The recursion protection for hard interrupt stacks is an unsigned int per
CPU variable initialized to -1 named __irq_count.
The irq stack switching is only done when the variable is -1, which creates
worse code than just checking for 0. When the stack switching happens it
uses this_cpu_add/sub(1), but there is no reason to do so. It simply can
use straight writes. This is a historical leftover from the low level ASM
code which used inc and jz to make a decision.
Rename it to hardirq_stack_inuse, make it a bool and use plain stores.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20210210002512.228830141@linutronix.de
efi_recover_from_page_fault() doesn't recover -- it does a special EFI
mini-oops. Rename it to make it clear that it crashes.
While renaming it, I noticed a blatant bug: a page fault oops in a
different thread happening concurrently with an EFI runtime service call
would be misinterpreted as an EFI page fault. Fix that.
This isn't quite exact. The situation could be improved by using a
special CS for calls into EFI.
[ bp: Massage commit message and simplify in interrupt check. ]
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/f43b1e80830dc78ed60ed8b0826f4f189254570c.1612924255.git.luto@kernel.org
POPF is a rather expensive operation, so don't use it for restoring
irq flags. Instead, test whether interrupts are enabled in the flags
parameter and enable interrupts via STI in that case.
This results in the restore_fl paravirt op to be no longer needed.
Suggested-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20210120135555.32594-7-jgross@suse.com
USERGS_SYSRET64 is used to return from a syscall via SYSRET, but
a Xen PV guest will nevertheless use the IRET hypercall, as there
is no sysret PV hypercall defined.
So instead of testing all the prerequisites for doing a sysret and
then mangling the stack for Xen PV again for doing an iret just use
the iret exit from the beginning.
This can easily be done via an ALTERNATIVE like it is done for the
sysenter compat case already.
It should be noted that this drops the optimization in Xen for not
restoring a few registers when returning to user mode, but it seems
as if the saved instructions in the kernel more than compensate for
this drop (a kernel build in a Xen PV guest was slightly faster with
this patch applied).
While at it remove the stale sysret32 remnants.
Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20210120135555.32594-6-jgross@suse.com
SWAPGS is used only for interrupts coming from user mode or for
returning to user mode. So there is no reason to use the PARAVIRT
framework, as it can easily be replaced by an ALTERNATIVE depending
on X86_FEATURE_XENPV.
There are several instances using the PV-aware SWAPGS macro in paths
which are never executed in a Xen PV guest. Replace those with the
plain swapgs instruction. For SWAPGS_UNSAFE_STACK the same applies.
Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Andy Lutomirski <luto@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210120135555.32594-5-jgross@suse.com
Intel Moorestown and Medfield are quite old Intel Atom based
32-bit platforms, which were in limited use in some Android phones,
tablets and consumer electronics more than eight years ago.
There are no bugs or problems ever reported outside from Intel
for breaking any of that platforms for years. It seems no real
users exists who run more or less fresh kernel on it. Commit
05f4434bc1 ("ASoC: Intel: remove mfld_machine") is also in align
with this theory.
Due to above and to reduce a burden of supporting outdated drivers,
remove the support for outdated platforms completely.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Intel Moorestown and Medfield are quite old Intel Atom based
32-bit platforms, which were in limited use in some Android phones,
tablets and consumer electronics more than eight years ago.
There are no bugs or problems ever reported outside from Intel
for breaking any of that platforms for years. It seems no real
users exists who run more or less fresh kernel on it. Commit
05f4434bc1 ("ASoC: Intel: remove mfld_machine") is also in align
with this theory.
Due to above and to reduce a burden of supporting outdated drivers,
remove the support for outdated platforms completely.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Hyper-V emulation is enabled in KVM unconditionally. This is bad at least
from security standpoint as it is an extra attack surface. Ideally, there
should be a per-VM capability explicitly enabled by VMM but currently it
is not the case and we can't mandate one without breaking backwards
compatibility. We can, however, check guest visible CPUIDs and only enable
Hyper-V emulation when "Hv#1" interface was exposed in
HYPERV_CPUID_INTERFACE.
