"used greatest stack depth: n bytes left" messages

"used greatest stack depth: n bytes left" messages

[Le Rouzic]

 Hi,

Since 2.6.23-RC5 we can get the following kind of messages in 
/var/log/messages:

Sep 11 12:22:40 frec kernel: loadkeys used greatest stack depth: 4568 
bytes left
Sep 11 12:22:40 frec kernel: rc.sysinit used greatest stack depth: 4448 
bytes left
Sep 11 12:22:40 frec kernel: readahead used greatest stack depth: 4264 
bytes left
Sep 11 13:38:36 frec kernel: mount.nfs4 used greatest stack depth: 4088 
bytes left
Sep 11 17:12:08 frec kernel: mount.nfs4 used greatest stack depth: 3512 
bytes left

Are there only WARNINGs?
Does that mean we need to act on some kernel parameter ? Which one?
Can we conclude anything is not normal?
How nfsv4 can be impacted in its behavior?



Cheers

-- 
-----------------------------------------------------------------
Company : Bull, Architect of an Open World TM (www.bull.com)
Name    : Aime Le Rouzic 
Mail    : Bull - BP 208 - 38432 Echirolles Cedex - France
E-Mail  : aime.le-rouzic@bull.net
Phone   : 33 (4) 76.29.75.51
Fax     : 33 (4) 76.29.75.18
----------------------------------------------------------------- 


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

Re: "used greatest stack depth: n bytes left" messages

[Kevin Coffman]

On 9/12/07, Le Rouzic <aime.le-rouzic@bull.net> wrote:
>
>
>
>  Hi,
>
> Since 2.6.23-RC5 we can get the following kind of messages in
> /var/log/messages:
>
> Sep 11 12:22:40 frec kernel: loadkeys used greatest stack depth: 4568
> bytes left
> Sep 11 12:22:40 frec kernel: rc.sysinit used greatest stack depth: 4448
> bytes left
> Sep 11 12:22:40 frec kernel: readahead used greatest stack depth: 4264
> bytes left
> Sep 11 13:38:36 frec kernel: mount.nfs4 used greatest stack depth: 4088
> bytes left
> Sep 11 17:12:08 frec kernel: mount.nfs4 used greatest stack depth: 3512
> bytes left
>
> Are there only WARNINGs?
> Does that mean we need to act on some kernel parameter ? Which one?
> Can we conclude anything is not normal?
> How nfsv4 can be impacted in its behavior?
>
> Cheers

You must have changed your kernel configuration to include stack debugging?
This message comes from kerne/exit.c and is #ifdef'd with
CONFIG_DEBUG_STACK_USAGE.  So it seems to be more of a debug warning.

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

Re: [NFS] "used greatest stack depth: n bytes left" messages

[Le Rouzic]

Kevin Coffman a =E9crit :

>On 9/12/07, Le Rouzic <aime.le-rouzic@bull.net> wrote:
>  =

>
>>
>> Hi,
>>
>>Since 2.6.23-RC5 we can get the following kind of messages in
>>/var/log/messages:
>>
>>Sep 11 12:22:40 frec kernel: loadkeys used greatest stack depth: 4568
>>bytes left
>>Sep 11 12:22:40 frec kernel: rc.sysinit used greatest stack depth: 4448
>>bytes left
>>Sep 11 12:22:40 frec kernel: readahead used greatest stack depth: 4264
>>bytes left
>>Sep 11 13:38:36 frec kernel: mount.nfs4 used greatest stack depth: 4088
>>bytes left
>>Sep 11 17:12:08 frec kernel: mount.nfs4 used greatest stack depth: 3512
>>bytes left
>>
>>Are there only WARNINGs?
>>Does that mean we need to act on some kernel parameter ? Which one?
>>Can we conclude anything is not normal?
>>How nfsv4 can be impacted in its behavior?
>>
>>Cheers
>>    =

>>
>
>You must have changed your kernel configuration to include stack debugging?
>This message comes from kerne/exit.c and is #ifdef'd with
>CONFIG_DEBUG_STACK_USAGE.  So it seems to be more of a debug warning.
>
>  =

>
   Hi,
  CONFIG_DEBUG_STACK_USAGE was already set in my previous kernel configurat=
ion
  but those messages are new  but like you wrote more for debug warning.
  Nevertheless the question was more to know if they mean some degradation
  happening. So far, it looks not.

  Cheers

 =


-- =

-----------------------------------------------------------------
Company : Bull, Architect of an Open World TM (www.bull.com)
Name    : Aime Le Rouzic =

Mail    : Bull - BP 208 - 38432 Echirolles Cedex - France
E-Mail  : aime.le-rouzic@bull.net
Phone   : 33 (4) 76.29.75.51
Fax     : 33 (4) 76.29.75.18
----------------------------------------------------------------- =

Re: [NFS] "used greatest stack depth: n bytes left" messages

[J. Bruce Fields]

On Wed, Sep 12, 2007 at 02:54:08PM +0200, Le Rouzic wrote:
>   CONFIG_DEBUG_STACK_USAGE was already set in my previous kernel configuration
>   but those messages are new  but like you wrote more for debug warning.
>   Nevertheless the question was more to know if they mean some degradation
>   happening. So far, it looks not.

Right, it doesn't necessarily mean anything, it's just telling you the
number of bytes that weren't used on the stack.  If you've got thousands
of bytes left:

> >>Sep 11 12:22:40 frec kernel: loadkeys used greatest stack depth: 4568
> >>bytes left
> >>Sep 11 12:22:40 frec kernel: rc.sysinit used greatest stack depth: 4448
> >>bytes left
> >>Sep 11 12:22:40 frec kernel: readahead used greatest stack depth: 4264
> >>bytes left
> >>Sep 11 13:38:36 frec kernel: mount.nfs4 used greatest stack depth: 4088
> >>bytes left
> >>Sep 11 17:12:08 frec kernel: mount.nfs4 used greatest stack depth: 3512
> >>bytes left

then nothing's coming anywhere near overflowing its stack, so that's
good.  (But do you know what stack size your kernel is using?  I guess
it's >4k, so probably 8k, suggesting some of these processes are using
over 4k, which seems like a lot.)

--b.