The description in the Intel SDM of how the divide configuration
register is used: "The APIC timer frequency will be the processor's bus
clock or core crystal clock frequency divided by the value specified in
the divide configuration register."
Observation of baremetal shown that when the TDCR is change, the TMCCT
does not change or make a big jump in value, but the rate at which it
count down change.
The patch update the emulation to APIC timer to so that a change to the
divide configuration would be reflected in the value of the counter and
when the next interrupt is triggered.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
[Fixed some whitespace and added a check for negative delta and running
timer. - Radim]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
If we take TSC-deadline mode timer out of the picture, the Intel SDM
does not say that the timer is disable when the timer mode is change,
either from one-shot to periodic or vice versa.
After this patch, the timer is no longer disarmed on change of mode, so
the counter (TMCCT) keeps counting down.
So what does a write to LVTT changes ? On baremetal, the change of mode
is probably taken into account only when the counter reach 0. When this
happen, LVTT is use to figure out if the counter should restard counting
down from TMICT (so periodic mode) or stop counting (if one-shot mode).
This patch is based on observation of the behavior of the APIC timer on
baremetal as well as check that they does not go against the description
written in the Intel SDM.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
[Fixed rate limiting of periodic timer.]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Extract the logic of limit lapic periodic timer frequency to a new function,
this function will be used by later patches.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
SDM 10.5.4.1 TSC-Deadline Mode mentioned that "Transitioning between TSC-Deadline
mode and other timer modes also disarms the timer". So the APIC Timer Initial Count
Register for one-shot/periodic mode should be reset. This patch do it.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
[Removed unnecessary definition of APIC_LVT_TIMER_MASK.]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
KVM doesn't expose the PLE capability to the L1 hypervisor, however,
ple_window still shows the default value on L1 hypervisor. This patch
fixes it by clearing all the PLE related module parameter if there is
no PLE capability.
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
When KVM emulates an exit from L2 to L1, it loads L1 CR4 into the
guest CR4. Before this CR4 loading, the guest CR4 refers to L2
CR4. Because these two CR4's are in different levels of guest, we
should vmx_set_cr4() rather than kvm_set_cr4() here. The latter, which
is used to handle guest writes to its CR4, checks the guest change to
CR4 and may fail if the change is invalid.
The failure may cause trouble. Consider we start
a L1 guest with non-zero L1 PCID in use,
(i.e. L1 CR4.PCIDE == 1 && L1 CR3.PCID != 0)
and
a L2 guest with L2 PCID disabled,
(i.e. L2 CR4.PCIDE == 0)
and following events may happen:
1. If kvm_set_cr4() is used in load_vmcs12_host_state() to load L1 CR4
into guest CR4 (in VMCS01) for L2 to L1 exit, it will fail because
of PCID check. As a result, the guest CR4 recorded in L0 KVM (i.e.
vcpu->arch.cr4) is left to the value of L2 CR4.
2. Later, if L1 attempts to change its CR4, e.g., clearing VMXE bit,
kvm_set_cr4() in L0 KVM will think L1 also wants to enable PCID,
because the wrong L2 CR4 is used by L0 KVM as L1 CR4. As L1
CR3.PCID != 0, L0 KVM will inject GP to L1 guest.
Fixes: 4704d0befb ("KVM: nVMX: Exiting from L2 to L1")
Cc: qemu-stable@nongnu.org
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
is_last_gpte() is not equivalent to the pseudo-code given in commit
6bb69c9b69 ("KVM: MMU: simplify last_pte_bitmap") because an incorrect
value of last_nonleaf_level may override the result even if level == 1.
It is critical for is_last_gpte() to return true on level == 1 to
terminate page walks. Otherwise memory corruption may occur as level
is used as an index to various data structures throughout the page
walking code. Even though the actual bug would be wherever the MMU is
initialized (as in the previous patch), be defensive and ensure here
that is_last_gpte() returns the correct value.
This patch is also enough to fix CVE-2017-12188.
Fixes: 6bb69c9b69
Cc: stable@vger.kernel.org
Cc: Andy Honig <ahonig@google.com>
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
[Panic if walk_addr_generic gets an incorrect level; this is a serious
bug and it's not worth a WARN_ON where the recovery path might hide
further exploitable issues; suggested by Andrew Honig. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The function updates context->root_level but didn't call
update_last_nonleaf_level so the previous and potentially wrong value
was used for page walks. For example, a zero value of last_nonleaf_level
would allow a potential out-of-bounds access in arch/x86/mmu/paging_tmpl.h's
walk_addr_generic function (CVE-2017-12188).
Fixes: 155a97a3d7
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The rework of the posted interrupt handling broke building without
support for the local APIC:
ERROR: "boot_cpu_physical_apicid" [arch/x86/kvm/kvm-intel.ko] undefined!
That configuration is probably not particularly useful anyway, so
we can avoid the randconfig failures by adding a Kconfig dependency.
Fixes: 8b306e2f3c ("KVM: VMX: avoid double list add with VT-d posted interrupts")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
When compiling the kernel with the '-frecord-gcc-switches' flag, objtool
complains:
arch/x86/kvm/emulate.o: warning: objtool: .GCC.command.line+0x0: special: can't find new instruction
And also the kernel fails to link.
The problem is that the 'kvm_fastop_exception' code gets placed into the
throwaway '.GCC.command.line' section instead of '.text'.
Exception fixup code is conventionally placed in the '.fixup' section,
so put it there where it belongs.
Reported-and-tested-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Currently, in PREEMPT_COUNT=n kernel, kvm_async_pf_task_wait() could call
schedule() to reschedule in some cases. This could result in
accidentally ending the current RCU read-side critical section early,
causing random memory corruption in the guest, or otherwise preempting
the currently running task inside between preempt_disable and
preempt_enable.
The difficulty to handle this well is because we don't know whether an
async PF delivered in a preemptible section or RCU read-side critical section
for PREEMPT_COUNT=n, since preempt_disable()/enable() and rcu_read_lock/unlock()
are both no-ops in that case.
To cure this, we treat any async PF interrupting a kernel context as one
that cannot be preempted, preventing kvm_async_pf_task_wait() from choosing
the schedule() path in that case.
To do so, a second parameter for kvm_async_pf_task_wait() is introduced,
so that we know whether it's called from a context interrupting the
kernel, and the parameter is set properly in all the callsites.
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Wanpeng Li <wanpeng.li@hotmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
that was finally triggered by PCID support.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)
iQEcBAABAgAGBQJZzmJqAAoJEL/70l94x66D99oH/R4hOMfzDxFOgW3LnaCQJvwo
n1+tH3as0dfdkpggZ+UmJuKnbVJ0625+qozenrdYkKtk1YiyIIQWG3vdsz4HBfzp
CYK2NVVymf0dg8DQaluz6iB1R28ke12PggzyFv01s1QyENBDA8J38pslZarPM2OA
tnpRKC6B59/VmRD0PWS6yRmTXY+HfzWlWg4JMraq2VdybbEXJhh8BNfjjNn30DkZ
SW8kHq60AUd5Arhb3cmiPiXZCQ7odqF2u2mEcBmnA9hAacaGEheSzKCUOaEIjmZV
5/jTyG1tZkN7CbrG81ryuoa8A6qTOSyHxo1QkzAmE/p+s2IzGfzzLqmtfIsAWkE=
=1lM1
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull kvm fixes from Paolo Bonzini:
"Mixed bugfixes. Perhaps the most interesting one is a latent bug that
was finally triggered by PCID support"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
kvm/x86: Handle async PF in RCU read-side critical sections
KVM: nVMX: Fix nested #PF intends to break L1's vmlauch/vmresume
KVM: VMX: use cmpxchg64
KVM: VMX: simplify and fix vmx_vcpu_pi_load
KVM: VMX: avoid double list add with VT-d posted interrupts
KVM: VMX: extract __pi_post_block
KVM: PPC: Book3S HV: Check for updated HDSISR on P9 HDSI exception
KVM: nVMX: fix HOST_CR3/HOST_CR4 cache
------------[ cut here ]------------
WARNING: CPU: 4 PID: 5280 at /home/kernel/linux/arch/x86/kvm//vmx.c:11394 nested_vmx_vmexit+0xc2b/0xd70 [kvm_intel]
CPU: 4 PID: 5280 Comm: qemu-system-x86 Tainted: G W OE 4.13.0+ #17
RIP: 0010:nested_vmx_vmexit+0xc2b/0xd70 [kvm_intel]
Call Trace:
? emulator_read_emulated+0x15/0x20 [kvm]
? segmented_read+0xae/0xf0 [kvm]
vmx_inject_page_fault_nested+0x60/0x70 [kvm_intel]
? vmx_inject_page_fault_nested+0x60/0x70 [kvm_intel]
x86_emulate_instruction+0x733/0x810 [kvm]
vmx_handle_exit+0x2f4/0xda0 [kvm_intel]
? kvm_arch_vcpu_ioctl_run+0xd2f/0x1c60 [kvm]
kvm_arch_vcpu_ioctl_run+0xdab/0x1c60 [kvm]
? kvm_arch_vcpu_load+0x62/0x230 [kvm]
kvm_vcpu_ioctl+0x340/0x700 [kvm]
? kvm_vcpu_ioctl+0x340/0x700 [kvm]
? __fget+0xfc/0x210
do_vfs_ioctl+0xa4/0x6a0
? __fget+0x11d/0x210
SyS_ioctl+0x79/0x90
entry_SYSCALL_64_fastpath+0x23/0xc2
A nested #PF is triggered during L0 emulating instruction for L2. However, it
doesn't consider we should not break L1's vmlauch/vmresme. This patch fixes
it by queuing the #PF exception instead ,requesting an immediate VM exit from
L2 and keeping the exception for L1 pending for a subsequent nested VM exit.
