireless-testing from a couple of
> days ago, an X86 system, and this patch:
> >>>>> http://osdir.com/ml/linux-wireless/2009-05/msg01241.html
> >>>>> ath9k can bring up a mesh point
> interface and doesn't crash for me,
> >>>>> but there are no mesh beacons, so mesh
> isn't working. ?On ath5k, there
> >>>>> are also no beacons and mesh doesn't
> work, but there's no crash.
> >>>>>
> >>>>
> >>>>
> >>>> Hmm, try adding NL80211_IFTYPE_MESH_POINT
> here ->
> >>>>
> >>>> http://git.kernel.org/?p=linux/kernel/git/linville/wireless-testing.git;a=blob;f=drivers/net/wireless/ath/ath5k/pcu.c;h=ec35503f6a40e6b6a1cc4323287b31e7cdd3a428;hb=HEAD#l733
> >>>
> >>> It's already there, in ath5k_hw_set_opmode.
> ?We'll need to trace
> >>> what's going on and see why we don't get
> beacons. ?Last night I had an
> >>> ath5k MP suddenly start beaconing but I
> haven't been able to reproduce
> >>> that yet.
> >>
> >> 736 ? ? ? ? /* When in AP mode zero timer0 to
> start TSF */
> >> 737 ? ? ? ? if (ah->ah_op_mode ==
> NL80211_IFTYPE_AP)
> >> 738 ? ? ? ? ? ? ? ? ath5k_hw_reg_write(ah,
> 0, AR5K_TIMER0);
> >>
> >> We need to do this for NL80211_IFTYPE_MESH_POINT
> or DBA won't fire up
> >> and beacon queue wont start...
> >
> > Thanks Nik, that gets beaconing going but only after a
> issuing a scan
> > (we have the same problem on rt2x00).
> >
>
> Cool, i'll send a patch for this asap ;-)
>
> Thanks for testing !
>
>
> --
> GPG ID: 0xD21DB2DB
> As you read this post global entropy rises. Have Fun ;-)
> Nick
>