Move kvm_intr_is_single_vcpu() to lapic.c, drop its export, and make its
"fast" helper local to lapic.c. kvm_intr_is_single_vcpu() is only usable
if the local APIC is in-kernel, i.e. it most definitely belongs in the
local APIC code.
No functional change intended.
Fixes: cf04ec393e ("KVM: x86: Dedup AVIC vs. PI code for identifying target vCPU")
Link: https://lore.kernel.org/r/20250919003303.1355064-4-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM x86 CET virtualization support for 6.18
Add support for virtualizing Control-flow Enforcement Technology (CET) on
Intel (Shadow Stacks and Indirect Branch Tracking) and AMD (Shadow Stacks).
CET is comprised of two distinct features, Shadow Stacks (SHSTK) and Indirect
Branch Tracking (IBT), that can be utilized by software to help provide
Control-flow integrity (CFI). SHSTK defends against backward-edge attacks
(a.k.a. Return-oriented programming (ROP)), while IBT defends against
forward-edge attacks (a.k.a. similarly CALL/JMP-oriented programming (COP/JOP)).
Attackers commonly use ROP and COP/JOP methodologies to redirect the control-
flow to unauthorized targets in order to execute small snippets of code,
a.k.a. gadgets, of the attackers choice. By chaining together several gadgets,
an attacker can perform arbitrary operations and circumvent the system's
defenses.
SHSTK defends against backward-edge attacks, which execute gadgets by modifying
the stack to branch to the attacker's target via RET, by providing a second
stack that is used exclusively to track control transfer operations. The
shadow stack is separate from the data/normal stack, and can be enabled
independently in user and kernel mode.
When SHSTK is is enabled, CALL instructions push the return address on both the
data and shadow stack. RET then pops the return address from both stacks and
compares the addresses. If the return addresses from the two stacks do not
match, the CPU generates a Control Protection (#CP) exception.
IBT defends against backward-edge attacks, which branch to gadgets by executing
indirect CALL and JMP instructions with attacker controlled register or memory
state, by requiring the target of indirect branches to start with a special
marker instruction, ENDBRANCH. If an indirect branch is executed and the next
instruction is not an ENDBRANCH, the CPU generates a #CP. Note, ENDBRANCH
behaves as a NOP if IBT is disabled or unsupported.
From a virtualization perspective, CET presents several problems. While SHSTK
and IBT have two layers of enabling, a global control in the form of a CR4 bit,
and a per-feature control in user and kernel (supervisor) MSRs (U_CET and S_CET
respectively), the {S,U}_CET MSRs can be context switched via XSAVES/XRSTORS.
Practically speaking, intercepting and emulating XSAVES/XRSTORS is not a viable
option due to complexity, and outright disallowing use of XSTATE to context
switch SHSTK/IBT state would render the features unusable to most guests.
To limit the overall complexity without sacrificing performance or usability,
simply ignore the potential virtualization hole, but ensure that all paths in
KVM treat SHSTK/IBT as usable by the guest if the feature is supported in
hardware, and the guest has access to at least one of SHSTK or IBT. I.e. allow
userspace to advertise one of SHSTK or IBT if both are supported in hardware,
even though doing so would allow a misbehaving guest to use the unadvertised
feature.
Fully emulating SHSTK and IBT would also require significant complexity, e.g.
to track and update branch state for IBT, and shadow stack state for SHSTK.
Given that emulating large swaths of the guest code stream isn't necessary on
modern CPUs, punt on emulating instructions that meaningful impact or consume
SHSTK or IBT. However, instead of doing nothing, explicitly reject emulation
of such instructions so that KVM's emulator can't be abused to circumvent CET.
Disable support for SHSTK and IBT if KVM is configured such that emulation of
arbitrary guest instructions may be required, specifically if Unrestricted
Guest (Intel only) is disabled, or if KVM will emulate a guest.MAXPHYADDR that
is smaller than host.MAXPHYADDR.
Lastly disable SHSTK support if shadow paging is enabled, as the protections
for the shadow stack are novel (shadow stacks require Writable=0,Dirty=1, so
that they can't be directly modified by software), i.e. would require
non-trivial support in the Shadow MMU.
Note, AMD CPUs currently only support SHSTK. Explicitly disable IBT support
so that KVM doesn't over-advertise if AMD CPUs add IBT, and virtualizing IBT
in SVM requires KVM modifications.
KVM x86 changes for 6.18
- Don't (re)check L1 intercepts when completing userspace I/O to fix a flaw
where a misbehaving usersepace (a.k.a. syzkaller) could swizzle L1's
intercepts and trigger a variety of WARNs in KVM.
- Emulate PERF_CNTR_GLOBAL_STATUS_SET for PerfMonV2 guests, as the MSR is
supposed to exist for v2 PMUs.
- Allow Centaur CPU leaves (base 0xC000_0000) for Zhaoxin CPUs.
- Clean up KVM's vector hashing code for delivering lowest priority IRQs.
- Clean up the fastpath handler code to only handle IPIs and WRMSRs that are
actually "fast", as opposed to handling those that KVM _hopes_ are fast, and
in the process of doing so add fastpath support for TSC_DEADLINE writes on
AMD CPUs.
- Clean up a pile of PMU code in anticipation of adding support for mediated
vPMUs.
- Add support for the immediate forms of RDMSR and WRMSRNS, sans full
emulator support (KVM should never need to emulate the MSRs outside of
forced emulation and other contrived testing scenarios).
- Clean up the MSR APIs in preparation for CET and FRED virtualization, as
well as mediated vPMU support.
- Rejecting a fully in-kernel IRQCHIP if EOIs are protected, i.e. for TDX VMs,
as KVM can't faithfully emulate an I/O APIC for such guests.
- KVM_REQ_MSR_FILTER_CHANGED into a generic RECALC_INTERCEPTS in preparation
for mediated vPMU support, as KVM will need to recalculate MSR intercepts in
response to PMU refreshes for guests with mediated vPMUs.
- Misc cleanups and minor fixes.
KVM SVM changes for 6.18
- Require a minimum GHCB version of 2 when starting SEV-SNP guests via
KVM_SEV_INIT2 so that invalid GHCB versions result in immediate errors
instead of latent guest failures.
- Add support for Secure TSC for SEV-SNP guests, which prevents the untrusted
host from tampering with the guest's TSC frequency, while still allowing the
the VMM to configure the guest's TSC frequency prior to launch.
- Mitigate the potential for TOCTOU bugs when accessing GHCB fields by
wrapping all accesses via READ_ONCE().
- Validate the XCR0 provided by the guest (via the GHCB) to avoid tracking a
bogous XCR0 value in KVM's software model.
- Save an SEV guest's policy if and only if LAUNCH_START fully succeeds to
avoid leaving behind stale state (thankfully not consumed in KVM).
- Explicitly reject non-positive effective lengths during SNP's LAUNCH_UPDATE
instead of subtly relying on guest_memfd to do the "heavy" lifting.
- Reload the pre-VMRUN TSC_AUX on #VMEXIT for SEV-ES guests, not the host's
desired TSC_AUX, to fix a bug where KVM could clobber a different vCPU's
TSC_AUX due to hardware not matching the value cached in the user-return MSR
infrastructure.
- Enable AVIC by default for Zen4+ if x2AVIC (and other prereqs) is supported,
and clean up the AVIC initialization code along the way.
KVM x86 MMU changes for 6.18
- Recover possible NX huge pages within the TDP MMU under read lock to
reduce guest jitter when restoring NX huge pages.
- Return -EAGAIN during prefault if userspace concurrently deletes/moves the
relevant memslot to fix an issue where prefaulting could deadlock with the
memslot update.
- Don't retry in TDX's anti-zero-step mitigation if the target memslot is
invalid, i.e. is being deleted or moved, to fix a deadlock scenario similar
to the aforementioned prefaulting case.
Drop X86_CR4_CET from CR4_RESERVED_BITS and instead mark CET as reserved
if and only if IBT *and* SHSTK are unsupported, i.e. allow CR4.CET to be
set if IBT or SHSTK is supported. This creates a virtualization hole if
the CPU supports both IBT and SHSTK, but the kernel or vCPU model only
supports one of the features. However, it's entirely legal for a CPU to
have only one of IBT or SHSTK, i.e. the hole is a flaw in the architecture,
not in KVM.
More importantly, so long as KVM is careful to initialize and context
switch both IBT and SHSTK state (when supported in hardware) if either
feature is exposed to the guest, a misbehaving guest can only harm itself.
E.g. VMX initializes host CET VMCS fields based solely on hardware
capabilities.
Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Signed-off-by: Mathias Krause <minipli@grsecurity.net>
Tested-by: Mathias Krause <minipli@grsecurity.net>
Tested-by: John Allen <john.allen@amd.com>
Tested-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Signed-off-by: Chao Gao <chao.gao@intel.com>
[sean: split to separate patch, write changelog]
Reviewed-by: Binbin Wu <binbin.wu@linux.intel.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Link: https://lore.kernel.org/r/20250919223258.1604852-24-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Add PFERR_SS_MASK, a.k.a. Shadow Stack access, and WARN if KVM attempts to
check permissions for a Shadow Stack access as KVM hasn't been taught to
understand the magic Writable=0,Dirty=1 combination that is required for
Shadow Stack accesses, and likely will never learn. There are no plans to
support Shadow Stacks with the Shadow MMU, and the emulator rejects all
instructions that affect Shadow Stacks, i.e. it should be impossible for
KVM to observe a #PF due to a shadow stack access.
Reviewed-by: Binbin Wu <binbin.wu@linux.intel.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Link: https://lore.kernel.org/r/20250919223258.1604852-22-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Maintain per-guest valid XSS bits and check XSS validity against them
rather than against KVM capabilities. This is to prevent bits that are
supported by KVM but not supported for a guest from being set.
Opportunistically return KVM_MSR_RET_UNSUPPORTED on IA32_XSS MSR accesses
if guest CPUID doesn't enumerate X86_FEATURE_XSAVES. Since
KVM_MSR_RET_UNSUPPORTED takes care of host_initiated cases, drop the
host_initiated check.
Signed-off-by: Chao Gao <chao.gao@intel.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Reviewed-by: Binbin Wu <binbin.wu@linux.intel.com>
Link: https://lore.kernel.org/r/20250919223258.1604852-7-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Merge the queue of SVM changes for 6.18 to pick up the KVM-defined GHCB
helpers so that kvm_ghcb_get_xss() can be used to virtualize CET for
SEV-ES+ guests.
In the user return MSR support, the cached value is always the hardware
value of the specific MSR. Therefore, add a helper to retrieve the
cached value, which can replace the need for RDMSR, for example, to
allow SEV-ES guests to restore the correct host hardware value without
using RDMSR.
Cc: stable@vger.kernel.org
Signed-off-by: Hou Wenlong <houwenlong.hwl@antgroup.com>
[sean: drop "cache" from the name, make it a one-liner, tag for stable]
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Link: https://lore.kernel.org/r/20250923153738.1875174-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Use __kvm_set_xcr() to propagate XCR0 changes from the GHCB to KVM's
software model in order to validate the new XCR0 against KVM's view of
the supported XCR0. Allowing garbage is thankfully mostly benign, as
kvm_load_{guest,host}_xsave_state() bail early for vCPUs with protected
state, xstate_required_size() will simply provide garbage back to the
guest, and attempting to save/restore the bad value via KVM_{G,S}ET_XCRS
will only harm the guest (setting XCR0 will fail).
However, allowing the guest to put junk into a field that KVM assumes is
valid is a CVE waiting to happen. And as a bonus, using the proper API
eliminates the ugly open coding of setting arch.cpuid_dynamic_bits_dirty.
Simply ignore bad values, as either the guest managed to get an
unsupported value into hardware, or the guest is misbehaving and providing
pure garbage. In either case, KVM can't fix the broken guest.
Note, using __kvm_set_xcr() also avoids recomputing dynamic CPUID bits
if XCR0 isn't actually changing (relatively to KVM's previous snapshot).
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Fixes: 291bd20d5d ("KVM: SVM: Add initial support for a VMGEXIT VMEXIT")
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/r/20250919223258.1604852-4-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Rework the MSR_FILTER_CHANGED request into a more generic RECALC_INTERCEPTS
request, and expand the responsibilities of vendor code to recalculate all
intercepts that vary based on userspace input, e.g. instruction intercepts
that are tied to guest CPUID.
Providing a generic recalc request will allow the upcoming mediated PMU
support to trigger a recalc when PMU features, e.g. PERF_CAPABILITIES, are
set by userspace, without having to make multiple calls to/from PMU code.
As a bonus, using a request will effectively coalesce recalcs, e.g. will
reduce the number of recalcs for normal usage from 3+ to 1 (vCPU create,
set CPUID, set PERF_CAPABILITIES (Intel only), set filter).
The downside is that MSR filter changes that are done in isolation will do
a small amount of unnecessary work, but that's already a relatively slow
path, and the cost of recalculating instruction intercepts is negligible.
Tested-by: Xudong Hao <xudong.hao@intel.com>
Link: https://lore.kernel.org/r/20250806195706.1650976-25-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reject KVM_CREATE_IRQCHIP if the VM type has protected EOIs, i.e. if KVM
can't intercept EOI and thus can't faithfully emulate level-triggered
interrupts that are routed through the I/O APIC. For TDX VMs, the
TDX-Module owns the VMX EOI-bitmap and configures all IRQ vectors to have
the CPU accelerate EOIs, i.e. doesn't allow KVM to intercept any EOIs.
KVM already requires a split irqchip[1], but does so during vCPU creation,
which is both too late to allow userspace to fallback to a split irqchip
and a less-than-stellar experience for userspace since an -EINVAL on
KVM_VCPU_CREATE is far harder to debug/triage than failure exactly on
KVM_CREATE_IRQCHIP. And of course, allowing an action that ultimately
fails is arguably a bug regardless of the impact on userspace.
Link: https://lore.kernel.org/lkml/20250222014757.897978-11-binbin.wu@linux.intel.com [1]
Link: https://lore.kernel.org/lkml/aK3vZ5HuKKeFuuM4@google.com
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sagi Shahar <sagis@google.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Reviewed-by: Binbin Wu <binbin.wu@linux.intel.com>
Acked-by: Kai Huang <kai.huang@intel.com>
Link: https://lore.kernel.org/r/20250827011726.2451115-1-sagis@google.com
[sean: massage shortlog+changelog, relocate setting has_protected_eoi]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Rework kvm_mmu_max_mapping_level() to consult guest_memfd for all mappings,
not just private mappings, so that hugepage support plays nice with the
upcoming support for backing non-private memory with guest_memfd.
In addition to getting the max order from guest_memfd for gmem-only
memslots, update TDX's hook to effectively ignore shared mappings, as TDX's
restrictions on page size only apply to Secure EPT mappings. Do nothing
for SNP, as RMP restrictions apply to both private and shared memory.
Suggested-by: Ackerley Tng <ackerleytng@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Fuad Tabba <tabba@google.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Message-ID: <20250729225455.670324-16-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Enable KVM_GUEST_MEMFD for all KVM x86 64-bit builds, i.e. for "default"
VM types when running on 64-bit KVM. This will allow using guest_memfd
to back non-private memory for all VM shapes, by supporting mmap() on
guest_memfd.
Opportunistically clean up various conditionals that become tautologies
once x86 selects KVM_GUEST_MEMFD more broadly. Specifically, because
SW protected VMs, SEV, and TDX are all 64-bit only, private memory no
longer needs to take explicit dependencies on KVM_GUEST_MEMFD, because
it is effectively a prerequisite.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Fuad Tabba <tabba@google.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-ID: <20250729225455.670324-10-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Rename the Kconfig option CONFIG_KVM_PRIVATE_MEM to
CONFIG_KVM_GUEST_MEMFD. The original name implied that the feature only
supported "private" memory. However, CONFIG_KVM_PRIVATE_MEM enables
guest_memfd in general, which is not exclusively for private memory.
Subsequent patches in this series will add guest_memfd support for
non-CoCo VMs, whose memory is not private.
Renaming the Kconfig option to CONFIG_KVM_GUEST_MEMFD more accurately
reflects its broader scope as the main Kconfig option for all
guest_memfd-backed memory. This provides clearer semantics for the
option and avoids confusion as new features are introduced.
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Reviewed-by: Gavin Shan <gshan@redhat.com>
Reviewed-by: Shivank Garg <shivankg@amd.com>
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Co-developed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Fuad Tabba <tabba@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-ID: <20250729225455.670324-2-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Advertise support for the immediate form of MSR instructions to userspace
if the instructions are supported by the underlying CPU, and KVM is using
VMX, i.e. is running on an Intel-compatible CPU.
For SVM, explicitly clear X86_FEATURE_MSR_IMM to ensure KVM doesn't over-
report support if AMD-compatible CPUs ever implement the immediate forms,
as SVM will likely require explicit enablement in KVM.
Signed-off-by: Xin Li (Intel) <xin@zytor.com>
[sean: massage changelog]
Link: https://lore.kernel.org/r/20250805202224.1475590-7-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Add support for the immediate forms of RDMSR and WRMSRNS (currently
Intel-only). The immediate variants are only valid in 64-bit mode, and
use a single general purpose register for the data (the register is also
encoded in the instruction, i.e. not implicit like regular RDMSR/WRMSR).
The immediate variants are primarily motivated by performance, not code
size: by having the MSR index in an immediate, it is available *much*
earlier in the CPU pipeline, which allows hardware much more leeway about
how a particular MSR is handled.
Intel VMX support for the immediate forms of MSR accesses communicates
exit information to the host as follows:
1) The immediate form of RDMSR uses VM-Exit Reason 84.
2) The immediate form of WRMSRNS uses VM-Exit Reason 85.
3) For both VM-Exit reasons 84 and 85, the Exit Qualification field is
set to the MSR index that triggered the VM-Exit.
4) Bits 3 ~ 6 of the VM-Exit Instruction Information field are set to
the register encoding used by the immediate form of the instruction,
i.e. the destination register for RDMSR, and the source for WRMSRNS.
5) The VM-Exit Instruction Length field records the size of the
immediate form of the MSR instruction.
To deal with userspace RDMSR exits, stash the destination register in a
new kvm_vcpu_arch field, similar to cui_linear_rip, pio, etc.
Alternatively, the register could be saved in kvm_run.msr or re-retrieved
from the VMCS, but the former would require sanitizing the value to ensure
userspace doesn't clobber the value to an out-of-bounds index, and the
latter would require a new one-off kvm_x86_ops hook.
Don't bother adding support for the instructions in KVM's emulator, as the
only way for RDMSR/WRMSR to be encountered is if KVM is emulating large
swaths of code due to invalid guest state, and a vCPU cannot have invalid
guest state while in 64-bit mode.
Signed-off-by: Xin Li (Intel) <xin@zytor.com>
[sean: minor tweaks, massage and expand changelog]
Link: https://lore.kernel.org/r/20250805202224.1475590-5-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Calculate and track PMCs that are counting instructions/branches retired
when the PMC's event selector (or fixed counter control) is modified
instead evaluating the event selector on-demand. Immediately recalc a
PMC's configuration on writes to avoid false negatives/positives when
KVM skips an emulated WRMSR, which is guaranteed to occur before the
main run loop processes KVM_REQ_PMU.
Out of an abundance of caution, and because it's relatively cheap, recalc
reprogrammed PMCs in kvm_pmu_handle_event() as well. Recalculating in
response to KVM_REQ_PMU _should_ be unnecessary, but for now be paranoid
to avoid introducing easily-avoidable bugs in edge cases. The code can be
removed in the future if necessary, e.g. in the unlikely event that the
overhead of recalculating to-be-emulated PMCs is noticeable.
Note! Deliberately don't check the PMU event filters, as doing so could
result in KVM consuming stale information.
Tracking which PMCs are counting branches/instructions will allow grabbing
SRCU in the fastpath VM-Exit handlers if and only if a PMC event might be
triggered (to consult the event filters), and will also allow the upcoming
mediated PMU to do the right thing with respect to counting instructions
(the mediated PMU won't be able to update PMCs in the VM-Exit fastpath).
Reviewed-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
Link: https://lore.kernel.org/r/20250805190526.1453366-12-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Track possible NX huge pages for the TDP MMU separately from Shadow MMUs
in anticipation of doing recovery for the TDP MMU while holding mmu_lock
for read instead of write.
Use a small structure to hold the list of pages along with the number of
pages/entries in the list, as relying on kvm->stat.nx_lpage_splits to
calculating the number of pages to recover would result in over-zapping
when both TDP and Shadow MMUs are active.
Suggested-by: Sean Christopherson <seanjc@google.com>
Suggested-by: David Matlack <dmatlack@google.com>
Signed-off-by: Vipin Sharma <vipinsh@google.com>
Co-developed-by: James Houghton <jthoughton@google.com>
Signed-off-by: James Houghton <jthoughton@google.com>
Link: https://lore.kernel.org/r/20250707224720.4016504-2-jthoughton@google.com
[sean: rewrite changelog, use #ifdef instead of dummy KVM_TDP_MMU #define]
Signed-off-by: Sean Christopherson <seanjc@google.com>
KVM x86 MMU changes for 6.17
- Exempt nested EPT from the the !USER + CR0.WP logic, as EPT doesn't interact
with CR0.WP.
- Move the TDX hardware setup code to tdx.c to better co-locate TDX code
and eliminate a few global symbols.
- Dynamically allocation the shadow MMU's hashed page list, and defer
allocating the hashed list until it's actually needed (the TDP MMU doesn't
use the list).
KVM x86 misc changes for 6.17
- Prevert the host's DEBUGCTL.FREEZE_IN_SMM (Intel only) when running the
guest. Failure to honor FREEZE_IN_SMM can bleed host state into the guest.
- Explicitly check vmcs12.GUEST_DEBUGCTL on nested VM-Enter (Intel only) to
prevent L1 from running L2 with features that KVM doesn't support, e.g. BTF.
- Intercept SPEC_CTRL on AMD if the MSR shouldn't exist according to the
vCPU's CPUID model.
- Rework the MSR interception code so that the SVM and VMX APIs are more or
less identical.
- Recalculate all MSR intercepts from the "source" on MSR filter changes, and
drop the dedicated "shadow" bitmaps (and their awful "max" size defines).
- WARN and reject loading kvm-amd.ko instead of panicking the kernel if the
nested SVM MSRPM offsets tracker can't handle an MSR.
- Advertise support for LKGS (Load Kernel GS base), a new instruction that's
loosely related to FRED, but is supported and enumerated independently.
- Fix a user-triggerable WARN that syzkaller found by stuffing INIT_RECEIVED,
a.k.a. WFS, and then putting the vCPU into VMX Root Mode (post-VMXON). Use
the same approach KVM uses for dealing with "impossible" emulation when
running a !URG guest, and simply wait until KVM_RUN to detect that the vCPU
has architecturally impossible state.
- Add KVM_X86_DISABLE_EXITS_APERFMPERF to allow disabling interception of
APERF/MPERF reads, so that a "properly" configured VM can "virtualize"
APERF/MPERF (with many caveats).
- Reject KVM_SET_TSC_KHZ if vCPUs have been created, as changing the "default"
frequency is unsupported for VMs with a "secure" TSC, and there's no known
use case for changing the default frequency for other VM types.
KVM VFIO device assignment cleanups for 6.17
Kill off kvm_arch_{start,end}_assignment() and x86's associated tracking now
that KVM no longer uses assigned_device_count as a bad heuristic for "VM has
an irqbypass producer" or for "VM has access to host MMIO".
KVM MMIO Stale Data mitigation cleanup for 6.17
Rework KVM's mitigation for the MMIO State Data vulnerability to track
whether or not a vCPU has access to (host) MMIO based on the MMU that will be
used when running in the guest. The current approach doesn't actually detect
whether or not a guest has access to MMIO, and is prone to false negatives (and
to a lesser extent, false positives), as KVM_DEV_VFIO_FILE_ADD is optional, and
obviously only covers VFIO devices.
KVM IRQ changes for 6.17
- Rework irqbypass to track/match producers and consumers via an xarray
instead of a linked list. Using a linked list leads to O(n^2) insertion
times, which is hugely problematic for use cases that create large numbers
of VMs. Such use cases typically don't actually use irqbypass, but
eliminating the pointless registration is a future problem to solve as it
likely requires new uAPI.
- Track irqbypass's "token" as "struct eventfd_ctx *" instead of a "void *",
to avoid making a simple concept unnecessarily difficult to understand.
- Add CONFIG_KVM_IOAPIC for x86 to allow disabling support for I/O APIC, PIC,
and PIT emulation at compile time.
- Drop x86's irq_comm.c, and move a pile of IRQ related code into irq.c.
- Fix a variety of flaws and bugs in the AVIC device posted IRQ code.
- Inhibited AVIC if a vCPU's ID is too big (relative to what hardware
supports) instead of rejecting vCPU creation.
- Extend enable_ipiv module param support to SVM, by simply leaving IsRunning
clear in the vCPU's physical ID table entry.
- Disable IPI virtualization, via enable_ipiv, if the CPU is affected by
erratum #1235, to allow (safely) enabling AVIC on such CPUs.
- Dedup x86's device posted IRQ code, as the vast majority of functionality
can be shared verbatime between SVM and VMX.
- Harden the device posted IRQ code against bugs and runtime errors.
- Use vcpu_idx, not vcpu_id, for GA log tag/metadata, to make lookups O(1)
instead of O(n).
- Generate GA Log interrupts if and only if the target vCPU is blocking, i.e.
only if KVM needs a notification in order to wake the vCPU.
- Decouple device posted IRQs from VFIO device assignment, as binding a VM to
a VFIO group is not a requirement for enabling device posted IRQs.
- Clean up and document/comment the irqfd assignment code.
- Disallow binding multiple irqfds to an eventfd with a priority waiter, i.e.
ensure an eventfd is bound to at most one irqfd through the entire host,
and add a selftest to verify eventfd:irqfd bindings are globally unique.
Pull KVM fixes from Paolo Bonzini:
"Many patches, pretty much all of them small, that accumulated while I
was on vacation.
ARM:
- Remove the last leftovers of the ill-fated FPSIMD host state
mapping at EL2 stage-1
- Fix unexpected advertisement to the guest of unimplemented S2 base
granule sizes
- Gracefully fail initialising pKVM if the interrupt controller isn't
GICv3
- Also gracefully fail initialising pKVM if the carveout allocation
fails
- Fix the computing of the minimum MMIO range required for the host
on stage-2 fault
- Fix the generation of the GICv3 Maintenance Interrupt in nested
mode
x86:
- Reject SEV{-ES} intra-host migration if one or more vCPUs are
actively being created, so as not to create a non-SEV{-ES} vCPU in
an SEV{-ES} VM
- Use a pre-allocated, per-vCPU buffer for handling de-sparsification
of vCPU masks in Hyper-V hypercalls; fixes a "stack frame too
large" issue
- Allow out-of-range/invalid Xen event channel ports when configuring
IRQ routing, to avoid dictating a specific ioctl() ordering to
userspace
- Conditionally reschedule when setting memory attributes to avoid
soft lockups when userspace converts huge swaths of memory to/from
private
- Add back MWAIT as a required feature for the MONITOR/MWAIT selftest
- Add a missing field in struct sev_data_snp_launch_start that
resulted in the guest-visible workarounds field being filled at the
wrong offset
- Skip non-canonical address when processing Hyper-V PV TLB flushes
to avoid VM-Fail on INVVPID
- Advertise supported TDX TDVMCALLs to userspace
- Pass SetupEventNotifyInterrupt arguments to userspace
- Fix TSC frequency underflow"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: x86: avoid underflow when scaling TSC frequency
KVM: arm64: Remove kvm_arch_vcpu_run_map_fp()
KVM: arm64: Fix handling of FEAT_GTG for unimplemented granule sizes
KVM: arm64: Don't free hyp pages with pKVM on GICv2
KVM: arm64: Fix error path in init_hyp_mode()
KVM: arm64: Adjust range correctly during host stage-2 faults
KVM: arm64: nv: Fix MI line level calculation in vgic_v3_nested_update_mi()
KVM: x86/hyper-v: Skip non-canonical addresses during PV TLB flush
KVM: SVM: Add missing member in SNP_LAUNCH_START command structure
Documentation: KVM: Fix unexpected unindent warnings
KVM: selftests: Add back the missing check of MONITOR/MWAIT availability
KVM: Allow CPU to reschedule while setting per-page memory attributes
KVM: x86/xen: Allow 'out of range' event channel ports in IRQ routing table.
KVM: x86/hyper-v: Use preallocated per-vCPU buffer for de-sparsified vCPU masks
KVM: SVM: Initialize vmsa_pa in VMCB to INVALID_PAGE if VMSA page is NULL
KVM: SVM: Reject SEV{-ES} intra host migration if vCPU creation is in-flight
KVM: TDX: Report supported optional TDVMCALLs in TDX capabilities
KVM: TDX: Exit to userspace for SetupEventNotifyInterrupt
Pull CPU speculation fixes from Borislav Petkov:
"Add the mitigation logic for Transient Scheduler Attacks (TSA)
TSA are new aspeculative side channel attacks related to the execution
timing of instructions under specific microarchitectural conditions.
In some cases, an attacker may be able to use this timing information
to infer data from other contexts, resulting in information leakage.
Add the usual controls of the mitigation and integrate it into the
existing speculation bugs infrastructure in the kernel"
* tag 'tsa_x86_bugs_for_6.16' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/process: Move the buffer clearing before MONITOR
x86/microcode/AMD: Add TSA microcode SHAs
KVM: SVM: Advertise TSA CPUID bits to guests
x86/bugs: Add a Transient Scheduler Attacks mitigation
x86/bugs: Rename MDS machinery to something more generic
Drop kvm_arch_{start,end}_assignment() and all associated code now that
KVM x86 no longer consumes assigned_device_count. Tracking whether or not
a VFIO-assigned device is formally associated with a VM is fundamentally
flawed, as such an association is optional for general usage, i.e. is prone
to false negatives. E.g. prior to commit 2edd9cb79f ("kvm: detect
assigned device via irqbypass manager"), device passthrough via VFIO would
fail to enable IRQ bypass if userspace omitted the formal VFIO<=>KVM
binding.
And device drivers that *need* the VFIO<=>KVM connection, e.g. KVM-GT,
shouldn't be relying on generic x86 tracking infrastructure.
Cc: Jim Mattson <jmattson@google.com>
Link: https://lore.kernel.org/r/20250523011756.3243624-6-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Enforce the MMIO State Data mitigation if KVM has ever mapped host MMIO
into the VM, not if the VM has an assigned device. VFIO is but one of
many ways to map host MMIO into a KVM guest, and even within VFIO,
formally attaching a device to a VM via KVM_DEV_VFIO_FILE_ADD is entirely
optional.
Track whether or not the guest can access host MMIO on a per-MMU basis,
i.e. based on whether or not the vCPU has a mapping to host MMIO. For
simplicity, track MMIO mappings in "special" rools (those without a
kvm_mmu_page) at the VM level, as only Intel CPUs are vulnerable, and so
only legacy 32-bit shadow paging is affected, i.e. lack of precise
tracking is a complete non-issue.
Make the per-MMU and per-VM flags sticky. Detecting when *all* MMIO
mappings have been removed would be absurdly complex. And in practice,
removing MMIO from a guest will be done by deleting the associated memslot,
which by default will force KVM to re-allocate all roots. Special roots
will forever be mitigated, but as above, the affected scenarios are not
expected to be performance sensitive.
Use a VMX_RUN flag to communicate the need for a buffers flush to
vmx_vcpu_enter_exit() so that kvm_vcpu_can_access_host_mmio() and all its
dependencies don't need to be marked __always_inline, e.g. so that KASAN
doesn't trigger a noinstr violation.
Cc: Pawan Gupta <pawan.kumar.gupta@linux.intel.com>
Cc: Borislav Petkov <bp@alien8.de>
Fixes: 8cb861e9e3 ("x86/speculation/mmio: Add mitigation for Processor MMIO Stale Data")
Tested-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com>
Link: https://lore.kernel.org/r/20250523011756.3243624-4-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Allocate VM structs via kvzalloc(), i.e. try to use a contiguous physical
allocation before falling back to __vmalloc(), to avoid the overhead of
establishing the virtual mappings. For non-debug builds, The SVM and VMX
(and TDX) structures are now just below 7000 bytes in the worst case
scenario (see below), i.e. are order-1 allocations, and will likely remain
that way for quite some time.
Add compile-time assertions in vendor code to ensure the size of the
structures, sans the memslot hash tables, are order-0 allocations, i.e.
are less than 4KiB. There's nothing fundamentally wrong with a larger
kvm_{svm,vmx,tdx} size, but given that the size of the structure (without
the memslots hash tables) is below 2KiB after 18+ years of existence,
more than doubling the size would be quite notable.
Add sanity checks on the memslot hash table sizes, partly to ensure they
aren't resized without accounting for the impact on VM structure size, and
partly to document that the majority of the size of VM structures comes
from the memslots.
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Link: https://lore.kernel.org/r/20250523001138.3182794-4-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Dynamically allocate the (massive) array of hashed lists used to track
shadow pages, as the array itself is 32KiB, i.e. is an order-3 allocation
all on its own, and is *exactly* an order-3 allocation. Dynamically
allocating the array will allow allocating "struct kvm" using kvmalloc(),
and will also allow deferring allocation of the array until it's actually
needed, i.e. until the first shadow root is allocated.
Opportunistically use kvmalloc() for the hashed lists, as an order-3
allocation is (stating the obvious) less likely to fail than an order-4
allocation, and the overhead of vmalloc() is undesirable given that the
size of the allocation is fixed.
Cc: Vipin Sharma <vipinsh@google.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Link: https://lore.kernel.org/r/20250523001138.3182794-3-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Use a preallocated per-vCPU bitmap for tracking the unpacked set of vCPUs
being targeted for Hyper-V's paravirt TLB flushing. If KVM_MAX_NR_VCPUS
is set to 4096 (which is allowed even for MAXSMP=n builds), putting the
vCPU mask on-stack pushes kvm_hv_flush_tlb() past the default FRAME_WARN
limit.
arch/x86/kvm/hyperv.c:2001:12: error: stack frame size (1288) exceeds limit (1024)
in 'kvm_hv_flush_tlb' [-Werror,-Wframe-larger-than]
2001 | static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc)
| ^
1 error generated.
Note, sparse_banks was given the same treatment by commit 7d5e88d301
("KVM: x86: hyper-v: Use preallocated buffer in 'struct kvm_vcpu_hv'
instead of on-stack 'sparse_banks'"), for the exact same reason.
Reported-by: Abinash Lalotra <abinashsinghlalotra@gmail.com>
Closes: https://lore.kernel.org/all/20250613111023.786265-1-abinashsinghlalotra@gmail.com
Link: https://lore.kernel.org/all/aEylI-O8kFnFHrOH@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Rename kvm_set_msi_irq() to kvm_msi_to_lapic_irq() to better capture what
it actually does, e.g. it's _really_ easy to conflate kvm_set_msi_irq()
with kvm_set_msi().
Opportunistically delete the public declaration and export, as they are
no longer used/needed.
No functional change intended.
Link: https://lore.kernel.org/r/20250611224604.313496-64-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Use a dedicated counter to track the number of IRQs that can utilize IRQ
bypass instead of piggybacking the assigned device count. As evidenced by
commit 2edd9cb79f ("kvm: detect assigned device via irqbypass manager"),
it's possible for a device to be able to post IRQs to a vCPU without said
device being assigned to a VM.