Note, VMMs are free to act in any sequence they like, e.g. they can try
to set MSRs first and CPUIDs later so we still need to allow the host
to read/write Hyper-V specific MSRs unconditionally.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210126134816.1880136-14-vkuznets@redhat.com>
[Add selftest vcpu_set_hv_cpuid API to avoid breaking xen_vmcall_test. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Hyper-V context is only needed for guests which use Hyper-V emulation in
KVM (e.g. Windows/Hyper-V guests). 'struct kvm_vcpu_hv' is, however, quite
big, it accounts for more than 1/4 of the total 'struct kvm_vcpu_arch'
which is also quite big already. This all looks like a waste.
Allocate 'struct kvm_vcpu_hv' dynamically. This patch does not bring any
(intentional) functional change as we still allocate the context
unconditionally but it paves the way to doing that only when needed.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210126134816.1880136-13-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Current KVM_USER_MEM_SLOTS limits are arch specific (512 on Power, 509 on x86,
32 on s390, 16 on MIPS) but they don't really need to be. Memory slots are
allocated dynamically in KVM when added so the only real limitation is
'id_to_index' array which is 'short'. We don't have any other
KVM_MEM_SLOTS_NUM/KVM_USER_MEM_SLOTS-sized statically defined structures.
Low KVM_USER_MEM_SLOTS can be a limiting factor for some configurations.
In particular, when QEMU tries to start a Windows guest with Hyper-V SynIC
enabled and e.g. 256 vCPUs the limit is hit as SynIC requires two pages per
vCPU and the guest is free to pick any GFN for each of them, this fragments
memslots as QEMU wants to have a separate memslot for each of these pages
(which are supposed to act as 'overlay' pages).
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210127175731.2020089-3-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
kvm_get_dr and emulator_get_dr except an in-range value for the register
number so they cannot fail. Change the return type to void.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The TDP MMU assumes that it can do atomic accesses to 64-bit PTEs.
Rather than just disabling it, compile it out completely so that it
is possible to use for example 64-bit xchg.
To limit the number of stubs, wrap all accesses to tdp_mmu_enabled
or tdp_mmu_page with a function. Calls to all other functions in
tdp_mmu.c are eliminated and do not even reach the linker.
Reviewed-by: Sean Christopherson <seanjc@google.com>
Tested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Pull syscall entry fixes from Borislav Petkov:
- For syscall user dispatch, separate prctl operation from syscall
redirection range specification before the API has been made official
in 5.11.
- Ensure tasks using the generic syscall code do trap after returning
from a syscall when single-stepping is requested.
* tag 'core_urgent_for_v5.11_rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
entry: Use different define for selector variable in SUD
entry: Ensure trap after single-step on system call return
Pull x86 fixes from Borislav Petkov:
"I hope this is the last batch of x86/urgent updates for this round:
- Remove superfluous EFI PGD range checks which lead to those
assertions failing with certain kernel configs and LLVM.
- Disable setting breakpoints on facilities involved in #DB exception
handling to avoid infinite loops.
- Add extra serialization to non-serializing MSRs (IA32_TSC_DEADLINE
and x2 APIC MSRs) to adhere to SDM's recommendation and avoid any
theoretical issues.
- Re-add the EPB MSR reading on turbostat so that it works on older
kernels which don't have the corresponding EPB sysfs file.
- Add Alder Lake to the list of CPUs which support split lock.
- Fix %dr6 register handling in order to be able to set watchpoints
with gdb again.
- Disable CET instrumentation in the kernel so that gcc doesn't add
ENDBR64 to kernel code and thus confuse tracing"
* tag 'x86_urgent_for_v5.11_rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/efi: Remove EFI PGD build time checks
x86/debug: Prevent data breakpoints on cpu_dr7
x86/debug: Prevent data breakpoints on __per_cpu_offset
x86/apic: Add extra serialization for non-serializing MSRs
tools/power/turbostat: Fallback to an MSR read for EPB
x86/split_lock: Enable the split lock feature on another Alder Lake CPU
x86/debug: Fix DR6 handling
x86/build: Disable CET instrumentation in the kernel
Commit 2991552447 ("entry: Drop usage of TIF flags in the generic syscall
code") introduced a bug on architectures using the generic syscall entry
code, in which processes stopped by PTRACE_SYSCALL do not trap on syscall
return after receiving a TIF_SINGLESTEP.
The reason is that the meaning of TIF_SINGLESTEP flag is overloaded to
cause the trap after a system call is executed, but since the above commit,
the syscall call handler only checks for the SYSCALL_WORK flags on the exit
work.
