| Age | Commit message (Collapse) | Author |
|---|
|
Use tdp_mmu_clear_spte_bits() when clearing the Accessed bit in TDP MMU
SPTEs so as to use an atomic-AND instead of XCHG to clear the A-bit.
Similar to the D-bit story, this will allow KVM to bypass
__handle_changed_spte() by ensuring only the A-bit is modified.
Link: https://lore.kernel.org/all/Y9HcHRBShQgjxsQb@google.com
Signed-off-by: Vipin Sharma <vipinsh@google.com>
Reviewed-by: David Matlack <dmatlack@google.com>
[sean: massage changelog]
Link: https://lore.kernel.org/r/20230321220021.2119033-9-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Remove bool parameter "record_dirty_log" from __tdp_mmu_set_spte() and
refactor the code as this variable is always set to true by its caller.
Signed-off-by: Vipin Sharma <vipinsh@google.com>
Reviewed-by: David Matlack <dmatlack@google.com>
Link: https://lore.kernel.org/r/20230321220021.2119033-8-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Drop everything except marking the PFN dirty and the relevant tracepoint
parts of __handle_changed_spte() when clearing the dirty status of gfns in
the TDP MMU. Clearing only the Dirty (or Writable) bit doesn't affect
the SPTEs shadow-present status, whether or not the SPTE is a leaf, or
change the SPTE's PFN. I.e. other than marking the PFN dirty, none of the
functional updates handled by __handle_changed_spte() are relevant.
Losing __handle_changed_spte()'s sanity checks does mean that a bug could
theoretical go unnoticed, but that scenario is extremely unlikely, e.g.
would effectively require a misconfigured or a locking bug elsewhere.
Opportunistically remove a comment blurb from __handle_changed_spte()
about all modifications to TDP MMU SPTEs needing to invoke said function,
that "rule" hasn't been true since fast page fault support was added for
the TDP MMU (and perhaps even before).
Tested on a VM (160 vCPUs, 160 GB memory) and found that performance of
clear dirty log stage improved by ~40% in dirty_log_perf_test (with the
full optimization applied).
Before optimization:
--------------------
Iteration 1 clear dirty log time: 3.638543593s
Iteration 2 clear dirty log time: 3.145032742s
Iteration 3 clear dirty log time: 3.142340358s
Clear dirty log over 3 iterations took 9.925916693s. (Avg 3.308638897s/iteration)
After optimization:
-------------------
Iteration 1 clear dirty log time: 2.318988110s
Iteration 2 clear dirty log time: 1.794470164s
Iteration 3 clear dirty log time: 1.791668628s
Clear dirty log over 3 iterations took 5.905126902s. (Avg 1.968375634s/iteration)
Link: https://lore.kernel.org/all/Y9hXmz%2FnDOr1hQal@google.com
Signed-off-by: Vipin Sharma <vipinsh@google.com>
Reviewed-by: David Matlack <dmatlack@google.com>
[sean: split the switch to atomic-AND to a separate patch]
Link: https://lore.kernel.org/r/20230321220021.2119033-7-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Drop the unnecessary call to handle access-tracking changes when clearing
the dirty status of TDP MMU SPTEs. Neither the Dirty bit nor the Writable
bit has any impact on the accessed state of a page, i.e. clearing only
the aforementioned bits doesn't make an accessed SPTE suddently not
accessed.
Signed-off-by: Vipin Sharma <vipinsh@google.com>
[sean: split to separate patch, write changelog]
Link: https://lore.kernel.org/r/20230321220021.2119033-6-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Optimize the clearing of dirty state in TDP MMU SPTEs by doing an
atomic-AND (on SPTEs that have volatile bits) instead of the full XCHG
that currently ends up being invoked (see kvm_tdp_mmu_write_spte()).
Clearing _only_ the bit in question will allow KVM to skip the many
irrelevant checks in __handle_changed_spte() by avoiding any collateral
damage due to the XCHG writing all SPTE bits, e.g. the XCHG could race
with fast_page_fault() setting the W-bit and the CPU setting the D-bit,
and thus incorrectly drop the CPU's D-bit update.
Link: https://lore.kernel.org/all/Y9hXmz%2FnDOr1hQal@google.com
Signed-off-by: Vipin Sharma <vipinsh@google.com>
Reviewed-by: David Matlack <dmatlack@google.com>
[sean: split the switch to atomic-AND to a separate patch]
Link: https://lore.kernel.org/r/20230321220021.2119033-5-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Deduplicate the guts of the TDP MMU's clearing of dirty status by
snapshotting whether to check+clear the Dirty bit vs. the Writable bit,
which is the only difference between the two flavors of dirty tracking.
Note, kvm_ad_enabled() is just a wrapper for shadow_accessed_mask, i.e.
is constant after kvm-{intel,amd}.ko is loaded.
Link: https://lore.kernel.org/all/Yz4Qi7cn7TWTWQjj@google.com
Signed-off-by: Vipin Sharma <vipinsh@google.com>
[sean: split to separate patch, apply to dirty log, write changelog]
Link: https://lore.kernel.org/r/20230321220021.2119033-4-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Use the constant-after-module-load kvm_ad_enabled() to check if SPTEs in
the TDP MMU need to be write-protected when clearing accessed/dirty status
instead of manually checking every SPTE. The per-SPTE A/D enabling is
specific to nested EPT MMUs, i.e. when KVM is using EPT A/D bits but L1 is
not, and so cannot happen in the TDP MMU (which is non-nested only).
Keep the original code as sanity checks buried under MMU_WARN_ON().
MMU_WARN_ON() is more or less useless at the moment, but there are plans
to change that.
Link: https://lore.kernel.org/all/Yz4Qi7cn7TWTWQjj@google.com
Signed-off-by: Vipin Sharma <vipinsh@google.com>
[sean: split to separate patch, apply to dirty path, write changelog]
Link: https://lore.kernel.org/r/20230321220021.2119033-3-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Move conditions in kvm_tdp_mmu_write_spte() to check if an SPTE should
be written atomically or not to a separate function.
This new function, kvm_tdp_mmu_spte_need_atomic_write(), will be used
in future commits to optimize clearing bits in SPTEs.
Signed-off-by: Vipin Sharma <vipinsh@google.com>
Reviewed-by: David Matlack <dmatlack@google.com>
Reviewed-by: Ben Gardon <bgardon@google.com>
Link: https://lore.kernel.org/r/20230321220021.2119033-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
When introduced, IRQFD resampling worked on POWER8 with XICS. However
KVM on POWER9 has never implemented it - the compatibility mode code
("XICS-on-XIVE") misses the kvm_notify_acked_irq() call and the native
XIVE mode does not handle INTx in KVM at all.
This moved the capability support advertising to platforms and stops
advertising it on XIVE, i.e. POWER9 and later.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Anup Patel <anup@brainfault.org>
Acked-by: Nicholas Piggin <npiggin@gmail.com>
Message-Id: <20220504074807.3616813-1-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Don't report an error code to L1 when synthesizing a nested VM-Exit and
L2 is in Real Mode. Per Intel's SDM, regarding the error code valid bit:
This bit is always 0 if the VM exit occurred while the logical processor
was in real-address mode (CR0.PE=0).
The bug was introduced by a recent fix for AMD's Paged Real Mode, which
moved the error code suppression from the common "queue exception" path
to the "inject exception" path, but missed VMX's "synthesize VM-Exit"
path.
Fixes: b97f07458373 ("KVM: x86: determine if an exception has an error code only when injecting it.")
Cc: stable@vger.kernel.org
Cc: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20230322143300.2209476-3-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
When injecting an exception into a vCPU in Real Mode, suppress the error
code by clearing the flag that tracks whether the error code is valid, not
by clearing the error code itself. The "typo" was introduced by recent
fix for SVM's funky Paged Real Mode.
Opportunistically hoist the logic above the tracepoint so that the trace
is coherent with respect to what is actually injected (this was also the
behavior prior to the buggy commit).
Fixes: b97f07458373 ("KVM: x86: determine if an exception has an error code only when injecting it.")
Cc: stable@vger.kernel.org
Cc: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20230322143300.2209476-2-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
Clear vcpu->mmio_needed when injecting an exception from the emulator to
squash a (legitimate) warning about vcpu->mmio_needed being true at the
start of KVM_RUN without a callback being registered to complete the
userspace MMIO exit. Suppressing the MMIO write exit is inarguably wrong
from an architectural perspective, but it is the least awful hack-a-fix
due to shortcomings in KVM's uAPI, not to mention that KVM already
suppresses MMIO writes in this scenario.
Outside of REP string instructions, KVM doesn't provide a way to resume
an instruction at the exact point where it was "interrupted" if said
instruction partially completed before encountering an MMIO access. For
MMIO reads, KVM immediately exits to userspace upon detecting MMIO as
userspace provides the to-be-read value in a buffer, and so KVM can safely
(more or less) restart the instruction from the beginning. When the
emulator re-encounters the MMIO read, KVM will service the MMIO by getting
the value from the buffer instead of exiting to userspace, i.e. KVM won't
put the vCPU into an infinite loop.
On an emulated MMIO write, KVM finishes the instruction before exiting to
userspace, as exiting immediately would ultimately hang the vCPU due to
the aforementioned shortcoming of KVM not being able to resume emulation
in the middle of an instruction.