Leave the calls to kvm_arch_{start,end}_assignment() alone for the moment
to avoid regressing the MMIO stale data mitigation. KVM is abusing the
assigned device count when applying mmio_stale_data_clear, and it's not at
all clear if vDPA devices rely on this behavior. This will hopefully be
cleaned up in the future, as the number of assigned devices is a terrible
heuristic for detecting if a VM has access to host MMIO.
Link: https://lore.kernel.org/r/20250611224604.313496-55-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Move enable_ipiv to common x86 so that it can be reused by SVM to control
IPI virtualization when AVIC is enabled. SVM doesn't actually provide a
way to truly disable IPI virtualization, but KVM can get close enough by
skipping the necessary table programming.
Link: https://lore.kernel.org/r/20250611224604.313496-18-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Inhibit AVIC with a new "ID too big" flag if userspace creates a vCPU with
an ID that is too big, but otherwise allow vCPU creation to succeed.
Rejecting KVM_CREATE_VCPU with EINVAL violates KVM's ABI as KVM advertises
that the max vCPU ID is 4095, but disallows creating vCPUs with IDs bigger
than 254 (AVIC) or 511 (x2AVIC).
Alternatively, KVM could advertise an accurate value depending on which
AVIC mode is in use, but that wouldn't really solve the underlying problem,
e.g. would be a breaking change if KVM were to ever try and enable AVIC or
x2AVIC by default.
Cc: Maxim Levitsky <mlevitsk@redhat.com>
Tested-by: Sairaj Kodilkar <sarunkod@amd.com>
Link: https://lore.kernel.org/r/20250611224604.313496-14-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
When updating IRTEs in response to a GSI routing or IRQ bypass change,
pass the new/current routing information along with the associated irqfd.
This will allow KVM x86 to harden, simplify, and deduplicate its code.
Since adding/removing a bypass producer is now conveniently protected with
irqfds.lock, i.e. can't run concurrently with kvm_irq_routing_update(),
use the routing information cached in the irqfd instead of looking up
the information in the current GSI routing tables.
Opportunistically convert an existing printk() to pr_info() and put its
string onto a single line (old code that strictly adhered to 80 chars).
Link: https://lore.kernel.org/r/20250611224604.313496-5-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Add a Kconfig to allow building KVM without support for emulating a I/O
APIC, PIC, and PIT, which is desirable for deployments that effectively
don't support a fully in-kernel IRQ chip, i.e. never expect any VMM to
create an in-kernel I/O APIC. E.g. compiling out support eliminates a few
thousand lines of guest-facing code and gives security folks warm fuzzies.
As a bonus, wrapping relevant paths with CONFIG_KVM_IOAPIC #ifdefs makes
it much easier for readers to understand which bits and pieces exist
specifically for fully in-kernel IRQ chips.
Opportunistically convert all two in-kernel uses of __KVM_HAVE_IOAPIC to
CONFIG_KVM_IOAPIC, e.g. rather than add a second #ifdef to generate a stub
for kvm_arch_post_irq_routing_update().
Acked-by: Kai Huang <kai.huang@intel.com>
Link: https://lore.kernel.org/r/20250611213557.294358-15-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Don't bother clearing the PIT's IRQ line status when destroying the PIT,
as userspace can't possibly rely on KVM to lower the IRQ line in any sane
use case, and it's not at all obvious that clearing the PIT's IRQ line is
correct/desirable in kvm_create_pit()'s error path.
When called from kvm_arch_pre_destroy_vm(), the entire VM is being torn
down and thus {kvm_pic,kvm_ioapic}.irq_states are unreachable.
As for the error path in kvm_create_pit(), the only way the PIT's bit in
irq_states can be set is if userspace raises the associated IRQ before
KVM_CREATE_PIT{2} completes. Forcefully clearing the bit would clobber
userspace's input, nonsensical though that input may be. Not to mention
that no known VMM will continue on if PIT creation fails.
Acked-by: Kai Huang <kai.huang@intel.com>
Link: https://lore.kernel.org/r/20250611213557.294358-12-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Hardcode the PIT's source IRQ ID to '2' instead of "finding" that bit 2
is always the first available bit in irq_sources_bitmap. Bits 0 and 1 are
set/reserved by kvm_arch_init_vm(), i.e. long before kvm_create_pit() can
be invoked, and KVM allows at most one in-kernel PIT instance, i.e. it's
impossible for the PIT to find a different free bit (there are no other
users of kvm_request_irq_source_id().
Delete the now-defunct irq_sources_bitmap and all its associated code.
Acked-by: Kai Huang <kai.huang@intel.com>
Link: https://lore.kernel.org/r/20250611213557.294358-11-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Drop the superfluous and confusing kvm_set_pic_irq() => kvm_pic_set_irq()
wrapper, and instead wire up ->set() directly to its final destination.
Opportunistically move the declaration kvm_pic_set_irq() to irq.h to
start gathering more of the in-kernel APIC/IO-APIC logic in irq.{c,h}.
No functional change intended.
Acked-by: Kai Huang <kai.huang@intel.com>
Link: https://lore.kernel.org/r/20250611213557.294358-3-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
