Re: ath9k: BUG kmalloc-8192: Poison overwritten

["Justin P. Mattock" <justinmattock@gmail.com>]

On 05/13/10 23:36, Luis R. Rodriguez wrote:
> On Thu, May 13, 2010 at 11:31 PM, Justin P. Mattock
> <justinmattock@gmail.com>  wrote:
>> On 05/13/10 23:20, Bruno Randolf wrote:
>>>
>>> On Friday 14 May 2010 15:16:22 Luis R. Rodriguez wrote:
>>>>
>>>> On Thu, May 13, 2010 at 9:44 PM, Justin P. Mattock
>>>>
>>>> <justinmattock@gmail.com>    wrote:
>>>>>
>>>>> On 05/13/10 21:01, Luis R. Rodriguez wrote:
>>>>>>
>>>>>> On Thu, May 13, 2010 at 7:14 PM, Justin P. Mattock
>>>>>>
>>>>>> <justinmattock@gmail.com>     wrote:
>>>>>>>
>>>>>>> what I can try, is(not at the convention, on eth0
>>>>>>> at the moment), but when I get back to the convention
>>>>>>> center place I can try your patch as well as the
>>>>>>> modprobe option, to see if I can get any signs of
>>>>>>> a recreation(if so I'll bisect there).
>>>>>>
>>>>>> The debug info I just need upon load of the module, I don't need you
>>>>>> to run the debug stuff to try to reproduce. The debug print upon load
>>>>>> will tell us the rxbuf size and cache line size.
>>>>>>
>>>>>>    Luis
>>>>>
>>>>> o.k. it's not pretty due
>>>>> to loads of avc's for SELinux:
>>>>> (I run a rootless system).
>>>>>
>>>>> [   84.172649] ath9k: Driver unloaded
>>>>> [  100.675300] audit_printk_skb: 6 callbacks suppressed
>>>>> [  100.675306] type=1400 audit(1273811633.675:20): avc:  denied  {
>>>>> search
>>>>> } for  pid=2168 comm="modprobe" name="modules" dev=sda3 ino=2500
>>>>> scontext=name:staff_r:staff_sudo_t:s0
>>>>> tcontext=system_u:object_r:modules_object_t:s0 tclass=dir
>>>>> [  100.675408] type=1400 audit(1273811633.675:20): avc:  denied  {
>>>>> search
>>>>> } for  pid=2168 comm="modprobe" name="2.6.34-rc7-00057-gcdfda35"
>>>>> dev=sda3 ino=524392 scontext=name:staff_r:staff_sudo_t:s0
>>>>> tcontext=name:object_r:modules_object_t:s0 tclass=dir
>>>>> [  100.675552] type=1400 audit(1273811633.675:20): avc:  denied  { read
>>>>> }
>>>>> for  pid=2168 comm="modprobe" name="modules.dep.bin" dev=sda3 ino=525251
>>>>> scontext=name:staff_r:staff_sudo_t:s0
>>>>> tcontext=name:object_r:modules_object_t:s0 tclass=file
>>>>> [  100.675598] type=1400 audit(1273811633.675:20): avc:  denied  { open
>>>>> }
>>>>> for  pid=2168 comm="modprobe" name="modules.dep.bin" dev=sda3 ino=525251
>>>>> scontext=name:staff_r:staff_sudo_t:s0
>>>>> tcontext=name:object_r:modules_object_t:s0 tclass=file
>>>>> [  100.675748] type=1300 audit(1273811633.675:20): arch=c000003e
>>>>> syscall=2 success=yes exit=3 a0=60d140 a1=0 a2=1b6 a3=0 items=0
>>>>> ppid=2080 pid=2168 auid=1000 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0
>>>>> sgid=0 fsgid=0 tty=pts0 ses=1 comm="modprobe" exe="/sbin/modprobe"
>>>>> subj=name:staff_r:staff_sudo_t:s0 key=(null)
>>>>> [  100.675902] type=1400 audit(1273811633.675:21): avc:  denied  {