Split the meaning of TIF_SINGLESTEP such that it only means single-step
mode, and create a new type of SYSCALL_WORK to request a trap immediately
after a syscall in single-step mode. In the current implementation, the
SYSCALL_WORK flag shadows the TIF_SINGLESTEP flag for simplicity.
Update x86 to flip this bit when a tracer enables single stepping.
Fixes: 2991552447 ("entry: Drop usage of TIF flags in the generic syscall code")
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Kyle Huey <me@kylehuey.com>
Link: https://lore.kernel.org/r/87h7mtc9pr.fsf_-_@collabora.com
Comment says "by preventing anything executable" which is not true. Even
PROT_NONE mapping can't be installed at (1<<47 - 4096).
mmap(0x7ffffffff000, 4096, PROT_NONE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = -1 ENOMEM
[ bp: Fixup to the moved location in page_64_types.h. ]
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Andy Lutomirski <luto@kernel.org>
Link: https://lkml.kernel.org/r/20200305181719.GA5490@avx2
Jan Kiszka reported that the x2apic_wrmsr_fence() function uses a plain
MFENCE while the Intel SDM (10.12.3 MSR Access in x2APIC Mode) calls for
MFENCE; LFENCE.
Short summary: we have special MSRs that have weaker ordering than all
the rest. Add fencing consistent with current SDM recommendations.
This is not known to cause any issues in practice, only in theory.
Longer story below:
The reason the kernel uses a different semantic is that the SDM changed
(roughly in late 2017). The SDM changed because folks at Intel were
auditing all of the recommended fences in the SDM and realized that the
x2apic fences were insufficient.
Why was the pain MFENCE judged insufficient?
WRMSR itself is normally a serializing instruction. No fences are needed
because the instruction itself serializes everything.
But, there are explicit exceptions for this serializing behavior written
into the WRMSR instruction documentation for two classes of MSRs:
IA32_TSC_DEADLINE and the X2APIC MSRs.
Back to x2apic: WRMSR is *not* serializing in this specific case.
But why is MFENCE insufficient? MFENCE makes writes visible, but
only affects load/store instructions. WRMSR is unfortunately not a
load/store instruction and is unaffected by MFENCE. This means that a
non-serializing WRMSR could be reordered by the CPU to execute before
the writes made visible by the MFENCE have even occurred in the first
place.
This means that an x2apic IPI could theoretically be triggered before
there is any (visible) data to process.
Does this affect anything in practice? I honestly don't know. It seems
quite possible that by the time an interrupt gets to consume the (not
yet) MFENCE'd data, it has become visible, mostly by accident.
To be safe, add the SDM-recommended fences for all x2apic WRMSRs.
This also leaves open the question of the _other_ weakly-ordered WRMSR:
MSR_IA32_TSC_DEADLINE. While it has the same ordering architecture as
the x2APIC MSRs, it seems substantially less likely to be a problem in
practice. While writes to the in-memory Local Vector Table (LVT) might
theoretically be reordered with respect to a weakly-ordered WRMSR like
TSC_DEADLINE, the SDM has this to say:
In x2APIC mode, the WRMSR instruction is used to write to the LVT
entry. The processor ensures the ordering of this write and any
subsequent WRMSR to the deadline; no fencing is required.
But, that might still leave xAPIC exposed. The safest thing to do for
now is to add the extra, recommended LFENCE.
[ bp: Massage commit message, fix typos, drop accidentally added
newline to tools/arch/x86/include/asm/barrier.h. ]
Reported-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/20200305174708.F77040DD@viggo.jf.intel.com
Rename cr3_lm_rsvd_bits to reserved_gpa_bits, and use it for all GPA
legality checks. AMD's APM states:
If the C-bit is an address bit, this bit is masked from the guest
physical address when it is translated through the nested page tables.
Thus, any access that can conceivably be run through NPT should ignore
the C-bit when checking for validity.
For features that KVM emulates in software, e.g. MTRRs, there is no
clear direction in the APM for how the C-bit should be handled. For
such cases, follow the SME behavior inasmuch as possible, since SEV is
is essentially a VM-specific variant of SME. For SME, the APM states:
In this case the upper physical address bits are treated as reserved
when the feature is enabled except where otherwise indicated.
