Pull KVM updates from Paolo Bonzini:
"s390:
- implement diag318
x86:
- Report last CPU for debugging
- Emulate smaller MAXPHYADDR in the guest than in the host
- .noinstr and tracing fixes from Thomas
- nested SVM page table switching optimization and fixes
Generic:
- Unify shadow MMU cache data structures across architectures"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (127 commits)
KVM: SVM: Fix sev_pin_memory() error handling
KVM: LAPIC: Set the TDCR settable bits
KVM: x86: Specify max TDP level via kvm_configure_mmu()
KVM: x86/mmu: Rename max_page_level to max_huge_page_level
KVM: x86: Dynamically calculate TDP level from max level and MAXPHYADDR
KVM: VXM: Remove temporary WARN on expected vs. actual EPTP level mismatch
KVM: x86: Pull the PGD's level from the MMU instead of recalculating it
KVM: VMX: Make vmx_load_mmu_pgd() static
KVM: x86/mmu: Add separate helper for shadow NPT root page role calc
KVM: VMX: Drop a duplicate declaration of construct_eptp()
KVM: nSVM: Correctly set the shadow NPT root level in its MMU role
KVM: Using macros instead of magic values
MIPS: KVM: Fix build error caused by 'kvm_run' cleanup
KVM: nSVM: remove nonsensical EXITINFO1 adjustment on nested NPF
KVM: x86: Add a capability for GUEST_MAXPHYADDR < HOST_MAXPHYADDR support
KVM: VMX: optimize #PF injection when MAXPHYADDR does not match
KVM: VMX: Add guest physical address check in EPT violation and misconfig
KVM: VMX: introduce vmx_need_pf_intercept
KVM: x86: update exception bitmap on CPUID changes
KVM: x86: rename update_bp_intercept to update_exception_bitmap
...
Pull x86 fsgsbase from Thomas Gleixner:
"Support for FSGSBASE. Almost 5 years after the first RFC to support
it, this has been brought into a shape which is maintainable and
actually works.
This final version was done by Sasha Levin who took it up after Intel
dropped the ball. Sasha discovered that the SGX (sic!) offerings out
there ship rogue kernel modules enabling FSGSBASE behind the kernels
back which opens an instantanious unpriviledged root hole.
The FSGSBASE instructions provide a considerable speedup of the
context switch path and enable user space to write GSBASE without
kernel interaction. This enablement requires careful handling of the
exception entries which go through the paranoid entry path as they
can no longer rely on the assumption that user GSBASE is positive (as
enforced via prctl() on non FSGSBASE enabled systemn).
All other entries (syscalls, interrupts and exceptions) can still just
utilize SWAPGS unconditionally when the entry comes from user space.
Converting these entries to use FSGSBASE has no benefit as SWAPGS is
only marginally slower than WRGSBASE and locating and retrieving the
kernel GSBASE value is not a free operation either. The real benefit
of RD/WRGSBASE is the avoidance of the MSR reads and writes.
The changes come with appropriate selftests and have held up in field
testing against the (sanitized) Graphene-SGX driver"
* tag 'x86-fsgsbase-2020-08-04' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (21 commits)
x86/fsgsbase: Fix Xen PV support
x86/ptrace: Fix 32-bit PTRACE_SETREGS vs fsbase and gsbase
selftests/x86/fsgsbase: Add a missing memory constraint
selftests/x86/fsgsbase: Fix a comment in the ptrace_write_gsbase test
selftests/x86: Add a syscall_arg_fault_64 test for negative GSBASE
selftests/x86/fsgsbase: Test ptracer-induced GS base write with FSGSBASE
selftests/x86/fsgsbase: Test GS selector on ptracer-induced GS base write
Documentation/x86/64: Add documentation for GS/FS addressing mode
x86/elf: Enumerate kernel FSGSBASE capability in AT_HWCAP2
x86/cpu: Enable FSGSBASE on 64bit by default and add a chicken bit
x86/entry/64: Handle FSGSBASE enabled paranoid entry/exit
x86/entry/64: Introduce the FIND_PERCPU_BASE macro
x86/entry/64: Switch CR3 before SWAPGS in paranoid entry
x86/speculation/swapgs: Check FSGSBASE in enabling SWAPGS mitigation
x86/process/64: Use FSGSBASE instructions on thread copy and ptrace
x86/process/64: Use FSBSBASE in switch_to() if available
x86/process/64: Make save_fsgs_for_kvm() ready for FSGSBASE
x86/fsgsbase/64: Enable FSGSBASE instructions in helper functions
x86/fsgsbase/64: Add intrinsics for FSGSBASE instructions
x86/cpu: Add 'unsafe_fsgsbase' to enable CR4.FSGSBASE
...
