From: David Woodhouse <dwmw2@infradead.org>
commit 7e2175ebd695f17860c5bd4ad7616cce12ed4591 upstream.
In commit b043138246 ("x86/KVM: Make sure KVM_VCPU_FLUSH_TLB flag is
not missed") we switched to using a gfn_to_pfn_cache for accessing the
guest steal time structure in order to allow for an atomic xchg of the
preempted field. This has a couple of problems.
Firstly, kvm_map_gfn() doesn't work at all for IOMEM pages when the
atomic flag is set, which it is in kvm_steal_time_set_preempted(). So a
guest vCPU using an IOMEM page for its steal time would never have its
preempted field set.
Secondly, the gfn_to_pfn_cache is not invalidated in all cases where it
should have been. There are two stages to the GFN->PFN conversion;
first the GFN is converted to a userspace HVA, and then that HVA is
looked up in the process page tables to find the underlying host PFN.
Correct invalidation of the latter would require being hooked up to the
MMU notifiers, but that doesn't happen---so it just keeps mapping and
unmapping the *wrong* PFN after the userspace page tables change.
In the !IOMEM case at least the stale page *is* pinned all the time it's
cached, so it won't be freed and reused by anyone else while still
receiving the steal time updates. The map/unmap dance only takes care
of the KVM administrivia such as marking the page dirty.
Until the gfn_to_pfn cache handles the remapping automatically by
integrating with the MMU notifiers, we might as well not get a
kernel mapping of it, and use the perfectly serviceable userspace HVA
that we already have. We just need to implement the atomic xchg on
the userspace address with appropriate exception handling, which is
fairly trivial.
Cc: stable@vger.kernel.org
Fixes: b043138246 ("x86/KVM: Make sure KVM_VCPU_FLUSH_TLB flag is not missed")
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Message-Id: <3645b9b889dac6438394194bb5586a46b68d581f.camel@infradead.org>
[I didn't entirely agree with David's assessment of the
usefulness of the gfn_to_pfn cache, and integrated the outcome
of the discussion in the above commit message. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[risbhat@amazon.com: Use the older mark_page_dirty_in_slot api without
kvm argument]
Signed-off-by: Rishabh Bhatnagar <risbhat@amazon.com>
Tested-by: Allen Pais <apais@linux.microsoft.com>
Acked-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 6cd88243c7e03845a450795e134b488fc2afb736 upstream.
If a vCPU is outside guest mode and is scheduled out, it might be in the
process of making a memory access. A problem occurs if another vCPU uses
the PV TLB flush feature during the period when the vCPU is scheduled
out, and a virtual address has already been translated but has not yet
been accessed, because this is equivalent to using a stale TLB entry.
To avoid this, only report a vCPU as preempted if sure that the guest
is at an instruction boundary. A rescheduling request will be delivered
to the host physical CPU as an external interrupt, so for simplicity
consider any vmexit *not* instruction boundary except for external
interrupts.
It would in principle be okay to report the vCPU as preempted also
if it is sleeping in kvm_vcpu_block(): a TLB flush IPI will incur the
vmentry/vmexit overhead unnecessarily, and optimistic spinning is
also unlikely to succeed. However, leave it for later because right
now kvm_vcpu_check_block() is doing memory accesses. Even
though the TLB flush issue only applies to virtual memory address,
it's very much preferrable to be conservative.
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[OP: use VCPU_STAT() for debugfs entries]
Signed-off-by: Ovidiu Panait <ovidiu.panait@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 683412ccf61294d727ead4a73d97397396e69a6b upstream.
Flush the CPU caches when memory is reclaimed from an SEV guest (where
reclaim also includes it being unmapped from KVM's memslots). Due to lack
of coherency for SEV encrypted memory, failure to flush results in silent
data corruption if userspace is malicious/broken and doesn't ensure SEV
guest memory is properly pinned and unpinned.
Cache coherency is not enforced across the VM boundary in SEV (AMD APM
vol.2 Section 15.34.7). Confidential cachelines, generated by confidential
VM guests have to be explicitly flushed on the host side. If a memory page
containing dirty confidential cachelines was released by VM and reallocated
to another user, the cachelines may corrupt the new user at a later time.
KVM takes a shortcut by assuming all confidential memory remain pinned
until the end of VM lifetime. Therefore, KVM does not flush cache at
mmu_notifier invalidation events. Because of this incorrect assumption and
the lack of cache flushing, malicous userspace can crash the host kernel:
creating a malicious VM and continuously allocates/releases unpinned
confidential memory pages when the VM is running.
Add cache flush operations to mmu_notifier operations to ensure that any
physical memory leaving the guest VM get flushed. In particular, hook
mmu_notifier_invalidate_range_start and mmu_notifier_release events and
flush cache accordingly. The hook after releasing the mmu lock to avoid
contention with other vCPUs.
