Re: [Qemu-devel] [PATCH v2 09/10] intel-iommu: don't unmap all for shadow page table

2018-05-18 Thread Auger Eric
Hi Peter, On 05/18/2018 08:06 AM, Peter Xu wrote: > On Thu, May 17, 2018 at 07:23:33PM +0200, Auger Eric wrote: >> Hi Peter, >> >> On 05/04/2018 05:08 AM, Peter Xu wrote: >>> IOMMU replay was carried out before in many use cases, e.g., context >>> cache invalidations, domain flushes. We used this

Re: [Qemu-devel] [PATCH v2 09/10] intel-iommu: don't unmap all for shadow page table

2018-05-17 Thread Peter Xu
On Thu, May 17, 2018 at 07:23:33PM +0200, Auger Eric wrote: > Hi Peter, > > On 05/04/2018 05:08 AM, Peter Xu wrote: > > IOMMU replay was carried out before in many use cases, e.g., context > > cache invalidations, domain flushes. We used this mechanism to sync the > > shadow page table by firstly

Re: [Qemu-devel] [PATCH v2 09/10] intel-iommu: don't unmap all for shadow page table

2018-05-17 Thread Auger Eric
Hi Peter, On 05/04/2018 05:08 AM, Peter Xu wrote: > IOMMU replay was carried out before in many use cases, e.g., context > cache invalidations, domain flushes. We used this mechanism to sync the > shadow page table by firstly (1) unmap the whole address space, then > (2) walk the page table to re

[Qemu-devel] [PATCH v2 09/10] intel-iommu: don't unmap all for shadow page table

2018-05-03 Thread Peter Xu
IOMMU replay was carried out before in many use cases, e.g., context cache invalidations, domain flushes. We used this mechanism to sync the shadow page table by firstly (1) unmap the whole address space, then (2) walk the page table to remap what's in the table. This is very dangerous. The prob