This reverts commit 71883a62fcd6c70639fa12cda733378b4d997409.
The above commit contains an optimization to kvm_zap_gfn_range which uses gfn-limited TLB flushes, if enabled. If using these limited flushes, kvm_zap_gfn_range passes lock_flush_tlb=false to slot_handle_level_range which creates a race when the function unlocks to call cond_resched. See an example of this race below:
CPU 0 CPU 1 CPU 3 // zap_direct_gfn_range mmu_lock() // *ptep == pte_1 *ptep = 0 if (lock_flush_tlb) flush_tlbs() mmu_unlock() // In invalidate range // MMU notifier mmu_lock() if (pte != 0) *ptep = 0 flush = true if (flush) flush_remote_tlbs() mmu_unlock() return // Host MM reallocates // page previously // backing guest memory. // Guest accesses // invalid page // through pte_1 // in its TLB!!
Tested: Ran all kvm-unit-tests on a Intel Haswell machine with and without this patch. The patch introduced no new failures.
Signed-off-by: Ben Gardon bgardon@google.com Cc: stable@vger.kernel.org --- arch/x86/kvm/mmu.c | 16 +++------------- 1 file changed, 3 insertions(+), 13 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 8d43b7c0f56fd..4cda5ee488454 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -5660,13 +5660,8 @@ void kvm_zap_gfn_range(struct kvm *kvm, gfn_t gfn_start, gfn_t gfn_end) { struct kvm_memslots *slots; struct kvm_memory_slot *memslot; - bool flush_tlb = true; - bool flush = false; int i;
- if (kvm_available_flush_tlb_with_range()) - flush_tlb = false; - spin_lock(&kvm->mmu_lock); for (i = 0; i < KVM_ADDRESS_SPACE_NUM; i++) { slots = __kvm_memslots(kvm, i); @@ -5678,17 +5673,12 @@ void kvm_zap_gfn_range(struct kvm *kvm, gfn_t gfn_start, gfn_t gfn_end) if (start >= end) continue;
- flush |= slot_handle_level_range(kvm, memslot, - kvm_zap_rmapp, PT_PAGE_TABLE_LEVEL, - PT_MAX_HUGEPAGE_LEVEL, start, - end - 1, flush_tlb); + slot_handle_level_range(kvm, memslot, kvm_zap_rmapp, + PT_PAGE_TABLE_LEVEL, PT_MAX_HUGEPAGE_LEVEL, + start, end - 1, true); } }
- if (flush) - kvm_flush_remote_tlbs_with_address(kvm, gfn_start, - gfn_end - gfn_start + 1); - spin_unlock(&kvm->mmu_lock); }