[LEDE-DEV] mt7621 cpu stalls - need help testing

[bart van zoest]

Hi John and the rest,

The kernel version got bumped to 4.9 so I wondered if there is a 4.9
version of the patch?
Tried searching your staging tree but I couldn't find any.

Kind regards,
Bart

On Wed, Jul 5, 2017 at 9:54 PM, Jaap Buurman <jaapbuurman at gmail.com> wrote:
> Dear all,
>
> I just had time to test it myself and I can confirm that SQM cake is
> still crashing the router. Hopefully we do have a nice lead now where
> to look for the issue. If any more testing is needed in the future,
> please let me know. Thank you very much again for your time and
> effort, John!
>
> Kind regards,
>
> Jaap
>
> On Sat, Jul 1, 2017 at 11:29 AM, Jaap Buurman <jaapbuurman at gmail.com> wrote:
>> Dear John,
>>
>> I haven't had time to test it yet myself, but in the previously
>> mentioned topic by Bart van Zoest there have already been two reports
>> of crashing devices with SQM cake enabled unfortunately.
>>
>> One of the posters provided a stack trace, which hopefully will be
>> useful in debugging the issue:
>>
>>
>> Applied blogic's patches to both current snapshot and 17.01, and
>> tested with both fq_codel and cake.
>> After fidling for a bit with the limits it looked like the bug was
>> fixed, but sadly no. About 300mbit ingress/egress does the trick, and
>> after ~15/20minutes and some heavy downloading, reboots and crashes
>> happen.
>>
>> [ 945.720000] INFO: rcu_sched detected stalls on CPUs/tasks:
>> [ 945.730000] 1-...: (21 GPs behind) idle=a22/0/0 softirq=25119/25120 fqs=1
>> [ 945.740000] (detected by 3, t=6004 jiffies, g=3848, c=3847, q=313)
>> [ 945.750000] Task dump for CPU 1:
>> [ 945.760000] swapper/1 R running 0 0 1 0x00100000
>> [ 945.770000] Stack : 00000000 87c4b180 000000dc ffffffff 000000c2
>> 00000000 804db2a4 80490000
>> [ 945.770000] 8048874c 00000001 00000001 80488540 80488724 80490000
>> 80490000 8000c0e0
>> [ 945.770000] 1100fc03 00000001 87c70000 87c71ec0 80490000 8000c410
>> 1100fc03 00000001
>> [ 945.770000] 804db2a4 80490000 804db2a4 8005ed68 80490000 8001b2f8
>> 1100fc03 00000000
>> [ 945.770000] 00000004 804884a0 000000a0 8001b300 c939c939 c939c939
>> c939c939 c939c939
>> [ 945.770000] ...
>> [ 945.840000] Call Trace:
>> [ 945.850000] [<8000be98>] __schedule+0x574/0x758
>> [ 945.860000] [<8000c0e0>] schedule+0x64/0x7c
>> [ 945.870000] [<8000c410>] schedule_preempt_disabled+0x10/0x1c
>> [ 945.880000] [<8005ed68>] cpu_startup_entry+0x11c/0x1b8
>> [ 945.890000] [<8001b300>] start_secondary+0x440/0x470
>> [ 945.900000]
>> [ 945.900000] rcu_sched kthread starved for 6019 jiffies! g3848 c3847
>> f0x0 s3 ->state=0x1
>>
>>
>> Hopefully this information will be helpful! I believe we are on the
>> right track, since with 300mbit it would normally crash within seconds
>> for me. For this poster it took 15-20 minutes, which is quite
>> impressive already.
>>
>> Yours sincerely,
>>
>> Jaap
>>
>> On Fri, Jun 30, 2017 at 1:47 PM, bart van zoest <bartvanzoest at gmail.com> wrote:
>>> Hi John and the rest,
>>>
>>> This is great news! Hopefully this will solve the problems of the
>>> people owning mt7621 with SQM QoS!
>>> I have compiled a build with your revised patch for the D-Link
>>> DIR-860L B1 located at
>>> https://forum.lede-project.org/t/optimized-build-for-the-d-link-dir-860l/948
>>> for the people wanting to test it.
>>> Hopefully, there will be testers and feedback soon!
>>>
>>> Regards,
>>> Bart
>>>
>>> On Thu, Jun 29, 2017 at 8:56 PM, John Crispin <john at phrozen.org> wrote:
>>>>
>>>>
>>>> On 29/06/17 20:14, Jaap Buurman wrote:
>>>>>
>>>>> Dear John,
>>>>>
>>>>> This patch sounds very very promising! I will compile and test this
>>>>> first come this weekend. Thank you so very much for having a look at
>>>>> this issue :)
>>>>>
>>>>> Yours sincerely,
>>>>>
>>>>> Jaap
>>>>>
>>>>> _______________________________________________
>>>>> Lede-dev mailing list
>>>>> Lede-dev at lists.infradead.org
>>>>> http://lists.infradead.org/mailman/listinfo/lede-dev
>>>>
>>>> i pushed a bad patch, please use this instead
>>>>
>>>> https://git.lede-project.org/?p=lede/blogic/staging.git;a=commit;h=c05efda56aecea0f0f52a000a3ce271775b5fb24
>>>>
>>>> i'll provide a v4.9 version tomorrow
>>>>
>>>>     John
>>>>
>>>>
>>>> _______________________________________________
>>>> Lede-dev mailing list
>>>> Lede-dev at lists.infradead.org
>>>> http://lists.infradead.org/mailman/listinfo/lede-dev