KVM fixes for 6.14 part 1
- Reject Hyper-V SEND_IPI hypercalls if the local APIC isn't being emulated
by KVM to fix a NULL pointer dereference.
- Enter guest mode (L2) from KVM's perspective before initializing the vCPU's
nested NPT MMU so that the MMU is properly tagged for L2, not L1.
- Load the guest's DR6 outside of the innermost .vcpu_run() loop, as the
guest's value may be stale if a VM-Exit is handled in the fastpath.
The kernel's initcall infrastructure lacks the ability to express
dependencies between initcalls, whereas the modules infrastructure
automatically handles dependencies via symbol loading. Ensure the
PSP SEV driver is initialized before proceeding in sev_hardware_setup()
if KVM is built-in as the dependency isn't handled by the initcall
infrastructure.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Ashish Kalra <ashish.kalra@amd.com>
Message-ID: <f78ddb64087df27e7bcb1ae0ab53f55aa0804fab.1739226950.git.ashish.kalra@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Move the conditional loading of hardware DR6 with the guest's DR6 value
out of the core .vcpu_run() loop to fix a bug where KVM can load hardware
with a stale vcpu->arch.dr6.
When the guest accesses a DR and host userspace isn't debugging the guest,
KVM disables DR interception and loads the guest's values into hardware on
VM-Enter and saves them on VM-Exit. This allows the guest to access DRs
at will, e.g. so that a sequence of DR accesses to configure a breakpoint
only generates one VM-Exit.
For DR0-DR3, the logic/behavior is identical between VMX and SVM, and also
identical between KVM_DEBUGREG_BP_ENABLED (userspace debugging the guest)
and KVM_DEBUGREG_WONT_EXIT (guest using DRs), and so KVM handles loading
DR0-DR3 in common code, _outside_ of the core kvm_x86_ops.vcpu_run() loop.
But for DR6, the guest's value doesn't need to be loaded into hardware for
KVM_DEBUGREG_BP_ENABLED, and SVM provides a dedicated VMCB field whereas
VMX requires software to manually load the guest value, and so loading the
guest's value into DR6 is handled by {svm,vmx}_vcpu_run(), i.e. is done
_inside_ the core run loop.
Unfortunately, saving the guest values on VM-Exit is initiated by common
x86, again outside of the core run loop. If the guest modifies DR6 (in
hardware, when DR interception is disabled), and then the next VM-Exit is
a fastpath VM-Exit, KVM will reload hardware DR6 with vcpu->arch.dr6 and
clobber the guest's actual value.
The bug shows up primarily with nested VMX because KVM handles the VMX
preemption timer in the fastpath, and the window between hardware DR6
being modified (in guest context) and DR6 being read by guest software is
orders of magnitude larger in a nested setup. E.g. in non-nested, the
VMX preemption timer would need to fire precisely between #DB injection
and the #DB handler's read of DR6, whereas with a KVM-on-KVM setup, the
window where hardware DR6 is "dirty" extends all the way from L1 writing
DR6 to VMRESUME (in L1).
L1's view:
==========
<L1 disables DR interception>
CPU 0/KVM-7289 [023] d.... 2925.640961: kvm_entry: vcpu 0
A: L1 Writes DR6
CPU 0/KVM-7289 [023] d.... 2925.640963: <hack>: Set DRs, DR6 = 0xffff0ff1
B: CPU 0/KVM-7289 [023] d.... 2925.640967: kvm_exit: vcpu 0 reason EXTERNAL_INTERRUPT intr_info 0x800000ec
D: L1 reads DR6, arch.dr6 = 0
CPU 0/KVM-7289 [023] d.... 2925.640969: <hack>: Sync DRs, DR6 = 0xffff0ff0
CPU 0/KVM-7289 [023] d.... 2925.640976: kvm_entry: vcpu 0
L2 reads DR6, L1 disables DR interception
CPU 0/KVM-7289 [023] d.... 2925.640980: kvm_exit: vcpu 0 reason DR_ACCESS info1 0x0000000000000216
CPU 0/KVM-7289 [023] d.... 2925.640983: kvm_entry: vcpu 0
CPU 0/KVM-7289 [023] d.... 2925.640983: <hack>: Set DRs, DR6 = 0xffff0ff0
L2 detects failure
CPU 0/KVM-7289 [023] d.... 2925.640987: kvm_exit: vcpu 0 reason HLT
L1 reads DR6 (confirms failure)
CPU 0/KVM-7289 [023] d.... 2925.640990: <hack>: Sync DRs, DR6 = 0xffff0ff0
L0's view:
==========
L2 reads DR6, arch.dr6 = 0
CPU 23/KVM-5046 [001] d.... 3410.005610: kvm_exit: vcpu 23 reason DR_ACCESS info1 0x0000000000000216
CPU 23/KVM-5046 [001] ..... 3410.005610: kvm_nested_vmexit: vcpu 23 reason DR_ACCESS info1 0x0000000000000216
L2 => L1 nested VM-Exit
CPU 23/KVM-5046 [001] ..... 3410.005610: kvm_nested_vmexit_inject: reason: DR_ACCESS ext_inf1: 0x0000000000000216
CPU 23/KVM-5046 [001] d.... 3410.005610: kvm_entry: vcpu 23
CPU 23/KVM-5046 [001] d.... 3410.005611: kvm_exit: vcpu 23 reason VMREAD
CPU 23/KVM-5046 [001] d.... 3410.005611: kvm_entry: vcpu 23
CPU 23/KVM-5046 [001] d.... 3410.005612: kvm_exit: vcpu 23 reason VMREAD
CPU 23/KVM-5046 [001] d.... 3410.005612: kvm_entry: vcpu 23
L1 writes DR7, L0 disables DR interception
CPU 23/KVM-5046 [001] d.... 3410.005612: kvm_exit: vcpu 23 reason DR_ACCESS info1 0x0000000000000007
CPU 23/KVM-5046 [001] d.... 3410.005613: kvm_entry: vcpu 23
L0 writes DR6 = 0 (arch.dr6)
CPU 23/KVM-5046 [001] d.... 3410.005613: <hack>: Set DRs, DR6 = 0xffff0ff0
A: <L1 writes DR6 = 1, no interception, arch.dr6 is still '0'>
B: CPU 23/KVM-5046 [001] d.... 3410.005614: kvm_exit: vcpu 23 reason PREEMPTION_TIMER
CPU 23/KVM-5046 [001] d.... 3410.005614: kvm_entry: vcpu 23
C: L0 writes DR6 = 0 (arch.dr6)
CPU 23/KVM-5046 [001] d.... 3410.005614: <hack>: Set DRs, DR6 = 0xffff0ff0
L1 => L2 nested VM-Enter
CPU 23/KVM-5046 [001] d.... 3410.005616: kvm_exit: vcpu 23 reason VMRESUME
L0 reads DR6, arch.dr6 = 0
Reported-by: John Stultz <jstultz@google.com>
Closes: https://lkml.kernel.org/r/CANDhNCq5_F3HfFYABqFGCA1bPd_%2BxgNj-iDQhH4tDk%2Bwi8iZZg%40mail.gmail.com
Fixes: 375e28ffc0 ("KVM: X86: Set host DR6 only on VMX and for KVM_DEBUGREG_WONT_EXIT")
Fixes: d67668e9dd ("KVM: x86, SVM: isolate vcpu->arch.dr6 from vmcb->save.dr6")
Cc: stable@vger.kernel.org
Cc: Jim Mattson <jmattson@google.com>
Tested-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/r/20250125011833.3644371-1-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
When preparing vmcb02 for nested VMRUN (or state restore), "enter" guest
mode prior to initializing the MMU for nested NPT so that guest_mode is
set in the MMU's role. KVM's model is that all L2 MMUs are tagged with
guest_mode, as the behavior of hypervisor MMUs tends to be significantly
different than kernel MMUs.
