Document fast page fault and mmu-lock in locking.txt
Signed-off-by: Xiao Guangrong <xiaoguangrong@xxxxxxxxxxxxxxxxxx>
---
Documentation/virtual/kvm/locking.txt | 136 ++++++++++++++++++++++++++++++++-
1 files changed, 135 insertions(+), 1 deletions(-)
diff --git a/Documentation/virtual/kvm/locking.txt b/Documentation/virtual/kvm/locking.txt
index 3b4cd3b..a4a485d 100644
--- a/Documentation/virtual/kvm/locking.txt
+++ b/Documentation/virtual/kvm/locking.txt
@@ -6,7 +6,135 @@ KVM Lock Overview
(to be written)
-2. Reference
+2: Exception
+------------
+
+Fast page fault:
+
+Fast page fault is the fast path which fixes the guest page fault out of
+the mmu-lock on x86. Currently, the page fault can be fast only if the
+shadow page table is present and it is caused by write-protect, that means
+we just need change the W bit of the spte.
+
+What we use to avoid all the race is the SPTE_HOST_WRITEABLE bit,
+SPTE_MMU_WRITEABLE bit on the spte:
+- SPTE_HOST_WRITEABLE means the gfn is writable on host.
+- SPTE_MMU_WRITEABLE means the gfn is writable on mmu. The bit is set when
+ the gfn is writable on guest mmu and it is not write-protected by shadow
+ page write-protection.
+
+On fast page fault path, we will use cmpxchg to atomically set the spte W
+bit if spte.SPTE_HOST_WRITEABLE = 1 and spte.SPTE_WRITE_PROTECT = 1, this
+is safe because whenever changing these bits can be detected by cmpxchg.
+
+But we need carefully check these cases:
+1): The mapping from gfn to pfn
+The mapping from gfn to pfn may be changed since we can only ensure the pfn
+is not changed during cmpxchg. This is a ABA problem, for example, below case
+will happen:
+
+At the beginning:
+gpte = gfn1
+gfn1 is mapped to pfn1 on host
+spte is the shadow page table entry corresponding with gpte and
+spte = pfn1
+
+ VCPU 0 VCPU0
+on fast page fault path:
+
+ old_spte = *spte;
+ pfn1 is swapped out:
+ spte = 0;
+
+ pfn1 is re-alloced for gfn2.
+
+ gpte is changed to point to
+ gfn2 by the guest:
+ spte = pfn1;
+
+ if (cmpxchg(spte, old_spte, old_spte+W)
+ mark_page_dirty(vcpu->kvm, gfn1)
+ OOPS!!!
+
+We dirty-log for gfn1, that means gfn2 is lost in dirty-bitmap.
+
+For direct sp, we can easily avoid it since the spte of direct sp is fixed
+to gfn. For indirect sp, before we do cmpxchg, we call gfn_to_pfn_atomic()
+to pin gfn to pfn, because after gfn_to_pfn_atomic():
+- We have held the refcount of pfn that means the pfn can not be freed and
+ be reused for another gfn.
+- The pfn is writable that means it can not be shared between different gfns
+ by KSM.
+
+Then, we can ensure the dirty bitmaps is correctly set for a gfn.
+
+2): Dirty bit tracking
+In the origin code, the spte can be fast updated (non-atomically) if the
+spte is read-only and the Accessed bit has already been set since the
+Accessed bit and Dirty bit can not be lost.
+
+But it is not true after fast page fault since the spte can be marked
+writable between reading spte and updating spte. Like below case:
+
+At the beginning:
+spte.W = 0
+spte.Accessed = 1
+
+ VCPU 0 VCPU0
+In mmu_spte_clear_track_bits():
+
+ old_spte = *spte;
+
+ /* 'if' condition is satisfied. */
+ if (old_spte.Accssed == 1 &&
+ old_spte.W == 0)
+ spte = 0ull;
+ on fast page fault path:
+ spte.W = 1
+ memory write on the spte:
+ spte.Dirty = 1
+
+
+ else
+ old_spte = xchg(spte, 0ull)
+
+
+ if (old_spte.Accssed == 1)
+ kvm_set_pfn_accessed(spte.pfn);
+ if (old_spte.Dirty == 1)
+ kvm_set_pfn_dirty(spte.pfn);
+ OOPS!!!
+
+The Dirty bit is lost in this case.
+
+In order to avoid this kind of issue, we always treat the spte as "volatile"
+if it can be updated out of mmu-lock, see spte_has_volatile_bits(), it means,
+the spte is always atomicly updated in this case.
+
+3): flush tlbs due to spte updated
+If the spte is updated from writable to readonly, we should flush all TLBs,
+otherwise rmap_write_protect will find a read-only spte, even though the
+writable spte might be cached on a CPU's TLB.
+
+As mentioned before, the spte can be updated to writeable out of mmu-lock on
+fast page fault path, in order to easily audit the path, we see if TLBs need
+be flushed caused by this reason in mmu_spte_update() since this is a common
+function to update spte (present -> present).
+
+See mmu_spte_update():
+
+| if (!spte_has_volatile_bits(old_spte))
+| __update_clear_spte_fast(sptep, new_spte);
+| else
+| old_spte = __update_clear_spte_slow(sptep, new_spte);
+|
+| if (is_writable_pte(old_spte) && !is_writable_pte(new_spte))
+| ret = true;
+
+Since the spte is "volatile" if it can be updated out of mmu-lock, we always
+atomicly update the spte, the race caused by fast page fault can be avoided.
+
+3. Reference
------------
Name: kvm_lock
@@ -23,3 +151,9 @@ Arch: x86
Protects: - kvm_arch::{last_tsc_write,last_tsc_nsec,last_tsc_offset}
- tsc offset in vmcb
Comment: 'raw' because updating the tsc offsets must not be preempted.
+
+Name: kvm->mmu_lock
+Type: spinlock_t
+Arch: any
+Protects: -shadow page/shadow tlb entry
+Comment: it is a spinlock since it is used in mmu notifier.
--
1.7.7.6
--
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
