# Commit e4e9d2d4e76bd8fe229c124bd57fc6ba824271b3 # Date 2015-07-07 11:37:26 +0200 # Author Andrew Cooper # Committer Jan Beulich x86/p2m-ept: don't unmap the EPT pagetable while it is still in use The call to iommu_pte_flush() between the two hunks uses &ept_entry->epte which is a pointer into the mapped page. It is eventually passed to `clflush` instruction which will suffer a pagefault if the virtual mapping has fallen out of the TLB. (XEN) ----[ Xen-4.5.0-xs102594-d x86_64 debug=y Not tainted ]---- (XEN) CPU: 7 (XEN) RIP: e008:[] cacheline_flush+0x4/0x9 (XEN) Xen call trace: (XEN) [] cacheline_flush+0x4/0x9 (XEN) [] __iommu_flush_cache+0x4a/0x6a (XEN) [] iommu_pte_flush+0x2b/0xd5 (XEN) [] ept_set_entry+0x4bc/0x61f (XEN) [] p2m_set_entry+0xd1/0x112 (XEN) [] clear_mmio_p2m_entry+0x1a0/0x200 (XEN) [] unmap_mmio_regions+0x49/0x73 (XEN) [] do_domctl+0x15bd/0x1edb (XEN) [] syscall_enter+0xeb/0x145 (XEN) (XEN) Pagetable walk from ffff820040004ae0: (XEN) L4[0x104] = 00000008668a5063 ffffffffffffffff (XEN) L3[0x001] = 00000008668a3063 ffffffffffffffff (XEN) L2[0x000] = 000000086689c063 ffffffffffffffff (XEN) L1[0x004] = 000000056f078063 000000000007f678 (XEN) (XEN) **************************************** (XEN) Panic on CPU 7: (XEN) FATAL PAGE FAULT (XEN) [error_code=0000] (XEN) Faulting linear address: ffff820040004ae0 (XEN) **************************************** Signed-off-by: Andrew Cooper Reviewed-by: George Dunlap Reviewed-by: Jan Beulich --- a/xen/arch/x86/mm/p2m-ept.c +++ b/xen/arch/x86/mm/p2m-ept.c @@ -764,8 +764,6 @@ ept_set_entry(struct p2m_domain *p2m, un p2m->max_mapped_pfn = gfn + (1UL << order) - 1; out: - unmap_domain_page(table); - if ( needs_sync != sync_off ) ept_sync_domain(p2m); @@ -788,6 +786,8 @@ out: } } + unmap_domain_page(table); + /* Release the old intermediate tables, if any. This has to be the last thing we do, after the ept_sync_domain() and removal from the iommu tables, so as to avoid a potential