Practically speaking, the bug is relatively benign, as KVM only directly
queries role.guest_mode in kvm_mmu_free_guest_mode_roots() and
kvm_mmu_page_ad_need_write_protect(), which SVM doesn't use, and in paths
that are optimizations (mmu_page_zap_pte() and
shadow_mmu_try_split_huge_pages()).
And while the role is incorprated into shadow page usage, because nested
NPT requires KVM to be using NPT for L1, reusing shadow pages across L1
and L2 is impossible as L1 MMUs will always have direct=1, while L2 MMUs
will have direct=0.
Hoist the TLB processing and setting of HF_GUEST_MASK to the beginning
of the flow instead of forcing guest_mode in the MMU, as nothing in
nested_vmcb02_prepare_control() between the old and new locations touches
TLB flush requests or HF_GUEST_MASK, i.e. there's no reason to present
inconsistent vCPU state to the MMU.
Fixes: 69cb877487 ("KVM: nSVM: move MMU setup to nested_prepare_vmcb_control")
Cc: stable@vger.kernel.org
Reported-by: Yosry Ahmed <yosry.ahmed@linux.dev>
Reviewed-by: Yosry Ahmed <yosry.ahmed@linux.dev>
Link: https://lore.kernel.org/r/20250130010825.220346-1-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Advertise support for Hyper-V's SEND_IPI and SEND_IPI_EX hypercalls if and
only if the local API is emulated/virtualized by KVM, and explicitly reject
said hypercalls if the local APIC is emulated in userspace, i.e. don't rely
on userspace to opt-in to KVM_CAP_HYPERV_ENFORCE_CPUID.
Rejecting SEND_IPI and SEND_IPI_EX fixes a NULL-pointer dereference if
Hyper-V enlightenments are exposed to the guest without an in-kernel local
APIC:
dump_stack+0xbe/0xfd
__kasan_report.cold+0x34/0x84
kasan_report+0x3a/0x50
__apic_accept_irq+0x3a/0x5c0
kvm_hv_send_ipi.isra.0+0x34e/0x820
kvm_hv_hypercall+0x8d9/0x9d0
kvm_emulate_hypercall+0x506/0x7e0
__vmx_handle_exit+0x283/0xb60
vmx_handle_exit+0x1d/0xd0
vcpu_enter_guest+0x16b0/0x24c0
vcpu_run+0xc0/0x550
kvm_arch_vcpu_ioctl_run+0x170/0x6d0
kvm_vcpu_ioctl+0x413/0xb20
__se_sys_ioctl+0x111/0x160
do_syscal1_64+0x30/0x40
entry_SYSCALL_64_after_hwframe+0x67/0xd1
Note, checking the sending vCPU is sufficient, as the per-VM irqchip_mode
can't be modified after vCPUs are created, i.e. if one vCPU has an
in-kernel local APIC, then all vCPUs have an in-kernel local APIC.
Reported-by: Dongjie Zou <zoudongjie@huawei.com>
Fixes: 214ff83d44 ("KVM: x86: hyperv: implement PV IPI send hypercalls")
Fixes: 2bc39970e9 ("x86/kvm/hyper-v: Introduce KVM_GET_SUPPORTED_HV_CPUID")
Cc: stable@vger.kernel.org
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Link: https://lore.kernel.org/r/20250118003454.2619573-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
When waking a VM's NX huge page recovery thread, ensure the thread is
actually alive before trying to wake it. Now that the thread is spawned
on-demand during KVM_RUN, a VM without a recovery thread is reachable via
the related module params.
BUG: kernel NULL pointer dereference, address: 0000000000000040
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015
RIP: 0010:vhost_task_wake+0x5/0x10
Call Trace:
<TASK>
set_nx_huge_pages+0xcc/0x1e0 [kvm]
param_attr_store+0x8a/0xd0
module_attr_store+0x1a/0x30
kernfs_fop_write_iter+0x12f/0x1e0
vfs_write+0x233/0x3e0
ksys_write+0x60/0xd0
do_syscall_64+0x5b/0x160
entry_SYSCALL_64_after_hwframe+0x4b/0x53
RIP: 0033:0x7f3b52710104
</TASK>
Modules linked in: kvm_intel kvm
CR2: 0000000000000040
Fixes: 931656b9e2 ("kvm: defer huge page recovery vhost task to later")
Cc: stable@vger.kernel.org
Cc: Keith Busch <kbusch@kernel.org>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-ID: <20250124234623.3609069-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The only statement in a kvm_arch_post_init_vm implementation
can be moved into the x86 kvm_arch_init_vm. Do so and remove all
traces from architecture-independent code.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
SYNTHESIZED_F() generally is used together with setup_force_cpu_cap(),
i.e. when it makes sense to present the feature even if cpuid does not
have it *and* the VM is not able to see the difference. For example,
it can be used when mitigations on the host automatically protect
the guest as well.
The "SYNTHESIZED_F(SRSO_USER_KERNEL_NO)" line came in as a conflict
resolution between the CPUID overhaul from the KVM tree and support
for the feature in the x86 tree. Using it right now does not hurt,
or make a difference for that matter, because there is no
setup_force_cpu_cap(X86_FEATURE_SRSO_USER_KERNEL_NO). However, it
is a little less future proof in case such a setup_force_cpu_cap()
appears later, for a case where the kernel somehow is not vulnerable
but the guest would have to apply the mitigation.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Pull kvm updates from Paolo Bonzini:
"Loongarch:
- Clear LLBCTL if secondary mmu mapping changes
- Add hypercall service support for usermode VMM
x86:
- Add a comment to kvm_mmu_do_page_fault() to explain why KVM
performs a direct call to kvm_tdp_page_fault() when RETPOLINE is
enabled
- Ensure that all SEV code is compiled out when disabled in Kconfig,
even if building with less brilliant compilers
- Remove a redundant TLB flush on AMD processors when guest CR4.PGE
changes
- Use str_enabled_disabled() to replace open coded strings
- Drop kvm_x86_ops.hwapic_irr_update() as KVM updates hardware's
APICv cache prior to every VM-Enter
- Overhaul KVM's CPUID feature infrastructure to track all vCPU
capabilities instead of just those where KVM needs to manage state
and/or explicitly enable the feature in hardware. Along the way,
refactor the code to make it easier to add features, and to make it
more self-documenting how KVM is handling each feature
- Rework KVM's handling of VM-Exits during event vectoring; this
plugs holes where KVM unintentionally puts the vCPU into infinite
loops in some scenarios (e.g. if emulation is triggered by the
exit), and brings parity between VMX and SVM
- Add pending request and interrupt injection information to the
kvm_exit and kvm_entry tracepoints respectively
- Fix a relatively benign flaw where KVM would end up redoing RDPKRU
when loading guest/host PKRU, due to a refactoring of the kernel
helpers that didn't account for KVM's pre-checking of the need to
do WRPKRU
- Make the completion of hypercalls go through the complete_hypercall
function pointer argument, no matter if the hypercall exits to
userspace or not.