Collecting the various relavant SME snippets in the APM and cross-
referencing the omissions with Linux kernel code, this leaves MTTRs and
APIC_BASE as the only flows that KVM emulates that should _not_ ignore
the C-bit.
Note, this means the reserved bit checks in the page tables are
technically broken. This will be remedied in a future patch.
Although the page table checks are technically broken, in practice, it's
all but guaranteed to be irrelevant. NPT is required for SEV, i.e.
shadowing page tables isn't needed in the common case. Theoretically,
the checks could be in play for nested NPT, but it's extremely unlikely
that anyone is running nested VMs on SEV, as doing so would require L1
to expose sensitive data to L0, e.g. the entire VMCB. And if anyone is
running nested VMs, L0 can't read the guest's encrypted memory, i.e. L1
would need to put its NPT in shared memory, in which case the C-bit will
never be set. Or, L1 could use shadow paging, but again, if L0 needs to
read page tables, e.g. to load PDPTRs, the memory can't be encrypted if
L1 has any expectation of L0 doing the right thing.
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Brijesh Singh <brijesh.singh@amd.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20210204000117.3303214-8-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It turns out that we can't handle event channels *entirely* in userspace
by delivering them as ExtINT, because KVM is a bit picky about when it
accepts ExtINT interrupts from a legacy PIC. The in-kernel local APIC
has to have LVT0 configured in APIC_MODE_EXTINT and unmasked, which
isn't necessarily the case for Xen guests especially on secondary CPUs.
To cope with this, add kvm_xen_get_interrupt() which checks the
evtchn_pending_upcall field in the Xen vcpu_info, and delivers the Xen
upcall vector (configured by KVM_XEN_ATTR_TYPE_UPCALL_VECTOR) if it's
set regardless of LAPIC LVT0 configuration. This gives us the minimum
support we need for completely userspace-based implementation of event
channels.
This does mean that vcpu_enter_guest() needs to check for the
evtchn_pending_upcall flag being set, because it can't rely on someone
having set KVM_REQ_EVENT unless we were to add some way for userspace to
do so manually.
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Allow the Xen emulated guest the ability to register secondary
vcpu time information. On Xen guests this is used in order to be
mapped to userspace and hence allow vdso gettimeofday to work.
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
The vcpu info supersedes the per vcpu area of the shared info page and
the guest vcpus will use this instead.
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Xen added this in 2015 (Xen 4.6). On x86_64 and Arm it fills what was
previously a 32-bit hole in the generic shared_info structure; on
i386 it had to go at the end of struct arch_shared_info.
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Add KVM_XEN_ATTR_TYPE_SHARED_INFO to allow hypervisor to know where the
guest's shared info page is.
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Add a new exit reason for emulator to handle Xen hypercalls.
Since this means KVM owns the ABI, dispense with the facility for the
VMM to provide its own copy of the hypercall pages; just fill them in
directly using VMCALL/VMMCALL as we do for the Hyper-V hypercall page.
This behaviour is enabled by a new INTERCEPT_HCALL flag in the
KVM_XEN_HVM_CONFIG ioctl structure, and advertised by the same flag
being returned from the KVM_CAP_XEN_HVM check.
Rename xen_hvm_config() to kvm_xen_write_hypercall_page() and move it
to the nascent xen.c while we're at it, and add a test case.
Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
To prepare for handling page faults in parallel, change the TDP MMU
page fault handler to use atomic operations to set SPTEs so that changes
are not lost if multiple threads attempt to modify the same SPTE.
Reviewed-by: Peter Feiner <pfeiner@google.com>
Signed-off-by: Ben Gardon <bgardon@google.com>
Message-Id: <20210202185734.1680553-21-bgardon@google.com>
[Document new locking rules. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add a read / write lock to be used in place of the MMU spinlock on x86.
The rwlock will enable the TDP MMU to handle page faults, and other
operations in parallel in future commits.
Reviewed-by: Peter Feiner <pfeiner@google.com>
Signed-off-by: Ben Gardon <bgardon@google.com>
Message-Id: <20210202185734.1680553-19-bgardon@google.com>
[Introduce virt/kvm/mmu_lock.h - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Drop kvm_cpu_vmxoff() in favor of the kernel's cpu_vmxoff(). Modify the
latter to return -EIO on fault so that KVM can invoke
kvm_spurious_fault() when appropriate. In addition to the obvious code
reuse, dropping kvm_cpu_vmxoff() also eliminates VMX's last usage of the
__ex()/__kvm_handle_fault_on_reboot() macros, thus helping pave the way
toward dropping them entirely.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20201231002702.2223707-7-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Explicitly tell the compiler that VMXOFF modifies flags (like all VMX
instructions), and mark memory as clobbered since VMXOFF must not be
reordered and also may have memory side effects (though the kernel
really shouldn't be accessing the root VMCS anyways).
Practically speaking, adding the clobbers is most likely a nop; the
primary motivation is to properly document VMXOFF's behavior.
For the flags clobber, both Clang and GCC automatically mark flags as
clobbered; this is noted in commit 4b1e54786e ("KVM/x86: Use assembly
instruction mnemonics instead of .byte streams"), which intentionally
removed the previous clobber. But, neither Clang nor GCC documents
this behavior, and there's no downside to including the clobber.
For the memory clobber, the RFLAGS.IF and CR4.VMXE manipulations that
immediately follow VMXOFF have compiler barriers of their own, i.e.
VMXOFF can't get reordered after clearing CR4.VMXE, which is really
what's of interest.
Cc: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: David P. Reed <dpreed@deepplum.com>
[sean: rewrote changelog, dropped comment adjustments]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20201231002702.2223707-4-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Silently ignore all faults on VMXOFF in the reboot flows as such faults
are all but guaranteed to be due to the CPU not being in VMX root.
Because (a) VMXOFF may be executed in NMI context, e.g. after VMXOFF but
before CR4.VMXE is cleared, (b) there's no way to query the CPU's VMX
state without faulting, and (c) the whole point is to get out of VMX
root, eating faults is the simplest way to achieve the desired behaior.
Technically, VMXOFF can fault (or fail) for other reasons, but all other
fault and failure scenarios are mode related, i.e. the kernel would have
to magically end up in RM, V86, compat mode, at CPL>0, or running with
the SMI Transfer Monitor active. The kernel is beyond hosed if any of
those scenarios are encountered; trying to do something fancy in the
error path to handle them cleanly is pointless.
Fixes: 1e9931146c ("x86: asm/virtext.h: add cpu_vmxoff() inline function")
Reported-by: David P. Reed <dpreed@deepplum.com>
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20201231002702.2223707-2-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Convert kvm_x86_ops to use static calls. Note that all kvm_x86_ops are
covered here except for 'pmu_ops and 'nested ops'.
Here are some numbers running cpuid in a loop of 1 million calls averaged
over 5 runs, measured in the vm (lower is better).
Intel Xeon 3000MHz:
|default |mitigations=off
-------------------------------------
vanilla |.671s |.486s
static call|.573s(-15%)|.458s(-6%)
AMD EPYC 2500MHz:
|default |mitigations=off
-------------------------------------
vanilla |.710s |.609s
static call|.664s(-6%) |.609s(0%)
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Sean Christopherson <seanjc@google.com>
Signed-off-by: Jason Baron <jbaron@akamai.com>
Message-Id: <e057bf1b8a7ad15652df6eeba3f907ae758d3399.1610680941.git.jbaron@akamai.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Use static calls to improve kvm_x86_ops performance. Introduce the
definitions that will be used by a subsequent patch to actualize the
savings. Add a new kvm-x86-ops.h header that can be used for the
definition of static calls. This header is also intended to be
used to simplify the defition of svm_kvm_ops and vmx_x86_ops.
Note that all functions in kvm_x86_ops are covered here except for
'pmu_ops' and 'nested ops'. I think they can be covered by static
calls in a simlilar manner, but were omitted from this series to
reduce scope and because I don't think they have as large of a
performance impact.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Sean Christopherson <seanjc@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Jason Baron <jbaron@akamai.com>
Message-Id: <e5cc82ead7ab37b2dceb0837a514f3f8bea4f8d1.1610680941.git.jbaron@akamai.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
New AMD CPUs have a change that checks #VMEXIT intercept on special SVM
instructions before checking their EAX against reserved memory region.
This change is indicated by CPUID_0x8000000A_EDX[28]. If it is 1, #VMEXIT
is triggered before #GP. KVM doesn't need to intercept and emulate #GP
faults as #GP is supposed to be triggered.