Calculate the desired TDP level on the fly using the max TDP level and
MAXPHYADDR instead of doing the same when CPUID is updated. This avoids
the hidden dependency on cpuid_maxphyaddr() in vmx_get_tdp_level() and
also standardizes the "use 5-level paging iff MAXPHYADDR > 48" behavior
across x86.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200716034122.5998-8-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Use the shadow_root_level from the current MMU as the root level for the
PGD, i.e. for VMX's EPTP. This eliminates the weird dependency between
VMX and the MMU where both must independently calculate the same root
level for things to work correctly. Temporarily keep VMX's calculation
of the level and use it to WARN if the incoming level diverges.
Opportunistically refactor kvm_mmu_load_pgd() to avoid indentation hell,
and rename a 'cr3' param in the load_mmu_pgd prototype that managed to
survive the cr3 purge.
No functional change intended.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200716034122.5998-6-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Use the generic infrastructure to check for and handle pending work before
transitioning into guest mode.
This now handles TIF_NOTIFY_RESUME as well which was ignored so
far. Handling it is important as this covers task work and task work will
be used to offload the heavy lifting of POSIX CPU timers to thread context.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20200722220520.979724969@linutronix.de
This patch adds a new capability KVM_CAP_SMALLER_MAXPHYADDR which
allows userspace to query if the underlying architecture would
support GUEST_MAXPHYADDR < HOST_MAXPHYADDR and hence act accordingly
(e.g. qemu can decide if it should warn for -cpu ..,phys-bits=X)
The complications in this patch are due to unexpected (but documented)
behaviour we see with NPF vmexit handling in AMD processor. If
SVM is modified to add guest physical address checks in the NPF
and guest #PF paths, we see the followning error multiple times in
the 'access' test in kvm-unit-tests:
test pte.p pte.36 pde.p: FAIL: pte 2000021 expected 2000001
Dump mapping: address: 0x123400000000
------L4: 24c3027
------L3: 24c4027
------L2: 24c5021
------L1: 1002000021
This is because the PTE's accessed bit is set by the CPU hardware before
the NPF vmexit. This is handled completely by hardware and cannot be fixed
in software.
Therefore, availability of the new capability depends on a boolean variable
allow_smaller_maxphyaddr which is set individually by VMX and SVM init
routines. On VMX it's always set to true, on SVM it's only set to true
when NPT is not enabled.
CC: Tom Lendacky <thomas.lendacky@amd.com>
CC: Babu Moger <babu.moger@amd.com>
Signed-off-by: Mohammed Gamal <mgamal@redhat.com>
Message-Id: <20200710154811.418214-10-mgamal@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Ignore non-present page faults, since those cannot have reserved
bits set.
When running access.flat with "-cpu Haswell,phys-bits=36", the
number of trapped page faults goes down from 8872644 to 3978948.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200710154811.418214-9-mgamal@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Check guest physical address against its maximum, which depends on the
guest MAXPHYADDR. If the guest's physical address exceeds the
maximum (i.e. has reserved bits set), inject a guest page fault with
PFERR_RSVD_MASK set.
This has to be done both in the EPT violation and page fault paths, as
there are complications in both cases with respect to the computation
of the correct error code.
For EPT violations, unfortunately the only possibility is to emulate,
because the access type in the exit qualification might refer to an
access to a paging structure, rather than to the access performed by
the program.
Trapping page faults instead is needed in order to correct the error code,
but the access type can be obtained from the original error code and
passed to gva_to_gpa. The corrections required in the error code are
subtle. For example, imagine that a PTE for a supervisor page has a reserved
bit set. On a supervisor-mode access, the EPT violation path would trigger.
However, on a user-mode access, the processor will not notice the reserved
bit and not include PFERR_RSVD_MASK in the error code.
Co-developed-by: Mohammed Gamal <mgamal@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200710154811.418214-8-mgamal@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We would like to introduce a callback to update the #PF intercept
when CPUID changes. Just reuse update_bp_intercept since VMX is
already using update_exception_bitmap instead of a bespoke function.