This should actually work all the time, making vmx_inject_page_fault_nested
totally unnecessary. However, that's not working yet, so this patch can work
around the issue in the meanwhile.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The simplify part: do not touch pi_desc.nv, we can set it when the
VCPU is first created. Likewise, pi_desc.sn is only handled by
vmx_vcpu_pi_load, do not touch it in __pi_post_block.
The fix part: do not check kvm_arch_has_assigned_device, instead
check the SN bit to figure out whether vmx_vcpu_pi_put ran before.
This matches what the previous patch did in pi_post_block.
Cc: Huangweidong <weidong.huang@huawei.com>
Cc: Gonglei <arei.gonglei@huawei.com>
Cc: wangxin <wangxinxin.wang@huawei.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Tested-by: Longpeng (Mike) <longpeng2@huawei.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In some cases, for example involving hot-unplug of assigned
devices, pi_post_block can forget to remove the vCPU from the
blocked_vcpu_list. When this happens, the next call to
pi_pre_block corrupts the list.
Fix this in two ways. First, check vcpu->pre_pcpu in pi_pre_block
and WARN instead of adding the element twice in the list. Second,
always do the list removal in pi_post_block if vcpu->pre_pcpu is
set (not -1).
The new code keeps interrupts disabled for the whole duration of
pi_pre_block/pi_post_block. This is not strictly necessary, but
easier to follow. For the same reason, PI.ON is checked only
after the cmpxchg, and to handle it we just call the post-block
code. This removes duplication of the list removal code.
Cc: Huangweidong <weidong.huang@huawei.com>
Cc: Gonglei <arei.gonglei@huawei.com>
Cc: wangxin <wangxinxin.wang@huawei.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Tested-by: Longpeng (Mike) <longpeng2@huawei.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Rename this function to better express that it's all about
initializing the FPU state of a task which goes hand in hand
with the fpu::initialized field.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Eric Biggers <ebiggers3@gmail.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Yu-cheng Yu <yu-cheng.yu@intel.com>
Link: http://lkml.kernel.org/r/20170923130016.21448-33-mingo@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull x86 fixes from Ingo Molnar:
"Another round of CR3/PCID related fixes (I think this addresses all
but one of the known problems with PCID support), an objtool fix plus
a Clang fix that (finally) solves all Clang quirks to build a bootable
x86 kernel as-is"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/asm: Fix inline asm call constraints for Clang
objtool: Handle another GCC stack pointer adjustment bug
x86/mm/32: Load a sane CR3 before cpu_init() on secondary CPUs
x86/mm/32: Move setup_clear_cpu_cap(X86_FEATURE_PCID) earlier
x86/mm/64: Stop using CR3.PCID == 0 in ASID-aware code
x86/mm: Factor out CR3-building code
For inline asm statements which have a CALL instruction, we list the
stack pointer as a constraint to convince GCC to ensure the frame
pointer is set up first:
static inline void foo()
{
register void *__sp asm(_ASM_SP);
asm("call bar" : "+r" (__sp))
}
Unfortunately, that pattern causes Clang to corrupt the stack pointer.
The fix is easy: convert the stack pointer register variable to a global
variable.
It should be noted that the end result is different based on the GCC
version. With GCC 6.4, this patch has exactly the same result as
before:
defconfig defconfig-nofp distro distro-nofp
before 9820389 9491555 8816046 8516940
after 9820389 9491555 8816046 8516940
With GCC 7.2, however, GCC's behavior has changed. It now changes its
behavior based on the conversion of the register variable to a global.
That somehow convinces it to *always* set up the frame pointer before
inserting *any* inline asm. (Therefore, listing the variable as an
output constraint is a no-op and is no longer necessary.) It's a bit
overkill, but the performance impact should be negligible. And in fact,
there's a nice improvement with frame pointers disabled:
defconfig defconfig-nofp distro distro-nofp
before 9796316 9468236 9076191 8790305
after 9796957 9464267 9076381 8785949
So in summary, while listing the stack pointer as an output constraint
is no longer necessary for newer versions of GCC, it's still needed for
older versions.
Suggested-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Reported-by: Matthias Kaehlcke <mka@chromium.org>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Dmitriy Vyukov <dvyukov@google.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Miguel Bernal Marin <miguel.bernal.marin@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/3db862e970c432ae823cf515c52b54fec8270e0e.1505942196.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
For nested virt we maintain multiple VMCS that can run on a vCPU. So it is
incorrect to keep vmcs_host_cr3 and vmcs_host_cr4, whose purpose is caching
the value of the rarely changing HOST_CR3 and HOST_CR4 VMCS fields, in
vCPU-wide data structures.
Hyper-V nested on KVM runs into this consistently for me with PCID enabled.
CR3 is updated with a new value, unlikely(cr3 != vmx->host_state.vmcs_host_cr3)
fires, and the currently loaded VMCS is updated. Then we switch from L2 to
L1 and the next exit reverts CR3 to its old value.
Fixes: d6e41f1151 ("x86/mm, KVM: Teach KVM's VMX code that CR3 isn't a constant")
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
WARN_ON_ONCE(pi_test_sn(&vmx->pi_desc)) in kvm_vcpu_trigger_posted_interrupt()
intends to detect the violation of invariant that VT-d PI notification
event is not suppressed when vcpu is in the guest mode. Because the
two checks for the target vcpu mode and the target suppress field
cannot be performed atomically, the target vcpu mode may change in
between. If that does happen, WARN_ON_ONCE() here may raise false
alarms.
As the previous patch fixed the real invariant breaker, remove this
WARN_ON_ONCE() to avoid false alarms, and document the allowed cases
instead.
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Reported-by: "Ramamurthy, Venkatesh" <venkatesh.ramamurthy@intel.com>
Reported-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Fixes: 28b835d60f ("KVM: Update Posted-Interrupts Descriptor when vCPU is preempted")
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
In kvm_vcpu_trigger_posted_interrupt() and pi_pre_block(), KVM
assumes that PI notification events should not be suppressed when the
target vCPU is not blocked.
vmx_update_pi_irte() sets the SN field before changing an interrupt
from posting to remapping, but it does not check the vCPU mode.
Therefore, the change of SN field may break above the assumption.
Besides, I don't see reasons to suppress notification events here, so
remove the changes of SN field to avoid race condition.
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Reported-by: "Ramamurthy, Venkatesh" <venkatesh.ramamurthy@intel.com>
Reported-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Fixes: 28b835d60f ("KVM: Update Posted-Interrupts Descriptor when vCPU is preempted")
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Routine check_cr_write() will trigger emulator_get_cpuid()->
kvm_cpuid() to get maxphyaddr, and NULL is passed as values
for ebx/ecx/edx. This is problematic because kvm_cpuid() will
dereference these pointers.
Fixes: d1cd3ce900 ("KVM: MMU: check guest CR3 reserved bits based on its physical address width.")
Reported-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Yu Zhang <yu.c.zhang@linux.intel.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
When emulating a nested VM-entry from L1 to L2, several control field
validation checks are deferred to the hardware. Should one of these
validation checks fail, vcpu_vmx_run will set the vmx->fail flag. When
this happens, the L2 guest state is not loaded (even in part), and
execution should continue in L1 with the next instruction after the
VMLAUNCH/VMRESUME.
The VMCS12 is not modified (except for the VM-instruction error
field), the VMCS12 MSR save/load lists are not processed, and the CPU
state is not loaded from the VMCS12 host area. Moreover, the vmcs02
exit reason is stale, so it should not be consulted for any reason.
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
On an early VMLAUNCH/VMRESUME failure (i.e. one which sets the
VM-instruction error field of the current VMCS), the launch state of
the current VMCS is not set to "launched," and the VM-exit information
fields of the current VMCS (including IDT-vectoring information and
exit reason) are stale.
On a late VMLAUNCH/VMRESUME failure (i.e. one which sets the high bit
of the exit reason field), the launch state of the current VMCS is not
set to "launched," and only two of the VM-exit information fields of
the current VMCS are modified (exit reason and exit
qualification). The remaining VM-exit information fields of the
current VMCS (including IDT-vectoring information, in particular) are
stale.
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
After a successful VM-entry, RFLAGS is cleared, with the exception of
bit 1, which is always set. This is handled by load_vmcs12_host_state.
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The value of the guest_irq argument to vmx_update_pi_irte() is
ultimately coming from a KVM_IRQFD API call. Do not BUG() in
vmx_update_pi_irte() if the value is out-of bounds. (Especially,
since KVM as a whole seems to hang after that.)
Instead, print a message only once if we find that we don't have a
route for a certain IRQ (which can be out-of-bounds or within the
array).
This fixes CVE-2017-1000252.