Co-developed-by: Bandan Das <bsd@redhat.com>
Signed-off-by: Bandan Das <bsd@redhat.com>
Signed-off-by: Wei Huang <wei.huang2@amd.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20210126081831.570253-4-wei.huang2@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
DR6_INIT contains the 1-reserved bits as well as the bit that is cleared
to 0 when the condition (e.g. RTM) happens. The value can be used to
initialize dr6 and also be the XOR mask between the #DB exit
qualification (or payload) and DR6.
Concerning that DR6_INIT is used as initial value only once, rename it
to DR6_ACTIVE_LOW and apply it in other places, which would make the
incoming changes for bus lock debug exception more simple.
Signed-off-by: Chenyi Qiang <chenyi.qiang@intel.com>
Message-Id: <20210202090433.13441-2-chenyi.qiang@intel.com>
[Define DR6_FIXED_1 from DR6_ACTIVE_LOW and DR6_VOLATILE. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Virtual Machine can exploit bus locks to degrade the performance of
system. Bus lock can be caused by split locked access to writeback(WB)
memory or by using locks on uncacheable(UC) memory. The bus lock is
typically >1000 cycles slower than an atomic operation within a cache
line. It also disrupts performance on other cores (which must wait for
the bus lock to be released before their memory operations can
complete).
To address the threat, bus lock VM exit is introduced to notify the VMM
when a bus lock was acquired, allowing it to enforce throttling or other
policy based mitigations.
A VMM can enable VM exit due to bus locks by setting a new "Bus Lock
Detection" VM-execution control(bit 30 of Secondary Processor-based VM
execution controls). If delivery of this VM exit was preempted by a
higher priority VM exit (e.g. EPT misconfiguration, EPT violation, APIC
access VM exit, APIC write VM exit, exception bitmap exiting), bit 26 of
exit reason in vmcs field is set to 1.
In current implementation, the KVM exposes this capability through
KVM_CAP_X86_BUS_LOCK_EXIT. The user can get the supported mode bitmap
(i.e. off and exit) and enable it explicitly (disabled by default). If
bus locks in guest are detected by KVM, exit to user space even when
current exit reason is handled by KVM internally. Set a new field
KVM_RUN_BUS_LOCK in vcpu->run->flags to inform the user space that there
is a bus lock detected in guest.
Document for Bus Lock VM exit is now available at the latest "Intel
Architecture Instruction Set Extensions Programming Reference".
Document Link:
https://software.intel.com/content/www/us/en/develop/download/intel-architecture-instruction-set-extensions-programming-reference.html
Co-developed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Signed-off-by: Xiaoyao Li <xiaoyao.li@intel.com>
Signed-off-by: Chenyi Qiang <chenyi.qiang@intel.com>
Message-Id: <20201106090315.18606-4-chenyi.qiang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Remove the update_pte() shadow paging logic, which was obsoleted by
commit 4731d4c7a0 ("KVM: MMU: out of sync shadow core"), but never
removed. As pointed out by Yu, KVM never write protects leaf page
tables for the purposes of shadow paging, and instead marks their
associated shadow page as unsync so that the guest can write PTEs at
will.
The update_pte() path, which predates the unsync logic, optimizes COW
scenarios by refreshing leaf SPTEs when they are written, as opposed to
zapping the SPTE, restarting the guest, and installing the new SPTE on
the subsequent fault. Since KVM no longer write-protects leaf page
tables, update_pte() is unreachable and can be dropped.
Reported-by: Yu Zhang <yu.c.zhang@intel.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20210115004051.4099250-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add AVX version of the Vector Neural Network (VNNI) Instructions.
A processor supports AVX VNNI instructions if CPUID.0x07.0x1:EAX[4] is
present. The following instructions are available when this feature is
present.
1. VPDPBUS: Multiply and Add Unsigned and Signed Bytes
2. VPDPBUSDS: Multiply and Add Unsigned and Signed Bytes with Saturation
3. VPDPWSSD: Multiply and Add Signed Word Integers
4. VPDPWSSDS: Multiply and Add Signed Integers with Saturation
The only in-kernel usage of this is kvm passthrough. The CPU feature
flag is shown as "avx_vnni" in /proc/cpuinfo.
This instruction is currently documented in the latest "extensions"
manual (ISE). It will appear in the "main" manual (SDM) in the future.
