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

Re: [Xen-devel] [PATCH RFC] x86/ioreq server: Optimize p2m cleaning up code in p2m_finish_type_change().





On 4/5/2017 11:11 PM, George Dunlap wrote:
On 05/04/17 16:10, George Dunlap wrote:
On 05/04/17 09:59, Yu Zhang wrote:
Previously, p2m_finish_type_change() is triggered to iterate and
clean up the p2m table when an ioreq server unmaps from memory type
HVMMEM_ioreq_server. And the current iteration number is set to 256
And after these iterations, hypercall pre-emption is checked.

But it is likely that no p2m change is performed for the just finished
iterations, which means p2m_finish_type_change() will return quite
soon. So in such scenario, we can allow the p2m iteration to continue,
without checking the hypercall pre-emption.
Suppose you have a guest with 128TiB of RAM, and the ioreq_server p2m
entries are at the very end of RAM.  Won't this run for several minutes
before even allowing preemption?
Sorry, this should be GiB.  But I think you get my point. :-)

Yep. I got it.
I'd better reconsider it - my head is quite dizzy now. Maybe together with your generic p2m change solution in 4.10. :-)

Thanks
Yu

  -George




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

 


Rackspace

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