For the vast majority of _emulated_ instructions, deferring the userspace
exit doesn't cause problems as very few x86 instructions (again ignoring
string operations) generate multiple writes. But for instructions that
generate multiple writes, e.g. PUSHA (multiple pushes onto the stack),
deferring the exit effectively results in only the final write triggering
an exit to userspace. KVM does support multiple MMIO "fragments", but
only for page splits; if an instruction performs multiple distinct MMIO
writes, the number of fragments gets reset when the next MMIO write comes
along and any previous MMIO writes are dropped.
Circling back to the warning, if a deferred MMIO write coincides with an
exception, e.g. in this case a #SS due to PUSHA underflowing the stack
after queueing a write to an MMIO page on a previous push, KVM injects
the exceptions and leaves the deferred MMIO pending without registering a
callback, thus triggering the splat.
Sweep the problem under the proverbial rug as dropping MMIO writes is not
unique to the exception scenario (see above), i.e. instructions like PUSHA
are fundamentally broken with respect to MMIO, and have been since KVM's
inception.
Reported-by: zhangjianguo <zhangjianguo18@huawei.com>
Reported-by: syzbot+760a73552f47a8cd0fd9@syzkaller.appspotmail.com
Reported-by: syzbot+8accb43ddc6bd1f5713a@syzkaller.appspotmail.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20230322141220.2206241-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
KVM irqfd based emulation of level-triggered interrupts doesn't work
quite correctly in some cases, particularly in the case of interrupts
that are handled in a Linux guest as oneshot interrupts (IRQF_ONESHOT).
Such an interrupt is acked to the device in its threaded irq handler,
i.e. later than it is acked to the interrupt controller (EOI at the end
of hardirq), not earlier.
Linux keeps such interrupt masked until its threaded handler finishes,
to prevent the EOI from re-asserting an unacknowledged interrupt.
However, with KVM + vfio (or whatever is listening on the resamplefd)
we always notify resamplefd at the EOI, so vfio prematurely unmasks the
host physical IRQ, thus a new physical interrupt is fired in the host.
This extra interrupt in the host is not a problem per se. The problem is
that it is unconditionally queued for injection into the guest, so the
guest sees an extra bogus interrupt. [*]
There are observed at least 2 user-visible issues caused by those
extra erroneous interrupts for a oneshot irq in the guest:
1. System suspend aborted due to a pending wakeup interrupt from
ChromeOS EC (drivers/platform/chrome/cros_ec.c).
2. Annoying "invalid report id data" errors from ELAN0000 touchpad
(drivers/input/mouse/elan_i2c_core.c), flooding the guest dmesg
every time the touchpad is touched.
The core issue here is that by the time when the guest unmasks the IRQ,
the physical IRQ line is no longer asserted (since the guest has
acked the interrupt to the device in the meantime), yet we
unconditionally inject the interrupt queued into the guest by the
previous resampling. So to fix the issue, we need a way to detect that
the IRQ is no longer pending, and cancel the queued interrupt in this
case.
With IOAPIC we are not able to probe the physical IRQ line state
directly (at least not if the underlying physical interrupt controller
is an IOAPIC too), so in this patch we use irqfd resampler for that.
Namely, instead of injecting the queued interrupt, we just notify the
resampler that this interrupt is done. If the IRQ line is actually
already deasserted, we are done. If it is still asserted, a new
interrupt will be shortly triggered through irqfd and injected into the
guest.
In the case if there is no irqfd resampler registered for this IRQ, we
cannot fix the issue, so we keep the existing behavior: immediately
unconditionally inject the queued interrupt.
This patch fixes the issue for x86 IOAPIC only. In the long run, we can
fix it for other irqchips and other architectures too, possibly taking
advantage of reading the physical state of the IRQ line, which is
possible with some other irqchips (e.g. with arm64 GIC, maybe even with
the legacy x86 PIC).
[*] In this description we assume that the interrupt is a physical host
interrupt forwarded to the guest e.g. by vfio. Potentially the same
issue may occur also with a purely virtual interrupt from an
emulated device, e.g. if the guest handles this interrupt, again, as
a oneshot interrupt.
Signed-off-by: Dmytro Maluka <dmy@semihalf.com>
Link: https://lore.kernel.org/kvm/31420943-8c5f-125c-a5ee-d2fde2700083@semihalf.com/
Link: https://lore.kernel.org/lkml/87o7wrug0w.wl-maz@kernel.org/
Message-Id: <20230322204344.50138-3-dmy@semihalf.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
The Hyper-V "EnlightenedNptTlb" enlightenment is always enabled when KVM
is running on top of Hyper-V and Hyper-V exposes support for it (which
is always). On AMD CPUs this enlightenment results in ASID invalidations
not flushing TLB entries derived from the NPT. To force the underlying
(L0) hypervisor to rebuild its shadow page tables, an explicit hypercall
is needed.
The original KVM implementation of Hyper-V's "EnlightenedNptTlb" on SVM
only added remote TLB flush hooks. This worked out fine for a while, as
sufficient remote TLB flushes where being issued in KVM to mask the
problem. Since v5.17, changes in the TDP code reduced the number of
flushes and the out-of-sync TLB prevents guests from booting
successfully.
Split svm_flush_tlb_current() into separate callbacks for the 3 cases
(guest/all/current), and issue the required Hyper-V hypercall when a
Hyper-V TLB flush is needed. The most important case where the TLB flush
was missing is when loading a new PGD, which is followed by what is now
svm_flush_tlb_current().
Cc: stable@vger.kernel.org # v5.17+
Fixes: 1e0c7d40758b ("KVM: SVM: hyper-v: Remote TLB flush for SVM")
Link: https://lore.kernel.org/lkml/43980946-7bbf-dcef-7e40-af904c456250@linux.microsoft.com/
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Jeremi Piotrowski <jpiotrowski@linux.microsoft.com>
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20230324145233.4585-1-jpiotrowski@linux.microsoft.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
|
|
To be able to trace invocations of smp_send_reschedule(), rename the
arch-specific definitions of it to arch_smp_send_reschedule() and wrap it
into an smp_send_reschedule() that contains a tracepoint.
Changes to include the declaration of the tracepoint were driven by the
following coccinelle script:
@func_use@
@@
smp_send_reschedule(...);
@include@
@@
#include <trace/events/ipi.h>
@no_include depends on func_use && !include@
@@
#include <...>
+
+ #include <trace/events/ipi.h>
[csky bits]
[riscv bits]
Signed-off-by: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Guo Ren <guoren@kernel.org>
Acked-by: Palmer Dabbelt <palmer@rivosinc.com>
Link: https://lore.kernel.org/r/20230307143558.294354-6-vschneid@redhat.com
|
|
If !guest_cpuid_has(vcpu, X86_FEATURE_PCID), CR4.PCIDE would have been in
vcpu->arch.cr4_guest_rsvd_bits and failed earlier kvm_is_valid_cr4() check.
Remove this meaningless check.
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Signed-off-by: Robert Hoo <robert.hu@linux.intel.com>
Fixes: 4683d758f48e ("KVM: x86: Supplement __cr4_reserved_bits() with X86_FEATURE_PCID check")
Link: https://lore.kernel.org/r/20230308072936.1293101-1-robert.hu@intel.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Yell loudly if KVM attempts to load CS outside of Real Mode without an
accompanying control transfer type, i.e. on X86_TRANSFER_NONE. KVM uses
X86_TRANSFER_NONE when emulating IRET and exceptions/interrupts for Real
Mode, but IRET emulation for Protected Mode is non-existent. WARN instead
of trying to pass in a less-wrong type, e.g. X86_TRANSFER_RET, as
emulating IRET goes even beyond emulating FAR RET (which KVM also doesn't
fully support).
Reported-by: Hou Wenlong <houwenlong.hwl@antgroup.com>
Link: https://lore.kernel.org/r/20230216202254.671772-1-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Use a common X86_FEATURE_* flag for AMD's PSFD, and suppress it from
/proc/cpuinfo via the standard method of an empty string instead of
hacking in a one-off "private" #define in KVM. The request that led to
KVM defining its own flag was really just that the feature not show up
in /proc/cpuinfo, and additional patches+discussions in the interim have
clarified that defining flags in cpufeatures.h purely so that KVM can
advertise features to userspace is ok so long as the kernel already uses
a word to track the associated CPUID leaf.
No functional change intended.
Link: https://lore.kernel.org/all/d1b1e0da-29f0-c443-6c86-9549bbe1c79d@redhat.como
Link: https://lore.kernel.org/all/YxGZH7aOXQF7Pu5q@nazgul.tnic
Link: https://lore.kernel.org/all/Y3O7UYWfOLfJkwM%2F@zn.tnic
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Link: https://lore.kernel.org/r/20230124194519.2893234-1-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Allow L1 to use vNMI to accelerate its injection of NMI to L2 by
propagating vNMI int_ctl bits from/to vmcb12 to/from vmcb02.
To handle both the case where vNMI is enabled for L1 and L2, and where
vNMI is enabled for L1 but _not_ L2, move pending L1 vNMIs to nmi_pending
on nested VM-Entry and raise KVM_REQ_EVENT, i.e. rely on existing code to
route the NMI to the correct domain.
On nested VM-Exit, reverse the process and set/clear V_NMI_PENDING for L1
based one whether nmi_pending is zero or non-zero. There is no need to
consider vmcb02 in this case, as V_NMI_PENDING can be set in vmcb02 if
vNMI is disabled for L2, and if vNMI is enabled for L2, then L1 and L2
have different NMI contexts.
Co-developed-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Santosh Shukla <santosh.shukla@amd.com>
Link: https://lore.kernel.org/r/20230227084016.3368-12-santosh.shukla@amd.com
[sean: massage changelog to match the code]
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Add support for SVM's Virtual NMIs implementation, which adds proper
tracking of virtual NMI blocking, and an intr_ctrl flag that software can
set to mark a virtual NMI as pending. Pending virtual NMIs are serviced
by hardware if/when virtual NMIs become unblocked, i.e. act more or less
like real NMIs.
Introduce two new kvm_x86_ops callbacks so to support SVM's vNMI, as KVM
needs to treat a pending vNMI as partially injected. Specifically, if
two NMIs (for L1) arrive concurrently in KVM's software model, KVM's ABI
is to inject one and pend the other. Without vNMI, KVM manually tracks
the pending NMI and uses NMI windows to detect when the NMI should be
injected.
With vNMI, the pending NMI is simply stuffed into the VMCB and handed
off to hardware. This means that KVM needs to be able to set a vNMI
pending on-demand, and also query if a vNMI is pending, e.g. to honor the
"at most one NMI pending" rule and to preserve all NMIs across save and
restore.
Warn if KVM attempts to open an NMI window when vNMI is fully enabled,
as the above logic should prevent KVM from ever getting to
kvm_check_and_inject_events() with two NMIs pending _in software_, and
the "at most one NMI pending" logic should prevent having an NMI pending
in hardware and an NMI pending in software if NMIs are also blocked, i.e.
if KVM can't immediately inject the second NMI.
Signed-off-by: Santosh Shukla <Santosh.Shukla@amd.com>
Co-developed-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Link: https://lore.kernel.org/r/20230227084016.3368-11-santosh.shukla@amd.com
[sean: rewrite shortlog and changelog, massage code comments]
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Use the asynchronous NMI queue to handle pending NMIs coming in from
userspace during KVM_SET_VCPU_EVENTS so that all of KVM's logic for
handling multiple NMIs goes through process_nmi(). This will simplify
supporting SVM's upcoming "virtual NMI" functionality, which will need
changes KVM manages pending NMIs.
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Change return type of is_long_mode() to bool to avoid implicit cast,
as literally every user of is_long_mode() treats its return value as a
boolean.
Signed-off-by: Binbin Wu <binbin.wu@linux.intel.com>
Link: https://lore.kernel.org/r/20230322045824.22970-5-binbin.wu@linux.intel.com
Reviewed-by: Kai Huang <kai.huang@intel.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Save all pending NMIs in KVM_GET_VCPU_EVENTS, and queue KVM_REQ_NMI if one
or more NMIs are pending after KVM_SET_VCPU_EVENTS in order to re-evaluate
pending NMIs with respect to NMI blocking.
KVM allows multiple NMIs to be pending in order to faithfully emulate bare
metal handling of simultaneous NMIs (on bare metal, truly simultaneous
NMIs are impossible, i.e. one will always arrive first and be consumed).
Support for simultaneous NMIs botched the save/restore though. KVM only
saves one pending NMI, but allows userspace to restore 255 pending NMIs
as kvm_vcpu_events.nmi.pending is a u8, and KVM's internal state is stored
in an unsigned int.
Fixes: 7460fb4a3400 ("KVM: Fix simultaneous NMIs")
Signed-off-by: Santosh Shukla <Santosh.Shukla@amd.com>
Link: https://lore.kernel.org/r/20230227084016.3368-8-santosh.shukla@amd.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Tweak the code and comment that deals with concurrent NMIs to explicitly
call out that x86 allows exactly one pending NMI, but that KVM needs to
temporarily allow two pending NMIs in order to workaround the fact that
the target vCPU cannot immediately recognize an incoming NMI, unlike bare
metal.
No functional change intended.
Signed-off-by: Santosh Shukla <Santosh.Shukla@amd.com>
Link: https://lore.kernel.org/r/20230227084016.3368-7-santosh.shukla@amd.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Don't raise KVM_REQ_EVENT if no NMIs are pending at the end of
process_nmi(). Finishing process_nmi() without a pending NMI will become
much more likely when KVM gains support for AMD's vNMI, which allows
pending vNMIs in hardware, i.e. doesn't require explicit injection.
Signed-off-by: Santosh Shukla <Santosh.Shukla@amd.com>
Link: https://lore.kernel.org/r/20230227084016.3368-6-santosh.shukla@amd.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
SEV-ES guests don't use IRET interception for the detection of
an end of a NMI.
Therefore it makes sense to create a wrapper to avoid repeating
the check for the SEV-ES.
No functional change is intended.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
[Renamed iret intercept API of style svm_{clr,set}_iret_intercept()]
Signed-off-by: Santosh Shukla <Santosh.Shukla@amd.com>
Link: https://lore.kernel.org/r/20230227084016.3368-5-santosh.shukla@amd.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
If L1 doesn't intercept interrupts, then KVM will use vmcb02's V_IRQ
to detect an interrupt window for L1 IRQs. On a subsequent nested
VM-Exit, KVM might need to copy the current V_IRQ from vmcb02 to vmcb01
to continue waiting for an interrupt window, i.e. if there is still a
pending IRQ for L1.
Raise KVM_REQ_EVENT on nested exit if L1 isn't intercepting IRQs to ensure
that KVM will re-enable interrupt window detection if needed.
Note that this is a theoretical bug because KVM already raises
KVM_REQ_EVENT on each nested VM exit, because the nested VM exit resets
RFLAGS and kvm_set_rflags() raises the KVM_REQ_EVENT unconditionally.
Explicitly raise KVM_REQ_EVENT for the interrupt window case to avoid
having an unnecessary dependency on kvm_set_rflags(), and to document
the scenario.
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
[santosh: reworded description as per Sean's v2 comment]
Signed-off-by: Santosh Shukla <Santosh.Shukla@amd.com>
Link: https://lore.kernel.org/r/20230227084016.3368-4-santosh.shukla@amd.com
[sean: further massage changelog and comment]
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Disable intercept of virtual interrupts (used to detect interrupt windows)
if the saved host (L1) RFLAGS.IF is '0', as the effective RFLAGS.IF for L1
interrupts will never be set while L2 is running (L2's RFLAGS.IF doesn't
affect L1 IRQs when virtual interrupts are enabled).
Suggested-by: Sean Christopherson <seanjc@google.com>
Link: https://lkml.kernel.org/r/Y9hybI65So5X2LFg%40google.com
Signed-off-by: Santosh Shukla <Santosh.Shukla@amd.com>
Link: https://lore.kernel.org/r/20230227084016.3368-3-santosh.shukla@amd.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
intercepting VINTR
Don't sync vmcb02 V_IRQ back to vmcb12 if KVM (L0) is intercepting
virtual interrupts in order to request an interrupt window, as KVM
has usurped vmcb02's int_ctl. If an interrupt window opens before
the next VM-Exit, svm_clear_vintr() will restore vmcb12's int_ctl.
If no window opens, V_IRQ will be correctly preserved in vmcb12's
int_ctl (because it was never recognized while L2 was running).
Suggested-by: Sean Christopherson <seanjc@google.com>
Link: https://lkml.kernel.org/r/Y9hybI65So5X2LFg%40google.com
Signed-off-by: Santosh Shukla <Santosh.Shukla@amd.com>
Link: https://lore.kernel.org/r/20230227084016.3368-2-santosh.shukla@amd.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Use kvm_is_cr4_bit_set() to query SMAP and SMEP when determining whether
or not AMD's SMAP+SEV errata prevents KVM from emulating an instruction.
This eliminates an implicit cast from ulong to bool and makes the code
slightly more readable.
Note, any overhead from making multiple calls to kvm_read_cr4_bits() is
negligible, not to mention the code is question is encountered only in
rare situations, i.e. is not a remotely hot path.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Binbin Wu <binbin.wu@linux.intel.com>
Link: https://lore.kernel.org/r/20230322045824.22970-4-binbin.wu@linux.intel.com
[sean: keep local smap/smep variables, massage changelog]
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Convert is_{pae,pse,paging}() to use kvm_is_cr{0,4}_bit_set() and return
bools. Returning an "int" requires not one, but two implicit casts, first
from "unsigned long" to "int", and then again to a "bool". Both casts are
more than a bit dangerous; the ulong=>int casts would drop a bit on 64-bit
kernels _if_ the bits in question weren't in the lower 32 bits, and the
int=>bool cast can result in false negatives/positives, e.g. see commit
0c928ff26bd6 ("KVM: SVM: Fix benign "bool vs. int" comparison in
svm_set_cr0()").
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Binbin Wu <binbin.wu@linux.intel.com>
Link: https://lore.kernel.org/r/20230322045824.22970-3-binbin.wu@linux.intel.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Add helpers to check if a specific CR0/CR4 bit is set to avoid a plethora
of implicit casts from the "unsigned long" return of kvm_read_cr*_bits(),
and to make each caller's intent more obvious.
Defer converting helpers that do truly ugly casts from "unsigned long" to
"int", e.g. is_pse(), to a future commit so that their conversion is more
isolated.
Opportunistically drop the superfluous pcid_enabled from kvm_set_cr3();
the local variable is used only once, immediately after its declaration.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Binbin Wu <binbin.wu@linux.intel.com>
Link: https://lore.kernel.org/r/20230322045824.22970-2-binbin.wu@linux.intel.com
[sean: move "obvious" conversions to this commit, massage changelog]
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Explicitly convert the return from is_paging() to a bool when comparing
against old_paging, which is also a boolean. is_paging() sneakily uses
kvm_read_cr0_bits() and returns an int, i.e. returns X86_CR0_PG or 0, not
1 or 0.
Luckily, the bug is benign as it only results in a false positive, not a
false negative, i.e. only causes a spurious refresh of CR4 when paging is
enabled in both the old and new.
Cc: Maxim Levitsky <mlevitsk@redhat.com>
Fixes: c53bbe2145f5 ("KVM: x86: SVM: don't passthrough SMAP/SMEP/PKE bits in !NPT && !gCR0.PG case")
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Guests like grsecurity that make heavy use of CR0.WP to implement kernel
level W^X will suffer from the implied VMEXITs.
With EPT there is no need to intercept a guest change of CR0.WP, so
simply make it a guest owned bit if we can do so.
This implies that a read of a guest's CR0.WP bit might need a VMREAD.
However, the only potentially affected user seems to be kvm_init_mmu()
which is a heavy operation to begin with. But also most callers already
cache the full value of CR0 anyway, so no additional VMREAD is needed.
The only exception is nested_vmx_load_cr3().
This change is VMX-specific, as SVM has no such fine grained control
register intercept control.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Mathias Krause <minipli@grsecurity.net>
Link: https://lore.kernel.org/r/20230322013731.102955-7-minipli@grsecurity.net
Co-developed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Make use of the kvm_read_cr{0,4}_bits() helper functions when we only
want to know the state of certain bits instead of the whole register.
This not only makes the intent cleaner, it also avoids a potential
VMREAD in case the tested bits aren't guest owned.
Signed-off-by: Mathias Krause <minipli@grsecurity.net>
Link: https://lore.kernel.org/r/20230322013731.102955-5-minipli@grsecurity.net
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
If paging is disabled, there are no permission bits to emulate.
Micro-optimize this case to avoid unnecessary work.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Mathias Krause <minipli@grsecurity.net>
Link: https://lore.kernel.org/r/20230322013731.102955-4-minipli@grsecurity.net
Co-developed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
There is no need to unload the MMU roots with TDP enabled when only
CR0.WP has changed -- the paging structures are still valid, only the
permission bitmap needs to be updated.
One heavy user of toggling CR0.WP is grsecurity's KERNEXEC feature to
implement kernel W^X.
The optimization brings a huge performance gain for this case as the
following micro-benchmark running 'ssdd 10 50000' from rt-tests[1] on a
grsecurity L1 VM shows (runtime in seconds, lower is better):
legacy TDP shadow
kvm-x86/next@d8708b 8.43s 9.45s 70.3s
+patch 5.39s 5.63s 70.2s
For legacy MMU this is ~36% faster, for TDP MMU even ~40% faster. Also
TDP and legacy MMU now both have a similar runtime which vanishes the
need to disable TDP MMU for grsecurity.
Shadow MMU sees no measurable difference and is still slow, as expected.
[1] https://git.kernel.org/pub/scm/utils/rt-tests/rt-tests.git
Signed-off-by: Mathias Krause <minipli@grsecurity.net>
Link: https://lore.kernel.org/r/20230322013731.102955-3-minipli@grsecurity.net
Co-developed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Fix a small comment typo in make_spte().
Signed-off-by: Mathias Krause <minipli@grsecurity.net>
Link: https://lore.kernel.org/r/20230322013731.102955-6-minipli@grsecurity.net
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Most of the time, calls to get_guest_pgd result in calling
kvm_read_cr3 (the exception is only nested TDP). Hardcode
the default instead of using the get_cr3 function, avoiding
a retpoline if they are enabled.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Mathias Krause <minipli@grsecurity.net>
Link: https://lore.kernel.org/r/20230322013731.102955-2-minipli@grsecurity.net
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
nested_vmx_setup_ctls_msrs() is used to set up the various VMX MSR
controls for nested VMX. But it is a bit lengthy, just add helpers
to setup the configuration of VMX MSRs.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Yu Zhang <yu.c.zhang@linux.intel.com>
Link: https://lore.kernel.org/r/20230119141946.585610-2-yu.c.zhang@linux.intel.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
nested_vmx_setup_ctls_msrs() initializes the vmcs_conf.nested,
which stores the global VMX MSR configurations when nested is
supported, regardless of any particular CPUID settings for one
VM.
Commit 6defc591846d ("KVM: nVMX: include conditional controls
in /dev/kvm KVM_GET_MSRS") added the some feature flags for
secondary proc-based controls, so that those features can be
available in KVM_GET_MSRS. Yet this commit did not remove the
obsolete comments in nested_vmx_setup_ctls_msrs().
Just fix the comments, and no functional change intended.