Fixes: efc644048e ("KVM: x86: Update IRTE for posted-interrupts")
Signed-off-by: Jan H. Schönherr <jschoenh@amazon.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If L1 does not specify the "use TPR shadow" VM-execution control in
vmcs12, then L0 must specify the "CR8-load exiting" and "CR8-store
exiting" VM-execution controls in vmcs02. Failure to do so will give
the L2 VM unrestricted read/write access to the hardware CR8.
This fixes CVE-2017-12154.
Signed-off-by: Jim Mattson <jmattson@google.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
qemu-system-x86-8600 [004] d..1 7205.687530: kvm_entry: vcpu 2
qemu-system-x86-8600 [004] .... 7205.687532: kvm_exit: reason EXCEPTION_NMI rip 0xffffffffa921297d info ffffeb2c0e44e018 80000b0e
qemu-system-x86-8600 [004] .... 7205.687532: kvm_page_fault: address ffffeb2c0e44e018 error_code 0
qemu-system-x86-8600 [004] .... 7205.687620: kvm_try_async_get_page: gva = 0xffffeb2c0e44e018, gfn = 0x427e4e
qemu-system-x86-8600 [004] .N.. 7205.687628: kvm_async_pf_not_present: token 0x8b002 gva 0xffffeb2c0e44e018
kworker/4:2-7814 [004] .... 7205.687655: kvm_async_pf_completed: gva 0xffffeb2c0e44e018 address 0x7fcc30c4e000
qemu-system-x86-8600 [004] .... 7205.687703: kvm_async_pf_ready: token 0x8b002 gva 0xffffeb2c0e44e018
qemu-system-x86-8600 [004] d..1 7205.687711: kvm_entry: vcpu 2
After running some memory intensive workload in guest, I catch the kworker
which completes the GUP too quickly, and queues an "Page Ready" #PF exception
after the "Page not Present" exception before the next vmentry as the above
trace which will result in #DF injected to guest.
This patch fixes it by clearing the queue for "Page not Present" if "Page Ready"
occurs before the next vmentry since the GUP has already got the required page
and shadow page table has already been fixed by "Page Ready" handler.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Fixes: 7c90705bf2 ("KVM: Inject asynchronous page fault into a PV guest if page is swapped out.")
[Changed indentation and added clearing of injected. - Radim]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Don't block vCPU if there is pending exception.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
SVM AVIC hardware accelerates guest write to APIC_EOI register
(for edge-trigger interrupt), which means it does not trap to KVM.
So, only enable SVM AVIC only in split irqchip mode.
(e.g. launching qemu w/ option '-machine kernel_irqchip=split').
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Fixes: 44a95dae1d ("KVM: x86: Detect and Initialize AVIC support")
[Removed pr_debug - Radim.]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Modify struct kvm_x86_ops.arch.apicv_active() to take struct kvm_vcpu
pointer as parameter in preparation to subsequent changes.
Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Preparing the base code for subsequent changes. This does not change
existing logic.
Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Clang resolves __builtin_constant_p() to false even if the expression is
constant in the end. The only purpose of that expression was to
differentiate a case where the following expression couldn't be checked
at compile-time, so we can just remove the check.
Clang handles the following two correctly. Turn it into BUG_ON if there
are any more problems with this.
Fixes: d6321d4933 ("KVM: x86: generalize guest_cpuid_has_ helpers")
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
When user space sets kvm_run->immediate_exit, KVM is supposed to
return quickly. However, when a vCPU is in KVM_MP_STATE_UNINITIALIZED,
the value is not considered and the vCPU blocks.
Fix that oversight.
Fixes: 460df4c1fc ("KVM: race-free exit from KVM_RUN without POSIX signals")
Signed-off-by: Jan H. Schönherr <jschoenh@amazon.de>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
KVM API says that KVM_RUN will return with -EINTR when a signal is
pending. However, if a vCPU is in KVM_MP_STATE_UNINITIALIZED, then
the return value is unconditionally -EAGAIN.
Copy over some code from vcpu_run(), so that the case of a pending
signal results in the expected return value.
Signed-off-by: Jan H. Schönherr <jschoenh@amazon.de>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Common:
- improve heuristic for boosting preempted spinlocks by ignoring VCPUs
in user mode
ARM:
- fix for decoding external abort types from guests
- added support for migrating the active priority of interrupts when
running a GICv2 guest on a GICv3 host
- minor cleanup
PPC:
- expose storage keys to userspace
- merge powerpc/topic/ppc-kvm branch that contains
find_linux_pte_or_hugepte and POWER9 thread management cleanup
- merge kvm-ppc-fixes with a fix that missed 4.13 because of vacations
- fixes
s390:
- merge of topic branch tlb-flushing from the s390 tree to get the
no-dat base features
- merge of kvm/master to avoid conflicts with additional sthyi fixes
- wire up the no-dat enhancements in KVM
- multiple epoch facility (z14 feature)
- Configuration z/Architecture Mode
- more sthyi fixes
- gdb server range checking fix
- small code cleanups
x86:
- emulate Hyper-V TSC frequency MSRs
- add nested INVPCID
- emulate EPTP switching VMFUNC
- support Virtual GIF
- support 5 level page tables
- speedup nested VM exits by packing byte operations
- speedup MMIO by using hardware provided physical address
- a lot of fixes and cleanups, especially nested
-----BEGIN PGP SIGNATURE-----
iQEcBAABCAAGBQJZspE1AAoJEED/6hsPKofoDcMIALT11n+LKV50QGwQdg2W1GOt
aChbgnj/Kegit3hQlDhVNb8kmdZEOZzSL81Lh0VPEr7zXU8QiWn2snbizDPv8sde
MpHhcZYZZ0YrpoiZKjl8yiwcu88OWGn2qtJ7OpuTS5hvEGAfxMncp0AMZho6fnz/
ySTwJ9GK2MTgBw39OAzCeDOeoYn4NKYMwjJGqBXRhNX8PG/1wmfqv0vPrd6wfg31
KJ58BumavwJjr8YbQ1xELm9rpQrAmaayIsG0R1dEUqCbt5a1+t2gt4h2uY7tWcIv
ACt2bIze7eF3xA+OpRs+eT+yemiH3t9btIVmhCfzUpnQ+V5Z55VMSwASLtTuJRQ=
=R8Ry
-----END PGP SIGNATURE-----
Merge tag 'kvm-4.14-1' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM updates from Radim Krčmář:
"First batch of KVM changes for 4.14
Common:
- improve heuristic for boosting preempted spinlocks by ignoring
VCPUs in user mode
ARM:
- fix for decoding external abort types from guests
- added support for migrating the active priority of interrupts when
running a GICv2 guest on a GICv3 host
- minor cleanup
PPC:
- expose storage keys to userspace
- merge kvm-ppc-fixes with a fix that missed 4.13 because of
vacations
- fixes
s390:
- merge of kvm/master to avoid conflicts with additional sthyi fixes
- wire up the no-dat enhancements in KVM
- multiple epoch facility (z14 feature)
- Configuration z/Architecture Mode
- more sthyi fixes
- gdb server range checking fix
- small code cleanups
x86:
- emulate Hyper-V TSC frequency MSRs
- add nested INVPCID
- emulate EPTP switching VMFUNC
- support Virtual GIF
- support 5 level page tables
- speedup nested VM exits by packing byte operations
- speedup MMIO by using hardware provided physical address
- a lot of fixes and cleanups, especially nested"
* tag 'kvm-4.14-1' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (67 commits)
KVM: arm/arm64: Support uaccess of GICC_APRn
KVM: arm/arm64: Extract GICv3 max APRn index calculation
KVM: arm/arm64: vITS: Drop its_ite->lpi field
KVM: arm/arm64: vgic: constify seq_operations and file_operations
KVM: arm/arm64: Fix guest external abort matching
KVM: PPC: Book3S HV: Fix memory leak in kvm_vm_ioctl_get_htab_fd
KVM: s390: vsie: cleanup mcck reinjection
KVM: s390: use WARN_ON_ONCE only for checking
KVM: s390: guestdbg: fix range check
KVM: PPC: Book3S HV: Report storage key support to userspace
KVM: PPC: Book3S HV: Fix case where HDEC is treated as 32-bit on POWER9
KVM: PPC: Book3S HV: Fix invalid use of register expression
KVM: PPC: Book3S HV: Fix H_REGISTER_VPA VPA size validation
KVM: PPC: Book3S HV: Fix setting of storage key in H_ENTER
KVM: PPC: e500mc: Fix a NULL dereference
KVM: PPC: e500: Fix some NULL dereferences on error
KVM: PPC: Book3S HV: Protect updates to spapr_tce_tables list
KVM: s390: we are always in czam mode
KVM: s390: expose no-DAT to guest and migration support
KVM: s390: sthyi: remove invalid guest write access
...
This fix was intended for 4.13, but didn't get in because both
maintainers were on vacation.
Paul Mackerras:
"It adds mutual exclusion between list_add_rcu and list_del_rcu calls
on the kvm->arch.spapr_tce_tables list. Without this, userspace could
potentially trigger corruption of the list and cause a host crash or
worse."
Pull x86 apic updates from Thomas Gleixner:
"This update provides:
- Cleanup of the IDT management including the removal of the extra
tracing IDT. A first step to cleanup the vector management code.
- The removal of the paravirt op adjust_exception_frame. This is a
XEN specific issue, but merged through this branch to avoid nasty
merge collisions
- Prevent dmesg spam about the TSC DEADLINE bug, when the CPU has
disabled the TSC DEADLINE timer in CPUID.
