| Age | Commit message (Collapse) | Author |
|---|
|
The third nybble of AMD's event select overlaps with Intel's IN_TX and
IN_TXCP bits. Therefore, we can't use AMD64_RAW_EVENT_MASK on Intel
platforms that support TSX.
Declare a raw_event_mask in the kvm_pmu structure, initialize it in
the vendor-specific pmu_refresh() functions, and use that mask for
PERF_TYPE_RAW configurations in reprogram_gp_counter().
Fixes: 710c47651431 ("KVM: x86/pmu: Use AMD64_RAW_EVENT_MASK for PERF_TYPE_RAW")
Signed-off-by: Jim Mattson <jmattson@google.com>
Message-Id: <20220308012452.3468611-1-jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Don't actually set a request bit in vcpu->requests when making a request
purely to force a vCPU to exit the guest. Logging a request but not
actually consuming it would cause the vCPU to get stuck in an infinite
loop during KVM_RUN because KVM would see the pending request and bail
from VM-Enter to service the request.
Note, it's currently impossible for KVM to set KVM_REQ_GPC_INVALIDATE as
nothing in KVM is wired up to set guest_uses_pa=true. But, it'd be all
too easy for arch code to introduce use of kvm_gfn_to_pfn_cache_init()
without implementing handling of the request, especially since getting
test coverage of MMU notifier interaction with specific KVM features
usually requires a directed test.
Opportunistically rename gfn_to_pfn_cache_invalidate_start()'s wake_vcpus
to evict_vcpus. The purpose of the request is to get vCPUs out of guest
mode, it's supposed to _avoid_ waking vCPUs that are blocking.
Opportunistically rename KVM_REQ_GPC_INVALIDATE to be more specific as to
what it wants to accomplish, and to genericize the name so that it can
used for similar but unrelated scenarios, should they arise in the future.
Add a comment and documentation to explain why the "no action" request
exists.
Add compile-time assertions to help detect improper usage. Use the inner
assertless helper in the one s390 path that makes requests without a
hardcoded request.
Cc: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220223165302.3205276-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Re-introduce zapping only leaf SPTEs in kvm_zap_gfn_range() and
kvm_tdp_mmu_unmap_gfn_range(), this time without losing a pending TLB
flush when processing multiple roots (including nested TDP shadow roots).
Dropping the TLB flush resulted in random crashes when running Hyper-V
Server 2019 in a guest with KSM enabled in the host (or any source of
mmu_notifier invalidations, KSM is just the easiest to force).
This effectively revert commits 873dd122172f8cce329113cfb0dfe3d2344d80c0
and fcb93eb6d09dd302cbef22bd95a5858af75e4156, and thus restores commit
cf3e26427c08ad9015956293ab389004ac6a338e, plus this delta on top:
bool kvm_tdp_mmu_zap_leafs(struct kvm *kvm, int as_id, gfn_t start, gfn_t end,
struct kvm_mmu_page *root;
for_each_tdp_mmu_root_yield_safe(kvm, root, as_id)
- flush = tdp_mmu_zap_leafs(kvm, root, start, end, can_yield, false);
+ flush = tdp_mmu_zap_leafs(kvm, root, start, end, can_yield, flush);
return flush;
}
Cc: Ben Gardon <bgardon@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tested-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20220325230348.2587437-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
As guest_irq is coming from KVM_IRQFD API call, it may trigger
crash in svm_update_pi_irte() due to out-of-bounds:
crash> bt
PID: 22218 TASK: ffff951a6ad74980 CPU: 73 COMMAND: "vcpu8"
#0 [ffffb1ba6707fa40] machine_kexec at ffffffff8565b397
#1 [ffffb1ba6707fa90] __crash_kexec at ffffffff85788a6d
#2 [ffffb1ba6707fb58] crash_kexec at ffffffff8578995d
#3 [ffffb1ba6707fb70] oops_end at ffffffff85623c0d
#4 [ffffb1ba6707fb90] no_context at ffffffff856692c9
#5 [ffffb1ba6707fbf8] exc_page_fault at ffffffff85f95b51
#6 [ffffb1ba6707fc50] asm_exc_page_fault at ffffffff86000ace
[exception RIP: svm_update_pi_irte+227]
RIP: ffffffffc0761b53 RSP: ffffb1ba6707fd08 RFLAGS: 00010086
RAX: ffffb1ba6707fd78 RBX: ffffb1ba66d91000 RCX: 0000000000000001
RDX: 00003c803f63f1c0 RSI: 000000000000019a RDI: ffffb1ba66db2ab8
RBP: 000000000000019a R8: 0000000000000040 R9: ffff94ca41b82200
R10: ffffffffffffffcf R11: 0000000000000001 R12: 0000000000000001
R13: 0000000000000001 R14: ffffffffffffffcf R15: 000000000000005f
ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018
#7 [ffffb1ba6707fdb8] kvm_irq_routing_update at ffffffffc09f19a1 [kvm]
#8 [ffffb1ba6707fde0] kvm_set_irq_routing at ffffffffc09f2133 [kvm]
#9 [ffffb1ba6707fe18] kvm_vm_ioctl at ffffffffc09ef544 [kvm]
RIP: 00007f143c36488b RSP: 00007f143a4e04b8 RFLAGS: 00000246
RAX: ffffffffffffffda RBX: 00007f05780041d0 RCX: 00007f143c36488b
RDX: 00007f05780041d0 RSI: 000000004008ae6a RDI: 0000000000000020
RBP: 00000000000004e8 R8: 0000000000000008 R9: 00007f05780041e0
R10: 00007f0578004560 R11: 0000000000000246 R12: 00000000000004e0
R13: 000000000000001a R14: 00007f1424001c60 R15: 00007f0578003bc0
ORIG_RAX: 0000000000000010 CS: 0033 SS: 002b
Vmx have been fix this in commit 3a8b0677fc61 (KVM: VMX: Do not BUG() on
out-of-bounds guest IRQ), so we can just copy source from that to fix
this.
Co-developed-by: Yi Liu <liu.yi24@zte.com.cn>
Signed-off-by: Yi Liu <liu.yi24@zte.com.cn>
Signed-off-by: Yi Wang <wang.yi59@zte.com.cn>
Message-Id: <20220309113025.44469-1-wang.yi59@zte.com.cn>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
If kvm->arch.tdp_mmu_zap_wq cannot be created, the failure has
to be propagated up to kvm_mmu_init_vm and kvm_arch_init_vm.
kvm_arch_init_vm also has to undo all the initialization, so
group all the MMU initialization code at the beginning and
handle cleaning up of kvm_page_track_init.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Setting non-zero values to SYNIC/STIMER MSRs activates certain features,
this should not happen when KVM_CAP_HYPERV_SYNIC{,2} was not activated.
Note, it would've been better to forbid writing anything to SYNIC/STIMER
MSRs, including zeroes, however, at least QEMU tries clearing
HV_X64_MSR_STIMER0_CONFIG without SynIC. HV_X64_MSR_EOM MSR is somewhat
'special' as writing zero there triggers an action, this also should not
happen when SynIC wasn't activated.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20220325132140.25650-4-vkuznets@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
kvm_irq_delivery_to_apic_fast()
When kvm_irq_delivery_to_apic_fast() is called with APIC_DEST_SELF
shorthand, 'src' must not be NULL. Crash the VM with KVM_BUG_ON()
instead of crashing the host.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20220325132140.25650-3-vkuznets@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
When KVM_CAP_HYPERV_SYNIC{,2} is activated, KVM already checks for
irqchip_in_kernel() so normally SynIC irqs should never be set. It is,
however, possible for a misbehaving VMM to write to SYNIC/STIMER MSRs
causing erroneous behavior.
The immediate issue being fixed is that kvm_irq_delivery_to_apic()
(kvm_irq_delivery_to_apic_fast()) crashes when called with
'irq.shorthand = APIC_DEST_SELF' and 'src == NULL'.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20220325132140.25650-2-vkuznets@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Clang warns:
arch/x86/kvm/cpuid.c:739:2: error: unannotated fall-through between switch labels [-Werror,-Wimplicit-fallthrough]
default:
^
arch/x86/kvm/cpuid.c:739:2: note: insert 'break;' to avoid fall-through
default:
^
break;
1 error generated.
Clang is a little more pedantic than GCC, which does not warn when
falling through to a case that is just break or return. Clang's version
is more in line with the kernel's own stance in deprecated.rst, which
states that all switch/case blocks must end in either break,
fallthrough, continue, goto, or return. Add the missing break to silence
the warning.
Fixes: f144c49e8c39 ("KVM: x86: synthesize CPUID leaf 0x80000021h if useful")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Message-Id: <20220322152906.112164-1-nathan@kernel.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Instead of using array_size, use a function that takes care of the
multiplication. While at it, switch to kvcalloc since this allocation
should not be very large.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
KVM_CAP_DISABLE_QUIRKS is irrevocably broken. The capability does not
advertise the set of quirks which may be disabled to userspace, so it is
impossible to predict the behavior of KVM. Worse yet,
KVM_CAP_DISABLE_QUIRKS will tolerate any value for cap->args[0], meaning
it fails to reject attempts to set invalid quirk bits.
The only valid workaround for the quirky quirks API is to add a new CAP.
Actually advertise the set of quirks that can be disabled to userspace
so it can predict KVM's behavior. Reject values for cap->args[0] that
contain invalid bits.
Finally, add documentation for the new capability and describe the
existing quirks.
Signed-off-by: Oliver Upton <oupton@google.com>
Message-Id: <20220301060351.442881-5-oupton@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Non constant TSC is a nightmare on bare metal already, but with
virtualization it becomes a complete disaster because the workarounds
are horrible latency wise. That's also a preliminary for running RT in
a guest on top of a RT host.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Message-Id: <Yh5eJSG19S2sjZfy@linutronix.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Guests X86_BUG_NULL_SEG if and only if the host has them. Use the info
from static_cpu_has_bug to form the 0x80000021 CPUID leaf that was
defined for Zen3. Userspace can then set the bit even on older CPUs
that do not have the bug, such as Zen2.
Do the same for X86_FEATURE_LFENCE_RDTSC as well, since various processors
have had very different ways of detecting it and not all of them are
available to userspace.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
CPUID leaf 0x80000021 defines some features (or lack of bugs) of AMD
processors. Expose the ones that make sense via KVM_GET_SUPPORTED_CPUID.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
KVM_X86_OP_OPTIONAL_RET0 can only be used with 32-bit return values on 32-bit
systems, because unsigned long is only 32-bits wide there and 64-bit values
are returned in edx:eax.
Reported-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
This reverts commit cf3e26427c08ad9015956293ab389004ac6a338e.
Multi-vCPU Hyper-V guests started crashing randomly on boot with the
latest kvm/queue and the problem can be bisected the problem to this
particular patch. Basically, I'm not able to boot e.g. 16-vCPU guest
successfully anymore. Both Intel and AMD seem to be affected. Reverting
the commit saves the day.
Reported-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Since "KVM: x86/mmu: Zap only TDP MMU leafs in kvm_zap_gfn_range()"
is going to be reverted, it's not going to be true anymore that
the zap-page flow does not free any 'struct kvm_mmu_page'. Introduce
an early flush before tdp_mmu_zap_leafs() returns, to preserve
bisectability.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD
KVM/arm64 updates for 5.18
- Proper emulation of the OSLock feature of the debug architecture
- Scalibility improvements for the MMU lock when dirty logging is on
- New VMID allocator, which will eventually help with SVA in VMs
- Better support for PMUs in heterogenous systems
- PSCI 1.1 support, enabling support for SYSTEM_RESET2
- Implement CONFIG_DEBUG_LIST at EL2
- Make CONFIG_ARM64_ERRATUM_2077057 default y
- Reduce the overhead of VM exit when no interrupt is pending
- Remove traces of 32bit ARM host support from the documentation
- Updated vgic selftests
- Various cleanups, doc updates and spelling fixes
|
|
Various spelling mistakes in comments.
Detected with the help of Coccinelle.
Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220318103729.157574-24-Julia.Lawall@inria.fr
|
|
We currently deal with a set of booleans for VM features,
while they could be better represented as set of flags
contained in an unsigned long, similarily to what we are
doing on the CPU side.
Signed-off-by: Marc Zyngier <maz@kernel.org>
[Oliver: Flag-ify the 'ran_once' boolean]
Signed-off-by: Oliver Upton <oupton@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220311174001.605719-2-oupton@google.com
|
|
KVM/riscv changes for 5.18
- Prevent KVM_COMPAT from being selected
- Refine __kvm_riscv_switch_to() implementation
- RISC-V SBI v0.3 support
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into HEAD
KVM: s390: Fix, test and feature for 5.18 part 2
- memop selftest
- fix SCK locking
- adapter interruptions virtualization for secure guests
|
|
When handling the SCK instruction, the kvm lock is taken, even though
the vcpu lock is already being held. The normal locking order is kvm
lock first and then vcpu lock. This is can (and in some circumstances
does) lead to deadlocks.
The function kvm_s390_set_tod_clock is called both by the SCK handler
and by some IOCTLs to set the clock. The IOCTLs will not hold the vcpu
lock, so they can safely take the kvm lock. The SCK handler holds the
vcpu lock, but will also somehow need to acquire the kvm lock without
relinquishing the vcpu lock.
The solution is to factor out the code to set the clock, and provide
two wrappers. One is called like the original function and does the
locking, the other is called kvm_s390_try_set_tod_clock and uses
trylock to try to acquire the kvm lock. This new wrapper is then used
in the SCK handler. If locking fails, -EAGAIN is returned, which is
eventually propagated to userspace, thus also freeing the vcpu lock and
allowing for forward progress.
This is not the most efficient or elegant way to solve this issue, but
the SCK instruction is deprecated and its performance is not critical.
The goal of this patch is just to provide a simple but correct way to
fix the bug.
Fixes: 6a3f95a6b04c ("KVM: s390: Intercept SCK instruction")
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Link: https://lore.kernel.org/r/20220301143340.111129-1-imbrenda@linux.ibm.com
Cc: stable@vger.kernel.org
Signed-off-by: Christian Borntraeger <borntraeger@linux.ibm.com>
|
|
The SBI v0.3 specification extends SBI HSM extension by adding SBI HSM
suspend call and related HART states. This patch extends the KVM RISC-V
HSM implementation to provide KVM guest a minimal SBI HSM suspend call
which is equivalent to a WFI instruction.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
The wait for interrupt (WFI) instruction emulation can share the VCPU
halt logic with SBI HSM suspend emulation so this patch adds a common
kvm_riscv_vcpu_wfi() function for this purpose.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
We add defines related to SBI HSM suspend call and also update HSM states
naming as-per the latest SBI specification.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
The SBI v0.3 specification defines SRST (System Reset) extension which
provides a standard poweroff and reboot interface. This patch implements
SRST extension for the KVM Guest.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
We rename kvm_sbi_system_shutdown() to kvm_riscv_vcpu_sbi_system_reset()
and move it to vcpu_sbi.c so that it can be shared by SBI v0.1 shutdown
and SBI v0.3 SRST extension.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
We upgrade SBI spec version implemented by KVM RISC-V to v0.3 so
that Guest kernel can probe and use SBI extensions added by the
SBI v0.3 specification.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
Kernel uses __kvm_riscv_switch_to() and __kvm_switch_return() to switch
the context of host kernel and guest kernel. Several CSRs belonging to the
context will be read and written during the context switch. To ensure
atomic read-modify-write control of CSR and ordering of CSR accesses, some
hardware blocks flush the pipeline when writing a CSR. In this
circumstance, grouping CSR executions together as much as possible can
reduce the performance impact of the pipeline. Therefore, this commit
reorders the CSR instructions to enhance the context switch performance..
Signed-off-by: Vincent Chen <vincent.chen@sifive.com>
Suggested-by: Hsinyi Lee <hsinyi.lee@sifive.com>
Suggested-by: Fu-Ching Yang <fu-ching.yang@sifive.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
Eliminate the following coccicheck warning:
./arch/riscv/kvm/vcpu_sbi_v01.c:117:2-3: Unneeded semicolon
Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
|
|
* kvm-arm64/psci-1.1:
: .
: Limited PSCI-1.1 support from Will Deacon:
:
: This small series exposes the PSCI SYSTEM_RESET2 call to guests, which
: allows the propagation of a "reset_type" and a "cookie" back to the VMM.
: Although Linux guests only ever pass 0 for the type ("SYSTEM_WARM_RESET"),
: the vendor-defined range can be used by a bootloader to provide additional
: information about the reset, such as an error code.
: .
KVM: arm64: Really propagate PSCI SYSTEM_RESET2 arguments to userspace
Signed-off-by: Marc Zyngier <maz@kernel.org>
|
|
Commit d43583b890e7 ("KVM: arm64: Expose PSCI SYSTEM_RESET2 call to the
guest") hooked up the SYSTEM_RESET2 PSCI call for guests but failed to
preserve its arguments for userspace, instead overwriting them with
zeroes via smccc_set_retval(). As Linux only passes zeroes for these
arguments, this appeared to be working for Linux guests. Oh well.
Don't call smccc_set_retval() for a SYSTEM_RESET2 heading to userspace
and instead set X0 (and only X0) explicitly to PSCI_RET_INTERNAL_FAILURE
just in case the vCPU re-enters the guest.
Fixes: d43583b890e7 ("KVM: arm64: Expose PSCI SYSTEM_RESET2 call to the guest")
Reported-by: Andrew Walbran <qwandor@google.com>
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220309181308.982-1-will@kernel.org
|
|
* kvm-arm64/misc-5.18:
: .
: Misc fixes for KVM/arm64 5.18:
:
: - Drop unused kvm parameter to kvm_psci_version()
:
: - Implement CONFIG_DEBUG_LIST at EL2
:
: - Make CONFIG_ARM64_ERRATUM_2077057 default y
:
: - Only do the interrupt dance if we have exited because of an interrupt
:
: - Remove traces of 32bit ARM host support from the documentation
: .
Documentation: KVM: Update documentation to indicate KVM is arm64-only
KVM: arm64: Only open the interrupt window on exit due to an interrupt
KVM: arm64: Enable Cortex-A510 erratum 2077057 by default
Signed-off-by: Marc Zyngier <maz@kernel.org>
|
|
Expand KVM's mask for the AVIC host physical ID to the full 12 bits defined
by the architecture. The number of bits consumed by hardware is model
specific, e.g. early CPUs ignored bits 11:8, but there is no way for KVM
to enumerate the "true" size. So, KVM must allow using all bits, else it
risks rejecting completely legal x2APIC IDs on newer CPUs.
This means KVM relies on hardware to not assign x2APIC IDs that exceed the
"true" width of the field, but presumably hardware is smart enough to tie
the width to the max x2APIC ID. KVM also relies on hardware to support at
least 8 bits, as the legacy xAPIC ID is writable by software. But, those
assumptions are unavoidable due to the lack of any way to enumerate the
"true" width.
Cc: stable@vger.kernel.org
Cc: Maxim Levitsky <mlevitsk@redhat.com>
Suggested-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Sean Christopherson <seanjc@google.com>
Fixes: 44a95dae1d22 ("KVM: x86: Detect and Initialize AVIC support")
Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Message-Id: <20220211000851.185799-1-suravee.suthikulpanit@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Disallow calling tdp_mmu_set_spte_atomic() with a REMOVED "old" SPTE.
This solves a conundrum introduced by commit 3255530ab191 ("KVM: x86/mmu:
Automatically update iter->old_spte if cmpxchg fails"); if the helper
doesn't update old_spte in the REMOVED case, then theoretically the
caller could get stuck in an infinite loop as it will fail indefinitely
on the REMOVED SPTE. E.g. until recently, clear_dirty_gfn_range() didn't
check for a present SPTE and would have spun until getting rescheduled.
In practice, only the page fault path should "create" a new SPTE, all
other paths should only operate on existing, a.k.a. shadow present,
SPTEs. Now that the page fault path pre-checks for a REMOVED SPTE in all
cases, require all other paths to indirectly pre-check by verifying the
target SPTE is a shadow-present SPTE.
Note, this does not guarantee the actual SPTE isn't REMOVED, nor is that
scenario disallowed. The invariant is only that the caller mustn't
invoke tdp_mmu_set_spte_atomic() if the SPTE was REMOVED when last
observed by the caller.
Cc: David Matlack <dmatlack@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220226001546.360188-25-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Explicitly check for a REMOVED leaf SPTE prior to attempting to map
the final SPTE when handling a TDP MMU fault. Functionally, this is a
nop as tdp_mmu_set_spte_atomic() will eventually detect the frozen SPTE.
Pre-checking for a REMOVED SPTE is a minor optmization, but the real goal
is to allow tdp_mmu_set_spte_atomic() to have an invariant that the "old"
SPTE is never a REMOVED SPTE.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Message-Id: <20220226001546.360188-24-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Zap defunct roots, a.k.a. roots that have been invalidated after their
last reference was initially dropped, asynchronously via the existing work
queue instead of forcing the work upon the unfortunate task that happened
to drop the last reference.
If a vCPU task drops the last reference, the vCPU is effectively blocked
by the host for the entire duration of the zap. If the root being zapped
happens be fully populated with 4kb leaf SPTEs, e.g. due to dirty logging
being active, the zap can take several hundred seconds. Unsurprisingly,
most guests are unhappy if a vCPU disappears for hundreds of seconds.
E.g. running a synthetic selftest that triggers a vCPU root zap with
~64tb of guest memory and 4kb SPTEs blocks the vCPU for 900+ seconds.
Offloading the zap to a worker drops the block time to <100ms.
There is an important nuance to this change. If the same work item
was queued twice before the work function has run, it would only
execute once and one reference would be leaked. Therefore, now that
queueing and flushing items is not anymore protected by kvm->slots_lock,
kvm_tdp_mmu_invalidate_all_roots() has to check root->role.invalid and
skip already invalid roots. On the other hand, kvm_mmu_zap_all_fast()
must return only after those skipped roots have been zapped as well.
These two requirements can be satisfied only if _all_ places that
change invalid to true now schedule the worker before releasing the
mmu_lock. There are just two, kvm_tdp_mmu_put_root() and
kvm_tdp_mmu_invalidate_all_roots().
Co-developed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Message-Id: <20220226001546.360188-23-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
When zapping a TDP MMU root, perform the zap in two passes to avoid
zapping an entire top-level SPTE while holding RCU, which can induce RCU
stalls. In the first pass, zap SPTEs at PG_LEVEL_1G, and then
zap top-level entries in the second pass.
With 4-level paging, zapping a PGD that is fully populated with 4kb leaf
SPTEs take up to ~7 or so seconds (time varies based on kernel config,
number of (v)CPUs, etc...). With 5-level paging, that time can balloon
well into hundreds of seconds.
Before remote TLB flushes were omitted, the problem was even worse as
waiting for all active vCPUs to respond to the IPI introduced significant
overhead for VMs with large numbers of vCPUs.
By zapping 1gb SPTEs (both shadow pages and hugepages) in the first pass,
the amount of work that is done without dropping RCU protection is
strictly bounded, with the worst case latency for a single operation
being less than 100ms.
Zapping at 1gb in the first pass is not arbitrary. First and foremost,
KVM relies on being able to zap 1gb shadow pages in a single shot when
when repacing a shadow page with a hugepage. Zapping a 1gb shadow page
that is fully populated with 4kb dirty SPTEs also triggers the worst case
latency due writing back the struct page accessed/dirty bits for each 4kb
page, i.e. the two-pass approach is guaranteed to work so long as KVM can
cleany zap a 1gb shadow page.
rcu: INFO: rcu_sched self-detected stall on CPU
rcu: 52-....: (20999 ticks this GP) idle=7be/1/0x4000000000000000
softirq=15759/15759 fqs=5058
(t=21016 jiffies g=66453 q=238577)
NMI backtrace for cpu 52
Call Trace:
...
mark_page_accessed+0x266/0x2f0
kvm_set_pfn_accessed+0x31/0x40
handle_removed_tdp_mmu_page+0x259/0x2e0
__handle_changed_spte+0x223/0x2c0
handle_removed_tdp_mmu_page+0x1c1/0x2e0
__handle_changed_spte+0x223/0x2c0
handle_removed_tdp_mmu_page+0x1c1/0x2e0
__handle_changed_spte+0x223/0x2c0
zap_gfn_range+0x141/0x3b0
kvm_tdp_mmu_zap_invalidated_roots+0xc8/0x130
kvm_mmu_zap_all_fast+0x121/0x190
kvm_mmu_invalidate_zap_pages_in_memslot+0xe/0x10
kvm_page_track_flush_slot+0x5c/0x80
kvm_arch_flush_shadow_memslot+0xe/0x10
kvm_set_memslot+0x172/0x4e0
__kvm_set_memory_region+0x337/0x590
kvm_vm_ioctl+0x49c/0xf80
Reported-by: David Matlack <dmatlack@google.com>
Cc: Ben Gardon <bgardon@google.com>
Cc: Mingwei Zhang <mizhang@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Message-Id: <20220226001546.360188-22-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Allow yielding when zapping SPTEs after the last reference to a valid
root is put. Because KVM must drop all SPTEs in response to relevant
mmu_notifier events, mark defunct roots invalid and reset their refcount
prior to zapping the root. Keeping the refcount elevated while the zap
is in-progress ensures the root is reachable via mmu_notifier until the
zap completes and the last reference to the invalid, defunct root is put.
Allowing kvm_tdp_mmu_put_root() to yield fixes soft lockup issues if the
root in being put has a massive paging structure, e.g. zapping a root
that is backed entirely by 4kb pages for a guest with 32tb of memory can
take hundreds of seconds to complete.
watchdog: BUG: soft lockup - CPU#49 stuck for 485s! [max_guest_memor:52368]
RIP: 0010:kvm_set_pfn_dirty+0x30/0x50 [kvm]
__handle_changed_spte+0x1b2/0x2f0 [kvm]
handle_removed_tdp_mmu_page+0x1a7/0x2b8 [kvm]
__handle_changed_spte+0x1f4/0x2f0 [kvm]
handle_removed_tdp_mmu_page+0x1a7/0x2b8 [kvm]
__handle_changed_spte+0x1f4/0x2f0 [kvm]
tdp_mmu_zap_root+0x307/0x4d0 [kvm]
kvm_tdp_mmu_put_root+0x7c/0xc0 [kvm]
kvm_mmu_free_roots+0x22d/0x350 [kvm]
kvm_mmu_reset_context+0x20/0x60 [kvm]
kvm_arch_vcpu_ioctl_set_sregs+0x5a/0xc0 [kvm]
kvm_vcpu_ioctl+0x5bd/0x710 [kvm]
__se_sys_ioctl+0x77/0xc0
__x64_sys_ioctl+0x1d/0x20
do_syscall_64+0x44/0xa0
entry_SYSCALL_64_after_hwframe+0x44/0xae
KVM currently doesn't put a root from a non-preemptible context, so other
than the mmu_notifier wrinkle, yielding when putting a root is safe.
Yield-unfriendly iteration uses for_each_tdp_mmu_root(), which doesn't
take a reference to each root (it requires mmu_lock be held for the
entire duration of the walk).
tdp_mmu_next_root() is used only by the yield-friendly iterator.
tdp_mmu_zap_root_work() is explicitly yield friendly.
kvm_mmu_free_roots() => mmu_free_root_page() is a much bigger fan-out,
but is still yield-friendly in all call sites, as all callers can be
traced back to some combination of vcpu_run(), kvm_destroy_vm(), and/or
kvm_create_vm().
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220226001546.360188-21-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Use the system worker threads to zap the roots invalidated
by the TDP MMU's "fast zap" mechanism, implemented by
kvm_tdp_mmu_invalidate_all_roots().
At this point, apart from allowing some parallelism in the zapping of
roots, the workqueue is a glorified linked list: work items are added and
flushed entirely within a single kvm->slots_lock critical section. However,
the workqueue fixes a latent issue where kvm_mmu_zap_all_invalidated_roots()
assumes that it owns a reference to all invalid roots; therefore, no
one can set the invalid bit outside kvm_mmu_zap_all_fast(). Putting the
invalidated roots on a linked list... erm, on a workqueue ensures that
tdp_mmu_zap_root_work() only puts back those extra references that
kvm_mmu_zap_all_invalidated_roots() had gifted to it.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Defer TLB flushes to the caller when freeing TDP MMU shadow pages instead
of immediately flushing. Because the shadow pages are freed in an RCU
callback, so long as at least one CPU holds RCU, all CPUs are protected.
For vCPUs running in the guest, i.e. consuming TLB entries, KVM only
needs to ensure the caller services the pending TLB flush before dropping
its RCU protections. I.e. use the caller's RCU as a proxy for all vCPUs
running in the guest.
Deferring the flushes allows batching flushes, e.g. when installing a
1gb hugepage and zapping a pile of SPs. And when zapping an entire root,
deferring flushes allows skipping the flush entirely (because flushes are
not needed in that case).
Avoiding flushes when zapping an entire root is especially important as
synchronizing with other CPUs via IPI after zapping every shadow page can
cause significant performance issues for large VMs. The issue is
exacerbated by KVM zapping entire top-level entries without dropping
RCU protection, which can lead to RCU stalls even when zapping roots
backing relatively "small" amounts of guest memory, e.g. 2tb. Removing
the IPI bottleneck largely mitigates the RCU issues, though it's likely
still a problem for 5-level paging. A future patch will further address
the problem by zapping roots in multiple passes to avoid holding RCU for
an extended duration.
Reviewed-by: Ben Gardon <bgardon@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220226001546.360188-20-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
When yielding in the TDP MMU iterator, service any pending TLB flush
before dropping RCU protections in anticipation of using the caller's RCU
"lock" as a proxy for vCPUs in the guest.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Message-Id: <20220226001546.360188-19-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Zap only leaf SPTEs in the TDP MMU's zap_gfn_range(), and rename various
functions accordingly. When removing mappings for functional correctness
(except for the stupid VFIO GPU passthrough memslots bug), zapping the
leaf SPTEs is sufficient as the paging structures themselves do not point
at guest memory and do not directly impact the final translation (in the
TDP MMU).
Note, this aligns the TDP MMU with the legacy/full MMU, which zaps only
the rmaps, a.k.a. leaf SPTEs, in kvm_zap_gfn_range() and
kvm_unmap_gfn_range().
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Message-Id: <20220226001546.360188-18-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Now that all callers of zap_gfn_range() hold mmu_lock for write, drop
support for zapping with mmu_lock held for read. That all callers hold
mmu_lock for write isn't a random coincidence; now that the paths that
need to zap _everything_ have their own path, the only callers left are
those that need to zap for functional correctness. And when zapping is
required for functional correctness, mmu_lock must be held for write,
otherwise the caller has no guarantees about the state of the TDP MMU
page tables after it has run, e.g. the SPTE(s) it zapped can be
immediately replaced by a vCPU faulting in a page.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Message-Id: <20220226001546.360188-17-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Add a dedicated helper for zapping a TDP MMU root, and use it in the three
flows that do "zap_all" and intentionally do not do a TLB flush if SPTEs
are zapped (zapping an entire root is safe if and only if it cannot be in
use by any vCPU). Because a TLB flush is never required, unconditionally
pass "false" to tdp_mmu_iter_cond_resched() when potentially yielding.
Opportunistically document why KVM must not yield when zapping roots that
are being zapped by kvm_tdp_mmu_put_root(), i.e. roots whose refcount has
reached zero, and further harden the flow to detect improper KVM behavior
with respect to roots that are supposed to be unreachable.
In addition to hardening zapping of roots, isolating zapping of roots
will allow future simplification of zap_gfn_range() by having it zap only
leaf SPTEs, and by removing its tricky "zap all" heuristic. By having
all paths that truly need to free _all_ SPs flow through the dedicated
root zapper, the generic zapper can be freed of those concerns.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Message-Id: <20220226001546.360188-16-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Don't flush the TLBs when zapping all TDP MMU pages, as the only time KVM
uses the slow version of "zap everything" is when the VM is being
destroyed or the owning mm has exited. In either case, KVM_RUN is
unreachable for the VM, i.e. the guest TLB entries cannot be consumed.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Message-Id: <20220226001546.360188-15-seanjc@google.com>
Reviewed-by: Mingwei Zhang <mizhang@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
When recovering a potential hugepage that was shattered for the iTLB
multihit workaround, precisely zap only the target page instead of
iterating over the TDP MMU to find the SP that was passed in. This will
allow future simplification of zap_gfn_range() by having it zap only
leaf SPTEs.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220226001546.360188-14-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Refactor __tdp_mmu_set_spte() to work with raw values instead of a
tdp_iter objects so that a future patch can modify SPTEs without doing a
walk, and without having to synthesize a tdp_iter.
No functional change intended.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Message-Id: <20220226001546.360188-13-seanjc@google.com>
Reviewed-by: Mingwei Zhang <mizhang@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
WARN if the new_spte being set by __tdp_mmu_set_spte() is a REMOVED_SPTE,
which is called out by the comment as being disallowed but not actually
checked. Keep the WARN on the old_spte as well, because overwriting a
REMOVED_SPTE in the non-atomic path is also disallowed (as evidence by
lack of splats with the existing WARN).
Fixes: 08f07c800e9d ("KVM: x86/mmu: Flush TLBs after zap in TDP MMU PF handler")
Cc: Ben Gardon <bgardon@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Message-Id: <20220226001546.360188-12-seanjc@google.com>
Reviewed-by: Mingwei Zhang <mizhang@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