While at it, remove an unnecessary assignment in the SVM version,
which is already done in the caller (kvm_arch_vcpu_ioctl_set_guest_debug)
and has nothing to do with the exception bitmap.
Reviewed-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Entering guest mode is more or less the same as returning to user
space. From an instrumentation point of view both leave kernel mode and the
transition to guest or user mode reenables interrupts on the host. In user
mode an interrupt is served directly and in guest mode it causes a VM exit
which then handles or reinjects the interrupt.
The transition from guest mode or user mode to kernel mode disables
interrupts, which needs to be recorded in instrumentation to set the
correct state again.
This is important for e.g. latency analysis because otherwise the execution
time in guest or user mode would be wrongly accounted as interrupt disabled
and could trigger false positives.
Add hardirq tracing to guest enter/exit functions in the same way as it
is done in the user mode enter/exit code, respecting the RCU requirements.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200708195321.822002354@linutronix.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Context tracking for KVM happens way too early in the vcpu_run()
code. Anything after guest_enter_irqoff() and before guest_exit_irqoff()
cannot use RCU and should also be not instrumented.
The current way of doing this covers way too much code. Move it closer to
the actual vmenter/exit code.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200708195321.724574345@linutronix.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
To avoid complex and in some cases incorrect logic in
kvm_spec_ctrl_test_value, just try the guest's given value on the host
processor instead, and if it doesn't #GP, allow the guest to set it.
One such case is when host CPU supports STIBP mitigation
but doesn't support IBRS (as is the case with some Zen2 AMD cpus),
and in this case we were giving guest #GP when it tried to use STIBP
The reason why can can do the host test is that IA32_SPEC_CTRL msr is
passed to the guest, after the guest sets it to a non zero value
for the first time (due to performance reasons),
and as as result of this, it is pointless to emulate #GP condition on
this first access, in a different way than what the host CPU does.
This is based on a patch from Sean Christopherson, who suggested this idea.
Fixes: 6441fa6178 ("KVM: x86: avoid incorrect writes to host MSR_IA32_SPEC_CTRL")
Cc: stable@vger.kernel.org
Suggested-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20200708115731.180097-1-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The name of callback cpuid_update() is misleading that it's not about
updating CPUID settings of vcpu but updating the configurations of vcpu
based on the CPUIDs. So rename it to vcpu_after_set_cpuid().
Signed-off-by: Xiaoyao Li <xiaoyao.li@intel.com>
Message-Id: <20200709043426.92712-5-xiaoyao.li@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
More often than not, a failed VM-entry in an x86 production
environment is induced by a defective CPU. To help identify the bad
hardware, include the id of the last logical CPU to run a vCPU in the
information provided to userspace on a KVM exit for failed VM-entry or
for KVM internal errors not associated with emulation. The presence of
this additional information is indicated by a new capability,
KVM_CAP_LAST_CPU.
Signed-off-by: Jim Mattson <jmattson@google.com>
Reviewed-by: Oliver Upton <oupton@google.com>
Reviewed-by: Peter Shier <pshier@google.com>
Message-Id: <20200603235623.245638-5-jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Use the "common" KVM_POSSIBLE_CR*_GUEST_BITS defines to initialize the
CR0/CR4 guest host masks instead of duplicating most of the CR4 mask and
open coding the CR0 mask. SVM doesn't utilize the masks, i.e. the masks
are effectively VMX specific even if they're not named as such. This
avoids duplicate code, better documents the guest owned CR0 bit, and
eliminates the need for a build-time assertion to keep VMX and x86
synchronized.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200703040422.31536-3-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Mark CR4.TSD as being possibly owned by the guest as that is indeed the
case on VMX. Without TSD being tagged as possibly owned by the guest, a
targeted read of CR4 to get TSD could observe a stale value. This bug
is benign in the current code base as the sole consumer of TSD is the
emulator (for RDTSC) and the emulator always "reads" the entirety of CR4
when grabbing bits.
Add a build-time assertion in to ensure VMX doesn't hand over more CR4
bits without also updating x86.