Previously, the code assumed that KVM_HC_MAP_GPA_RANGE specifically
went to userspace, and all the others did not; the new code need
not special case KVM_HC_MAP_GPA_RANGE and in fact does not care at
all whether there was an exit to userspace or not
- As part of enabling TDX virtual machines, support support
separation of private/shared EPT into separate roots.
When TDX will be enabled, operations on private pages will need to
go through the privileged TDX Module via SEAMCALLs; as a result,
they are limited and relatively slow compared to reading a PTE.
The patches included in 6.14 allow KVM to keep a mirror of the
private EPT in host memory, and define entries in kvm_x86_ops to
operate on external page tables such as the TDX private EPT
- The recently introduced conversion of the NX-page reclamation
kthread to vhost_task moved the task under the main process. The
task is created as soon as KVM_CREATE_VM was invoked and this, of
course, broke userspace that didn't expect to see any child task of
the VM process until it started creating its own userspace threads.
In particular crosvm refuses to fork() if procfs shows any child
task, so unbreak it by creating the task lazily. This is arguably a
userspace bug, as there can be other kinds of legitimate worker
tasks and they wouldn't impede fork(); but it's not like userspace
has a way to distinguish kernel worker tasks right now. Should they
show as "Kthread: 1" in proc/.../status?
x86 - Intel:
- Fix a bug where KVM updates hardware's APICv cache of the highest
ISR bit while L2 is active, while ultimately results in a
hardware-accelerated L1 EOI effectively being lost
- Honor event priority when emulating Posted Interrupt delivery
during nested VM-Enter by queueing KVM_REQ_EVENT instead of
immediately handling the interrupt
- Rework KVM's processing of the Page-Modification Logging buffer to
reap entries in the same order they were created, i.e. to mark gfns
dirty in the same order that hardware marked the page/PTE dirty
- Misc cleanups
Generic:
- Cleanup and harden kvm_set_memory_region(); add proper lockdep
assertions when setting memory regions and add a dedicated API for
setting KVM-internal memory regions. The API can then explicitly
disallow all flags for KVM-internal memory regions
- Explicitly verify the target vCPU is online in kvm_get_vcpu() to
fix a bug where KVM would return a pointer to a vCPU prior to it
being fully online, and give kvm_for_each_vcpu() similar treatment
to fix a similar flaw
- Wait for a vCPU to come online prior to executing a vCPU ioctl, to
fix a bug where userspace could coerce KVM into handling the ioctl
on a vCPU that isn't yet onlined
- Gracefully handle xarray insertion failures; even though such
failures are impossible in practice after xa_reserve(), reserving
an entry is always followed by xa_store() which does not know (or
differentiate) whether there was an xa_reserve() before or not
RISC-V:
- Zabha, Svvptc, and Ziccrse extension support for guests. None of
them require anything in KVM except for detecting them and marking
them as supported; Zabha adds byte and halfword atomic operations,
while the others are markers for specific operation of the TLB and
of LL/SC instructions respectively
- Virtualize SBI system suspend extension for Guest/VM
- Support firmware counters which can be used by the guests to
collect statistics about traps that occur in the host
Selftests:
- Rework vcpu_get_reg() to return a value instead of using an
out-param, and update all affected arch code accordingly
- Convert the max_guest_memory_test into a more generic
mmu_stress_test. The basic gist of the "conversion" is to have the
test do mprotect() on guest memory while vCPUs are accessing said
memory, e.g. to verify KVM and mmu_notifiers are working as
intended
- Play nice with treewrite builds of unsupported architectures, e.g.
arm (32-bit), as KVM selftests' Makefile doesn't do anything to
ensure the target architecture is actually one KVM selftests
supports
- Use the kernel's $(ARCH) definition instead of the target triple
for arch specific directories, e.g. arm64 instead of aarch64,
mainly so as not to be different from the rest of the kernel
- Ensure that format strings for logging statements are checked by
the compiler even when the logging statement itself is disabled
- Attempt to whack the last LLC references/misses mole in the Intel
PMU counters test by adding a data load and doing CLFLUSH{OPT} on
the data instead of the code being executed. It seems that modern
Intel CPUs have learned new code prefetching tricks that bypass the
PMU counters
- Fix a flaw in the Intel PMU counters test where it asserts that
events are counting correctly without actually knowing what the
events count given the underlying hardware; this can happen if
Intel reuses a formerly microarchitecture-specific event encoding
as an architectural event, as was the case for Top-Down Slots"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (151 commits)
kvm: defer huge page recovery vhost task to later
KVM: x86/mmu: Return RET_PF* instead of 1 in kvm_mmu_page_fault()
KVM: Disallow all flags for KVM-internal memslots
KVM: x86: Drop double-underscores from __kvm_set_memory_region()
KVM: Add a dedicated API for setting KVM-internal memslots
KVM: Assert slots_lock is held when setting memory regions
KVM: Open code kvm_set_memory_region() into its sole caller (ioctl() API)
LoongArch: KVM: Add hypercall service support for usermode VMM
LoongArch: KVM: Clear LLBCTL if secondary mmu mapping is changed
KVM: SVM: Use str_enabled_disabled() helper in svm_hardware_setup()
KVM: VMX: read the PML log in the same order as it was written
KVM: VMX: refactor PML terminology
KVM: VMX: Fix comment of handle_vmx_instruction()
KVM: VMX: Reinstate __exit attribute for vmx_exit()
KVM: SVM: Use str_enabled_disabled() helper in sev_hardware_setup()
KVM: x86: Avoid double RDPKRU when loading host/guest PKRU
KVM: x86: Use LVT_TIMER instead of an open coded literal
RISC-V: KVM: Add new exit statstics for redirected traps
RISC-V: KVM: Update firmware counters for various events
RISC-V: KVM: Redirect instruction access fault trap to guest
...
Pull hyperv updates from Wei Liu:
- Introduce a new set of Hyper-V headers in include/hyperv and replace
the old hyperv-tlfs.h with the new headers (Nuno Das Neves)
- Fixes for the Hyper-V VTL mode (Roman Kisel)
- Fixes for cpu mask usage in Hyper-V code (Michael Kelley)
- Document the guest VM hibernation behaviour (Michael Kelley)
- Miscellaneous fixes and cleanups (Jacob Pan, John Starks, Naman Jain)
* tag 'hyperv-next-signed-20250123' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux:
Documentation: hyperv: Add overview of guest VM hibernation
hyperv: Do not overlap the hvcall IO areas in hv_vtl_apicid_to_vp_id()
hyperv: Do not overlap the hvcall IO areas in get_vtl()
hyperv: Enable the hypercall output page for the VTL mode
hv_balloon: Fallback to generic_online_page() for non-HV hot added mem
Drivers: hv: vmbus: Log on missing offers if any
Drivers: hv: vmbus: Wait for boot-time offers during boot and resume
uio_hv_generic: Add a check for HV_NIC for send, receive buffers setup
iommu/hyper-v: Don't assume cpu_possible_mask is dense
Drivers: hv: Don't assume cpu_possible_mask is dense
x86/hyperv: Don't assume cpu_possible_mask is dense
hyperv: Remove the now unused hyperv-tlfs.h files
hyperv: Switch from hyperv-tlfs.h to hyperv/hvhdk.h
hyperv: Add new Hyper-V headers in include/hyperv
hyperv: Clean up unnecessary #includes
hyperv: Move hv_connection_id to hyperv-tlfs.h
Some libraries want to ensure they are single threaded before forking,
so making the kernel's kvm huge page recovery process a vhost task of
the user process breaks those. The minijail library used by crosvm is
one such affected application.
Defer the task to after the first VM_RUN call, which occurs after the
parent process has forked all its jailed processes. This needs to happen
only once for the kvm instance, so introduce some general-purpose
infrastructure for that, too. It's similar in concept to pthread_once;
except it is actually usable, because the callback takes a parameter.
Cc: Sean Christopherson <seanjc@google.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Tested-by: Alyssa Ross <hi@alyssa.is>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Message-ID: <20250123153543.2769928-1-kbusch@meta.com>
[Move call_once API to include/linux. - Paolo]
Cc: stable@vger.kernel.org
Fixes: d96c77bd4e ("KVM: x86: switch hugepage recovery thread to vhost_task")
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Pull kthread updates from Frederic Weisbecker:
"Kthreads affinity follow either of 4 existing different patterns:
1) Per-CPU kthreads must stay affine to a single CPU and never
execute relevant code on any other CPU. This is currently handled
by smpboot code which takes care of CPU-hotplug operations.
Affinity here is a correctness constraint.
2) Some kthreads _have_ to be affine to a specific set of CPUs and
can't run anywhere else. The affinity is set through
kthread_bind_mask() and the subsystem takes care by itself to
handle CPU-hotplug operations. Affinity here is assumed to be a
correctness constraint.
3) Per-node kthreads _prefer_ to be affine to a specific NUMA node.
This is not a correctness constraint but merely a preference in
terms of memory locality. kswapd and kcompactd both fall into this
category. The affinity is set manually like for any other task and
CPU-hotplug is supposed to be handled by the relevant subsystem so
that the task is properly reaffined whenever a given CPU from the
node comes up. Also care should be taken so that the node affinity
doesn't cross isolated (nohz_full) cpumask boundaries.
4) Similar to the previous point except kthreads have a _preferred_
affinity different than a node. Both RCU boost kthreads and RCU
exp kworkers fall into this category as they refer to "RCU nodes"
from a distinctly distributed tree.
Currently the preferred affinity patterns (3 and 4) have at least 4
identified users, with more or less success when it comes to handle
CPU-hotplug operations and CPU isolation. Each of which do it in its
own ad-hoc way.
This is an infrastructure proposal to handle this with the following
API changes:
- kthread_create_on_node() automatically affines the created kthread
to its target node unless it has been set as per-cpu or bound with
kthread_bind[_mask]() before the first wake-up.
- kthread_affine_preferred() is a new function that can be called
right after kthread_create_on_node() to specify a preferred
affinity different than the specified node.
When the preferred affinity can't be applied because the possible
targets are offline or isolated (nohz_full), the kthread is affine to
the housekeeping CPUs (which means to all online CPUs most of the time
or only the non-nohz_full CPUs when nohz_full= is set).
kswapd, kcompactd, RCU boost kthreads and RCU exp kworkers have been
converted, along with a few old drivers.
Summary of the changes:
- Consolidate a bunch of ad-hoc implementations of
kthread_run_on_cpu()
- Introduce task_cpu_fallback_mask() that defines the default last
resort affinity of a task to become nohz_full aware
- Add some correctness check to ensure kthread_bind() is always
called before the first kthread wake up.
- Default affine kthread to its preferred node.
- Convert kswapd / kcompactd and remove their halfway working ad-hoc
affinity implementation
- Implement kthreads preferred affinity
- Unify kthread worker and kthread API's style
- Convert RCU kthreads to the new API and remove the ad-hoc affinity
implementation"
* tag 'kthread-for-6.14-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks:
kthread: modify kernel-doc function name to match code
rcu: Use kthread preferred affinity for RCU exp kworkers
treewide: Introduce kthread_run_worker[_on_cpu]()
kthread: Unify kthread_create_on_cpu() and kthread_create_worker_on_cpu() automatic format
rcu: Use kthread preferred affinity for RCU boost
kthread: Implement preferred affinity
mm: Create/affine kswapd to its preferred node
mm: Create/affine kcompactd to its preferred node
kthread: Default affine kthread to its preferred NUMA node
kthread: Make sure kthread hasn't started while binding it
sched,arm64: Handle CPU isolation on last resort fallback rq selection
arm64: Exclude nohz_full CPUs from 32bits el0 support
lib: test_objpool: Use kthread_run_on_cpu()
kallsyms: Use kthread_run_on_cpu()
soc/qman: test: Use kthread_run_on_cpu()
arm/bL_switcher: Use kthread_run_on_cpu()
As part of enabling TDX virtual machines, support support separation of
private/shared EPT into separate roots.
Confidential computing solutions almost invariably have concepts of
private and shared memory, but they may different a lot in the details.
In SEV, for example, the bit is handled more like a permission bit as
far as the page tables are concerned: the private/shared bit is not
included in the physical address.
For TDX, instead, the bit is more like a physical address bit, with
the host mapping private memory in one half of the address space and
shared in another. Furthermore, the two halves are mapped by different
EPT roots and only the shared half is managed by KVM; the private half
(also called Secure EPT in Intel documentation) gets managed by the
privileged TDX Module via SEAMCALLs.
As a result, the operations that actually change the private half of
the EPT are limited and relatively slow compared to reading a PTE. For
this reason the design for KVM is to keep a mirror of the private EPT in
host memory. This allows KVM to quickly walk the EPT and only perform the
slower private EPT operations when it needs to actually modify mid-level
private PTEs.
There are thus three sets of EPT page tables: external, mirror and
direct. In the case of TDX (the only user of this framework) the
first two cover private memory, whereas the third manages shared
memory:
external EPT - Hidden within the TDX module, modified via TDX module
calls.
mirror EPT - Bookkeeping tree used as an optimization by KVM, not
used by the processor.
direct EPT - Normal EPT that maps unencrypted shared memory.
Managed like the EPT of a normal VM.
Modifying external EPT
----------------------
Modifications to the mirrored page tables need to also perform the
same operations to the private page tables, which will be handled via
kvm_x86_ops. Although this prep series does not interact with the TDX
module at all to actually configure the private EPT, it does lay the
ground work for doing this.
In some ways updating the private EPT is as simple as plumbing PTE
modifications through to also call into the TDX module; however, the
locking is more complicated because inserting a single PTE cannot anymore
be done atomically with a single CMPXCHG. For this reason, the existing
FROZEN_SPTE mechanism is used whenever a call to the TDX module updates the
private EPT. FROZEN_SPTE acts basically as a spinlock on a PTE. Besides
protecting operation of KVM, it limits the set of cases in which the
TDX module will encounter contention on its own PTE locks.
Zapping external EPT
--------------------
While the framework tries to be relatively generic, and to be
understandable without knowing TDX much in detail, some requirements of
TDX sometimes leak; for example the private page tables also cannot be
zapped while the range has anything mapped, so the mirrored/private page
tables need to be protected from KVM operations that zap any non-leaf
PTEs, for example kvm_mmu_reset_context() or kvm_mmu_zap_all_fast().
For normal VMs, guest memory is zapped for several reasons: user
memory getting paged out by the guest, memslots getting deleted,
passthrough of devices with non-coherent DMA. Confidential computing
adds to these the conversion of memory between shared and privates. These
operations must not zap any private memory that is in use by the guest.
This is possible because the only zapping that is out of the control
of KVM/userspace is paging out userspace memory, which cannot apply to
guestmemfd operations. Thus a TDX VM will only zap private memory from
memslot deletion and from conversion between private and shared memory
which is triggered by the guest.
To avoid zapping too much memory, enums are introduced so that operations
can choose to target only private or shared memory, and thus only
direct or mirror EPT. For example:
Memslot deletion - Private and shared
MMU notifier based zapping - Shared only
Conversion to shared - Private only
Conversion to private - Shared only
Other cases of zapping will not be supported for KVM, for example
APICv update or non-coherent DMA status update; for the latter, TDX will
simply require that the CPU supports self-snoop and honor guest PAT
unconditionally for shared memory.
Make the completion of hypercalls go through the complete_hypercall
function pointer argument, no matter if the hypercall exits to
userspace or not. Previously, the code assumed that KVM_HC_MAP_GPA_RANGE
specifically went to userspace, and all the others did not; the new code
need not special case KVM_HC_MAP_GPA_RANGE and in fact does not care at
all whether there was an exit to userspace or not.
KVM x86 misc changes for 6.14:
- Overhaul KVM's CPUID feature infrastructure to track all vCPU capabilities
instead of just those where KVM needs to manage state and/or explicitly
enable the feature in hardware. Along the way, refactor the code to make
it easier to add features, and to make it more self-documenting how KVM
is handling each feature.
- Rework KVM's handling of VM-Exits during event vectoring; this plugs holes
where KVM unintentionally puts the vCPU into infinite loops in some scenarios
(e.g. if emulation is triggered by the exit), and brings parity between VMX
and SVM.
- Add pending request and interrupt injection information to the kvm_exit and
kvm_entry tracepoints respectively.
- Fix a relatively benign flaw where KVM would end up redoing RDPKRU when
loading guest/host PKRU, due to a refactoring of the kernel helpers that
didn't account for KVM's pre-checking of the need to do WRPKRU.
KVM VMX changes for 6.14:
- Fix a bug where KVM updates hardware's APICv cache of the highest ISR bit
while L2 is active, while ultimately results in a hardware-accelerated L1
EOI effectively being lost.
- Honor event priority when emulating Posted Interrupt delivery during nested
VM-Enter by queueing KVM_REQ_EVENT instead of immediately handling the
interrupt.
- Drop kvm_x86_ops.hwapic_irr_update() as KVM updates hardware's APICv cache
prior to every VM-Enter.
- Rework KVM's processing of the Page-Modification Logging buffer to reap
entries in the same order they were created, i.e. to mark gfns dirty in the
same order that hardware marked the page/PTE dirty.
- Misc cleanups.
KVM SVM changes for 6.14:
- Macrofy the SEV=n version of the sev_xxx_guest() helpers so that the code is
optimized away when building with less than brilliant compilers.
- Remove a now-redundant TLB flush when guest CR4.PGE changes.
- Use str_enabled_disabled() to replace open coded strings.
KVM x86 MMU changes for 6.14:
- Add a comment to kvm_mmu_do_page_fault() to explain why KVM performs a
direct call to kvm_tdp_page_fault() when RETPOLINE is enabled.
KVM kvm_set_memory_region() cleanups and hardening for 6.14:
- Add proper lockdep assertions when setting memory regions.
- Add a dedicated API for setting KVM-internal memory regions.
- Explicitly disallow all flags for KVM-internal memory regions.
Return RET_PF* (excluding RET_PF_EMULATE/RET_PF_CONTINUE/RET_PF_INVALID)
instead of 1 in kvm_mmu_page_fault().
The callers of kvm_mmu_page_fault() are KVM page fault handlers (i.e.,
npf_interception(), handle_ept_misconfig(), __vmx_handle_ept_violation(),
kvm_handle_page_fault()). They either check if the return value is > 0 (as
in npf_interception()) or pass it further to vcpu_run() to decide whether
to break out of the kernel loop and return to the user when r <= 0.
Therefore, returning any positive value is equivalent to returning 1.
Warn if r == RET_PF_CONTINUE (which should not be a valid value) to ensure
a positive return value.
This is a preparation to allow TDX's EPT violation handler to check the
RET_PF* value and retry internally for RET_PF_RETRY.
No functional changes are intended.
Signed-off-by: Yan Zhao <yan.y.zhao@intel.com>
Message-ID: <20250113021138.18875-1-yan.y.zhao@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Intel's PRM specifies that the CPU writes to the PML log 'backwards'
or in other words, it first writes entry 511, then entry 510 and so on.
I also confirmed on the bare metal that the CPU indeed writes the entries
in this order.
KVM on the other hand, reads the entries in the opposite order, from the
last written entry and towards entry 511 and dumps them in this order to
the dirty ring.
Usually this doesn't matter, except for one complex nesting case:
KVM reties the instructions that cause MMU faults.
This might cause an emulated PML log entry to be visible to L1's hypervisor
before the actual memory write was committed.
This happens when the L0 MMU fault is followed directly by the VM exit to
L1, for example due to a pending L1 interrupt or due to the L1's
'PML log full' event.
This problem doesn't have a noticeable real-world impact because this
write retry is not much different from the guest writing to the same page
multiple times, which is also not reflected in the dirty log. The users of
the dirty logging only rely on correct reporting of the clean pages, or
in other words they assume that if a page is clean, then no writes were
committed to it since the moment it was marked clean.
However KVM has a kvm_dirty_log_test selftest, a test that tests both
the clean and the dirty pages vs the memory contents, and can fail if it
detects a dirty page which has an old value at the offset 0 which the test
writes.
To avoid failure, the test has a workaround for this specific problem:
The test skips checking memory that belongs to the last dirty ring entry,
which it has seen, relying on the fact that as long as memory writes are
committed in-order, only the last entry can belong to a not yet committed
memory write.
However, since L1's KVM is reading the PML log in the opposite direction
that L0 wrote it, the last dirty ring entry often will be not the last
entry written by the L0.
To fix this, switch the order in which KVM reads the PML log.
Note that this issue is not present on the bare metal, because on the
bare metal, an update of the A/D bits of a present entry, PML logging and
the actual memory write are all done by the CPU without any hypervisor
intervention and pending interrupt evaluation, thus once a PML log and/or
vCPU kick happens, all memory writes that are in the PML log are
committed to memory.
The only exception to this rule is when the guest hits a not present EPT
entry, in which case KVM first reads (backward) the PML log, dumps it to
the dirty ring, and *then* sets up a SPTE entry with A/D bits set, and logs
this to the dirty ring, thus making the entry be the last one in the
dirty ring.
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Link: https://lore.kernel.org/r/20241219221034.903927-3-mlevitsk@redhat.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Fix a goof in handle_vmx_instruction()'s comment where it references the
non-existent nested_vmx_setup(); the function that overwrites the exit
handlers is nested_vmx_hardware_setup().
Note, this isn't a case of a stale comment, e.g. due to the function being
renamed. The comment has always been wrong.
Fixes: e4027cfafd ("KVM: nVMX: Set callbacks for nested functions during hardware setup")
Signed-off-by: Gao Shiyuan <gaoshiyuan@baidu.com>
Link: https://lore.kernel.org/r/20250103153814.73903-1-gaoshiyuan@baidu.com
[sean: massage changelog]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tag vmx_exit() with __exit now that it's no longer used by vmx_init().
Commit a7b9020b06 ("x86/l1tf: Handle EPT disabled state proper") dropped
the "__exit" attribute from vmx_exit() because vmx_init() was changed to
call vmx_exit().
