Re: [Xen-devel] Re: [Xen-ia64-devel] Weekly benchmark results [ww24]

[Steven Hand <Steven.Hand@xxxxxxxxxxxx>]

>==================
>STATUS
>==================
>I do the heavy load test of create/destroy.
>CREDIT scheduler(cshed_schedule) checks BUG_ON(!vcpu_running) at the end of co
de.
>It makes error.
>The reason is that
>atomic_inc(&v->pausecnt)@vcpu_pause() is called without lock.
>(spin_lock(&schedule_data[cpu].schedule_lock))
>This lock-less "pausecnt" makes vcpu_running state changing 
>during the lock of 
>spin_lock_irq(&schedule_data[cpu].schedule_lock)@__enter_schduler().
>The code of cshed_schedule exists within this lock. 

Thanks for this - we have also reproduced. 
The assertation was in fact bogus as you surmised. Have checked in a 
fix. 

cheers,

S.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel