Re: [Xen-ia64-devel] VP/P2M: domain_page_flush

On Wed, Apr 26, 2006 at 11:09:35AM +0200, Tristan Gingold wrote:
> Hi,
> 
> as you noticed domain_page_flush is a big hammer and may seriously impact 
> performance.  I suppose that's a reason why VHPT size is reduced.

That's right.


> It would be really better to only purge by range (like vcpu_ptc_ga).
> But can the mapping addresses be easily known ?
> 
> What do you think about this issue ?

The tuning ideas are discussed in the following thread.
Please see and comment on it!

Subject: [Xen-ia64-devel] [RFC] grant table API clean up and performance 
        tuning memo        
Message-ID: <20060324094620.GB20542%yamahata@xxxxxxxxxxxxx> 
http://lists.xensource.com/archives/html/xen-ia64-devel/2006-03/msg00445.html

-- 
yamahata

_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel

<Prev in Thread]	Current Thread	[Next in Thread>
[Xen-ia64-devel] VP/P2M: domain_page_flush, Tristan Gingold Re: [Xen-ia64-devel] VP/P2M: domain_page_flush, Isaku Yamahata <= Re: [Xen-ia64-devel] VP/P2M: domain_page_flush, Tristan Gingold

Previous by Date:

RE: [Xen-ia64-devel] [PATCH] emulate ldfp8 in mmio, Xu, Anthony

Next by Date:

Re: the P2M/VP patch merge plan (was Re: [Xen-ia64-devel] [PATCH][RFC][TAKE5] the P2M/VP patches), Isaku Yamahata

Previous by Thread:

[Xen-ia64-devel] VP/P2M: domain_page_flush, Tristan Gingold

Next by Thread:

Re: [Xen-ia64-devel] VP/P2M: domain_page_flush, Tristan Gingold

Indexes:

[Date] [Thread] [Top] [All Lists]