Cc: stable@vger.kernel.org
Suggested-by: Sean Christpherson <seanjc@google.com>
Reported-by: Mingwei Zhang <mizhang@google.com>
Signed-off-by: Mingwei Zhang <mizhang@google.com>
Message-Id: <20220421031407.2516575-4-mizhang@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[OP: applied kvm_arch_guest_memory_reclaimed() calls in kvm_set_memslot() and
kvm_mmu_notifier_invalidate_range_start();
OP: adjusted kvm_arch_guest_memory_reclaimed() to not use static_call_cond()]
Signed-off-by: Ovidiu Panait <ovidiu.panait@windriver.com>
Reviewed-by: Liam Merwick <liam.merwick@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 2c985527dd8d283e786ad7a67e532ef7f6f00fac ]
The mask value of fixed counter control register should be dynamic
adjusted with the number of fixed counters. This patch introduces a
variable that includes the reserved bits of fixed counter control
registers. This is a generic code refactoring.
Co-developed-by: Luwei Kang <luwei.kang@intel.com>
Signed-off-by: Luwei Kang <luwei.kang@intel.com>
Signed-off-by: Like Xu <like.xu@linux.intel.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Message-Id: <20220411101946.20262-6-likexu@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 95b065bf5c431c06c68056a03a5853b660640ecc ]
The third nybble of AMD's event select overlaps with Intel's IN_TX and
IN_TXCP bits. Therefore, we can't use AMD64_RAW_EVENT_MASK on Intel
platforms that support TSX.
Declare a raw_event_mask in the kvm_pmu structure, initialize it in
the vendor-specific pmu_refresh() functions, and use that mask for
PERF_TYPE_RAW configurations in reprogram_gp_counter().
Fixes: 710c47651431 ("KVM: x86/pmu: Use AMD64_RAW_EVENT_MASK for PERF_TYPE_RAW")
Signed-off-by: Jim Mattson <jmattson@google.com>
Message-Id: <20220308012452.3468611-1-jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c2fe3cd4604ac87c587db05d41843d667dc43815 ]
Split out VMX's checks on CR4.VMXE to a dedicated hook, .is_valid_cr4(),
and invoke the new hook from kvm_valid_cr4(). This fixes an issue where
KVM_SET_SREGS would return success while failing to actually set CR4.
Fixing the issue by explicitly checking kvm_x86_ops.set_cr4()'s return
in __set_sregs() is not a viable option as KVM has already stuffed a
variety of vCPU state.
Note, kvm_valid_cr4() and is_valid_cr4() have different return types and
inverted semantics. This will be remedied in a future patch.
Fixes: 5e1746d620 ("KVM: nVMX: Allow setting the VMXE bit in CR4")
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20201007014417.29276-5-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 982bae43f11c37b51d2f1961bb25ef7cac3746fa upstream.
Mark kvm_mmu_x86_module_init() with __init, the entire reason it exists
is to initialize variables when kvm.ko is loaded, i.e. it must never be
called after module initialization.
Fixes: 1d0e84806047 ("KVM: x86/mmu: Resolve nx_huge_pages when kvm.ko is loaded")
Cc: stable@vger.kernel.org
Reviewed-by: Kai Huang <kai.huang@intel.com>
Tested-by: Michael Roth <michael.roth@amd.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220803224957.1285926-2-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1d0e84806047f38027d7572adb4702ef7c09b317 upstream.
Resolve nx_huge_pages to true/false when kvm.ko is loaded, leaving it as
-1 is technically undefined behavior when its value is read out by
param_get_bool(), as boolean values are supposed to be '0' or '1'.
Alternatively, KVM could define a custom getter for the param, but the
auto value doesn't depend on the vendor module in any way, and printing
"auto" would be unnecessarily unfriendly to the user.
In addition to fixing the undefined behavior, resolving the auto value
also fixes the scenario where the auto value resolves to N and no vendor
module is loaded. Previously, -1 would result in Y being printed even
though KVM would ultimately disable the mitigation.
Rename the existing MMU module init/exit helpers to clarify that they're
invoked with respect to the vendor module, and add comments to document
why KVM has two separate "module init" flows.
=========================================================================
UBSAN: invalid-load in kernel/params.c:320:33
load of value 255 is not a valid value for type '_Bool'
CPU: 6 PID: 892 Comm: tail Not tainted 5.17.0-rc3+ #799
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015
Call Trace:
<TASK>
dump_stack_lvl+0x34/0x44
ubsan_epilogue+0x5/0x40
__ubsan_handle_load_invalid_value.cold+0x43/0x48
param_get_bool.cold+0xf/0x14
param_attr_show+0x55/0x80
module_attr_show+0x1c/0x30
sysfs_kf_seq_show+0x93/0xc0
seq_read_iter+0x11c/0x450
new_sync_read+0x11b/0x1a0
vfs_read+0xf0/0x190
ksys_read+0x5f/0xe0
do_syscall_64+0x3b/0xc0
entry_SYSCALL_64_after_hwframe+0x44/0xae
</TASK>
=========================================================================
Fixes: b8e8c8303f ("kvm: mmu: ITLB_MULTIHIT mitigation")
Cc: stable@vger.kernel.org
Reported-by: Bruno Goncalves <bgoncalv@redhat.com>
Reported-by: Jan Stancek <jstancek@redhat.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220331221359.3912754-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f7e570780efc5cec9b2ed1e0472a7da14e864fdb upstream.
