On Mon, Nov 15, 2010 at 05:55:25PM +0800, Xiao Guangrong wrote: > On 11/15/2010 05:30 PM, Avi Kivity wrote: > > >> Yeah, this 'retry' is unnecessary if the process is killed, but this > >> case is infrequent, the most case is the process keeps running and try > >> to access the fault address later. > > > > The problem is that if we retry in this case, we install an incorrect spte? > > > > ...... > > >> can avoid if the page mapping have been fixed. > > > > The guest may have changed page directories or other levels. > > > > ...... > > >> > Or another thread may have mmap()ed something else over the > >> > same address. > >> > >> The mmap virtual address is also visible for other threads since the > >> threads > >> have the same page table, so i think this case is the same as above? > > > > Again, don't we install the wrong spte in this case? > > > > I think it doesn't corrupts spte since we will walk guest page table again > and map it to shadow pages when we retry #PF. But if the page is not mapped by new process we can inject #PF into a guest. -- Gleb. -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