>>>>> getattr } for  pid=2168 comm="modprobe"
>>>>> path="/lib/modules/2.6.34-rc7-00057-gcdfda35/modules.dep.bin" dev=sda3
>>>>> ino=525251 scontext=name:staff_r:staff_sudo_t:s0
>>>>> tcontext=name:object_r:modules_object_t:s0 tclass=file
>>>>> [  100.676052] type=1300 audit(1273811633.675:21): arch=c000003e
>>>>> syscall=5 success=yes exit=0 a0=3 a1=7fffd2d1dd70 a2=7fffd2d1dd70 a3=0
>>>>> items=0 ppid=2080 pid=2168 auid=1000 uid=0 gid=0 euid=0 suid=0 fsuid=0
>>>>> egid=0 sgid=0 fsgid=0 tty=pts0 ses=1 comm="modprobe"
>>>>> exe="/sbin/modprobe"
>>>>> subj=name:staff_r:staff_sudo_t:s0 key=(null)
>>>>> [  100.698392] ath9k 0000:03:00.0: PCI INT A ->    GSI 17 (level, low) ->
>>>>> IRQ 17
>>>>> [  100.698409] ath9k 0000:03:00.0: setting latency timer to 64
>>>>> [  100.828787] ath: EEPROM regdomain: 0x64
>>>>> [  100.828790] ath: EEPROM indicates we should expect a direct regpair
>>>>> map [  100.828793] ath: Country alpha2 being used: 00
>>>>> [  100.828795] ath: Regpair used: 0x64
>>>>> [  100.848609] type=1400 audit(1273811633.678:22): avc:  denied  {
>>>>> search
>>>>> } for  pid=2168 comm="modprobe" name="ieee80211" dev=debugfs ino=18
>>>>> scontext=name:staff_r:staff_sudo_t:s0
>>>>> tcontext=system_u:object_r:debugfs_t:s0 tclass=dir
>>>>> [  100.848788] phy1: Selected rate control algorithm
>>>>> 'ath9k_rate_control'
>>>>> [  100.850035] Registered led device: ath9k-phy1::radio
>>>>> [  100.850488] Registered led device: ath9k-phy1::assoc
>>>>> [  100.851227] Registered led device: ath9k-phy1::tx
>>>>> [  100.851633] Registered led device: ath9k-phy1::rx
>>>>> [  100.851640] phy1: Atheros AR5418 MAC/BB Rev:2 AR5133 RF Rev:81
>>>>> mem=0xffffc900005a0000, irq=17
>>>>> [  100.852240] type=1300 audit(1273811633.678:22): arch=c000003e
>>>>> syscall=175 success=yes exit=0 a0=7f1e9d6ab010 a1=19580 a2=60d920 a3=0
>>>>> items=0 ppid=2080 pid=2168 auid=1000 uid=0 gid=0 euid=0 suid=0 fsuid=0
>>>>> egid=0 sgid=0 fsgid=0 tty=pts0 ses=1 comm="modprobe"
>>>>> exe="/sbin/modprobe"
>>>>> subj=name:staff_r:staff_sudo_t:s0 key=(null)
>>>>>
>>>>> but there you go dmesg of the debug info for you
>>>>
>>>> Justin, did you forget to use the debug parameter on modprobe ath9k?
>>>>
>>>> modprobe ath9k debug=0x0000020
>>>>
>>>> I do not see the output I expected.
>>>
>>> i think you have to do at least "ifconfig wlan0 up" to see it - (at least
>>> for
>>> ath5k).
>>>
>>> bruno
>>>
>>
>> didn't think todo so, I'll redu
>> again and see. BTW: the connection
>> over here is dodgy(hotel) should I at-least
>> find a stable access point?
>
> Up to you, just need that print line.
>
>               ath_print(common, ATH_DBG_CONFIG, "cachelsz %u rxbufsize %u\n",
>                                 common->cachelsz, common->rx_bufsize);
>
>    Luis
>

alright.. (in a few) I'll go ahead and enable
as much as possible, as well as the prink
ans see if I can grab the right info for you.

Justin P. Mattock