[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Xen-devel] Re: [patch 0/3]Enable CMCI (Corrected Machine Check Error Interrupt) for Intel CPUs


>-----Original Message-----
>From: Keir Fraser [mailto:keir.fraser@xxxxxxxxxxxxx] 
>Sent: 2008年12月24日 22:55
>To: Jiang, Yunhong; Ke, Liping
>Cc: xen-devel@xxxxxxxxxxxxxxxxxxx
>Subject: Re: [Xen-devel] Re: [patch 0/3]Enable CMCI (Corrected 
>Machine Check Error Interrupt) for Intel CPUs
>On 24/12/2008 14:51, "Jiang, Yunhong" <yunhong.jiang@xxxxxxxxx> wrote:
>>>> Yes, that's what I mean of "do similar thing in the
>>> cmci_cpu_down()". We
>>>> placed it in stop_machine_run() because it has rendezvous
>>> all CPU already :$
>>> Yeah, on_each_cpu() rendezvous is pretty cheap. May as well
>>> have cleaner
>>> partitioned code than avoid an extra rendezvous.
>> I'd have a bit clarification here. Simply harirqs disabled in the
>> on_each_cpu() is not enough. What we need is, firstly make 
>sure every CPU has
>> irq disabled, then every CPU begin update the CMCI status. 
>If this is what you
>> mean, we will take a patch for it later.
>Every CPU will have IRQs disabled on entry to the on_each_cpu() IPI
>function. I couldn't mean much else -- for example, it's not 
>even valid for
>the caller of on_each_cpu() to enter it with IRQs disabled.

Currently on_each_cpu() has no gurantee that the fn() will be called with all 
CPU has entered the IPI. For example, maybe on CPU has been working on the fn() 
while the IPI is still pending on other CPU.

> -- Keir
Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.