On Thu, Mar 4, 2010 at 11:35 PM, Himanshu Chauhan <[email protected]> wrote: > On Tue, Mar 02, 2010 at 11:44:25AM +0530, Vikash Kumar wrote: >> >> BUG: scheduling while atomic: RAID_LAZY_WRITE/102/0x00000205 >> no locks held by RAID_LAZY_WRITE/102. >> > > Almost *always* a problem when doing stuff which can sleep while holding a > spinlock. > As you also verified that :) > > - Himanshu >
Yes that's correct. That's why I don't hold any locks while Issuing IO. But what I missed was that callbacks generated by block IO layer or async XOR operation in linux actually disable preemption and so IO couldn't be issued from the callbacks. -Vikash -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to [email protected] Please read the FAQ at http://kernelnewbies.org/FAQ
