From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wei Wang2 Subject: Re: Why call iommu_unmap_page() in set_p2m_entry()? Date: Tue, 18 Mar 2008 18:29:45 +0100 Message-ID: <1205861385.18682.47.camel@gran.amd.com> References: <08DF4D958216244799FC84F3514D70F00124FB6E@pdsmsx415.ccr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <08DF4D958216244799FC84F3514D70F00124FB6E@pdsmsx415.ccr.corp.intel.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: "Han, Weidong" Cc: xen-devel@lists.xensource.com, Keir Fraser List-Id: xen-devel@lists.xenproject.org Hi Weidong, Thanks for pointing it out. iommu_unmap_page() is not intended for flushing all iotlb entries. As AMD-IOMMU also supports domain-selective unmapping and I seem to find related functions in vtd code,I hope this operation can also be generic. If not, I plan to put this into amd specific functions called by domain destroy routines. Wei On Thu, 2008-03-13 at 14:28 +0100, Han, Weidong wrote: > As my understanding, AMD IOMMU doesn't share its page tables with p2m > table. So it needs call iommu_unmap_page() to free pages when shut > down > or destroy domain. Now iommu_unmap_page() is only called in > set_p2m_entry(), I think it won't be executed when shut down or > destroy > domain. I'm not familiar with AMD IOMMU code, maybe I missed > something. > Can you explain why do that? Thanks! > > Randy (Weidong) > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel > > >