* [gentoo-user] unstoppable, undeciperable kernel call trace in log
@ 2012-06-30 21:57 Maxim Wexler
2012-07-01 1:25 ` Paul Hartman
0 siblings, 1 reply; 4+ messages in thread
From: Maxim Wexler @ 2012-06-30 21:57 UTC (permalink / raw
To: gentoo-user
Hi group,
Please take a look at this clip from /var/log/messages:
https://paste.lugons.org/show/2679/
Can somebody tell me what it means? It never changes, char by char. It
goes on forever, over and over again. At one time I wanted kernel
debugging, but I've returned to menuconfig and turned it off(I thought
).
How do I stop this pointless verbosity?
Max
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [gentoo-user] unstoppable, undeciperable kernel call trace in log
2012-06-30 21:57 [gentoo-user] unstoppable, undeciperable kernel call trace in log Maxim Wexler
@ 2012-07-01 1:25 ` Paul Hartman
2012-07-01 3:11 ` Maxim Wexler
0 siblings, 1 reply; 4+ messages in thread
From: Paul Hartman @ 2012-07-01 1:25 UTC (permalink / raw
To: gentoo-user
On Sat, Jun 30, 2012 at 4:57 PM, Maxim Wexler <maxim.wexler@gmail.com> wrote:
> Hi group,
>
> Please take a look at this clip from /var/log/messages:
>
> https://paste.lugons.org/show/2679/
>
> Can somebody tell me what it means? It never changes, char by char. It
> goes on forever, over and over again. At one time I wanted kernel
> debugging, but I've returned to menuconfig and turned it off(I thought
> ).
>
> How do I stop this pointless verbosity?
Hi,
I don't have an answer, but if you google the phrase "rcu_bh detected
stall on CPU" you can find many reports similar to yours and maybe
some information.
Do you have RCU_CPU_STALL_INFO=y in your kernel config?
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [gentoo-user] unstoppable, undeciperable kernel call trace in log
2012-07-01 1:25 ` Paul Hartman
@ 2012-07-01 3:11 ` Maxim Wexler
2012-07-01 3:46 ` Maxim Wexler
0 siblings, 1 reply; 4+ messages in thread
From: Maxim Wexler @ 2012-07-01 3:11 UTC (permalink / raw
To: gentoo-user
> Hi,
>
> I don't have an answer, but if you google the phrase "rcu_bh detected
> stall on CPU" you can find many reports similar to yours and maybe
> some information.
This might be usefu
lhttp://www.kernel.org/doc/Documentation/RCU/stallwarn.txt
"RCU bugs can often be debugged with the help of CONFIG_RCU_TRACE"
from <help> in menuconfig.
│ This option provides tracing in RCU which presents stats │
│ in debugfs for debugging RCU implementation.
Ok, so I've set the RCU_TRACE=y. Next question: How do I use debugfs,
never used it before?
>
> Do you have RCU_CPU_STALL_INFO=y in your kernel config?
no.
syzygy@lumby ~ $ grep RCU /usr/src/linux/.config
# RCU Subsystem
CONFIG_TREE_RCU=y
# CONFIG_PREEMPT_RCU is not set
# CONFIG_RCU_TRACE is not set
CONFIG_RCU_FANOUT=32
# CONFIG_RCU_FANOUT_EXACT is not set
# CONFIG_RCU_FAST_NO_HZ is not set
# CONFIG_TREE_RCU_TRACE is not set
# CONFIG_SPARSE_RCU_POINTER is not set
CONFIG_RCU_CPU_STALL_TIMEOUT=60
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [gentoo-user] unstoppable, undeciperable kernel call trace in log
2012-07-01 3:11 ` Maxim Wexler
@ 2012-07-01 3:46 ` Maxim Wexler
0 siblings, 0 replies; 4+ messages in thread
From: Maxim Wexler @ 2012-07-01 3:46 UTC (permalink / raw
To: gentoo-user
This is weird
syzygy@lumby ~ $ grep DEBUG_FS /usr/src/linux/.config
# CONFIG_DEBUG_FS is not set
But the entry in menuconfig ->kernel hacking looks like this:' -*-
Debug Filesystem', and can't be changed. Not '[*]' which is
changeable with the spacebar, but '-*-' which is not.
According to <help> it depends on TREE_RCU_TRACE=y OR <a whole bunch
of other stuff>.
But I have TREE_RCU_TRACE=y. So I should be good right? But then why
doesn't it select DEBUG_FS?
googling lead me to http://en.wikipedia.org/wiki/Debugfs, which says
do: 'mount -t debugfs none /sys/kernel/debug'. Ok, but there's no
debug file or dir under /sys/kernel. Do I make one? And will it make a
difference since according to grep DEBUG_FS is not set in the kernel
config?
I'll go and make the new kernel and see what happens.
On 6/30/12, Maxim Wexler <maxim.wexler@gmail.com> wrote:
>> Hi,
>>
>> I don't have an answer, but if you google the phrase "rcu_bh detected
>> stall on CPU" you can find many reports similar to yours and maybe
>> some information.
>
> This might be usefu
> lhttp://www.kernel.org/doc/Documentation/RCU/stallwarn.txt
>
> "RCU bugs can often be debugged with the help of CONFIG_RCU_TRACE"
>
> from <help> in menuconfig.
>
> │ This option provides tracing in RCU which presents stats
> │
> │ in debugfs for debugging RCU implementation.
>
> Ok, so I've set the RCU_TRACE=y. Next question: How do I use debugfs,
> never used it before?
>
>>
>> Do you have RCU_CPU_STALL_INFO=y in your kernel config?
>
> no.
>
> syzygy@lumby ~ $ grep RCU /usr/src/linux/.config
> # RCU Subsystem
> CONFIG_TREE_RCU=y
> # CONFIG_PREEMPT_RCU is not set
> # CONFIG_RCU_TRACE is not set
> CONFIG_RCU_FANOUT=32
> # CONFIG_RCU_FANOUT_EXACT is not set
> # CONFIG_RCU_FAST_NO_HZ is not set
> # CONFIG_TREE_RCU_TRACE is not set
> # CONFIG_SPARSE_RCU_POINTER is not set
> CONFIG_RCU_CPU_STALL_TIMEOUT=60
>
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2012-07-01 3:47 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-06-30 21:57 [gentoo-user] unstoppable, undeciperable kernel call trace in log Maxim Wexler
2012-07-01 1:25 ` Paul Hartman
2012-07-01 3:11 ` Maxim Wexler
2012-07-01 3:46 ` Maxim Wexler
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox