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

Re: [Xen-devel] TLB flushing


  • To: Ky Srinivasan <ksrinivasan@xxxxxxxxxx>, <xen-devel@xxxxxxxxxxxxxxxxxxx>
  • From: Keir Fraser <keir@xxxxxxxxxxxxx>
  • Date: Tue, 23 Jan 2007 19:09:58 +0000
  • Delivery-date: Tue, 23 Jan 2007 11:09:45 -0800
  • List-id: Xen developer discussion <xen-devel.lists.xensource.com>
  • Thread-index: Acc/IhLJUZOjNKsVEdubbQAX8io7RQ==
  • Thread-topic: [Xen-devel] TLB flushing

On 23/1/07 18:28, "Ky Srinivasan" <ksrinivasan@xxxxxxxxxx> wrote:

> Looking at the do_mmuext_op(), it looks like we are not handling shadow mode
> uniformly :
> For instance, for MMUEXT_INVLPG_LOCAL case, we deal with the case where the
> domain may be operating under shadow mode; however, for other flush options we
> don't deal with the case where the domain may be operating in shadow mode.
> What is the rationale for this?

Most (or perhaps even all) of the references to shadow mode in arch/x86/mm.c
are to handle PV guests running in shadow_mode_translate and/or
shadow_mode_refcounts. Those modes were never really a well-integrated or
-tested configuration for PV guests and are now deprecated (e.g., reference
to shadow mode in MMUEXT_INVLPG_LOCAL will be removed sometime soon). Really
most of arch/x86/mm.c needs to be renamed to arch/x86/mm/pv.c...

 -- Keir


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


 


Rackspace

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