The motivation of this renaming is to make these variables and related
helper functions less mmu_notifier bound and can also be used for non
mmu_notifier based page invalidation. mmu_invalidate_* was chosen to
better describe the purpose of 'invalidating' a page that those
variables are used for.
- mmu_notifier_seq/range_start/range_end are renamed to
mmu_invalidate_seq/range_start/range_end.
- mmu_notifier_retry{_hva} helper functions are renamed to
mmu_invalidate_retry{_hva}.
- mmu_notifier_count is renamed to mmu_invalidate_in_progress to
avoid confusion with mn_active_invalidate_count.
- While here, also update kvm_inc/dec_notifier_count() to
kvm_mmu_invalidate_begin/end() to match the change for
mmu_notifier_count.
No functional change intended.
Signed-off-by: Chao Peng <chao.p.peng@linux.intel.com>
Message-Id: <20220816125322.1110439-3-chao.p.peng@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Put the reference to any struct page mapped/tracked by a gfn=>pfn cache
upon inserting the pfn into its associated cache, as opposed to putting
the reference only when the cache is done using the pfn. In other words,
don't pin pages while they're in the cache. One of the major roles of
the gfn=>pfn cache is to play nicely with invalidation events, i.e. it
exists in large part so that KVM doesn't rely on pinning pages.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220429210025.3293691-9-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Rework the gfn=>pfn cache (gpc) refresh logic to address multiple races
between the cache itself, and between the cache and mmu_notifier events.
The existing refresh code attempts to guard against races with the
mmu_notifier by speculatively marking the cache valid, and then marking
it invalid if a mmu_notifier invalidation occurs. That handles the case
where an invalidation occurs between dropping and re-acquiring gpc->lock,
but it doesn't handle the scenario where the cache is refreshed after the
cache was invalidated by the notifier, but before the notifier elevates
mmu_notifier_count. The gpc refresh can't use the "retry" helper as its
invalidation occurs _before_ mmu_notifier_count is elevated and before
mmu_notifier_range_start is set/updated.
CPU0 CPU1
---- ----
gfn_to_pfn_cache_invalidate_start()
|
-> gpc->valid = false;
kvm_gfn_to_pfn_cache_refresh()
|
|-> gpc->valid = true;
hva_to_pfn_retry()
|
-> acquire kvm->mmu_lock
kvm->mmu_notifier_count == 0
mmu_seq == kvm->mmu_notifier_seq
drop kvm->mmu_lock
return pfn 'X'
acquire kvm->mmu_lock
kvm_inc_notifier_count()
drop kvm->mmu_lock()
kernel frees pfn 'X'
kvm_gfn_to_pfn_cache_check()
|
|-> gpc->valid == true
caller accesses freed pfn 'X'
Key off of mn_active_invalidate_count to detect that a pfncache refresh
needs to wait for an in-progress mmu_notifier invalidation. While
mn_active_invalidate_count is not guaranteed to be stable, it is
guaranteed to be elevated prior to an invalidation acquiring gpc->lock,
so either the refresh will see an active invalidation and wait, or the
invalidation will run after the refresh completes.
Speculatively marking the cache valid is itself flawed, as a concurrent
kvm_gfn_to_pfn_cache_check() would see a valid cache with stale pfn/khva
values. The KVM Xen use case explicitly allows/wants multiple users;
even though the caches are allocated per vCPU, __kvm_xen_has_interrupt()
can read a different vCPU (or vCPUs). Address this race by invalidating
the cache prior to dropping gpc->lock (this is made possible by fixing
the above mmu_notifier race).
Complicating all of this is the fact that both the hva=>pfn resolution
and mapping of the kernel address can sleep, i.e. must be done outside
of gpc->lock.
Fix the above races in one fell swoop, trying to fix each individual race
is largely pointless and essentially impossible to test, e.g. closing one
hole just shifts the focus to the other hole.
Fixes: 982ed0de47 ("KVM: Reinstate gfn_to_pfn_cache with invalidation support")
Cc: stable@vger.kernel.org
Cc: David Woodhouse <dwmw@amazon.co.uk>
Cc: Mingwei Zhang <mizhang@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220429210025.3293691-8-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Protect gfn=>pfn cache refresh with a mutex to fully serialize refreshes.
The refresh logic doesn't protect against
- concurrent unmaps, or refreshes with different GPAs (which may or may not
happen in practice, for example if a cache is only used under vcpu->mutex;
but it's allowed in the code)
- a false negative on the memslot generation. If the first refresh sees
a stale memslot generation, it will refresh the hva and generation before
moving on to the hva=>pfn translation. If it then drops gpc->lock, a
different user of the cache can come along, acquire gpc->lock, see that
the memslot generation is fresh, and skip the hva=>pfn update due to the
userspace address also matching (because it too was updated).
The refresh path can already sleep during hva=>pfn resolution, so wrap
the refresh with a mutex to ensure that any given refresh runs to
completion before other callers can start their refresh.
Cc: stable@vger.kernel.org
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220429210025.3293691-7-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Don't adjust the userspace address in the gfn=>pfn cache by the page
offset from the gpa. KVM should never use the user address directly, and
all KVM operations that translate a user address to something else
require the user address to be page aligned. Ignoring the offset will
allow the cache to reuse a gfn=>hva translation in the unlikely event
that the page offset of the gpa changes, but the gfn does not. And more
importantly, not having to (un)adjust the user address will simplify a
future bug fix.
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220429210025.3293691-6-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Put the struct page reference to pfn acquired by hva_to_pfn() when the
old and new pfns for a gfn=>pfn cache match. The cache already has a
reference via the old/current pfn, and will only put one reference when
the cache is done with the pfn.
Fixes: 982ed0de47 ("KVM: Reinstate gfn_to_pfn_cache with invalidation support")
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220429210025.3293691-5-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Drop the @pga param from __release_gpc() and rename the helper to make it
more obvious that the cache itself is not being released. The helper
will be reused by a future commit to release a pfn+khva combination that
is _never_ associated with the cache, at which point the current name
would go from slightly misleading to blatantly wrong.
No functional change intended.
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220429210025.3293691-4-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It isn't OK to cache the dirty status of a page in internal structures
for an indefinite period of time.
Any time a vCPU exits the run loop to userspace might be its last; the
VMM might do its final check of the dirty log, flush the last remaining
dirty pages to the destination and complete a live migration. If we
have internal 'dirty' state which doesn't get flushed until the vCPU
is finally destroyed on the source after migration is complete, then
we have lost data because that will escape the final copy.
This problem already exists with the use of kvm_vcpu_unmap() to mark
pages dirty in e.g. VMX nesting.
Note that the actual Linux MM already considers the page to be dirty
since we have a writeable mapping of it. This is just about the KVM
dirty logging.
For the nesting-style use cases (KVM_GUEST_USES_PFN) we will need to
track which gfn_to_pfn_caches have been used and explicitly mark the
corresponding pages dirty before returning to userspace. But we would
have needed external tracking of that anyway, rather than walking the
full list of GPCs to find those belonging to this vCPU which are dirty.
So let's rely *solely* on that external tracking, and keep it simple
rather than laying a tempting trap for callers to fall into.
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20220303154127.202856-3-dwmw2@infradead.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Replace the guest_uses_pa and kernel_map booleans in the PFN cache code
with a unified enum/bitmask. Using explicit names makes it easier to
review and audit call sites.
Opportunistically add a WARN to prevent passing garbage; instantating a
cache without declaring its usage is either buggy or pointless.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20220303154127.202856-2-dwmw2@infradead.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Don't actually set a request bit in vcpu->requests when making a request
purely to force a vCPU to exit the guest. Logging a request but not
actually consuming it would cause the vCPU to get stuck in an infinite
loop during KVM_RUN because KVM would see the pending request and bail
from VM-Enter to service the request.
Note, it's currently impossible for KVM to set KVM_REQ_GPC_INVALIDATE as
nothing in KVM is wired up to set guest_uses_pa=true. But, it'd be all
too easy for arch code to introduce use of kvm_gfn_to_pfn_cache_init()
without implementing handling of the request, especially since getting
test coverage of MMU notifier interaction with specific KVM features
usually requires a directed test.
Opportunistically rename gfn_to_pfn_cache_invalidate_start()'s wake_vcpus
to evict_vcpus. The purpose of the request is to get vCPUs out of guest
mode, it's supposed to _avoid_ waking vCPUs that are blocking.
Opportunistically rename KVM_REQ_GPC_INVALIDATE to be more specific as to
what it wants to accomplish, and to genericize the name so that it can
used for similar but unrelated scenarios, should they arise in the future.
Add a comment and documentation to explain why the "no action" request
exists.
Add compile-time assertions to help detect improper usage. Use the inner
assertless helper in the one s390 path that makes requests without a
hardcoded request.
Cc: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220223165302.3205276-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If the cache's user host virtual address becomes invalid, there
is still a path from kvm_gfn_to_pfn_cache_refresh() where __release_gpc()
could release the pfn but the gpc->pfn field has not been overwritten
with an error value. If this happens, kvm_gfn_to_pfn_cache_unmap will
call put_page again on the same page.
Cc: stable@vger.kernel.org
Fixes: 982ed0de47 ("KVM: Reinstate gfn_to_pfn_cache with invalidation support")
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This can be used in two modes. There is an atomic mode where the cached
mapping is accessed while holding the rwlock, and a mode where the
physical address is used by a vCPU in guest mode.
For the latter case, an invalidation will wake the vCPU with the new
KVM_REQ_GPC_INVALIDATE, and the architecture will need to refresh any
caches it still needs to access before entering guest mode again.
Only one vCPU can be targeted by the wake requests; it's simple enough
to make it wake all vCPUs or even a mask but I don't see a use case for
that additional complexity right now.
Invalidation happens from the invalidate_range_start MMU notifier, which
needs to be able to sleep in order to wake the vCPU and wait for it.
This means that revalidation potentially needs to "wait" for the MMU
operation to complete and the invalidate_range_end notifier to be
invoked. Like the vCPU when it takes a page fault in that period, we
just spin — fixing that in a future patch by implementing an actual
*wait* may be another part of shaving this particularly hirsute yak.
As noted in the comments in the function itself, the only case where
the invalidate_range_start notifier is expected to be called *without*
being able to sleep is when the OOM reaper is killing the process. In
that case, we expect the vCPU threads already to have exited, and thus
there will be nothing to wake, and no reason to wait. So we clear the
KVM_REQUEST_WAIT bit and send the request anyway, then complain loudly
if there actually *was* anything to wake up.
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Message-Id: <20211210163625.2886-3-dwmw2@infradead.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
