1.. SPDX-License-Identifier: GPL-2.0 2 3================= 4KVM Lock Overview 5================= 6 71. Acquisition Orders 8--------------------- 9 10The acquisition orders for mutexes are as follows: 11 12- kvm->lock is taken outside vcpu->mutex 13 14- kvm->lock is taken outside kvm->slots_lock and kvm->irq_lock 15 16- kvm->slots_lock is taken outside kvm->irq_lock, though acquiring 17 them together is quite rare. 18 19On x86, vcpu->mutex is taken outside kvm->arch.hyperv.hv_lock. 20 21Everything else is a leaf: no other lock is taken inside the critical 22sections. 23 242. Exception 25------------ 26 27Fast page fault: 28 29Fast page fault is the fast path which fixes the guest page fault out of 30the mmu-lock on x86. Currently, the page fault can be fast in one of the 31following two cases: 32 331. Access Tracking: The SPTE is not present, but it is marked for access 34 tracking i.e. the SPTE_SPECIAL_MASK is set. That means we need to 35 restore the saved R/X bits. This is described in more detail later below. 36 372. Write-Protection: The SPTE is present and the fault is 38 caused by write-protect. That means we just need to change the W bit of 39 the spte. 40 41What we use to avoid all the race is the SPTE_HOST_WRITEABLE bit and 42SPTE_MMU_WRITEABLE bit on the spte: 43 44- SPTE_HOST_WRITEABLE means the gfn is writable on host. 45- SPTE_MMU_WRITEABLE means the gfn is writable on mmu. The bit is set when 46 the gfn is writable on guest mmu and it is not write-protected by shadow 47 page write-protection. 48 49On fast page fault path, we will use cmpxchg to atomically set the spte W 50bit if spte.SPTE_HOST_WRITEABLE = 1 and spte.SPTE_WRITE_PROTECT = 1, or 51restore the saved R/X bits if VMX_EPT_TRACK_ACCESS mask is set, or both. This 52is safe because whenever changing these bits can be detected by cmpxchg. 53 54But we need carefully check these cases: 55 561) The mapping from gfn to pfn 57 58The mapping from gfn to pfn may be changed since we can only ensure the pfn 59is not changed during cmpxchg. This is a ABA problem, for example, below case 60will happen: 61 62+------------------------------------------------------------------------+ 63| At the beginning:: | 64| | 65| gpte = gfn1 | 66| gfn1 is mapped to pfn1 on host | 67| spte is the shadow page table entry corresponding with gpte and | 68| spte = pfn1 | 69+------------------------------------------------------------------------+ 70| On fast page fault path: | 71+------------------------------------+-----------------------------------+ 72| CPU 0: | CPU 1: | 73+------------------------------------+-----------------------------------+ 74| :: | | 75| | | 76| old_spte = *spte; | | 77+------------------------------------+-----------------------------------+ 78| | pfn1 is swapped out:: | 79| | | 80| | spte = 0; | 81| | | 82| | pfn1 is re-alloced for gfn2. | 83| | | 84| | gpte is changed to point to | 85| | gfn2 by the guest:: | 86| | | 87| | spte = pfn1; | 88+------------------------------------+-----------------------------------+ 89| :: | 90| | 91| if (cmpxchg(spte, old_spte, old_spte+W) | 92| mark_page_dirty(vcpu->kvm, gfn1) | 93| OOPS!!! | 94+------------------------------------------------------------------------+ 95 96We dirty-log for gfn1, that means gfn2 is lost in dirty-bitmap. 97 98For direct sp, we can easily avoid it since the spte of direct sp is fixed 99to gfn. For indirect sp, we disabled fast page fault for simplicity. 100 101A solution for indirect sp could be to pin the gfn, for example via 102kvm_vcpu_gfn_to_pfn_atomic, before the cmpxchg. After the pinning: 103 104- We have held the refcount of pfn that means the pfn can not be freed and 105 be reused for another gfn. 106- The pfn is writable and therefore it cannot be shared between different gfns 107 by KSM. 108 109Then, we can ensure the dirty bitmaps is correctly set for a gfn. 110 1112) Dirty bit tracking 112 113In the origin code, the spte can be fast updated (non-atomically) if the 114spte is read-only and the Accessed bit has already been set since the 115Accessed bit and Dirty bit can not be lost. 116 117But it is not true after fast page fault since the spte can be marked 118writable between reading spte and updating spte. Like below case: 119 120+------------------------------------------------------------------------+ 121| At the beginning:: | 122| | 123| spte.W = 0 | 124| spte.Accessed = 1 | 125+------------------------------------+-----------------------------------+ 126| CPU 0: | CPU 1: | 127+------------------------------------+-----------------------------------+ 