- Adjust a debug message in the ioapic code to print out the
information correctly"
* 'x86-apic-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (51 commits)
x86/idt: Fix the X86_TRAP_BP gate
x86/xen: Get rid of paravirt op adjust_exception_frame
x86/eisa: Add missing include
x86/idt: Remove superfluous ALIGNment
x86/apic: Silence "FW_BUG TSC_DEADLINE disabled due to Errata" on CPUs without the feature
x86/idt: Remove the tracing IDT leftovers
x86/idt: Hide set_intr_gate()
x86/idt: Simplify alloc_intr_gate()
x86/idt: Deinline setup functions
x86/idt: Remove unused functions/inlines
x86/idt: Move interrupt gate initialization to IDT code
x86/idt: Move APIC gate initialization to tables
x86/idt: Move regular trap init to tables
x86/idt: Move IST stack based traps to table init
x86/idt: Move debug stack init to table based
x86/idt: Switch early trap init to IDT tables
x86/idt: Prepare for table based init
x86/idt: Move early IDT setup out of 32-bit asm
x86/idt: Move early IDT handler setup to IDT code
x86/idt: Consolidate IDT invalidation
...
Pull x86 mm changes from Ingo Molnar:
"PCID support, 5-level paging support, Secure Memory Encryption support
The main changes in this cycle are support for three new, complex
hardware features of x86 CPUs:
- Add 5-level paging support, which is a new hardware feature on
upcoming Intel CPUs allowing up to 128 PB of virtual address space
and 4 PB of physical RAM space - a 512-fold increase over the old
limits. (Supercomputers of the future forecasting hurricanes on an
ever warming planet can certainly make good use of more RAM.)
Many of the necessary changes went upstream in previous cycles,
v4.14 is the first kernel that can enable 5-level paging.
This feature is activated via CONFIG_X86_5LEVEL=y - disabled by
default.
(By Kirill A. Shutemov)
- Add 'encrypted memory' support, which is a new hardware feature on
upcoming AMD CPUs ('Secure Memory Encryption', SME) allowing system
RAM to be encrypted and decrypted (mostly) transparently by the
CPU, with a little help from the kernel to transition to/from
encrypted RAM. Such RAM should be more secure against various
attacks like RAM access via the memory bus and should make the
radio signature of memory bus traffic harder to intercept (and
decrypt) as well.
This feature is activated via CONFIG_AMD_MEM_ENCRYPT=y - disabled
by default.
(By Tom Lendacky)
- Enable PCID optimized TLB flushing on newer Intel CPUs: PCID is a
hardware feature that attaches an address space tag to TLB entries
and thus allows to skip TLB flushing in many cases, even if we
switch mm's.
(By Andy Lutomirski)
All three of these features were in the works for a long time, and
it's coincidence of the three independent development paths that they
are all enabled in v4.14 at once"
* 'x86-mm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (65 commits)
x86/mm: Enable RCU based page table freeing (CONFIG_HAVE_RCU_TABLE_FREE=y)
x86/mm: Use pr_cont() in dump_pagetable()
x86/mm: Fix SME encryption stack ptr handling
kvm/x86: Avoid clearing the C-bit in rsvd_bits()
x86/CPU: Align CR3 defines
x86/mm, mm/hwpoison: Clear PRESENT bit for kernel 1:1 mappings of poison pages
acpi, x86/mm: Remove encryption mask from ACPI page protection type
x86/mm, kexec: Fix memory corruption with SME on successive kexecs
x86/mm/pkeys: Fix typo in Documentation/x86/protection-keys.txt
x86/mm/dump_pagetables: Speed up page tables dump for CONFIG_KASAN=y
x86/mm: Implement PCID based optimization: try to preserve old TLB entries using PCID
x86: Enable 5-level paging support via CONFIG_X86_5LEVEL=y
x86/mm: Allow userspace have mappings above 47-bit
x86/mm: Prepare to expose larger address space to userspace
x86/mpx: Do not allow MPX if we have mappings above 47-bit
x86/mm: Rename tasksize_32bit/64bit to task_size_32bit/64bit()
x86/xen: Redefine XEN_ELFNOTE_INIT_P2M using PUD_SIZE * PTRS_PER_PUD
x86/mm/dump_pagetables: Fix printout of p4d level
x86/mm/dump_pagetables: Generalize address normalization
x86/boot: Fix memremap() related build failure
...
Pull x86 asm updates from Ingo Molnar:
- Introduce the ORC unwinder, which can be enabled via
CONFIG_ORC_UNWINDER=y.
The ORC unwinder is a lightweight, Linux kernel specific debuginfo
implementation, which aims to be DWARF done right for unwinding.
Objtool is used to generate the ORC unwinder tables during build, so
the data format is flexible and kernel internal: there's no
dependency on debuginfo created by an external toolchain.
The ORC unwinder is almost two orders of magnitude faster than the
(out of tree) DWARF unwinder - which is important for perf call graph
profiling. It is also significantly simpler and is coded defensively:
there has not been a single ORC related kernel crash so far, even
with early versions. (knock on wood!)
But the main advantage is that enabling the ORC unwinder allows
CONFIG_FRAME_POINTERS to be turned off - which speeds up the kernel
measurably:
With frame pointers disabled, GCC does not have to add frame pointer
instrumentation code to every function in the kernel. The kernel's
.text size decreases by about 3.2%, resulting in better cache
utilization and fewer instructions executed, resulting in a broad
kernel-wide speedup. Average speedup of system calls should be
roughly in the 1-3% range - measurements by Mel Gorman [1] have shown
a speedup of 5-10% for some function execution intense workloads.
The main cost of the unwinder is that the unwinder data has to be
stored in RAM: the memory cost is 2-4MB of RAM, depending on kernel
config - which is a modest cost on modern x86 systems.
Given how young the ORC unwinder code is it's not enabled by default
- but given the performance advantages the plan is to eventually make
it the default unwinder on x86.
See Documentation/x86/orc-unwinder.txt for more details.
- Remove lguest support: its intended role was that of a temporary
proof of concept for virtualization, plus its removal will enable the
reduction (removal) of the paravirt API as well, so Rusty agreed to
its removal. (Juergen Gross)
- Clean up and fix FSGS related functionality (Andy Lutomirski)
- Clean up IO access APIs (Andy Shevchenko)
- Enhance the symbol namespace (Jiri Slaby)
* 'x86-asm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (47 commits)
objtool: Handle GCC stack pointer adjustment bug
x86/entry/64: Use ENTRY() instead of ALIGN+GLOBAL for stub32_clone()
x86/fpu/math-emu: Add ENDPROC to functions
x86/boot/64: Extract efi_pe_entry() from startup_64()
x86/boot/32: Extract efi_pe_entry() from startup_32()
x86/lguest: Remove lguest support
x86/paravirt/xen: Remove xen_patch()
objtool: Fix objtool fallthrough detection with function padding
x86/xen/64: Fix the reported SS and CS in SYSCALL
objtool: Track DRAP separately from callee-saved registers
objtool: Fix validate_branch() return codes
x86: Clarify/fix no-op barriers for text_poke_bp()
x86/switch_to/64: Rewrite FS/GS switching yet again to fix AMD CPUs
selftests/x86/fsgsbase: Test selectors 1, 2, and 3
x86/fsgsbase/64: Report FSBASE and GSBASE correctly in core dumps
x86/fsgsbase/64: Fully initialize FS and GS state in start_thread_common
x86/asm: Fix UNWIND_HINT_REGS macro for older binutils
x86/asm/32: Fix regs_get_register() on segment registers
x86/xen/64: Rearrange the SYSCALL entries
x86/asm/32: Remove a bunch of '& 0xffff' from pt_regs segment reads
...
Calls to mmu_notifier_invalidate_page() were replaced by calls to
mmu_notifier_invalidate_range() and are now bracketed by calls to
mmu_notifier_invalidate_range_start()/end()
Remove now useless invalidate_page callback.
Changed since v1 (Linus Torvalds)
- remove now useless kvm_arch_mmu_notifier_invalidate_page()
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Tested-by: Mike Galbraith <efault@gmx.de>
Tested-by: Adam Borowski <kilobyte@angband.pl>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: kvm@vger.kernel.org
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The first 32 bits of gate struct are the same for 32 and 64 bit kernels.
The 32-bit version uses desc_struct and no designated data structure,
so we need different accessors for 32 and 64 bit kernels.
Aside of that the macros which are necessary to build the 32-bit
gate descriptor are horrible to read.
Unify the gate structs and switch all code fiddling with it over.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20170828064957.861974317@linutronix.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The following commit:
d0ec49d4de ("kvm/x86/svm: Support Secure Memory Encryption within KVM")
uses __sme_clr() to remove the C-bit in rsvd_bits(). rsvd_bits() is
just a simple function to return some 1 bits. Applying a mask based
on properties of the host MMU is incorrect. Additionally, the masks
computed by __reset_rsvds_bits_mask also apply to guest page tables,
where the C bit is reserved since we don't emulate SME.
The fix is to clear the C-bit from rsvd_bits_mask array after it has been
populated from __reset_rsvds_bits_mask()
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: kvm@vger.kernel.org
Cc: paolo.bonzini@gmail.com
Fixes: d0ec49d ("kvm/x86/svm: Support Secure Memory Encryption within KVM")
Link: http://lkml.kernel.org/r/20170825205540.123531-1-brijesh.singh@amd.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
According to the SDM, if the "use TPR shadow" VM-execution control is
1, bits 11:0 of the virtual-APIC address must be 0 and the address
should set any bits beyond the processor's physical-address width.
