Re: ksoftirqd high cpu load on kernels 2.6.24 to 2.6.27-rc1-mm1
From: Dushan Tcholich
Date: Sun Aug 31 2008 - 04:52:03 EST
Hello
I found the culprit.
When using powertop I get:
Top causes for wakeups:
35,2% (251,0) ip : br_stp_enable_bridge (br_hello_timer_expired
So I tried to turn them off with:
brctl sethello br0 0
but the problem persisted.
If I do
brctl sethello br0 5
context switch rate drops 200 times and problem is gone.
I think that the command brctl sethello br0 0 doesn't turn off hello
messages, but sends them 250 times per second.
Thanks for your time
Dushan
On Sat, Aug 30, 2008 at 3:48 AM, Dushan Tcholich <dusanc@xxxxxxxxx> wrote:
> Hello again
>
>
> On Mon, Aug 11, 2008 at 9:53 AM, Dushan Tcholich <dusanc@xxxxxxxxx> wrote:
>> Hi
>>
>> On Sun, Aug 10, 2008 at 9:00 PM, Dushan Tcholich <dusanc@xxxxxxxxx> wrote:
>>> Hi
>>> Sorry for answering this late, but I was short on time and couldn't
>>> get reiser4 to work with 2.6.27-rc2
>>>
>>> On Thu, Aug 7, 2008 at 8:58 PM, Francois Romieu <romieu@xxxxxxxxxxxxx> wrote:
>>>> Dushan Tcholich <dusanc@xxxxxxxxx> :
>>>> [...]
>>>>> I googled a little and found out that oprofile is a little above my head.
>>>>> So as I thought that some driver or HW might be responsible for this I
>>>>> tried to disable various onboard HW and found out that if I disable
>>>>> onboard ethernet problem dissapears, so now I've added netdev and
>>>>> maintainer of R8169 driver to cc.
>>>>
>>>> Can you try 2.6.27-rc2 and send the content of /proc/interrupts, dmesg,
>>>> ifconfig as well as a capture of the strange output from top ?
>>>>
>>
>> I tried some more kernels and I had the same problem with 2.6.23.17
>> and 2.6.27-rc1-mm1, but I couldn't reproduce it with kernel from
>> sysresccd 1.0.1 http://www.sysresccd.org/ which is a patched version
>> of 2.6.24 i think .7 when I booted it to change fs.
>> Could it be that something in userspace is creating this?
>>
>
> I dug a little more and found out some new info.
> Unsolved bugreport with same symptoms:
> http://marc.info/?l=linux-kernel&m=119613299024398&w=2
> Problems appear if I start br0 interface, as context switch rate
> increases 200 times. If I start eth0 instead everything looks ok.
> The bugreport above had bridging enabled too.
>
> When using eth0 I get:
> vmstat -n 1 10
> procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
> r b swpd free buff cache si so bi bo in cs us sy id wa
> 2 0 17720 162288 25024 576852 0 0 21 22 6 9 10 4 85 0
> 0 0 17720 162196 25024 576880 0 0 0 0 13 473 0 0 99 0
> 0 0 17720 162196 25028 576880 0 0 0 4 36 1122 1 0 99 0
> 0 0 17720 162196 25028 576880 0 0 0 0 83 844 0 0 99 0
> 0 0 17720 162196 25028 576880 0 0 0 0 55 691 1 0 99 0
> 0 0 17720 162556 25028 576880 0 0 0 0 13 490 1 0 100 0
> 0 0 17720 162100 25028 576880 0 0 0 0 39 561 6 0 94 0
> 1 0 17720 162028 25028 576880 0 0 0 0 16 1030 4 0 96 0
> 0 0 17720 162028 25028 576880 0 0 0 0 40 597 1 0 99 0
> 0 0 17720 162028 25028 576880 0 0 0 0 12 512 2 0 97 0
>
> top
>
> top - 03:30:07 up 6 days, 5:40, 4 users, load average: 0.02, 0.11, 0.28
> Tasks: 149 total, 2 running, 147 sleeping, 0 stopped, 0 zombie
> Cpu(s): 12.1%us, 3.3%sy, 0.0%ni, 84.6%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
> Mem: 1033388k total, 871336k used, 162052k free, 25016k buffers
> Swap: 610460k total, 17720k used, 592740k free, 576852k cached
>
> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
>
> 4 root 15 -5 0 0 0 S 0 0.0 554:23.44 ksoftirqd/0
>
> If I use br0 I get:
> vmstat -n 1 10
> procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
> r b swpd free buff cache si so bi bo in cs us sy id wa
> 0 0 17720 162112 25048 577120 0 0 21 22 6 11 10 4 86 0
> 0 0 17720 162152 25048 577120 0 0 0 0 14 111082 0 3 97 0
> 0 0 17720 162152 25048 577120 0 0 0 0 23 107134 1 3 96 0
> 1 0 17720 160148 25048 577120 0 0 0 0 11 109888 2 3 95 0
> 0 0 17720 162032 25048 577120 0 0 0 0 33 108163 1 2 97 0
> 0 0 17720 162032 25048 577120 0 0 0 0 7 104642 2 2 95 0
> 0 0 17720 162020 25048 577120 0 0 0 0 41 109135 0 2 98 0
> 0 0 17720 162036 25048 577120 0 0 0 0 9 105133 0 3 96 0
> 0 0 17720 162020 25048 577120 0 0 0 0 42 107605 1 2 97 0
> 1 0 17720 162032 25048 577120 0 0 0 0 5 110768 0 2 98 0
>
> top
>
> top - 03:32:03 up 6 days, 5:41, 4 users, load average: 0.09, 0.10, 0.26
> Tasks: 148 total, 2 running, 146 sleeping, 0 stopped, 0 zombie
> Cpu(s): 4.8%us, 2.0%sy, 0.0%ni, 93.0%id, 0.0%wa, 0.0%hi, 0.2%si, 0.0%st
> Mem: 1033388k total, 871564k used, 161824k free, 25048k buffers
> Swap: 610460k total, 17720k used, 592740k free, 577120k cached
>
> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
> 11489 dusan 20 0 235m 85m 24m R 6 8.5 1:03.51 firefox
> 4 root 15 -5 0 0 0 S 7 0.0 554:24.04 ksoftirqd/0
>
> I start br0 like this in /etc/conf.d/net
>
> bridge_br0="eth0"
> config_eth0=( "null" )
> config_br0=( "192.168.1.3/24" )
> RC_NEED_br0="net.eth0"
> brctl_br0=( "setfd 0" "sethello 0" "stp off" )
> #routes_br0=( "default gw 192.168.1.3" )
> depend_br0() {
> need net.eth0
> }
>
> I start eth0 like this in /etc/conf.d/net
> config_eth0=( "192.168.1.3/24" )
>
>>> I've copied my root to ext3 partition and with vanilla 2.6.27-rc2 I got:
>>> -With my .config problem is still here
>>> -With only rtl8169 removed from config there is no problem
>>>
>>>> It seems rather benign though.
>>>>
>>> Well I wouldn't agree from power managment standpoint :). This nic is
>>> in a lot of laptops.
>>> 8% of 2.13GHz Core2Duo CPU is a lot :)
>>>
>>> Btw. should LKML be removed from cc?
>>> If you need any more help please ask.
>>> I hope I'm not harrasing you too much :)
>>>> --
>>>> Ueimor
>>>>
>>> Have a nice day
>>> Dushan
>>>
>> ...
>>
> Have a nice day
> Dushan
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/