Fixes: 52ce3c21ae ("x86,kvm,vmx: Don't trap writes to CR4.TSD")
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200703040422.31536-2-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Remove support for context switching between the guest's and host's
desired UMWAIT_CONTROL. Propagating the guest's value to hardware isn't
required for correct functionality, e.g. KVM intercepts reads and writes
to the MSR, and the latency effects of the settings controlled by the
MSR are not architecturally visible.
As a general rule, KVM should not allow the guest to control power
management settings unless explicitly enabled by userspace, e.g. see
KVM_CAP_X86_DISABLE_EXITS. E.g. Intel's SDM explicitly states that C0.2
can improve the performance of SMT siblings. A devious guest could
disable C0.2 so as to improve the performance of their workloads at the
detriment to workloads running in the host or on other VMs.
Wholesale removal of UMWAIT_CONTROL context switching also fixes a race
condition where updates from the host may cause KVM to enter the guest
with the incorrect value. Because updates are are propagated to all
CPUs via IPI (SMP function callback), the value in hardware may be
stale with respect to the cached value and KVM could enter the guest
with the wrong value in hardware. As above, the guest can't observe the
bad value, but it's a weird and confusing wart in the implementation.
Removal also fixes the unnecessary usage of VMX's atomic load/store MSR
lists. Using the lists is only necessary for MSRs that are required for
correct functionality immediately upon VM-Enter/VM-Exit, e.g. EFER on
old hardware, or for MSRs that need to-the-uop precision, e.g. perf
related MSRs. For UMWAIT_CONTROL, the effects are only visible in the
kernel via TPAUSE/delay(), and KVM doesn't do any form of delay in
vcpu_vmx_run(). Using the atomic lists is undesirable as they are more
expensive than direct RDMSR/WRMSR.
Furthermore, even if giving the guest control of the MSR is legitimate,
e.g. in pass-through scenarios, it's not clear that the benefits would
outweigh the overhead. E.g. saving and restoring an MSR across a VMX
roundtrip costs ~250 cycles, and if the guest diverged from the host
that cost would be paid on every run of the guest. In other words, if
there is a legitimate use case then it should be enabled by a new
per-VM capability.
Note, KVM still needs to emulate MSR_IA32_UMWAIT_CONTROL so that it can
correctly expose other WAITPKG features to the guest, e.g. TPAUSE,
UMWAIT and UMONITOR.
Fixes: 6e3ba4abce ("KVM: vmx: Emulate MSR IA32_UMWAIT_CONTROL")
Cc: stable@vger.kernel.org
Cc: Jingqi Liu <jingqi.liu@intel.com>
Cc: Tao Xu <tao3.xu@intel.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200623005135.10414-1-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Explicitly pass the L2 GPA to kvm_arch_write_log_dirty(), which for all
intents and purposes is vmx_write_pml_buffer(), instead of having the
latter pull the GPA from vmcs.GUEST_PHYSICAL_ADDRESS. If the dirty bit
update is the result of KVM emulation (rare for L2), then the GPA in the
VMCS may be stale and/or hold a completely unrelated GPA.
Fixes: c5f983f6e8 ("nVMX: Implement emulated Page Modification Logging")
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200622215832.22090-2-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Guest crashes are observed on a Cascade Lake system when 'perf top' is
launched on the host, e.g.
BUG: unable to handle kernel paging request at fffffe0000073038
PGD 7ffa7067 P4D 7ffa7067 PUD 7ffa6067 PMD 7ffa5067 PTE ffffffffff120
Oops: 0000 [#1] SMP PTI
CPU: 1 PID: 1 Comm: systemd Not tainted 4.18.0+ #380
...
Call Trace:
serial8250_console_write+0xfe/0x1f0
call_console_drivers.constprop.0+0x9d/0x120
console_unlock+0x1ea/0x460
Call traces are different but the crash is imminent. The problem was
blindly bisected to the commit 041bc42ce2 ("KVM: VMX: Micro-optimize
vmexit time when not exposing PMU"). It was also confirmed that the
issue goes away if PMU is exposed to the guest.
With some instrumentation of the guest we can see what is being switched
(when we do atomic_switch_perf_msrs()):
vmx_vcpu_run: switching 2 msrs
vmx_vcpu_run: switching MSR38f guest: 70000000d host: 70000000f
vmx_vcpu_run: switching MSR3f1 guest: 0 host: 2
The current guess is that PEBS (MSR_IA32_PEBS_ENABLE, 0x3f1) is to blame.
Regardless of whether PMU is exposed to the guest or not, PEBS needs to
be disabled upon switch.
This reverts commit 041bc42ce2.
Reported-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20200619094046.654019-1-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
save_fsgs_for_kvm() is invoked via
vcpu_enter_guest()
kvm_x86_ops.prepare_guest_switch(vcpu)
vmx_prepare_switch_to_guest()
save_fsgs_for_kvm()
with preemption disabled, but interrupts enabled.
The upcoming FSGSBASE based GS safe needs interrupts to be disabled. This
could be done in the helper function, but that function is also called from
switch_to() which has interrupts disabled already.
Disable interrupts inside save_fsgs_for_kvm() and rename the function to
current_save_fsgs() so it can be invoked from other places.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20200528201402.1708239-7-sashal@kernel.org
Pull x86 entry updates from Thomas Gleixner:
"The x86 entry, exception and interrupt code rework
This all started about 6 month ago with the attempt to move the Posix
CPU timer heavy lifting out of the timer interrupt code and just have
lockless quick checks in that code path. Trivial 5 patches.
This unearthed an inconsistency in the KVM handling of task work and
the review requested to move all of this into generic code so other
architectures can share.
Valid request and solved with another 25 patches but those unearthed
inconsistencies vs. RCU and instrumentation.
Digging into this made it obvious that there are quite some
inconsistencies vs. instrumentation in general. The int3 text poke
handling in particular was completely unprotected and with the batched
update of trace events even more likely to expose to endless int3
recursion.
In parallel the RCU implications of instrumenting fragile entry code
came up in several discussions.
The conclusion of the x86 maintainer team was to go all the way and
make the protection against any form of instrumentation of fragile and
dangerous code pathes enforcable and verifiable by tooling.
A first batch of preparatory work hit mainline with commit
d5f744f9a2 ("Pull x86 entry code updates from Thomas Gleixner")
That (almost) full solution introduced a new code section
'.noinstr.text' into which all code which needs to be protected from
instrumentation of all sorts goes into. Any call into instrumentable
code out of this section has to be annotated. objtool has support to
validate this.
Kprobes now excludes this section fully which also prevents BPF from
fiddling with it and all 'noinstr' annotated functions also keep
ftrace off. The section, kprobes and objtool changes are already
merged.
The major changes coming with this are:
- Preparatory cleanups
- Annotating of relevant functions to move them into the
noinstr.text section or enforcing inlining by marking them
__always_inline so the compiler cannot misplace or instrument
them.
- Splitting and simplifying the idtentry macro maze so that it is
now clearly separated into simple exception entries and the more
interesting ones which use interrupt stacks and have the paranoid
handling vs. CR3 and GS.
- Move quite some of the low level ASM functionality into C code:
- enter_from and exit to user space handling. The ASM code now
calls into C after doing the really necessary ASM handling and
the return path goes back out without bells and whistels in
ASM.
- exception entry/exit got the equivivalent treatment
- move all IRQ tracepoints from ASM to C so they can be placed as
appropriate which is especially important for the int3
recursion issue.
- Consolidate the declaration and definition of entry points between
32 and 64 bit. They share a common header and macros now.
- Remove the extra device interrupt entry maze and just use the
regular exception entry code.
- All ASM entry points except NMI are now generated from the shared
header file and the corresponding macros in the 32 and 64 bit
entry ASM.
- The C code entry points are consolidated as well with the help of
DEFINE_IDTENTRY*() macros. This allows to ensure at one central
point that all corresponding entry points share the same
semantics. The actual function body for most entry points is in an
instrumentable and sane state.
There are special macros for the more sensitive entry points, e.g.
INT3 and of course the nasty paranoid #NMI, #MCE, #DB and #DF.
They allow to put the whole entry instrumentation and RCU handling
into safe places instead of the previous pray that it is correct
approach.
- The INT3 text poke handling is now completely isolated and the
recursion issue banned. Aside of the entry rework this required
other isolation work, e.g. the ability to force inline bsearch.
- Prevent #DB on fragile entry code, entry relevant memory and
disable it on NMI, #MC entry, which allowed to get rid of the
nested #DB IST stack shifting hackery.
- A few other cleanups and enhancements which have been made
possible through this and already merged changes, e.g.
consolidating and further restricting the IDT code so the IDT
table becomes RO after init which removes yet another popular
attack vector
- About 680 lines of ASM maze are gone.
There are a few open issues:
- An escape out of the noinstr section in the MCE handler which needs
some more thought but under the aspect that MCE is a complete
trainwreck by design and the propability to survive it is low, this
was not high on the priority list.
- Paravirtualization
When PV is enabled then objtool complains about a bunch of indirect
calls out of the noinstr section. There are a few straight forward
ways to fix this, but the other issues vs. general correctness were
more pressing than parawitz.
- KVM
KVM is inconsistent as well. Patches have been posted, but they
have not yet been commented on or picked up by the KVM folks.
- IDLE
Pretty much the same problems can be found in the low level idle
code especially the parts where RCU stopped watching. This was
beyond the scope of the more obvious and exposable problems and is
on the todo list.
The lesson learned from this brain melting exercise to morph the
evolved code base into something which can be validated and understood
is that once again the violation of the most important engineering
principle "correctness first" has caused quite a few people to spend
valuable time on problems which could have been avoided in the first
place. The "features first" tinkering mindset really has to stop.
With that I want to say thanks to everyone involved in contributing to
this effort. Special thanks go to the following people (alphabetical
order): Alexandre Chartre, Andy Lutomirski, Borislav Petkov, Brian
Gerst, Frederic Weisbecker, Josh Poimboeuf, Juergen Gross, Lai
Jiangshan, Macro Elver, Paolo Bonzin,i Paul McKenney, Peter Zijlstra,
Vitaly Kuznetsov, and Will Deacon"
* tag 'x86-entry-2020-06-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (142 commits)
x86/entry: Force rcu_irq_enter() when in idle task
x86/entry: Make NMI use IDTENTRY_RAW
x86/entry: Treat BUG/WARN as NMI-like entries
x86/entry: Unbreak __irqentry_text_start/end magic
x86/entry: __always_inline CR2 for noinstr
lockdep: __always_inline more for noinstr
x86/entry: Re-order #DB handler to avoid *SAN instrumentation
x86/entry: __always_inline arch_atomic_* for noinstr
x86/entry: __always_inline irqflags for noinstr
x86/entry: __always_inline debugreg for noinstr
x86/idt: Consolidate idt functionality
x86/idt: Cleanup trap_init()
x86/idt: Use proper constants for table size
x86/idt: Add comments about early #PF handling
x86/idt: Mark init only functions __init
x86/entry: Rename trace_hardirqs_off_prepare()
x86/entry: Clarify irq_{enter,exit}_rcu()
x86/entry: Remove DBn stacks
x86/entry: Remove debug IDT frobbing
x86/entry: Optimize local_db_save() for virt
...
Convert #MC to IDTENTRY_MCE:
- Implement the C entry points with DEFINE_IDTENTRY_MCE
- Emit the ASM stub with DECLARE_IDTENTRY_MCE
- Remove the ASM idtentry in 64bit
- Remove the open coded ASM entry code in 32bit
- Fixup the XEN/PV code
- Remove the old prototypes
- Remove the error code from *machine_check_vector() as
it is always 0 and not used by any of the functions
it can point to. Fixup all the functions as well.
No functional change.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Acked-by: Andy Lutomirski <luto@kernel.org>
Link: https://lkml.kernel.org/r/20200505135314.334980426@linutronix.de
Syzbot reports the following issue:
WARNING: CPU: 0 PID: 6819 at arch/x86/kvm/x86.c:618
kvm_inject_emulated_page_fault+0x210/0x290 arch/x86/kvm/x86.c:618
...
Call Trace:
...
RIP: 0010:kvm_inject_emulated_page_fault+0x210/0x290 arch/x86/kvm/x86.c:618
...
nested_vmx_get_vmptr+0x1f9/0x2a0 arch/x86/kvm/vmx/nested.c:4638
handle_vmon arch/x86/kvm/vmx/nested.c:4767 [inline]
handle_vmon+0x168/0x3a0 arch/x86/kvm/vmx/nested.c:4728
vmx_handle_exit+0x29c/0x1260 arch/x86/kvm/vmx/vmx.c:6067
'exception' we're trying to inject with kvm_inject_emulated_page_fault()
comes from:
nested_vmx_get_vmptr()
kvm_read_guest_virt()
kvm_read_guest_virt_helper()
vcpu->arch.walk_mmu->gva_to_gpa()
but it is only set when GVA to GPA conversion fails. In case it doesn't but
we still fail kvm_vcpu_read_guest_page(), X86EMUL_IO_NEEDED is returned and
nested_vmx_get_vmptr() calls kvm_inject_emulated_page_fault() with zeroed
'exception'. This happen when the argument is MMIO.