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The host pkru is restored right after vcpu exit (commit 1be0e61), so
KVM_GET_XSAVE will return the host PKRU value instead. Fix this by
using the guest PKRU explicitly in fill_xsave and load_xsave. This
part is based on a patch by Junkang Fu.
The host PKRU data may also not match the value in vcpu->arch.guest_fpu.state,
because it could have been changed by userspace since the last time
it was saved, so skip loading it in kvm_load_guest_fpu.
Reported-by: Junkang Fu <junkang.fjk@alibaba-inc.com>
Cc: Yang Zhang <zy107165@alibaba-inc.com>
Fixes: 1be0e61c1f
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Move it to struct kvm_arch_vcpu, replacing guest_pkru_valid with a
simple comparison against the host value of the register. The write of
PKRU in addition can be skipped if the guest has not enabled the feature.
Once we do this, we need not test OSPKE in the host anymore, because
guest_CR4.PKE=1 implies host_CR4.PKE=1.
The static PKU test is kept to elide the code on older CPUs.
Suggested-by: Yang Zhang <zy107165@alibaba-inc.com>
Fixes: 1be0e61c1f
Cc: stable@vger.kernel.org
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If the host has protection keys disabled, we cannot read and write the
guest PKRU---RDPKRU and WRPKRU fail with #GP(0) if CR4.PKE=0. Block
the PKU cpuid bit in that case.
This ensures that guest_CR4.PKE=1 implies host_CR4.PKE=1.
Fixes: 1be0e61c1f
Cc: stable@vger.kernel.org
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
------------[ cut here ]------------
WARNING: CPU: 7 PID: 3861 at /home/kernel/ssd/kvm/arch/x86/kvm//vmx.c:11299 nested_vmx_vmexit+0x176e/0x1980 [kvm_intel]
CPU: 7 PID: 3861 Comm: qemu-system-x86 Tainted: G W OE 4.13.0-rc4+ #11
RIP: 0010:nested_vmx_vmexit+0x176e/0x1980 [kvm_intel]
Call Trace:
? kvm_multiple_exception+0x149/0x170 [kvm]
? handle_emulation_failure+0x79/0x230 [kvm]
? load_vmcs12_host_state+0xa80/0xa80 [kvm_intel]
? check_chain_key+0x137/0x1e0
? reexecute_instruction.part.168+0x130/0x130 [kvm]
nested_vmx_inject_exception_vmexit+0xb7/0x100 [kvm_intel]
? nested_vmx_inject_exception_vmexit+0xb7/0x100 [kvm_intel]
vmx_queue_exception+0x197/0x300 [kvm_intel]
kvm_arch_vcpu_ioctl_run+0x1b0c/0x2c90 [kvm]
? kvm_arch_vcpu_runnable+0x220/0x220 [kvm]
? preempt_count_sub+0x18/0xc0
? restart_apic_timer+0x17d/0x300 [kvm]
? kvm_lapic_restart_hv_timer+0x37/0x50 [kvm]
? kvm_arch_vcpu_load+0x1d8/0x350 [kvm]
kvm_vcpu_ioctl+0x4e4/0x910 [kvm]
? kvm_vcpu_ioctl+0x4e4/0x910 [kvm]
? kvm_dev_ioctl+0xbe0/0xbe0 [kvm]
The flag "nested_run_pending", which can override the decision of which should run
next, L1 or L2. nested_run_pending=1 means that we *must* run L2 next, not L1. This
is necessary in particular when L1 did a VMLAUNCH of L2 and therefore expects L2 to
be run (and perhaps be injected with an event it specified, etc.). Nested_run_pending
is especially intended to avoid switching to L1 in the injection decision-point.
This can be handled just like the other cases in vmx_check_nested_events, instead of
having a special case in vmx_queue_exception.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
vmx_complete_interrupts() assumes that the exception is always injected,
so it can be dropped by kvm_clear_exception_queue(). However,
an exception cannot be injected immediately if it is: 1) originally
destined to a nested guest; 2) trapped to cause a vmexit; 3) happening
right after VMLAUNCH/VMRESUME, i.e. when nested_run_pending is true.
This patch applies to exceptions the same algorithm that is used for
NMIs, replacing exception.reinject with "exception.injected" (equivalent
to nmi_injected).
exception.pending now represents an exception that is queued and whose
side effects (e.g., update RFLAGS.RF or DR7) have not been applied yet.
If exception.pending is true, the exception might result in a nested
vmexit instead, too (in which case the side effects must not be applied).
exception.injected instead represents an exception that is going to be
injected into the guest at the next vmentry.
Reported-by: Radim Krčmář <rkrcmar@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Use kvm_event_needs_reinjection() encapsulation.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
update_permission_bitmask currently does a 128-iteration loop to,
essentially, compute a constant array. Computing the 8 bits in parallel
reduces it to 16 iterations, and is enough to speed it up substantially
because many boolean operations in the inner loop become constants or
simplify noticeably.
Because update_permission_bitmask is actually the top item in the profile
for nested vmexits, this speeds up an L2->L1 vmexit by about ten thousand
clock cycles, or up to 30%:
before after
cpuid 35173 25954
vmcall 35122 27079
inl_from_pmtimer 52635 42675
inl_from_qemu 53604 44599
inl_from_kernel 38498 30798
outl_to_kernel 34508 28816
wr_tsc_adjust_msr 34185 26818
rd_tsc_adjust_msr 37409 27049
mmio-no-eventfd:pci-mem 50563 45276
mmio-wildcard-eventfd:pci-mem 34495 30823
mmio-datamatch-eventfd:pci-mem 35612 31071
portio-no-eventfd:pci-io 44925 40661
portio-wildcard-eventfd:pci-io 29708 27269
portio-datamatch-eventfd:pci-io 31135 27164
(I wrote a small C program to compare the tables for all values of CR0.WP,
CR4.SMAP and CR4.SMEP, and they match).
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch exposes 5 level page table feature to the VM.
At the same time, the canonical virtual address checking is
extended to support both 48-bits and 57-bits address width.
Signed-off-by: Yu Zhang <yu.c.zhang@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Extends the shadow paging code, so that 5 level shadow page
table can be constructed if VM is running in 5 level paging
mode.
Also extends the ept code, so that 5 level ept table can be
constructed if maxphysaddr of VM exceeds 48 bits. Unlike the
shadow logic, KVM should still use 4 level ept table for a VM
whose physical address width is less than 48 bits, even when
the VM is running in 5 level paging mode.
Signed-off-by: Yu Zhang <yu.c.zhang@linux.intel.com>
[Unconditionally reset the MMU context in kvm_cpuid_update.
Changing MAXPHYADDR invalidates the reserved bit bitmasks.
- Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Now we have 4 level page table and 5 level page table in 64 bits
long mode, let's rename the PT64_ROOT_LEVEL to PT64_ROOT_4LEVEL,
then we can use PT64_ROOT_5LEVEL for 5 level page table, it's
helpful to make the code more clear.
Also PT64_ROOT_MAX_LEVEL is defined as 4, so that we can just
redefine it to 5 whenever a replacement is needed for 5 level
paging.
Signed-off-by: Yu Zhang <yu.c.zhang@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Currently, KVM uses CR3_L_MODE_RESERVED_BITS to check the
reserved bits in CR3. Yet the length of reserved bits in
guest CR3 should be based on the physical address width
exposed to the VM. This patch changes CR3 check logic to
calculate the reserved bits at runtime.
Signed-off-by: Yu Zhang <yu.c.zhang@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Return false in kvm_cpuid() when it fails to find the cpuid
entry. Also, this routine(and its caller) is optimized with
a new argument - check_limit, so that the check_cpuid_limit()
fall back can be avoided.
Signed-off-by: Yu Zhang <yu.c.zhang@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A guest may not be configured to support XSAVES/XRSTORS, even when the host
does. If the guest does not support XSAVES/XRSTORS, clear the secondary
execution control so that the processor will raise #UD.
Also clear the "allowed-1" bit for XSAVES/XRSTORS exiting in the
IA32_VMX_PROCBASED_CTLS2 MSR, and pass through VMCS12's control in
the VMCS02.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A guest may not be configured to support RDSEED, even when the host
does. If the guest does not support RDSEED, intercept the instruction
and synthesize #UD. Also clear the "allowed-1" bit for RDSEED exiting
in the IA32_VMX_PROCBASED_CTLS2 MSR.
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A guest may not be configured to support RDRAND, even when the host
does. If the guest does not support RDRAND, intercept the instruction
and synthesize #UD. Also clear the "allowed-1" bit for RDRAND exiting
in the IA32_VMX_PROCBASED_CTLS2 MSR.
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Currently, secondary execution controls are divided in three groups:
- static, depending mostly on the module arguments or the processor
(vmx_secondary_exec_control)
- static, depending on CPUID (vmx_cpuid_update)
- dynamic, depending on nested VMX or local APIC state
Because walking CPUID is expensive, prepare_vmcs02 is using only
the first group. This however is unnecessarily complicated. Just
cache the static secondary execution controls, and then prepare_vmcs02
does not need to compute them every time. Computation of all static
secondary execution controls is now kept in a single function,
vmx_compute_secondary_exec_control.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Lguest seems to be rather unused these days. It has seen only patches
ensuring it still builds the last two years and its official state is
"Odd Fixes".