However, commit e32b120071 ("KVM: VMX: Do _all_ initialization before
exposing /dev/kvm to userspace") changed vmx_init() to call __vmx_exit()
instead of the "full" vmx_exit(). This made it possible to mark vmx_exit()
as "__exit" again, as it originally was, and enjoy the benefits that it
provides (the function can be discarded from memory in situations where it
cannot be called, like the module being built-in or module unloading being
disabled in the kernel).
Signed-off-by: Costas Argyris <costas.argyris@amd.com>
Link: https://lore.kernel.org/r/20250102154050.2403-1-costas.argyris@amd.com
[sean: massage changelog]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Use the raw wrpkru() helper when loading the guest/host's PKRU on switch
to/from guest context, as the write_pkru() wrapper incurs an unnecessary
rdpkru(). In both paths, KVM is guaranteed to have performed RDPKRU since
the last possible write, i.e. KVM has a fresh cache of the current value
in hardware.
This effectively restores KVM's behavior to that of KVM prior to commit
c806e88734 ("x86/pkeys: Provide *pkru() helpers"), which renamed the raw
helper from __write_pkru() => wrpkru(), and turned __write_pkru() into a
wrapper. Commit 577ff465f5 ("x86/fpu: Only write PKRU if it is different
from current") then added the extra RDPKRU to avoid an unnecessary WRPKRU,
but completely missed that KVM already optimized away pointless writes.
Reported-by: Adrian Hunter <adrian.hunter@intel.com>
Fixes: 577ff465f5 ("x86/fpu: Only write PKRU if it is different from current")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Link: https://lore.kernel.org/r/20241221011647.3747448-1-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Use LVT_TIMER instead of the literal '0' to clean up the apic_lvt_mask
lookup when emulating handling writes to APIC_LVTT.
No functional change intended.
Signed-off-by: Liam Ni <zhiguangni01@gmail.com>
[sean: manually regenerate patch (whitespace damaged), massage changelog]
Signed-off-by: Sean Christopherson <seanjc@google.com>
kthread_create() creates a kthread without running it yet. kthread_run()
creates a kthread and runs it.
On the other hand, kthread_create_worker() creates a kthread worker and
runs it.
This difference in behaviours is confusing. Also there is no way to
create a kthread worker and affine it using kthread_bind_mask() or
kthread_affine_preferred() before starting it.
Consolidate the behaviours and introduce kthread_run_worker[_on_cpu]()
that behaves just like kthread_run(). kthread_create_worker[_on_cpu]()
will now only create a kthread worker without starting it.
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Add a few sanity checks to prevent memslot GFNs from ever having alias bits
set.
Like other Coco technologies, TDX has the concept of private and shared
memory. For TDX the private and shared mappings are managed on separate
EPT roots. The private half is managed indirectly though calls into a
protected runtime environment called the TDX module, where the shared half
is managed within KVM in normal page tables.
For TDX, the shared half will be mapped in the higher alias, with a "shared
bit" set in the GPA. However, KVM will still manage it with the same
memslots as the private half. This means memslot looks ups and zapping
operations will be provided with a GFN without the shared bit set.
If these memslot GFNs ever had the bit that selects between the two aliases
it could lead to unexpected behavior in the complicated code that directs
faulting or zapping operations between the roots that map the two aliases.
As a safety measure, prevent memslots from being set at a GFN range that
contains the alias bit.
Also, check in the kvm_faultin_pfn() for the fault path. This later check
does less today, as the alias bits are specifically stripped from the GFN
being checked, however future code could possibly call in to the fault
handler in a way that skips this stripping. Since kvm_faultin_pfn() now
has many references to vcpu->kvm, extract it to local variable.
Link: https://lore.kernel.org/kvm/ZpbKqG_ZhCWxl-Fc@google.com/
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-19-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Don't zap valid mirror roots in kvm_tdp_mmu_zap_all(), which in effect
is only direct roots (invalid and valid).
For TDX, kvm_tdp_mmu_zap_all() is only called during MMU notifier
release. Since, mirrored EPT comes from guest mem, it will never be
mapped to userspace, and won't apply. But in addition to be unnecessary,
mirrored EPT is cleaned up in a special way during VM destruction.
Pass the KVM_INVALID_ROOTS bit into __for_each_tdp_mmu_root_yield_safe()
as well, to clean up invalid direct roots, as is the current behavior.
While at it, remove an obsolete reference to work item-based zapping.
Co-developed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-18-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Rename kvm_tdp_mmu_invalidate_all_roots() to
kvm_tdp_mmu_invalidate_roots(), and make it enum kvm_tdp_mmu_root_types
as an argument.
kvm_tdp_mmu_invalidate_roots() is called with different root types. For
kvm_mmu_zap_all_fast() it only operates on shared roots. But when tearing
down a VM it needs to invalidate all roots. Have the callers only
invalidate the required roots instead of all roots.
Within kvm_tdp_mmu_invalidate_roots(), respect the root type
passed by checking the root type in root iterator.
Suggested-by: Chao Gao <chao.gao@intel.com>
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-17-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Integrate hooks for mirroring page table operations for cases where TDX
will zap PTEs or free page tables.
Like other Coco technologies, TDX has the concept of private and shared
memory. For TDX the private and shared mappings are managed on separate
EPT roots. The private half is managed indirectly though calls into a
protected runtime environment called the TDX module, where the shared half
is managed within KVM in normal page tables.
Since calls into the TDX module are relatively slow, walking private page
tables by making calls into the TDX module would not be efficient. Because
of this, previous changes have taught the TDP MMU to keep a mirror root,
which is separate, unmapped TDP root that private operations can be
directed to. Currently this root is disconnected from the guest. Now add
plumbing to propagate changes to the "external" page tables being
mirrored. Just create the x86_ops for now, leave plumbing the operations
into the TDX module for future patches.
Add two operations for tearing down page tables, one for freeing page
tables (free_external_spt) and one for zapping PTEs (remove_external_spte).
Define them such that remove_external_spte will perform a TLB flush as
well. (in TDX terms "ensure there are no active translations").
TDX MMU support will exclude certain MMU operations, so only plug in the
mirroring x86 ops where they will be needed. For zapping/freeing, only
hook tdp_mmu_iter_set_spte() which is used for mapping and linking PTs.
Don't bother hooking tdp_mmu_set_spte_atomic() as it is only used for
zapping PTEs in operations unsupported by TDX: zapping collapsible PTEs and
kvm_mmu_zap_all_fast().
In previous changes to address races around concurrent populating using
tdp_mmu_set_spte_atomic(), a solution was introduced to temporarily set
FROZEN_SPTE in the mirrored page tables while performing the external
operations. Such a solution is not needed for the tear down paths in TDX
as these will always be performed with the mmu_lock held for write.
Sprinkle some KVM_BUG_ON()s to reflect this.
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Co-developed-by: Kai Huang <kai.huang@intel.com>
Signed-off-by: Kai Huang <kai.huang@intel.com>
Co-developed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Yan Zhao <yan.y.zhao@intel.com>
Co-developed-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-16-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Integrate hooks for mirroring page table operations for cases where TDX
will set PTEs or link page tables.
Like other Coco technologies, TDX has the concept of private and shared
memory. For TDX the private and shared mappings are managed on separate
EPT roots. The private half is managed indirectly through calls into a
protected runtime environment called the TDX module, where the shared half
is managed within KVM in normal page tables.