Paolo also noticed that nested_vmx_get_vmptr() is not the only place in
KVM code where kvm_read/write_guest_virt*() return result is mishandled.
VMX instructions along with INVPCID have the same issue. This was already
noticed before, e.g. see commit 541ab2aeb2 ("KVM: x86: work around
leak of uninitialized stack contents") but was never fully fixed.
KVM could've handled the request correctly by going to userspace and
performing I/O but there doesn't seem to be a good need for such requests
in the first place.
Introduce vmx_handle_memory_failure() as an interim solution.
Note, nested_vmx_get_vmptr() now has three possible outcomes: OK, PF,
KVM_EXIT_INTERNAL_ERROR and callers need to know if userspace exit is
needed (for KVM_EXIT_INTERNAL_ERROR) in case of failure. We don't seem
to have a good enum describing this tristate, just add "int *ret" to
nested_vmx_get_vmptr() interface to pass the information.
Reported-by: syzbot+2a7156e11dc199bdbd8a@syzkaller.appspotmail.com
Suggested-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20200605115906.532682-1-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Intel CPUs have a new alternative MSR range (starting from MSR_IA32_PMC0)
for GP counters that allows writing the full counter width. Enable this
range from a new capability bit (IA32_PERF_CAPABILITIES.FW_WRITE[bit 13]).
The guest would query CPUID to get the counter width, and sign extends
the counter values as needed. The traditional MSRs always limit to 32bit,
even though the counter internally is larger (48 or 57 bits).
When the new capability is set, use the alternative range which do not
have these restrictions. This lowers the overhead of perf stat slightly
because it has to do less interrupts to accumulate the counter value.
Signed-off-by: Like Xu <like.xu@linux.intel.com>
Message-Id: <20200529074347.124619-3-like.xu@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Currently, APF mechanism relies on the #PF abuse where the token is being
passed through CR2. If we switch to using interrupts to deliver page-ready
notifications we need a different way to pass the data. Extent the existing
'struct kvm_vcpu_pv_apf_data' with token information for page-ready
notifications.
While on it, rename 'reason' to 'flags'. This doesn't change the semantics
as we only have reasons '1' and '2' and these can be treated as bit flags
but KVM_PV_REASON_PAGE_READY is going away with interrupt based delivery
making 'reason' name misleading.
The newly introduced apf_put_user_ready() temporary puts both flags and
token information, this will be changed to put token only when we switch
to interrupt based notifications.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20200525144125.143875-3-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
vmx_load_mmu_pgd is delaying the write of GUEST_CR3 to prepare_vmcs02 as
an optimization, but this is only correct before the nested vmentry.
If userspace is modifying CR3 with KVM_SET_SREGS after the VM has
already been put in guest mode, the value of CR3 will not be updated.
Remove the optimization, which almost never triggers anyway.
Fixes: 04f11ef458 ("KVM: nVMX: Always write vmcs02.GUEST_CR3 during nested VM-Enter")
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In case an interrupt arrives after nested.check_events but before the
call to kvm_cpu_has_injectable_intr, we could end up enabling the interrupt
window even if the interrupt is actually going to be a vmexit. This is
useless rather than harmful, but it really complicates reasoning about
SVM's handling of the VINTR intercept. We'd like to never bother with
the VINTR intercept if V_INTR_MASKING=1 && INTERCEPT_INTR=1, because in
that case there is no interrupt window and we can just exit the nested
guest whenever we want.
This patch moves the opening of the interrupt window inside
inject_pending_event. This consolidates the check for pending
interrupt/NMI/SMI in one place, and makes KVM's usage of immediate
exits more consistent, extending it beyond just nested virtualization.
There are two functional changes here. They only affect corner cases,
but overall they simplify the inject_pending_event.