Remove it in order to be able to clean up the paravirt code.
Signed-off-by: Juergen Gross <jgross@suse.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: boris.ostrovsky@oracle.com
Cc: lguest@lists.ozlabs.org
Cc: rusty@rustcorp.com.au
Cc: xen-devel@lists.xenproject.org
Link: http://lkml.kernel.org/r/20170816173157.8633-3-jgross@suse.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Enable the Virtual GIF feature. This is done by setting bit 25 at position
60h in the vmcb.
With this feature enabled, the processor uses bit 9 at position 60h as the
virtual GIF when executing STGI/CLGI instructions.
Since the execution of STGI by the L1 hypervisor does not cause a return to
the outermost (L0) hypervisor, the enable_irq_window and enable_nmi_window
are modified.
The IRQ window will be opened even if GIF is not set, under the assumption
that on resuming the L1 hypervisor the IRQ will be held pending until the
processor executes the STGI instruction.
For the NMI window, the STGI intercept is set. This will assist in opening
the window only when GIF=1.
Signed-off-by: Janakarajan Natarajan <Janakarajan.Natarajan@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We already always set that type but don't check if it is supported. Also
for nVMX, we only support WB for now. Let's just require it.
Signed-off-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Don't use shifts, tag them correctly as EPTP and use better matching
names (PWL vs. GAW).
Signed-off-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
There is currently some confusion between nested and L1 GPAs. The
assignment to "direct" in kvm_mmu_page_fault tries to fix that, but
it is not enough. What this patch does is fence off the MMIO cache
completely when using shadow nested page tables, since we have neither
a GVA nor an L1 GPA to put in the cache. This also allows some
simplifications in kvm_mmu_page_fault and FNAME(page_fault).
The EPT misconfig likewise does not have an L1 GPA to pass to
kvm_io_bus_write, so that must be skipped for guest mode.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
[Changed comment to say "GPAs" instead of "L1's physical addresses", as
per David's review. - Radim]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
When a guest causes a page fault which requires emulation, the
vcpu->arch.gpa_available flag is set to indicate that cr2 contains a
valid GPA.
Currently, emulator_read_write_onepage() makes use of gpa_available flag
to avoid a guest page walk for a known MMIO regions. Lets not limit
the gpa_available optimization to just MMIO region. The patch extends
the check to avoid page walk whenever gpa_available flag is set.
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
[Fix EPT=0 according to Wanpeng Li's fix, plus ensure VMX also uses the
new code. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
[Moved "ret < 0" to the else brach, as per David's review. - Radim]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Calling handle_mmio_page_fault() has been unnecessary since commit
e9ee956e31 ("KVM: x86: MMU: Move handle_mmio_page_fault() call to
kvm_mmu_page_fault()", 2016-02-22).
handle_mmio_page_fault() can now be made static.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Host-initiated writes to the IA32_APIC_BASE MSR do not have to follow
local APIC state transition constraints, but the value written must be
valid.
Signed-off-by: Jim Mattson <jmattson@google.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Bailing out immediately if there is no available mmu page to alloc.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
watchdog: BUG: soft lockup - CPU#5 stuck for 22s! [warn_test:3089]
irq event stamp: 20532
hardirqs last enabled at (20531): [<ffffffff8e9b6908>] restore_regs_and_iret+0x0/0x1d
hardirqs last disabled at (20532): [<ffffffff8e9b7ae8>] apic_timer_interrupt+0x98/0xb0
softirqs last enabled at (8266): [<ffffffff8e9badc6>] __do_softirq+0x206/0x4c1
softirqs last disabled at (8253): [<ffffffff8e083918>] irq_exit+0xf8/0x100
CPU: 5 PID: 3089 Comm: warn_test Tainted: G OE 4.13.0-rc3+ #8
RIP: 0010:kvm_mmu_prepare_zap_page+0x72/0x4b0 [kvm]
Call Trace:
make_mmu_pages_available.isra.120+0x71/0xc0 [kvm]
kvm_mmu_load+0x1cf/0x410 [kvm]
kvm_arch_vcpu_ioctl_run+0x1316/0x1bf0 [kvm]
kvm_vcpu_ioctl+0x340/0x700 [kvm]
? kvm_vcpu_ioctl+0x340/0x700 [kvm]
? __fget+0xfc/0x210
do_vfs_ioctl+0xa4/0x6a0
? __fget+0x11d/0x210
SyS_ioctl+0x79/0x90
entry_SYSCALL_64_fastpath+0x23/0xc2
? __this_cpu_preempt_check+0x13/0x20
This can be reproduced readily by ept=N and running syzkaller tests since
many syzkaller testcases don't setup any memory regions. However, if ept=Y
rmode identity map will be created, then kvm_mmu_calculate_mmu_pages() will
extend the number of VM's mmu pages to at least KVM_MIN_ALLOC_MMU_PAGES
which just hide the issue.
I saw the scenario kvm->arch.n_max_mmu_pages == 0 && kvm->arch.n_used_mmu_pages == 1,
so there is one active mmu page on the list, kvm_mmu_prepare_zap_page() fails
to zap any pages, however prepare_zap_oldest_mmu_page() always returns true.
It incurs infinite loop in make_mmu_pages_available() which causes mmu->lock
softlockup.
This patch fixes it by setting the return value of prepare_zap_oldest_mmu_page()
according to whether or not there is mmu page zapped.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Let's reuse the function introduced with eptp switching.
We don't explicitly have to check against enable_ept_ad_bits, as this
is implicitly done when checking against nested_vmx_ept_caps in
valid_ept_address().
Signed-off-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
"virtual_vmload_vmsave" is what is going to land in /proc/cpuinfo now
as per v4.13-rc4, for a single feature bit which is clearly too long.
So rename it to what it is called in the processor manual.
"v_vmsave_vmload" is a bit shorter, after all.
We could go more aggressively here but having it the same as in the
processor manual is advantageous.
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Radim Krčmář <rkrcmar@redhat.com>
Cc: Janakarajan Natarajan <Janakarajan.Natarajan@amd.com>
Cc: Jörg Rödel <joro@8bytes.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: kvm-ML <kvm@vger.kernel.org>
Link: http://lkml.kernel.org/r/20170801185552.GA3743@nazgul.tnic
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This is the same as commit 147277540b ("kvm: svm: Add support for
additional SVM NPF error codes", 2016-11-23), but for Intel processors.
In this case, the exit qualification field's bit 8 says whether the
EPT violation occurred while translating the guest's final physical
address or rather while translating the guest page tables.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit 147277540b ("kvm: svm: Add support for additional SVM NPF error
codes", 2016-11-23) added a new error code to aid nested page fault
handling. The commit unprotects (kvm_mmu_unprotect_page) the page when
we get a NPF due to guest page table walk where the page was marked RO.
However, if an L0->L2 shadow nested page table can also be marked read-only
when a page is read only in L1's nested page table. If such a page
is accessed by L2 while walking page tables it can cause a nested
page fault (page table walks are write accesses). However, after
kvm_mmu_unprotect_page we may get another page fault, and again in an
endless stream.
To cover this use case, we qualify the new error_code check with
vcpu->arch.mmu_direct_map so that the error_code check would run on L1
guest, and not the L2 guest. This avoids hitting the above scenario.
Fixes: 147277540b
Cc: stable@vger.kernel.org
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Thomas Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reported by syzkaller:
The kvm-intel.unrestricted_guest=0
WARNING: CPU: 5 PID: 1014 at /home/kernel/data/kvm/arch/x86/kvm//x86.c:7227 kvm_arch_vcpu_ioctl_run+0x38b/0x1be0 [kvm]
CPU: 5 PID: 1014 Comm: warn_test Tainted: G W OE 4.13.0-rc3+ #8
RIP: 0010:kvm_arch_vcpu_ioctl_run+0x38b/0x1be0 [kvm]
Call Trace:
? put_pid+0x3a/0x50
? rcu_read_lock_sched_held+0x79/0x80
? kmem_cache_free+0x2f2/0x350
kvm_vcpu_ioctl+0x340/0x700 [kvm]
? kvm_vcpu_ioctl+0x340/0x700 [kvm]
? __fget+0xfc/0x210
do_vfs_ioctl+0xa4/0x6a0
? __fget+0x11d/0x210
SyS_ioctl+0x79/0x90
entry_SYSCALL_64_fastpath+0x23/0xc2
? __this_cpu_preempt_check+0x13/0x20
The syszkaller folks reported a residual mmio emulation request to userspace
due to vm86 fails to emulate inject real mode interrupt(fails to read CS) and
incurs a triple fault. The vCPU returns to userspace with vcpu->mmio_needed == true
and KVM_EXIT_SHUTDOWN exit reason. However, the syszkaller testcase constructs
several threads to launch the same vCPU, the thread which lauch this vCPU after
the thread whichs get the vcpu->mmio_needed == true and KVM_EXIT_SHUTDOWN will
trigger the warning.
#define _GNU_SOURCE
#include <pthread.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/wait.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <sys/mman.h>
#include <fcntl.h>
#include <unistd.h>
#include <linux/kvm.h>
#include <stdio.h>
int kvmcpu;
struct kvm_run *run;
void* thr(void* arg)
{
int res;
res = ioctl(kvmcpu, KVM_RUN, 0);
printf("ret1=%d exit_reason=%d suberror=%d\n",
res, run->exit_reason, run->internal.suberror);
return 0;
}
void test()
{
int i, kvm, kvmvm;
pthread_t th[4];
kvm = open("/dev/kvm", O_RDWR);
kvmvm = ioctl(kvm, KVM_CREATE_VM, 0);
kvmcpu = ioctl(kvmvm, KVM_CREATE_VCPU, 0);
run = (struct kvm_run*)mmap(0, 4096, PROT_READ|PROT_WRITE, MAP_SHARED, kvmcpu, 0);
srand(getpid());
for (i = 0; i < 4; i++) {
pthread_create(&th[i], 0, thr, 0);
usleep(rand() % 10000);
}
for (i = 0; i < 4; i++)
pthread_join(th[i], 0);
}
int main()
{
for (;;) {
int pid = fork();
if (pid < 0)
exit(1);
if (pid == 0) {
test();
exit(0);
}
int status;
while (waitpid(pid, &status, __WALL) != pid) {}
}
return 0;
}
This patch fixes it by resetting the vcpu->mmio_needed once we receive
the triple fault to avoid the residue.
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Tested-by: Dmitry Vyukov <dvyukov@google.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
get_cpl requires vcpu_load, so we must cache the result (whether the
vcpu was preempted when its cpl=0) in kvm_vcpu_arch.
Signed-off-by: Longpeng(Mike) <longpeng2@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If a vcpu exits due to request a user mode spinlock, then
the spinlock-holder may be preempted in user mode or kernel mode.
(Note that not all architectures trap spin loops in user mode,
only AMD x86 and ARM/ARM64 currently do).
But if a vcpu exits in kernel mode, then the holder must be
preempted in kernel mode, so we should choose a vcpu in kernel mode
as a more likely candidate for the lock holder.
This introduces kvm_arch_vcpu_in_kernel() to decide whether the
vcpu is in kernel-mode when it's preempted. kvm_vcpu_on_spin's
new argument says the same of the spinning VCPU.
Signed-off-by: Longpeng(Mike) <longpeng2@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add guest_cpuid_clear() and use it instead of kvm_find_cpuid_entry().
Also replace some uses of kvm_find_cpuid_entry() with guest_cpuid_has().
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch turns guest_cpuid_has_XYZ(cpuid) into guest_cpuid_has(cpuid,
X86_FEATURE_XYZ), which gets rid of many very similar helpers.
When seeing a X86_FEATURE_*, we can know which cpuid it belongs to, but
this information isn't in common code, so we recreate it for KVM.
Add some BUILD_BUG_ONs to make sure that it runs nicely.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
bit(X86_FEATURE_NRIPS) is 3 since 2ccd71f1b2 ("x86/cpufeature: Move
some of the scattered feature bits to x86_capability"), so we can
simplify the code.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When L2 uses vmfunc, L0 utilizes the associated vmexit to
emulate a switching of the ept pointer by reloading the
guest MMU.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Bandan Das <bsd@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Expose VMFUNC in MSRs and VMCS fields. No actual VMFUNCs are enabled.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Bandan Das <bsd@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Enable VMFUNC in the secondary execution controls. This simplifies the
changes necessary to expose it to nested hypervisors. VMFUNCs still
cause #UD when invoked.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Bandan Das <bsd@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Let's also just use the underlying functions directly here.
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
[Rebased on top of 9f744c5974 ("KVM: nVMX: do not pin the VMCS12")]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
nested_get_page() just sounds confusing. All we want is a page from G1.
This is even unrelated to nested.
Let's introduce kvm_vcpu_gpa_to_page() so we don't get too lengthy
lines.
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
[Squash pasto fix from Wanpeng Li. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Expose the "Enable INVPCID" secondary execution control to the guest
and properly reflect the exit reason.
In addition, before this patch the guest was always running with
INVPCID enabled, causing pcid.flat's "Test on INVPCID when disabled"
test to fail.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It has been experimentally confirmed that supporting these two MSRs is one
of the necessary conditions for nested Hyper-V to use the TSC page. Modern
Windows guests are noticeably slower when they fall back to reading
timestamps from the HV_X64_MSR_TIME_REF_COUNT MSR instead of using the TSC
page.
The newly supported MSRs are advertised with the AccessFrequencyRegs
partition privilege flag and CPUID.40000003H:EDX[8] "Support for
determining timer frequencies is available" (both outside of the scope of
this KVM patch).
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
------------[ cut here ]------------
WARNING: CPU: 5 PID: 2288 at arch/x86/kvm/vmx.c:11124 nested_vmx_vmexit+0xd64/0xd70 [kvm_intel]
CPU: 5 PID: 2288 Comm: qemu-system-x86 Not tainted 4.13.0-rc2+ #7
RIP: 0010:nested_vmx_vmexit+0xd64/0xd70 [kvm_intel]
Call Trace:
vmx_check_nested_events+0x131/0x1f0 [kvm_intel]
? vmx_check_nested_events+0x131/0x1f0 [kvm_intel]
kvm_arch_vcpu_ioctl_run+0x5dd/0x1be0 [kvm]
? vmx_vcpu_load+0x1be/0x220 [kvm_intel]
? kvm_arch_vcpu_load+0x62/0x230 [kvm]
kvm_vcpu_ioctl+0x340/0x700 [kvm]
? kvm_vcpu_ioctl+0x340/0x700 [kvm]
? __fget+0xfc/0x210
do_vfs_ioctl+0xa4/0x6a0
? __fget+0x11d/0x210
SyS_ioctl+0x79/0x90
do_syscall_64+0x8f/0x750
? trace_hardirqs_on_thunk+0x1a/0x1c
entry_SYSCALL64_slow_path+0x25/0x25
This can be reproduced by booting L1 guest w/ 'noapic' grub parameter, which
means that tells the kernel to not make use of any IOAPICs that may be present
in the system.
Actually external_intr variable in nested_vmx_vmexit() is the req_int_win
variable passed from vcpu_enter_guest() which means that the L0's userspace
requests an irq window. I observed the scenario (!kvm_cpu_has_interrupt(vcpu) &&
L0's userspace reqeusts an irq window) is true, so there is no interrupt which
L1 requires to inject to L2, we should not attempt to emualte "Acknowledge
interrupt on exit" for the irq window requirement in this scenario.
This patch fixes it by not attempt to emulate "Acknowledge interrupt on exit"
if there is no L1 requirement to inject an interrupt to L2.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
[Added code comment to make it obvious that the behavior is not correct.
We should do a userspace exit with open interrupt window instead of the
nested VM exit. This patch still improves the behavior, so it was
accepted as a (temporary) workaround.]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
The host physical addresses of L1's Virtual APIC Page and Posted
Interrupt descriptor are loaded into the VMCS02. The CPU may write
to these pages via their host physical address while L2 is running,
bypassing address-translation-based dirty tracking (e.g. EPT write
protection). Mark them dirty on every exit from L2 to prevent them
from getting out of sync with dirty tracking.
Also mark the virtual APIC page and the posted interrupt descriptor
dirty when KVM is virtualizing posted interrupt processing.
Signed-off-by: David Matlack <dmatlack@google.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
According to the Intel SDM, software cannot rely on the current VMCS to be
coherent after a VMXOFF or shutdown. So this is a valid way to handle VMCS12
flushes.
24.11.1 Software Use of Virtual-Machine Control Structures
...
If a logical processor leaves VMX operation, any VMCSs active on
that logical processor may be corrupted (see below). To prevent
such corruption of a VMCS that may be used either after a return
to VMX operation or on another logical processor, software should
execute VMCLEAR for that VMCS before executing the VMXOFF instruction
or removing power from the processor (e.g., as part of a transition
to the S3 and S4 power states).
...
This fixes a "suspicious rcu_dereference_check() usage!" warning during
kvm_vm_release() because nested_release_vmcs12() calls
kvm_vcpu_write_guest_page() without holding kvm->srcu.
Signed-off-by: David Matlack <dmatlack@google.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Since the current implementation of VMCS12 does a memcpy in and out
of guest memory, we do not need current_vmcs12 and current_vmcs12_page
anymore. current_vmptr is enough to read and write the VMCS12.
And David Matlack noted:
This patch also fixes dirty tracking (memslot->dirty_bitmap) of the
VMCS12 page by using kvm_write_guest. nested_release_page() only marks
the struct page dirty.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
[Added David Matlack's note and nested_release_page_clean() fix.]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
'lapic_irq' is a local variable and its 'level' field isn't
initialized, so 'level' is random, it doesn't matter but
makes UBSAN unhappy:
UBSAN: Undefined behaviour in .../lapic.c:...
load of value 10 is not a valid value for type '_Bool'
...
Call Trace:
[<ffffffff81f030b6>] dump_stack+0x1e/0x20
[<ffffffff81f03173>] ubsan_epilogue+0x12/0x55
[<ffffffff81f03b96>] __ubsan_handle_load_invalid_value+0x118/0x162
[<ffffffffa1575173>] kvm_apic_set_irq+0xc3/0xf0 [kvm]
[<ffffffffa1575b20>] kvm_irq_delivery_to_apic_fast+0x450/0x910 [kvm]
[<ffffffffa15858ea>] kvm_irq_delivery_to_apic+0xfa/0x7a0 [kvm]
[<ffffffffa1517f4e>] kvm_emulate_hypercall+0x62e/0x760 [kvm]
[<ffffffffa113141a>] handle_vmcall+0x1a/0x30 [kvm_intel]
[<ffffffffa114e592>] vmx_handle_exit+0x7a2/0x1fa0 [kvm_intel]
...