Since calls into the TDX module are relatively slow, walking private page
tables by making calls into the TDX module would not be efficient. Because
of this, previous changes have taught the TDP MMU to keep a mirror root,
which is separate, unmapped TDP root that private operations can be
directed to. Currently this root is disconnected from any actual guest
mapping. Now add plumbing to propagate changes to the "external" page
tables being mirrored. Just create the x86_ops for now, leave plumbing the
operations into the TDX module for future patches.
Add two operations for setting up external page tables, one for linking
new page tables and one for setting leaf PTEs. Don't add any op for
configuring the root PFN, as TDX handles this itself. Don't provide a
way to set permissions on the PTEs also, as TDX doesn't support it.
This results in MMU "mirroring" support that is very targeted towards TDX.
Since it is likely there will be no other user, the main benefit of making
the support generic is to keep TDX specific *looking* code outside of the
MMU. As a generic feature it will make enough sense from TDX's
perspective. For developers unfamiliar with TDX arch it can express the
general concepts such that they can continue to work in the code.
TDX MMU support will exclude certain MMU operations, so only plug in the
mirroring x86 ops where they will be needed. For setting/linking, only
hook tdp_mmu_set_spte_atomic() which is used for mapping and linking
PTs. Don't bother hooking tdp_mmu_iter_set_spte() as it is only used for
setting PTEs in operations unsupported by TDX: splitting huge pages and
write protecting. Sprinkle KVM_BUG_ON()s to document as code that these
paths are not supported for mirrored page tables. For zapping operations,
leave those for near future changes.
Many operations in the TDP MMU depend on atomicity of the PTE update.
While the mirror PTE on KVM's side can be updated atomically, the update
that happens inside the external operations (S-EPT updates via TDX module
call) can't happen atomically with the mirror update. The following race
could result during two vCPU's populating private memory:
* vcpu 1: atomically update 2M level mirror EPT entry to be present
* vcpu 2: read 2M level EPT entry that is present
* vcpu 2: walk down into 4K level EPT
* vcpu 2: atomically update 4K level mirror EPT entry to be present
* vcpu 2: set_exterma;_spte() to update 4K secure EPT entry => error
because 2M secure EPT entry is not populated yet
* vcpu 1: link_external_spt() to update 2M secure EPT entry
Prevent this by setting the mirror PTE to FROZEN_SPTE while the reflect
operations are performed. Only write the actual mirror PTE value once the
reflect operations have completed. When trying to set a PTE to present and
encountering a frozen SPTE, retry the fault.
By doing this the race is prevented as follows:
* vcpu 1: atomically update 2M level EPT entry to be FROZEN_SPTE
* vcpu 2: read 2M level EPT entry that is FROZEN_SPTE
* vcpu 2: find that the EPT entry is frozen
abandon page table walk to resume guest execution
* vcpu 1: link_external_spt() to update 2M secure EPT entry
* vcpu 1: atomically update 2M level EPT entry to be present (unfreeze)
* vcpu 2: resume guest execution
Depending on vcpu 1 state, vcpu 2 may result in EPT violation
again or make progress on guest execution
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Co-developed-by: Kai Huang <kai.huang@intel.com>
Signed-off-by: Kai Huang <kai.huang@intel.com>
Co-developed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Yan Zhao <yan.y.zhao@intel.com>
Co-developed-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-15-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Teach the MMU notifier callbacks how to check kvm_gfn_range.process to
filter which KVM MMU root types to operate on.
The private GPAs are backed by guest memfd. Such memory is not subjected
to MMU notifier callbacks because it can't be mapped into the host user
address space. Now kvm_gfn_range conveys info about which root to operate
on. Enhance the callback to filter the root page table type.
The KVM MMU notifier comes down to two functions.
kvm_tdp_mmu_unmap_gfn_range() and __kvm_tdp_mmu_age_gfn_range():
- invalidate_range_start() calls kvm_tdp_mmu_unmap_gfn_range()
- invalidate_range_end() doesn't call into arch code
- the other callbacks call __kvm_tdp_mmu_age_gfn_range()
For VM's without a private/shared split in the EPT, all operations
should target the normal(direct) root.
With the switch from for_each_tdp_mmu_root() to
__for_each_tdp_mmu_root() in kvm_tdp_mmu_handle_gfn(), there are no
longer any users of for_each_tdp_mmu_root(). Remove it.
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-14-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add the ability for the TDP MMU to maintain a mirror of a separate
mapping.
Like other Coco technologies, TDX has the concept of private and shared
memory. For TDX the private and shared mappings are managed on separate
EPT roots. The private half is managed indirectly through calls into a
protected runtime environment called the TDX module, where the shared half
is managed within KVM in normal page tables.
In order to handle both shared and private memory, KVM needs to learn to
handle faults and other operations on the correct root for the operation.
KVM could learn the concept of private roots, and operate on them by
calling out to operations that call into the TDX module. But there are two
problems with that:
1. Calls into the TDX module are relatively slow compared to the simple
accesses required to read a PTE managed directly by KVM.
2. Other Coco technologies deal with private memory completely differently
and it will make the code confusing when being read from their
perspective. Special operations added for TDX that set private or zap
private memory will have nothing to do with these other private memory
technologies. (SEV, etc).
To handle these, instead teach the TDP MMU about a new concept "mirror
roots". Such roots maintain page tables that are not actually mapped,
and are just used to traverse quickly to determine if the mid level page
tables need to be installed. When the memory be mirrored needs to actually
be changed, calls can be made to via x86_ops.
private KVM page fault |
| |
V |
private GPA | CPU protected EPTP
| | |
V | V
mirror PT root | external PT root
| | |
V | V
mirror PT --hook to propagate-->external PT
| | |
\--------------------+------\ |
| | |
| V V
| private guest page
|
|
non-encrypted memory | encrypted memory
|
Leave calling out to actually update the private page tables that are being
mirrored for later changes. Just implement the handling of MMU operations
on to mirrored roots.
In order to direct operations to correct root, add root types
KVM_DIRECT_ROOTS and KVM_MIRROR_ROOTS. Tie the usage of mirrored/direct
roots to private/shared with conditionals. It could also be implemented by
making the kvm_tdp_mmu_root_types and kvm_gfn_range_filter enum bits line
up such that conversion could be a direct assignment with a case. Don't do
this because the mapping of private to mirrored is confusing enough. So it
is worth not hiding the logic in type casting.
Cleanup the mirror root in kvm_mmu_destroy() instead of the normal place
in kvm_mmu_free_roots(), because the private root that is being cannot be
rebuilt like a normal root. It needs to persist for the lifetime of the VM.
The TDX module will also need to be provided with page tables to use for
the actual mapping being mirrored by the mirrored page tables. Allocate
these in the mapping path using the recently added
kvm_mmu_alloc_external_spt().
Don't support 2M page for now. This is avoided by forcing 4k pages in the
fault. Add a KVM_BUG_ON() to verify.
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Co-developed-by: Kai Huang <kai.huang@intel.com>
Signed-off-by: Kai Huang <kai.huang@intel.com>
Co-developed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Yan Zhao <yan.y.zhao@intel.com>
Co-developed-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-13-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Take the root as an argument of tdp_mmu_for_each_pte() instead of looking
it up in the mmu. With no other purpose of passing the mmu, drop it.
Future changes will want to change which root is used based on the context
of the MMU operation. So change the callers to pass in the root currently
used, mmu->root.hpa in a preparatory patch to make the later one smaller
and easier to review.
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-12-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Define an enum kvm_tdp_mmu_root_types to specify the KVM MMU root type [1]
so that the iterator on the root page table can consistently filter the
root page table type instead of only_valid.