128| In mmu_spte_clear_track_bits():: | | 129| | | 130| old_spte = *spte; | | 131| | | 132| | | 133| /* 'if' condition is satisfied. */| | 134| if (old_spte.Accessed == 1 && | | 135| old_spte.W == 0) | | 136| spte = 0ull; | | 137+------------------------------------+-----------------------------------+ 138| | on fast page fault path:: | 139| | | 140| | spte.W = 1 | 141| | | 142| | memory write on the spte:: | 143| | | 144| | spte.Dirty = 1 | 145+------------------------------------+-----------------------------------+ 146| :: | | 147| | | 148| else | | 149| old_spte = xchg(spte, 0ull) | | 150| if (old_spte.Accessed == 1) | | 151| kvm_set_pfn_accessed(spte.pfn);| | 152| if (old_spte.Dirty == 1) | | 153| kvm_set_pfn_dirty(spte.pfn); | | 154| OOPS!!! | | 155+------------------------------------+-----------------------------------+ 156 157The Dirty bit is lost in this case. 158 159In order to avoid this kind of issue, we always treat the spte as "volatile" 160if it can be updated out of mmu-lock, see spte_has_volatile_bits(), it means, 161the spte is always atomically updated in this case. 162 1633) flush tlbs due to spte updated 164 165If the spte is updated from writable to readonly, we should flush all TLBs, 166otherwise rmap_write_protect will find a read-only spte, even though the 167writable spte might be cached on a CPU's TLB. 168 169As mentioned before, the spte can be updated to writable out of mmu-lock on 170fast page fault path, in order to easily audit the path, we see if TLBs need 171be flushed caused by this reason in mmu_spte_update() since this is a common 172function to update spte (present -> present). 173 174Since the spte is "volatile" if it can be updated out of mmu-lock, we always 175atomically update the spte, the race caused by fast page fault can be avoided, 176See the comments in spte_has_volatile_bits() and mmu_spte_update(). 177 178Lockless Access Tracking: 179 180This is used for Intel CPUs that are using EPT but do not support the EPT A/D 181bits. In this case, when the KVM MMU notifier is called to track accesses to a 182page (via kvm_mmu_notifier_clear_flush_young), it marks the PTE as not-present 183by clearing the RWX bits in the PTE and storing the original R & X bits in 184some unused/ignored bits. In addition, the SPTE_SPECIAL_MASK is also set on the 185PTE (using the ignored bit 62). When the VM tries to access the page later on, 186a fault is generated and the fast page fault mechanism described above is used 187to atomically restore the PTE to a Present state. The W bit is not saved when 188the PTE is marked for access tracking and during restoration to the Present 189state, the W bit is set depending on whether or not it was a write access. If 190it wasn't, then the W bit will remain clear until a write access happens, at 191which time it will be set using the Dirty tracking mechanism described above. 192 1933. Reference 194------------ 195 196:Name: kvm_lock 197:Type: mutex 198:Arch: any 199:Protects: - vm_list 200 201:Name: kvm_count_lock 202:Type: raw_spinlock_t 203:Arch: any 204:Protects: - hardware virtualization enable/disable 205:Comment: 'raw' because hardware enabling/disabling must be atomic /wrt 206 migration. 207 208:Name: kvm_arch::tsc_write_lock 209:Type: raw_spinlock 210:Arch: x86 211:Protects: - kvm_arch::{last_tsc_write,last_tsc_nsec,last_tsc_offset} 212 - tsc offset in vmcb 213:Comment: 'raw' because updating the tsc offsets must not be preempted. 214 215:Name: kvm->mmu_lock 216:Type: spinlock_t 217:Arch: any 218:Protects: -shadow page/shadow tlb entry 219:Comment: it is a spinlock since it is used in mmu notifier. 220 221:Name: kvm->srcu 222:Type: srcu lock 223:Arch: any 224:Protects: - kvm->memslots 225 - kvm->buses 226:Comment: The srcu read lock must be held while accessing memslots (e.g. 227 when using gfn_to_* functions) and while accessing in-kernel 228 MMIO/PIO address->device structure mapping (kvm->buses). 229 The srcu index can be stored in kvm_vcpu->srcu_idx per vcpu 230 if it is needed by multiple functions. 231 232:Name: blocked_vcpu_on_cpu_lock 233:Type: spinlock_t 234:Arch: x86 235:Protects: blocked_vcpu_on_cpu 236:Comment: This is a per-CPU lock and it is used for VT-d posted-interrupts. 237 When VT-d posted-interrupts is supported and the VM has assigned 238 devices, we put the blocked vCPU on the list blocked_vcpu_on_cpu 239 protected by blocked_vcpu_on_cpu_lock, when VT-d hardware issues 240 wakeup notification event since external interrupts from the 241 assigned devices happens, we will find the vCPU on the list to 242 wakeup. 243