Signed-off-by: Kyung Min Park <kyung.min.park@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Message-Id: <20210105004909.42000-2-yang.zhong@intel.com>
Acked-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add perf core PMU support for the Intel Sapphire Rapids server, which is
the successor of the Intel Ice Lake server. The enabling code is based
on Ice Lake, but there are several new features introduced.
The event encoding is changed and simplified, e.g., the event codes
which are below 0x90 are restricted to counters 0-3. The event codes
which above 0x90 are likely to have no restrictions. The event
constraints, extra_regs(), and hardware cache events table are changed
accordingly.
A new Precise Distribution (PDist) facility is introduced, which
further minimizes the skid when a precise event is programmed on the GP
counter 0. Enable the Precise Distribution (PDist) facility with :ppp
event. For this facility to work, the period must be initialized with a
value larger than 127. Add spr_limit_period() to apply the limit for
:ppp event.
Two new data source fields, data block & address block, are added in the
PEBS Memory Info Record for the load latency event. To enable the
feature,
- An auxiliary event has to be enabled together with the load latency
event on Sapphire Rapids. A new flag PMU_FL_MEM_LOADS_AUX is
introduced to indicate the case. A new event, mem-loads-aux, is
exposed to sysfs for the user tool.
Add a check in hw_config(). If the auxiliary event is not detected,
return an unique error -ENODATA.
- The union perf_mem_data_src is extended to support the new fields.
- Ice Lake and earlier models do not support block information, but the
fields may be set by HW on some machines. Add pebs_no_block to
explicitly indicate the previous platforms which don't support the new
block fields. Accessing the new block fields are ignored on those
platforms.
A new store Latency facility is introduced, which leverages the PEBS
facility where it can provide additional information about sampled
stores. The additional information includes the data address, memory
auxiliary info (e.g. Data Source, STLB miss) and the latency of the
store access. To enable the facility, the new event (0x02cd) has to be
programed on the GP counter 0. A new flag PERF_X86_EVENT_PEBS_STLAT is
introduced to indicate the event. The store_latency_data() is introduced
to parse the memory auxiliary info.
The layout of access latency field of PEBS Memory Info Record has been
changed. Two latency, instruction latency (bit 15:0) and cache access
latency (bit 47:32) are recorded.
- The cache access latency is similar to previous memory access latency.
For loads, the latency starts by the actual cache access until the
data is returned by the memory subsystem.
For stores, the latency starts when the demand write accesses the L1
data cache and lasts until the cacheline write is completed in the
memory subsystem.
The cache access latency is stored in low 32bits of the sample type
PERF_SAMPLE_WEIGHT_STRUCT.
- The instruction latency starts by the dispatch of the load operation
for execution and lasts until completion of the instruction it belongs
to.
Add a new flag PMU_FL_INSTR_LATENCY to indicate the instruction
latency support. The instruction latency is stored in the bit 47:32
of the sample type PERF_SAMPLE_WEIGHT_STRUCT.
Extends the PERF_METRICS MSR to feature TMA method level 2 metrics. The
lower half of the register is the TMA level 1 metrics (legacy). The
upper half is also divided into four 8-bit fields for the new level 2
metrics. Expose all eight Topdown metrics events to user space.
The full description for the SPR features can be found at Intel
Architecture Instruction Set Extensions and Future Features
Programming Reference, 319433-041.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1611873611-156687-5-git-send-email-kan.liang@linux.intel.com
Intel Sapphire Rapids server will introduce 8 metrics events. Intel
Ice Lake only supports 4 metrics events. A perf tool user may mistakenly
use the unsupported events via RAW format on Ice Lake. The user can
still get a value from the unsupported Topdown metrics event once the
following Sapphire Rapids enabling patch is applied.
To enable the 8 metrics events on Intel Sapphire Rapids, the
INTEL_TD_METRIC_MAX has to be updated, which impacts the
is_metric_event(). The is_metric_event() is a generic function.
On Ice Lake, the newly added SPR metrics events will be mistakenly
accepted as metric events on creation. At runtime, the unsupported
Topdown metrics events will be updated.
Add a variable num_topdown_events in x86_pmu to indicate the available
number of the Topdown metrics event on the platform. Apply the number
into is_metric_event(). Only the supported Topdown metrics events
should be created as metrics events.
Apply the num_topdown_events in icl_update_topdown_event() as well. The
function can be reused by the following patch.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1611873611-156687-4-git-send-email-kan.liang@linux.intel.com
