[ELDK] facing possible kernel preemption

Dipanjan Dutta dipanjan.dutta at ccpu.com
Thu Mar 27 20:12:07 CET 2008


Hello All,

 

I have a module which I inserted into 2.6.16 kernel (ELDK 4.1).

 

After a while, I see the module not responding and see traces like:

(my module name is sa_inop)

 

+++++++++++++++++++++++++++++++++++++++++++++++

...

scheduling while atomic: sa_inop/0x00000101/311

ssIsr: vectNum = 33

Call Trace:

[CE677B30] [C000C800]  (unreliable)

[CE677B60] [C01CA418]

[CE677BB0] [C01C98CC]

[CE677BF0] [D11D9FA4]

[CE677E20] [C001BCAC]

[CE677E40] [C001BE24]

[CE677E70] [C001BF0C]

[CE677E80] [C001C048]

[CE677E90] [C017682C]

[CE677EA0] [C017E338]

[CE677EB0] [C0132F68]

[CE677ED0] [C00612B8]

[CE677EE0] [C0061390]

[CE677F10] [C006175C]

[CE677F40] [C00041D4]

scheduling while atomic: in.telnetd/0x00000101/275

Call Trace:

[CEE9B9C0] [C000C800]  (unreliable)

...

+++++++++++++++++++++++++++++++++++++++++++++++

 

Any idea, what could be the issue? I have a watchdog timer in the system
and whenever such print comes, the watchdog goes for a reboot.

 

Thanks,

Dipanjan




More information about the eldk mailing list