Fixes: 6defc591846d ("KVM: nVMX: include conditional controls in /dev/kvm KVM_GET_MSRS")
Signed-off-by: Yu Zhang <yu.c.zhang@linux.intel.com>
Link: https://lore.kernel.org/r/20230119141946.585610-1-yu.c.zhang@linux.intel.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Adjust a variety of functions in mmu.c to put the function return type on
the same line as the function declaration. As stated in the Linus
specification:
But the "on their own line" is complete garbage to begin with. That
will NEVER be a kernel rule. We should never have a rule that assumes
things are so long that they need to be on multiple lines.
We don't put function return types on their own lines either, even if
some other projects have that rule (just to get function names at the
beginning of lines or some other odd reason).
Leave the functions generated by BUILD_MMU_ROLE_REGS_ACCESSOR() as-is,
that code is basically illegible no matter how it's formatted.
No functional change intended.
Link: https://lore.kernel.org/mm-commits/CAHk-=wjS-Jg7sGMwUPpDsjv392nDOOs0CtUtVkp=S6Q7JzFJRw@mail.gmail.com
Signed-off-by: Ben Gardon <bgardon@google.com>
Link: https://lore.kernel.org/r/20230202182809.1929122-4-bgardon@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Assert that mmu_lock is held for write in __walk_slot_rmaps() instead of
hoping the function comment will magically prevent introducing bugs.
Signed-off-by: Ben Gardon <bgardon@google.com>
Link: https://lore.kernel.org/r/20230202182809.1929122-3-bgardon@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Replace "slot_handle_level" with "walk_slot_rmaps" to better capture what
the helpers are doing, and to slightly shorten the function names so that
each function's return type and attributes can be placed on the same line
as the function declaration.
No functional change intended.
Link: https://lore.kernel.org/mm-commits/CAHk-=wjS-Jg7sGMwUPpDsjv392nDOOs0CtUtVkp=S6Q7JzFJRw@mail.gmail.com
Signed-off-by: Ben Gardon <bgardon@google.com>
Link: https://lore.kernel.org/r/20230202182809.1929122-2-bgardon@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Use gfn_t instead of u64 for kvm_flush_remote_tlbs_range()'s parameters,
since gfn_t is the standard type for GFNs throughout KVM.
Opportunistically rename pages to nr_pages to make its role even more
obvious.
No functional change intended.
Signed-off-by: David Matlack <dmatlack@google.com>
Link: https://lore.kernel.org/r/20230126184025.2294823-6-dmatlack@google.com
[sean: convert pages to gfn_t too, and rename]
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Rename kvm_flush_remote_tlbs_with_address() to
kvm_flush_remote_tlbs_range(). This name is shorter, which reduces the
number of callsites that need to be broken up across multiple lines, and
more readable since it conveys a range of memory is being flushed rather
than a single address.
No functional change intended.
Signed-off-by: David Matlack <dmatlack@google.com>
Link: https://lore.kernel.org/r/20230126184025.2294823-5-dmatlack@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Collapse kvm_flush_remote_tlbs_with_range() and
kvm_flush_remote_tlbs_with_address() into a single function. This
eliminates some lines of code and a useless NULL check on the range
struct.
Opportunistically switch from ENOTSUPP to EOPNOTSUPP to make checkpatch
happy.
Signed-off-by: David Matlack <dmatlack@google.com>
Link: https://lore.kernel.org/r/20230126184025.2294823-4-dmatlack@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
The TEE subdriver for CCP, the amdtee driver and the i2c-designware-amdpsp
drivers all include `psp-sev.h` even though they don't use SEV
functionality.
Move the definition of `__psp_pa` into a common header to be included
by all of these drivers.
Reviewed-by: Jan Dabros <jsd@semihalf.com>
Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com> # For the drivers/i2c/busses/i2c-designware-amdpsp.c
Acked-by: Sumit Garg <sumit.garg@linaro.org> # For TEE subsystem bits
Acked-by: Tom Lendacky <thomas.lendacky@amd.com>
Acked-by: Sean Christopherson <seanjc@google.com> # KVM
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
|
|
Rework "struct pte_list_desc" and pte_list_{add|remove} to track the tail
count, i.e. number of PTEs in non-head descriptors, and to always keep all
tail descriptors full so that adding a new entry and counting the number
of entries is done in constant time instead of linear time.
No visible performace is changed in tests. But pte_list_add() is no longer
shown in the perf result for the COWed pages even the guest forks millions
of tasks.
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Link: https://lore.kernel.org/r/20230113122910.672417-1-jiangshanlai@gmail.com
[sean: reword shortlog, tweak changelog, add lots of comments, add BUG_ON()]
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
Sync the spte only when the spte is set and avoid the indirect branch.
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Link: https://lore.kernel.org/r/20230216235321.735214-5-jiangshanlai@gmail.com
[sean: add wrapper instead of open coding each check]
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
