RE: [Xen-devel] Re: [PATCH]Add a flag for shadow pages

["Jiang, Yunhong" <yunhong.jiang@xxxxxxxxx>]

Keir Fraser <mailto:keir.fraser@xxxxxxxxxxxxx> wrote:
> On 04/03/2009 11:57, "Keir Fraser" <keir.fraser@xxxxxxxxxxxxx> wrote:
> 
>> On 04/03/2009 09:56, "Keir Fraser" <keir.fraser@xxxxxxxxxxxxx> wrote:
>> 
>>> Another would be to create a new function page_get_reference_and_owner()
>>> which obtains a reference on a guest page and *returns* the (now known
>>> valid) domain pointer. Probably this is nicer actually. Then all existing
>>> users of page_get_owner() need checking to ensure they don't need to use
>>> the new more expensive function -- I think some are probably actually
>>> unsafe now that shadow pages clobber the domain field.
>> 
>> I'm working on this by the way. I'll clean up everything except shadow uses
>> of page_get_owner().
> 
> Changeset 19268. See get_page_from_l1e() for an example safe
> usage of new
> page_get_owner_and_reference() function.

Thanks for your really quick-hand implementation. I will update my another 
patch accordingly tomorrow.

So still one question to the two assertion (or to Tim??) in 
sh_rm_write_access_from_sl1p()/sh_put_ref(). What's the potential error to be 
protected by this checking? If page is a shadow, it's count_info will always be 
0, right? Or it is just a sanity checking?

I need change this is because, if we mark a page offline, then the count_info 
is not 0, even for shadow page. Can I just checking the count_mask here?

Thanks
Yunhong Jiang

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