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

Re: [Xen-devel] comments for c/s 25919 Xen mce



Christoph Egger wrote:
> On 09/21/12 08:44, Liu, Jinsong wrote:
> 
>> Christoph, Keir, Jan
>> 
>> I have a draft look at c/s 25919. It moves some mce_intel.c logic to
>> mce.c (and remove old mce.c logic). By draft reviewing the patch I
>> think it need more work to do, and currently it in fact would hung
>> at AMD platform (I have no AMD platform to test), i.e,
>> MACHINE_CHECK_SOFTIRQ --> mce_delayed_action() --> mce_action() -->
>> ASSERT(handler_num);     
>> 
>> For AMD mce it may need add (if any misunderstanding please point to
>> me) 1). add default handler which used at softirq context
> 
> 
> This is mcee_softirq().
> 
>> 2). add AMD vmce inject logic
> 
> 
> Yes, patch is sent.
> See http://lists.xen.org/archives/html/xen-devel/2012-09/msg01413.html
> 
>> 3). more test
> 
> 
> There are more patches in my queue.
> 
> Christoph
> 

So, Christoph, considering you have a patches queue and would change more 
mce/vmce, how about wait a moment and then based your patches on my vmce 
patches? My vmce patches have been posted for quite some time, and have already 
been reviewed-rebased-tested several rounds. I think they are basically stable 
now, currently only need tools maintainers to have review patch4/5 (live 
migration tools part).

... I know I have no right asking you to do so, but we need cooperate for 
mce/vmce, and if you agree I would be very much appreciated. As for your 
mce/vmce patches, I'd like to help test at Intel's platform to make sure it 
works fine.

Thanks,
Jinsong

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel


 


Rackspace

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