search for: guest_get_eff_l1e

Just found this patch in house which I thought sent out already. :-) Still one thing remaining is to special case for UP guest which suffers overhead from this change... ---- Remove lock on first guest table walk Existing shadow fault path grabs big lock before walking guest tables, to ensure consistency with shadow content lest concurrent change from other vcpu in a bad OS. But this lock

ging_prep ensures that an mfn is backing the paged-out gfn, and transitions to the next state in the paging state machine for this page. Foreign mappings of the gfn will now succeed. This is the key idea, as it allows the pager to now map the gfn and fill in its contents. Unfortunately, it also allows any other foreign mapper to map the gfn and read its contents. This is particularly dangerous

...fault(struct vcpu *v, u LOCK_BIGLOCK(d); - /* - * Attempt to read the PTE that maps the VA being accessed. By checking for - * PDE validity in the L2 we avoid many expensive fixups in __get_user(). - */ + /* Attempt to read the PTE that maps the VA being accessed. */ guest_get_eff_l1e(v, addr, &pte); - if ( !(l1e_get_flags(pte) & _PAGE_PRESENT) ) - goto bail; page = l1e_get_page(pte); /* We are looking only for read-only mappings of p.t. pages. */ if ( ((l1e_get_flags(pte) & (_PAGE_PRESENT|_PAGE_RW)) != _PAGE_PRESENT) || + !mfn_vali...

1: correct LDT checks 2: add address validity check to guest_map_l1e() 3: use {rd,wr}{fs,gs}base when available 4: check for canonical address before doing page walks Signed-off-by: Jan Beulich <jbeulich@suse.com>

The following two patches allow the parallel use of memsharing, xenpaging and xen-access by using an independent ring buffer for each feature. Please review. v2: - update mem_event_check_ring arguments, check domain rather than domain_id - check ring_full first because its value was just evaluated - check if ring buffer is initialized before calling mem_access_domctl/mem_paging_domctl

...table walk >From: Tim Deegan >Sent: 2008年2月21日 22:13 >Hi, > >So, the idea seems sound, and avoids the shadow lock altogether on a >bunch more pagefaults, which is nice. > >I think that since PV pagetables are guaranteed to be read-only above >l1e, the guest_map_l1e and guest_get_eff_l1e functions can be >allowed to >drop the shadow lock and call guest_walk_tables with shadow_op == 0. >That would mean that there are no callers left setting shadow_op to 1, >and then the shadow_op argument (and the whole mechanism for calling >remove_write_access from guest_walk_table...