Forcibly leave nested virtualization operation if userspace toggles SMM
state via KVM_SET_VCPU_EVENTS or KVM_SYNC_X86_EVENTS. If userspace
forces the vCPU out of SMM while it's post-VMXON and then injects an SMI,
vmx_enter_smm() will overwrite vmx->nested.smm.vmxon and end up with both
vmxon=false and smm.vmxon=false, but all other nVMX state allocated.
Don't attempt to gracefully handle the transition as (a) most transitions
are nonsencial, e.g. forcing SMM while L2 is running, (b) there isn't
sufficient information to handle all transitions, e.g. SVM wants access
to the SMRAM save state, and (c) KVM_SET_VCPU_EVENTS must precede
KVM_SET_NESTED_STATE during state restore as the latter disallows putting
the vCPU into L2 if SMM is active, and disallows tagging the vCPU as
being post-VMXON in SMM if SMM is not active.
Abuse of KVM_SET_VCPU_EVENTS manifests as a WARN and memory leak in nVMX
due to failure to free vmcs01's shadow VMCS, but the bug goes far beyond
just a memory leak, e.g. toggling SMM on while L2 is active puts the vCPU
in an architecturally impossible state.
WARNING: CPU: 0 PID: 3606 at free_loaded_vmcs arch/x86/kvm/vmx/vmx.c:2665 [inline]
WARNING: CPU: 0 PID: 3606 at free_loaded_vmcs+0x158/0x1a0 arch/x86/kvm/vmx/vmx.c:2656
Modules linked in:
CPU: 1 PID: 3606 Comm: syz-executor725 Not tainted 5.17.0-rc1-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:free_loaded_vmcs arch/x86/kvm/vmx/vmx.c:2665 [inline]
RIP: 0010:free_loaded_vmcs+0x158/0x1a0 arch/x86/kvm/vmx/vmx.c:2656
Code: <0f> 0b eb b3 e8 8f 4d 9f 00 e9 f7 fe ff ff 48 89 df e8 92 4d 9f 00
Call Trace:
<TASK>
kvm_arch_vcpu_destroy+0x72/0x2f0 arch/x86/kvm/x86.c:11123
kvm_vcpu_destroy arch/x86/kvm/../../../virt/kvm/kvm_main.c:441 [inline]
kvm_destroy_vcpus+0x11f/0x290 arch/x86/kvm/../../../virt/kvm/kvm_main.c:460
kvm_free_vcpus arch/x86/kvm/x86.c:11564 [inline]
kvm_arch_destroy_vm+0x2e8/0x470 arch/x86/kvm/x86.c:11676
kvm_destroy_vm arch/x86/kvm/../../../virt/kvm/kvm_main.c:1217 [inline]
kvm_put_kvm+0x4fa/0xb00 arch/x86/kvm/../../../virt/kvm/kvm_main.c:1250
kvm_vm_release+0x3f/0x50 arch/x86/kvm/../../../virt/kvm/kvm_main.c:1273
__fput+0x286/0x9f0 fs/file_table.c:311
task_work_run+0xdd/0x1a0 kernel/task_work.c:164
exit_task_work include/linux/task_work.h:32 [inline]
do_exit+0xb29/0x2a30 kernel/exit.c:806
do_group_exit+0xd2/0x2f0 kernel/exit.c:935
get_signal+0x4b0/0x28c0 kernel/signal.c:2862
arch_do_signal_or_restart+0x2a9/0x1c40 arch/x86/kernel/signal.c:868
handle_signal_work kernel/entry/common.c:148 [inline]
exit_to_user_mode_loop kernel/entry/common.c:172 [inline]
exit_to_user_mode_prepare+0x17d/0x290 kernel/entry/common.c:207
__syscall_exit_to_user_mode_work kernel/entry/common.c:289 [inline]
syscall_exit_to_user_mode+0x19/0x60 kernel/entry/common.c:300
do_syscall_64+0x42/0xb0 arch/x86/entry/common.c:86
entry_SYSCALL_64_after_hwframe+0x44/0xae
</TASK>
Cc: stable@vger.kernel.org
Reported-by: syzbot+8112db3ab20e70d50c31@syzkaller.appspotmail.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220125220358.2091737-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Backported-by: Tadeusz Struk <tadeusz.struk@linaro.org>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1ebfaa11ebb5b603a3c3f54b2e84fcf1030f5a14 upstream.
Prior to commit 0baedd7927 ("KVM: x86: make Hyper-V PV TLB flush use
tlb_flush_guest()"), kvm_hv_flush_tlb() was using 'KVM_REQ_TLB_FLUSH |
KVM_REQUEST_NO_WAKEUP' when making a request to flush TLBs on other vCPUs
and KVM_REQ_TLB_FLUSH is/was defined as:
(0 | KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP)
so KVM_REQUEST_WAIT was lost. Hyper-V TLFS, however, requires that
"This call guarantees that by the time control returns back to the
caller, the observable effects of all flushes on the specified virtual
processors have occurred." and without KVM_REQUEST_WAIT there's a small
chance that the vCPU making the TLB flush will resume running before
all IPIs get delivered to other vCPUs and a stale mapping can get read
there.