TDX KVM will operate on KVM page tables with specified types. Shared page
table, private page table, or both. Introduce an enum instead of bool
only_valid so that we can easily enhance page table types applicable to
shared, private, or both in addition to valid or not. Replace
only_valid=false with KVM_ANY_ROOTS and only_valid=true with
KVM_ANY_VALID_ROOTS. Use KVM_ANY_ROOTS and KVM_ANY_VALID_ROOTS to wrap
KVM_VALID_ROOTS to avoid further code churn when direct vs mirror root
concepts are introduced in future patches.
Link: https://lore.kernel.org/kvm/ZivazWQw1oCU8VBC@google.com/ [1]
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-11-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Extract tdp_mmu_root_match() to check if the root has given types and use
it for the root page table iterator. It checks only_invalid now.
TDX KVM operates on a shared page table only (Shared-EPT), a mirrored page
table only (Secure-EPT), or both based on the operation. KVM MMU notifier
operations only on shared page table. KVM guest_memfd invalidation
operations only on mirrored page table, and so on. Introduce a centralized
matching function instead of open coding matching logic in the iterator.
The next step is to extend the function to check whether the page is shared
or private
Link: https://lore.kernel.org/kvm/ZivazWQw1oCU8VBC@google.com/
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-10-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Teach the MMU to map guest GFNs at a massaged position on the TDP, to aid
in implementing TDX shared memory.
Like other Coco technologies, TDX has the concept of private and shared
memory. For TDX the private and shared mappings are managed on separate
EPT roots. The private half is managed indirectly through calls into a
protected runtime environment called the TDX module, where the shared half
is managed within KVM in normal page tables.
For TDX, the shared half will be mapped in the higher alias, with a "shared
bit" set in the GPA. However, KVM will still manage it with the same
memslots as the private half. This means memslot looks ups and zapping
operations will be provided with a GFN without the shared bit set.
So KVM will either need to apply or strip the shared bit before mapping or
zapping the shared EPT. Having GFNs sometimes have the shared bit and
sometimes not would make the code confusing.
So instead arrange the code such that GFNs never have shared bit set.
Create a concept of "direct bits", that is stripped from the fault
address when setting fault->gfn, and applied within the TDP MMU iterator.
Calling code will behave as if it is operating on the PTE mapping the GFN
(without shared bits) but within the iterator, the actual mappings will be
shifted using bits specific for the root. SPs will have the GFN set
without the shared bit. In the end the TDP MMU will behave like it is
mapping things at the GFN without the shared bit but with a strange page
table format where everything is offset by the shared bit.
Since TDX only needs to shift the mapping like this for the shared bit,
which is mapped as the normal TDP root, add a "gfn_direct_bits" field to
the kvm_arch structure for each VM with a default value of 0. It will
have the bit set at the position of the GPA shared bit in GFN through TD
specific initialization code. Keep TDX specific concepts out of the MMU
code by not naming it "shared".
Ranged TLB flushes (i.e. flush_remote_tlbs_range()) target specific GFN
ranges. In convention established above, these would need to target the
shifted GFN range. It won't matter functionally, since the actual
implementation will always result in a full flush for the only planned
user (TDX). For correctness reasons, future changes can provide a TDX
x86_ops.flush_remote_tlbs_range implementation to return -EOPNOTSUPP and
force the full flush for TDs.
This leaves one problem. Some operations use a concept of max GFN (i.e.
kvm_mmu_max_gfn()), to iterate over the whole TDP range. When applying the
direct mask to the start of the range, the iterator would end up skipping
iterating over the range not covered by the direct mask bit. For safety,
make sure the __tdp_mmu_zap_root() operation iterates over the full GFN
range supported by the underlying TDP format. Add a new iterator helper,
for_each_tdp_pte_min_level_all(), that iterates the entire TDP GFN range,
regardless of root.
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Co-developed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Yan Zhao <yan.y.zhao@intel.com>
Co-developed-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-9-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add a struct kvm argument to the TDP MMU iterators.
Future changes will want to change how the iterator behaves based on a
member of struct kvm. Change the signature and callers of the iterator
loop helpers in a separate patch to make the future one easier to review.
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-8-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The kvm_tdp_mmu_alloc_root() function currently always returns 0. This
allows for the caller, mmu_alloc_direct_roots(), to call
kvm_tdp_mmu_alloc_root() and also return 0 in one line:
return kvm_tdp_mmu_alloc_root(vcpu);
So it is useful even though the return value of kvm_tdp_mmu_alloc_root()
is always the same. However, in future changes, kvm_tdp_mmu_alloc_root()
will be called twice in mmu_alloc_direct_roots(). This will force the
first call to either awkwardly handle the return value that will always
be zero or ignore it. So change kvm_tdp_mmu_alloc_root() to return void.
Do it in a separate change so the future change will be cleaner.
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20240718211230.1492011-7-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Introduce a "is_mirror" member to the kvm_mmu_page_role union to identify
SPTEs associated with the mirrored EPT.
The TDX module maintains the private half of the EPT mapped in the TD in
its protected memory. KVM keeps a copy of the private GPAs in a mirrored
EPT tree within host memory. This "is_mirror" attribute enables vCPUs to
find and get the root page of mirrored EPT from the MMU root list for a
guest TD. This also allows KVM MMU code to detect changes in mirrored EPT
according to the "is_mirror" mmu page role and propagate the changes to
the private EPT managed by TDX module.
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Message-ID: <20240718211230.1492011-6-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add an external pointer to struct kvm_mmu_page for TDX's private page table
and add helper functions to allocate/initialize/free a private page table
page. TDX will only be supported with the TDP MMU. Because KVM TDP MMU
doesn't use unsync_children and write_flooding_count, pack them to have
room for a pointer and use a union to avoid memory overhead.
For private GPA, CPU refers to a private page table whose contents are
encrypted. The dedicated APIs to operate on it (e.g. updating/reading its
PTE entry) are used, and their cost is expensive.
When KVM resolves the KVM page fault, it walks the page tables. To reuse
the existing KVM MMU code and mitigate the heavy cost of directly walking
the private page table allocate two sets of page tables for the private
half of the GPA space.
For the page tables that KVM will walk, allocate them like normal and refer
to them as mirror page tables. Additionally allocate one more page for the
page tables the CPU will walk, and call them external page tables. Resolve
the KVM page fault with the existing code, and do additional operations
necessary for modifying the external page table in future patches.
The relationship of the types of page tables in this scheme is depicted
below:
KVM page fault |
| |
V |
-------------+---------- |
| | |
V V |
shared GPA private GPA |
| | |
V V |
shared PT root mirror PT root | private PT root
| | | |
V V | V
shared PT mirror PT --propagate--> external PT
| | | |
| \-----------------+------\ |
| | | |
V | V V
shared guest page | private guest page
|
non-encrypted memory | encrypted memory
|
PT - Page table
Shared PT - Visible to KVM, and the CPU uses it for shared mappings.
External PT - The CPU uses it, but it is invisible to KVM. TDX module
updates this table to map private guest pages.
Mirror PT - It is visible to KVM, but the CPU doesn't use it. KVM uses
it to propagate PT change to the actual private PT.
Add a helper kvm_has_mirrored_tdp() to trigger this behavior and wire it
to the TDX vm type.
Co-developed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Reviewed-by: Binbin Wu <binbin.wu@linux.intel.com>
Message-ID: <20240718211230.1492011-5-rick.p.edgecombe@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