- re-injection of still-pending events will also use req_immediate_exit
instead of using interrupt-window intercepts. This should have no impact
on performance on Intel since it simply replaces an interrupt-window
or NMI-window exit for a preemption-timer exit. On AMD, which has no
equivalent of the preemption time, it may incur some overhead but an
actual effect on performance should only be visible in pathological cases.
- kvm_arch_interrupt_allowed and kvm_vcpu_has_events will return true
if an interrupt, NMI or SMI is blocked by nested_run_pending. This
makes sense because entering the VM will allow it to make progress
and deliver the event.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The second "/* fall through */" in rmode_exception() makes code harder to
read. Replace it with "return" to indicate they are different cases, only
the #DB and #BP check vcpu->guest_debug, while others don't care. And this
also improves the readability.
Suggested-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
Message-Id: <1582080348-20827-1-git-send-email-linmiaohe@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Take a u32 for the index in has_emulated_msr() to match hardware, which
treats MSR indices as unsigned 32-bit values. Functionally, taking a
signed int doesn't cause problems with the current code base, but could
theoretically cause problems with 32-bit KVM, e.g. if the index were
checked via a less-than statement, which would evaluate incorrectly for
MSR indices with bit 31 set.
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200218234012.7110-3-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We can simply look at bits 52-53 to identify MMIO entries in KVM's page
tables. Therefore, there is no need to pass a mask to kvm_mmu_set_mmio_spte_mask.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Even though we might not allow the guest to use WAITPKG's new
instructions, we should tell KVM that the feature is supported by the
host CPU.
Note that vmx_waitpkg_supported checks that WAITPKG _can_ be set in
secondary execution controls as specified by VMX capability MSR, rather
that we actually enable it for a guest.
Cc: stable@vger.kernel.org
Fixes: e69e72faa3 ("KVM: x86: Add support for user wait instructions")
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20200523161455.3940-2-mlevitsk@redhat.com>
Reviewed-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The hrtimer used to emulate the VMX-preemption timer must be pinned to
the same logical processor as the vCPU thread to be interrupted if we
want to have any hope of adhering to the architectural specification
of the VMX-preemption timer. Even with this change, the emulated
VMX-preemption timer VM-exit occasionally arrives too late.
Signed-off-by: Jim Mattson <jmattson@google.com>
Reviewed-by: Peter Shier <pshier@google.com>
Reviewed-by: Oliver Upton <oupton@google.com>
Message-Id: <20200508203643.85477-4-jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch implements a fastpath for the preemption timer vmexit. The vmexit
can be handled quickly so it can be performed with interrupts off and going
back directly to the guest.
Testing on SKX Server.
cyclictest in guest(w/o mwait exposed, adaptive advance lapic timer is default -1):
5540.5ns -> 4602ns 17%
kvm-unit-test/vmexit.flat:
w/o avanced timer:
tscdeadline_immed: 3028.5 -> 2494.75 17.6%
tscdeadline: 5765.7 -> 5285 8.3%
w/ adaptive advance timer default -1:
tscdeadline_immed: 3123.75 -> 2583 17.3%
tscdeadline: 4663.75 -> 4537 2.7%
Tested-by: Haiwei Li <lihaiwei@tencent.com>
Cc: Haiwei Li <lihaiwei@tencent.com>
Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Message-Id: <1588055009-12677-8-git-send-email-wanpengli@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Replace the ad hoc test in vmx_set_hv_timer with a test in the caller,
start_hv_timer. This test is not Intel-specific and would be duplicated
when introducing the fast path for the TSC deadline MSR.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
While optimizing posted-interrupt delivery especially for the timer
fastpath scenario, I measured kvm_x86_ops.deliver_posted_interrupt()
to introduce substantial latency because the processor has to perform
all vmentry tasks, ack the posted interrupt notification vector,
read the posted-interrupt descriptor etc.
This is not only slow, it is also unnecessary when delivering an
interrupt to the current CPU (as is the case for the LAPIC timer) because
PIR->IRR and IRR->RVI synchronization is already performed on vmentry
Therefore skip kvm_vcpu_trigger_posted_interrupt in this case, and
instead do vmx_sync_pir_to_irr() on the EXIT_FASTPATH_REENTER_GUEST
fastpath as well.
Tested-by: Haiwei Li <lihaiwei@tencent.com>
Cc: Haiwei Li <lihaiwei@tencent.com>
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Message-Id: <1588055009-12677-6-git-send-email-wanpengli@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