Fix the issue by adding KVM_REQUEST_WAIT flag to KVM_REQ_TLB_FLUSH_GUEST:
kvm_hv_flush_tlb() is the sole caller which uses it for
kvm_make_all_cpus_request()/kvm_make_vcpus_request_mask() where
KVM_REQUEST_WAIT makes a difference.
Cc: stable@kernel.org
Fixes: 0baedd7927 ("KVM: x86: make Hyper-V PV TLB flush use tlb_flush_guest()")
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20211209102937.584397-1-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit 07ffaf343e34b555c9e7ea39a9c81c439a706f13 ]
Trigger a full TLB flush on behalf of the guest on nested VM-Enter and
VM-Exit when VPID is disabled for L2. kvm_mmu_new_pgd() syncs only the
current PGD, which can theoretically leave stale, unsync'd entries in a
previous guest PGD, which could be consumed if L2 is allowed to load CR3
with PCID_NOFLUSH=1.
Rename KVM_REQ_HV_TLB_FLUSH to KVM_REQ_TLB_FLUSH_GUEST so that it can
be utilized for its obvious purpose of emulating a guest TLB flush.
Note, there is no change the actual TLB flush executed by KVM, even
though the fast PGD switch uses KVM_REQ_TLB_FLUSH_CURRENT. When VPID is
disabled for L2, vpid02 is guaranteed to be '0', and thus
nested_get_vpid02() will return the VPID that is shared by L1 and L2.
Generate the request outside of kvm_mmu_new_pgd(), as getting the common
helper to correctly identify which requested is needed is quite painful.
E.g. using KVM_REQ_TLB_FLUSH_GUEST when nested EPT is in play is wrong as
a TLB flush from the L1 kernel's perspective does not invalidate EPT
mappings. And, by using KVM_REQ_TLB_FLUSH_GUEST, nVMX can do future
simplification by moving the logic into nested_vmx_transition_tlb_flush().
Fixes: 41fab65e7c ("KVM: nVMX: Skip MMU sync on nested VMX transition when possible")
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20210609234235.1244004-2-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit f71a53d1180d5ecc346f0c6a23191d837fe2871b upstream.
Restore CR4.LA57 to the mmu_role to fix an amusing edge case with nested
virtualization. When KVM (L0) is using TDP, CR4.LA57 is not reflected in
mmu_role.base.level because that tracks the shadow root level, i.e. TDP
level. Normally, this is not an issue because LA57 can't be toggled
while long mode is active, i.e. the guest has to first disable paging,
then toggle LA57, then re-enable paging, thus ensuring an MMU
reinitialization.
But if L1 is crafty, it can load a new CR4 on VM-Exit and toggle LA57
without having to bounce through an unpaged section. L1 can also load a
new CR3 on exit, i.e. it doesn't even need to play crazy paging games, a
single entry PML5 is sufficient. Such shenanigans are only problematic
if L0 and L1 use TDP, otherwise L1 and L2 share an MMU that gets
reinitialized on nested VM-Enter/VM-Exit due to mmu_role.base.guest_mode.
Note, in the L2 case with nested TDP, even though L1 can switch between
L2s with different LA57 settings, thus bypassing the paging requirement,
in that case KVM's nested_mmu will track LA57 in base.level.
This reverts commit 8053f924ca.
Fixes: 8053f924ca ("KVM: x86/mmu: Drop kvm_mmu_extended_role.cr4_la57 hack")
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20210622175739.3610207-6-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 5104d7ffcf24749939bea7fdb5378d186473f890 upstream.
Disable preemption when probing a user return MSR via RDSMR/WRMSR. If
the MSR holds a different value per logical CPU, the WRMSR could corrupt
the host's value if KVM is preempted between the RDMSR and WRMSR, and
then rescheduled on a different CPU.
Opportunistically land the helper in common x86, SVM will use the helper
in a future commit.
Fixes: 4be5341026 ("KVM: VMX: Initialize vmx->guest_msrs[] right after allocation")
Cc: stable@vger.kernel.org
Cc: Xiaoyao Li <xiaoyao.li@intel.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20210504171734.1434054-6-seanjc@google.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c5e2184d1544f9e56140791eff1a351bea2e63b9 upstream.
Remove the update_pte() shadow paging logic, which was obsoleted by
commit 4731d4c7a0 ("KVM: MMU: out of sync shadow core"), but never
removed. As pointed out by Yu, KVM never write protects leaf page
tables for the purposes of shadow paging, and instead marks their
associated shadow page as unsync so that the guest can write PTEs at
will.
The update_pte() path, which predates the unsync logic, optimizes COW
scenarios by refreshing leaf SPTEs when they are written, as opposed to
zapping the SPTE, restarting the guest, and installing the new SPTE on
the subsequent fault. Since KVM no longer write-protects leaf page
tables, update_pte() is unreachable and can be dropped.
Reported-by: Yu Zhang <yu.c.zhang@intel.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20210115004051.4099250-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit b318e8decf6b9ef1bcf4ca06fae6d6a2cb5d5c5c ]
Fix a plethora of issues with MSR filtering by installing the resulting
filter as an atomic bundle instead of updating the live filter one range
at a time. The KVM_X86_SET_MSR_FILTER ioctl() isn't truly atomic, as
the hardware MSR bitmaps won't be updated until the next VM-Enter, but
the relevant software struct is atomically updated, which is what KVM
really needs.
Similar to the approach used for modifying memslots, make arch.msr_filter
a SRCU-protected pointer, do all the work configuring the new filter
outside of kvm->lock, and then acquire kvm->lock only when the new filter
has been vetted and created. That way vCPU readers either see the old
filter or the new filter in their entirety, not some half-baked state.
Yuan Yao pointed out a use-after-free in ksm_msr_allowed() due to a
TOCTOU bug, but that's just the tip of the iceberg...
- Nothing is __rcu annotated, making it nigh impossible to audit the
code for correctness.
- kvm_add_msr_filter() has an unpaired smp_wmb(). Violation of kernel
coding style aside, the lack of a smb_rmb() anywhere casts all code
into doubt.
- kvm_clear_msr_filter() has a double free TOCTOU bug, as it grabs
count before taking the lock.
- kvm_clear_msr_filter() also has memory leak due to the same TOCTOU bug.
The entire approach of updating the live filter is also flawed. While
installing a new filter is inherently racy if vCPUs are running, fixing
the above issues also makes it trivial to ensure certain behavior is
deterministic, e.g. KVM can provide deterministic behavior for MSRs with
identical settings in the old and new filters. An atomic update of the
filter also prevents KVM from getting into a half-baked state, e.g. if
installing a filter fails, the existing approach would leave the filter
in a half-baked state, having already committed whatever bits of the
filter were already processed.
[*] https://lkml.kernel.org/r/20210312083157.25403-1-yaoyuan0329os@gmail.com
Fixes: 1a155254ff ("KVM: x86: Introduce MSR filtering")
Cc: stable@vger.kernel.org
Cc: Alexander Graf <graf@amazon.com>
Reported-by: Yuan Yao <yaoyuan0329os@gmail.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20210316184436.2544875-2-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
kvm_cpu_accept_dm_intr and kvm_vcpu_ready_for_interrupt_injection are
a hodge-podge of conditions, hacked together to get something that
more or less works. But what is actually needed is much simpler;
in both cases the fundamental question is, do we have a place to stash
an interrupt if userspace does KVM_INTERRUPT?
In userspace irqchip mode, that is !vcpu->arch.interrupt.injected.
Currently kvm_event_needs_reinjection(vcpu) covers it, but it is
unnecessarily restrictive.
In split irqchip mode it's a bit more complicated, we need to check
kvm_apic_accept_pic_intr(vcpu) (the IRQ window exit is basically an INTACK
cycle and thus requires ExtINTs not to be masked) as well as
!pending_userspace_extint(vcpu). However, there is no need to
check kvm_event_needs_reinjection(vcpu), since split irqchip keeps
pending ExtINT state separate from event injection state, and checking
kvm_cpu_has_interrupt(vcpu) is wrong too since ExtINT has higher
priority than APIC interrupts. In fact the latter fixes a bug:
when userspace requests an IRQ window vmexit, an interrupt in the
local APIC can cause kvm_cpu_has_interrupt() to be true and thus
kvm_vcpu_ready_for_interrupt_injection() to return false. When this
happens, vcpu_run does not exit to userspace but the interrupt window
vmexits keep occurring. The VM loops without any hope of making progress.
Once we try to fix these with something like
return kvm_arch_interrupt_allowed(vcpu) &&
- !kvm_cpu_has_interrupt(vcpu) &&
- !kvm_event_needs_reinjection(vcpu) &&
- kvm_cpu_accept_dm_intr(vcpu);
+ (!lapic_in_kernel(vcpu)
+ ? !vcpu->arch.interrupt.injected
+ : (kvm_apic_accept_pic_intr(vcpu)
+ && !pending_userspace_extint(v)));
we realize two things. First, thanks to the previous patch the complex
conditional can reuse !kvm_cpu_has_extint(vcpu). Second, the interrupt
window request in vcpu_enter_guest()
bool req_int_win =
dm_request_for_irq_injection(vcpu) &&
kvm_cpu_accept_dm_intr(vcpu);
should be kept in sync with kvm_vcpu_ready_for_interrupt_injection():
it is unnecessary to ask the processor for an interrupt window
if we would not be able to return to userspace. Therefore,
kvm_cpu_accept_dm_intr(vcpu) is basically !kvm_cpu_has_extint(vcpu)
ANDed with the existing check for masked ExtINT. It all makes sense:
- we can accept an interrupt from userspace if there is a place
to stash it (and, for irqchip split, ExtINTs are not masked).
Interrupts from userspace _can_ be accepted even if right now
EFLAGS.IF=0.
- in order to tell userspace we will inject its interrupt ("IRQ
window open" i.e. kvm_vcpu_ready_for_interrupt_injection), both
KVM and the vCPU need to be ready to accept the interrupt.
... and this is what the patch implements.
Reported-by: David Woodhouse <dwmw@amazon.co.uk>
Analyzed-by: David Woodhouse <dwmw@amazon.co.uk>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Nikos Tsironis <ntsironis@arrikto.com>
Reviewed-by: David Woodhouse <dwmw@amazon.co.uk>
Tested-by: David Woodhouse <dwmw@amazon.co.uk>
SEV guests fail to boot on a system that supports the PCID feature.
While emulating the RSM instruction, KVM reads the guest CR3
and calls kvm_set_cr3(). If the vCPU is in the long mode,
kvm_set_cr3() does a sanity check for the CR3 value. In this case,
it validates whether the value has any reserved bits set. The
reserved bit range is 63:cpuid_maxphysaddr(). When AMD memory
encryption is enabled, the memory encryption bit is set in the CR3
value. The memory encryption bit may fall within the KVM reserved
bit range, causing the KVM emulation failure.
Introduce a new field cr3_lm_rsvd_bits in kvm_vcpu_arch which will
cache the reserved bits in the CR3 value. This will be initialized
to rsvd_bits(cpuid_maxphyaddr(vcpu), 63).
If the architecture has any special bits(like AMD SEV encryption bit)
that needs to be masked from the reserved bits, should be cleared
in vendor specific kvm_x86_ops.vcpu_after_set_cpuid handler.
Fixes: a780a3ea62 ("KVM: X86: Fix reserved bits check for MOV to CR3")
Signed-off-by: Babu Moger <babu.moger@amd.com>
Message-Id: <160521947657.32054.3264016688005356563.stgit@bmoger-ubuntu>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The TDP MMU must be able to allocate paging structure root pages and track
the usage of those pages. Implement a similar, but separate system for root
page allocation to that of the x86 shadow paging implementation. When
future patches add synchronization model changes to allow for parallel
page faults, these pages will need to be handled differently from the
x86 shadow paging based MMU's root pages.
Tested by running kvm-unit-tests and KVM selftests on an Intel Haswell
machine. This series introduced no new failures.
This series can be viewed in Gerrit at:
https://linux-review.googlesource.com/c/virt/kvm/kvm/+/2538
Signed-off-by: Ben Gardon <bgardon@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The TDP MMU offers an alternative mode of operation to the x86 shadow
paging based MMU, optimized for running an L1 guest with TDP. The TDP MMU
will require new fields that need to be initialized and torn down. Add
hooks into the existing KVM MMU initialization process to do that
initialization / cleanup. Currently the initialization and cleanup
fucntions do not do very much, however more operations will be added in
future patches.
Tested by running kvm-unit-tests and KVM selftests on an Intel Haswell
machine. This series introduced no new failures.
This series can be viewed in Gerrit at:
https://linux-review.googlesource.com/c/virt/kvm/kvm/+/2538
Signed-off-by: Ben Gardon <bgardon@google.com>
Message-Id: <20201014182700.2888246-4-bgardon@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
As vcpu->arch.cpuid_entries is now allocated dynamically, the only
remaining use for KVM_MAX_CPUID_ENTRIES is to check KVM_SET_CPUID/
KVM_SET_CPUID2 input for sanity. Since it was reported that the
current limit (80) is insufficient for some CPUs, bump
KVM_MAX_CPUID_ENTRIES and use an arbitrary value '256' as the new
limit.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20201001130541.1398392-4-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The current limit for guest CPUID leaves (KVM_MAX_CPUID_ENTRIES, 80)
is reported to be insufficient but before we bump it let's switch to
allocating vcpu->arch.cpuid_entries[] array dynamically. Currently,
'struct kvm_cpuid_entry2' is 40 bytes so vcpu->arch.cpuid_entries is
3200 bytes which accounts for 1/4 of the whole 'struct kvm_vcpu_arch'
but having it pre-allocated (for all vCPUs which we also pre-allocate)
gives us no real benefits.
Another plus of the dynamic allocation is that we now do kvm_check_cpuid()
check before we assign anything to vcpu->arch.cpuid_nent/cpuid_entries so
no changes are made in case the check fails.
Opportunistically remove unneeded 'out' labels from
kvm_vcpu_ioctl_set_cpuid()/kvm_vcpu_ioctl_set_cpuid2() and return
directly whenever possible.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20201001130541.1398392-3-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
KVM unconditionally provides PV features to the guest, regardless of the
configured CPUID. An unwitting guest that doesn't check
KVM_CPUID_FEATURES before use could access paravirt features that
userspace did not intend to provide. Fix this by checking the guest's
CPUID before performing any paravirtual operations.
Introduce a capability, KVM_CAP_ENFORCE_PV_FEATURE_CPUID, to gate the
aforementioned enforcement. Migrating a VM from a host w/o this patch to
a host with this patch could silently change the ABI exposed to the
guest, warranting that we default to the old behavior and opt-in for
the new one.
Reviewed-by: Jim Mattson <jmattson@google.com>
Reviewed-by: Peter Shier <pshier@google.com>
Signed-off-by: Oliver Upton <oupton@google.com>
Change-Id: I202a0926f65035b872bfe8ad15307c026de59a98
Message-Id: <20200818152429.1923996-4-oupton@google.com>
Reviewed-by: Wanpeng Li <wanpengli@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It's not desireable to have all MSRs always handled by KVM kernel space. Some
MSRs would be useful to handle in user space to either emulate behavior (like
uCode updates) or differentiate whether they are valid based on the CPU model.
To allow user space to specify which MSRs it wants to see handled by KVM,
this patch introduces a new ioctl to push filter rules with bitmaps into
KVM. Based on these bitmaps, KVM can then decide whether to reject MSR access.
With the addition of KVM_CAP_X86_USER_SPACE_MSR it can also deflect the
denied MSR events to user space to operate on.
If no filter is populated, MSR handling stays identical to before.
Signed-off-by: Alexander Graf <graf@amazon.com>
Message-Id: <20200925143422.21718-8-graf@amazon.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In the following commits we will add pieces of MSR filtering.
To ensure that code compiles even with the feature half-merged, let's add
a few stubs and struct definitions before the real patches start.
Signed-off-by: Alexander Graf <graf@amazon.com>
Message-Id: <20200925143422.21718-4-graf@amazon.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
MSRs are weird. Some of them are normal control registers, such as EFER.
Some however are registers that really are model specific, not very
interesting to virtualization workloads, and not performance critical.
Others again are really just windows into package configuration.
Out of these MSRs, only the first category is necessary to implement in
kernel space. Rarely accessed MSRs, MSRs that should be fine tunes against
certain CPU models and MSRs that contain information on the package level
are much better suited for user space to process. However, over time we have
accumulated a lot of MSRs that are not the first category, but still handled
by in-kernel KVM code.
This patch adds a generic interface to handle WRMSR and RDMSR from user
space. With this, any future MSR that is part of the latter categories can
be handled in user space.
Furthermore, it allows us to replace the existing "ignore_msrs" logic with
something that applies per-VM rather than on the full system. That way you
can run productive VMs in parallel to experimental ones where you don't care
about proper MSR handling.
Signed-off-by: Alexander Graf <graf@amazon.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
Message-Id: <20200925143422.21718-3-graf@amazon.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Rename the "shared_msrs" mechanism, which is used to defer restoring
MSRs that are only consumed when running in userspace, to a more banal
but less likely to be confusing "user_return_msrs".
The "shared" nomenclature is confusing as it's not obvious who is
sharing what, e.g. reasonable interpretations are that the guest value
is shared by vCPUs in a VM, or that the MSR value is shared/common to
guest and host, both of which are wrong.
"shared" is also misleading as the MSR value (in hardware) is not
guaranteed to be shared/reused between VMs (if that's indeed the correct
interpretation of the name), as the ability to share values between VMs
is simply a side effect (albiet a very nice side effect) of deferring
restoration of the host value until returning from userspace.
"user_return" avoids the above confusion by describing the mechanism
itself instead of its effects.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200923180409.32255-2-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Extend the kvm_exit tracepoint to align it with kvm_nested_vmexit in
terms of what information is captured. On SVM, add interrupt info and
error code, while on VMX it add IDT vectoring and error code. This
sets the stage for macrofying the kvm_exit tracepoint definition so that
it can be reused for kvm_nested_vmexit without loss of information.
Opportunistically stuff a zero for VM_EXIT_INTR_INFO if the VM-Enter
failed, as the field is guaranteed to be invalid. Note, it'd be
possible to further filter the interrupt/exception fields based on the
VM-Exit reason, but the helper is intended only for tracepoints, i.e.
an extra VMREAD or two is a non-issue, the failed VM-Enter case is just
low hanging fruit.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200923201349.16097-5-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Replace the existing kvm_x86_ops.need_emulation_on_page_fault() with a
more generic is_emulatable(), and unconditionally call the new function
in x86_emulate_instruction().
KVM will use the generic hook to support multiple security related
technologies that prevent emulation in one way or another. Similar to
the existing AMD #NPF case where emulation of the current instruction is
not possible due to lack of information, AMD's SEV-ES and Intel's SGX
and TDX will introduce scenarios where emulation is impossible due to
the guest's register state being inaccessible. And again similar to the
existing #NPF case, emulation can be initiated by kvm_mmu_page_fault(),
i.e. outside of the control of vendor-specific code.
While the cause and architecturally visible behavior of the various
cases are different, e.g. SGX will inject a #UD, AMD #NPF is a clean
resume or complete shutdown, and SEV-ES and TDX "return" an error, the
impact on the common emulation code is identical: KVM must stop
emulation immediately and resume the guest.
Query is_emulatable() in handle_ud() as well so that the
force_emulation_prefix code doesn't incorrectly modify RIP before
calling emulate_instruction() in the absurdly unlikely scenario that
KVM encounters forced emulation in conjunction with "do not emulate".
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200915232702.15945-1-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The 'flags' field of 'struct mmu_notifier_range' is used to indicate
whether invalidate_range_{start,end}() are permitted to block. In the
case of kvm_mmu_notifier_invalidate_range_start(), this field is not
forwarded on to the architecture-specific implementation of
kvm_unmap_hva_range() and therefore the backend cannot sensibly decide
whether or not to block.
Add an extra 'flags' parameter to kvm_unmap_hva_range() so that
architectures are aware as to whether or not they are permitted to block.
Cc: <stable@vger.kernel.org>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Suzuki K Poulose <suzuki.poulose@arm.com>
Cc: James Morse <james.morse@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
Message-Id: <20200811102725.7121-2-will@kernel.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Calculate the desired TDP level on the fly using the max TDP level and
MAXPHYADDR instead of doing the same when CPUID is updated. This avoids
the hidden dependency on cpuid_maxphyaddr() in vmx_get_tdp_level() and
also standardizes the "use 5-level paging iff MAXPHYADDR > 48" behavior
across x86.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200716034122.5998-8-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Use the shadow_root_level from the current MMU as the root level for the
PGD, i.e. for VMX's EPTP. This eliminates the weird dependency between
VMX and the MMU where both must independently calculate the same root
level for things to work correctly. Temporarily keep VMX's calculation
of the level and use it to WARN if the incoming level diverges.
Opportunistically refactor kvm_mmu_load_pgd() to avoid indentation hell,
and rename a 'cr3' param in the load_mmu_pgd prototype that managed to
survive the cr3 purge.
No functional change intended.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200716034122.5998-6-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch adds a new capability KVM_CAP_SMALLER_MAXPHYADDR which
allows userspace to query if the underlying architecture would
support GUEST_MAXPHYADDR < HOST_MAXPHYADDR and hence act accordingly
(e.g. qemu can decide if it should warn for -cpu ..,phys-bits=X)
The complications in this patch are due to unexpected (but documented)
behaviour we see with NPF vmexit handling in AMD processor. If
SVM is modified to add guest physical address checks in the NPF
and guest #PF paths, we see the followning error multiple times in
the 'access' test in kvm-unit-tests:
test pte.p pte.36 pde.p: FAIL: pte 2000021 expected 2000001
Dump mapping: address: 0x123400000000
------L4: 24c3027
------L3: 24c4027
------L2: 24c5021
------L1: 1002000021
This is because the PTE's accessed bit is set by the CPU hardware before
the NPF vmexit. This is handled completely by hardware and cannot be fixed
in software.
Therefore, availability of the new capability depends on a boolean variable
allow_smaller_maxphyaddr which is set individually by VMX and SVM init
routines. On VMX it's always set to true, on SVM it's only set to true
when NPT is not enabled.
CC: Tom Lendacky <thomas.lendacky@amd.com>
CC: Babu Moger <babu.moger@amd.com>
Signed-off-by: Mohammed Gamal <mgamal@redhat.com>
Message-Id: <20200710154811.418214-10-mgamal@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We would like to introduce a callback to update the #PF intercept
when CPUID changes. Just reuse update_bp_intercept since VMX is
already using update_exception_bitmap instead of a bespoke function.
While at it, remove an unnecessary assignment in the SVM version,
which is already done in the caller (kvm_arch_vcpu_ioctl_set_guest_debug)
and has nothing to do with the exception bitmap.
Reviewed-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Move x86's 'struct kvm_mmu_memory_cache' to common code in anticipation
of moving the entire x86 implementation code to common KVM and reusing
it for arm64 and MIPS. Add a new architecture specific asm/kvm_types.h
to control the existence and parameters of the struct. The new header
is needed to avoid a chicken-and-egg problem with asm/kvm_host.h as all
architectures define instances of the struct in their vCPU structs.
Add an asm-generic version of kvm_types.h to avoid having empty files on
PPC and s390 in the long term, and for arm64 and mips in the short term.
Suggested-by: Christoffer Dall <christoffer.dall@arm.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200703023545.8771-15-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add a gfp_zero flag to 'struct kvm_mmu_memory_cache' and use it to
control __GFP_ZERO instead of hardcoding a call to kmem_cache_zalloc().
A future patch needs such a flag for the __get_free_page() path, as
gfn arrays do not need/want the allocator to zero the memory. Convert
the kmem_cache paths to __GFP_ZERO now so as to avoid a weird and
inconsistent API in the future.
No functional change intended.
Reviewed-by: Ben Gardon <bgardon@google.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200703023545.8771-11-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The name of callback cpuid_update() is misleading that it's not about
updating CPUID settings of vcpu but updating the configurations of vcpu
based on the CPUIDs. So rename it to vcpu_after_set_cpuid().
Signed-off-by: Xiaoyao Li <xiaoyao.li@intel.com>
Message-Id: <20200709043426.92712-5-xiaoyao.li@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Explicitly pass the L2 GPA to kvm_arch_write_log_dirty(), which for all
intents and purposes is vmx_write_pml_buffer(), instead of having the
latter pull the GPA from vmcs.GUEST_PHYSICAL_ADDRESS. If the dirty bit
update is the result of KVM emulation (rare for L2), then the GPA in the
VMCS may be stale and/or hold a completely unrelated GPA.
Fixes: c5f983f6e8 ("nVMX: Implement emulated Page Modification Logging")
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200622215832.22090-2-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