Signed-off-by: Longpeng(Mike) <longpeng2@huawei.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
When SMP VM start, AP may lost INIT because of receiving INIT between
kvm_vcpu_ioctl_x86_get/set_vcpu_events.
vcpu 0 vcpu 1
kvm_vcpu_ioctl_x86_get_vcpu_events
events->smi.latched_init = 0
send INIT to vcpu1
set vcpu1's pending_events
kvm_vcpu_ioctl_x86_set_vcpu_events
if (events->smi.latched_init == 0)
clear INIT in pending_events
This patch fixes it by just update SMM related flags if we are in SMM.
Thanks Peng Hao for the report and original commit message.
Reported-by: Peng Hao <peng.hao2@zte.com.cn>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
There are three issues in nested_vmx_check_exception:
1) it is not taking PFEC_MATCH/PFEC_MASK into account, as reported
by Wanpeng Li;
2) it should rebuild the interruption info and exit qualification fields
from scratch, as reported by Jim Mattson, because the values from the
L2->L0 vmexit may be invalid (e.g. if an emulated instruction causes
a page fault, the EPT misconfig's exit qualification is incorrect).
3) CR2 and DR6 should not be written for exception intercept vmexits
(CR2 only for AMD).
This patch fixes the first two and adds a comment about the last,
outlining the fix.
Cc: Jim Mattson <jmattson@google.com>
Cc: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Do this in the caller of nested_vmx_vmexit instead.
nested_vmx_check_exception was doing a vmwrite to the vmcs02's
VM_EXIT_INTR_ERROR_CODE field, so that prepare_vmcs12 would move
the field to vmcs12->vm_exit_intr_error_code. However that isn't
possible on pre-Haswell machines. Moving the vmcs12 write to the
callers fixes it.
Reported-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[Changed nested_vmx_reflect_vmexit() return type to (int)1 from (bool)1,
thanks to fengguang.wu@intel.com]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Run kvm-unit-tests/eventinj.flat in L1 w/ ept=0 on both L0 and L1:
Before NMI IRET test
Sending NMI to self
NMI isr running stack 0x461000
Sending nested NMI to self
After nested NMI to self
Nested NMI isr running rip=40038e
After iret
After NMI to self
FAIL: NMI
Commit 4c4a6f790e (KVM: nVMX: track NMI blocking state separately
for each VMCS) tracks NMI blocking state separately for vmcs01 and
vmcs02. However it is not enough:
- The L2 (kvm-unit-tests/eventinj.flat) generates NMI that will fault
on IRET, so the L2 can generate #PF which can be intercepted by L0.
- L0 walks L1's guest page table and sees the mapping is invalid, it
resumes the L1 guest and injects the #PF into L1. At this point the
vmcs02 has nmi_known_unmasked=true.
- L1 sets set bit 3 (blocking by NMI) in the interruptibility-state field
of vmcs12 (and fixes the shadow page table) before resuming L2 guest.
- L1 executes VMRESUME to resume L2, causing a vmexit to L0
- during VMRESUME emulation, prepare_vmcs02 sets bit 3 in the
interruptibility-state field of vmcs02, but nmi_known_unmasked is
still true.
- L2 immediately exits to L0 with another page fault, because L0 still has
not updated the NGVA->HPA page tables. However, nmi_known_unmasked is
true so vmx_recover_nmi_blocking does not do anything.
The fix is to update nmi_known_unmasked when preparing vmcs02 from vmcs12.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The PI vector for L0 and L1 must be different. If dest vcpu0
is in guest mode while vcpu1 is delivering a non-nested PI to
vcpu0, there wont't be any vmexit so that the non-nested interrupt
will be delayed.
Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This reverts the change of commit f85c758dbe,
as the behavior it modified was intended.
The VM is running in 32-bit PAE mode, and Table 4-7 of the Intel manual
says:
Table 4-7. Use of CR3 with PAE Paging
Bit Position(s) Contents
4:0 Ignored
31:5 Physical address of the 32-Byte aligned
page-directory-pointer table used for linear-address
translation
63:32 Ignored (these bits exist only on processors supporting
the Intel-64 architecture)
To placate the static checker, write the mask explicitly as an
unsigned long constant instead of using a 32-bit unsigned constant.
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Fixes: f85c758dbe
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If the SynIC timer message delivery fails due to SINT message slot being
busy, there's no point to attempt starting the timer again until we're
notified of the slot being released by the guest (via EOM or EOI).
Even worse, when a oneshot timer fails to deliver its message, its
re-arming with an expiration time in the past leads to immediate retry
of the delivery, and so on, without ever letting the guest vcpu to run
and release the slot, which results in a livelock.
To avoid that, only start the timer when there's no timer message
pending delivery. When there is, meaning the slot is busy, the
processing will be restarted upon notification from the guest that the
slot is released.
Signed-off-by: Roman Kagan <rkagan@virtuozzo.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
This can be reproduced by EPT=1, unrestricted_guest=N, emulate_invalid_state=Y
or EPT=0, the trace of kvm-unit-tests/taskswitch2.flat is like below, it tries
to emulate invalid guest state task-switch:
kvm_exit: reason TASK_SWITCH rip 0x0 info 40000058 0
kvm_emulate_insn: 42000:0:0f 0b (0x2)
kvm_emulate_insn: 42000:0:0f 0b (0x2) failed
kvm_inj_exception: #UD (0x0)
kvm_entry: vcpu 0
kvm_exit: reason TASK_SWITCH rip 0x0 info 40000058 0
kvm_emulate_insn: 42000:0:0f 0b (0x2)
kvm_emulate_insn: 42000:0:0f 0b (0x2) failed
kvm_inj_exception: #UD (0x0)
......................
It appears that the task-switch emulation updates rflags (and vm86
flag) only after the segments are loaded, causing vmx->emulation_required
to be set, when in fact invalid guest state emulation is not needed.
This patch fixes it by updating vmx->emulation_required after the
rflags (and vm86 flag) is updated in task-switch emulation.
Thanks Radim for moving the update to vmx__set_flags and adding Paolo's
suggestion for the check.
Suggested-by: Nadav Amit <nadav.amit@gmail.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Nadav Amit <nadav.amit@gmail.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
KVM tries to select 'TASKSTATS', which had additional dependencies:
warning: (KVM) selects TASKSTATS which has unmet direct dependencies (NET && MULTIUSER)
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Immediately following MOV-to-SS/POP-to-SS, VM-entry is
disallowed. This check comes after the check for a valid VMCS. When
this check fails, the instruction pointer should fall through to the
next instruction, the ALU flags should be set to indicate VMfailValid,
and the VM-instruction error should be set to 26 ("VM entry with
events blocked by MOV SS").
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
vmx_recover_nmi_blocking is using a cached value of the guest
interruptibility info, which is stored in vmx->nmi_known_unmasked.
vmx_recover_nmi_blocking is run for both normal and nested guests,
so the cached value must be per-VMCS.
This fixes eventinj.flat in a nested non-EPT environment. With EPT it
works, because the EPT violation handler doesn't have the
vmx->nmi_known_unmasked optimization (it is unnecessary because, unlike
vmx_recover_nmi_blocking, it can just look at the exit qualification).
Thanks to Wanpeng Li for debugging the testcase and providing an initial
patch.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
kvm_read_cr3() returns an unsigned long and gfn is a u64. We intended
to mask out the bottom 5 bits but because of the type issue we mask the
top 32 bits as well. I don't know if this is a real problem, but it
causes static checker warnings.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Update the KVM support to work with SME. The VMCB has a number of fields
where physical addresses are used and these addresses must contain the
memory encryption mask in order to properly access the encrypted memory.
Also, use the memory encryption mask when creating and using the nested
page tables.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Borislav Petkov <bp@suse.de>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brijesh Singh <brijesh.singh@amd.com>
Cc: Dave Young <dyoung@redhat.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Larry Woodman <lwoodman@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Toshimitsu Kani <toshi.kani@hpe.com>
Cc: kasan-dev@googlegroups.com
Cc: kvm@vger.kernel.org
Cc: linux-arch@vger.kernel.org
Cc: linux-doc@vger.kernel.org
Cc: linux-efi@vger.kernel.org
Cc: linux-mm@kvack.org
Link: http://lkml.kernel.org/r/89146eccfa50334409801ff20acd52a90fb5efcf.1500319216.git.thomas.lendacky@amd.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Hyper-V identifies vCPUs by Virtual Processor Index, which can be
queried via HV_X64_MSR_VP_INDEX msr. It is defined by the spec as a
sequential number which can't exceed the maximum number of vCPUs per VM.
APIC ids can be sparse and thus aren't a valid replacement for VP
indices.
Current KVM uses its internal vcpu index as VP_INDEX. However, to make
it predictable and persistent across VM migrations, the userspace has to
control the value of VP_INDEX.
This patch achieves that, by storing vp_index explicitly on vcpu, and
allowing HV_X64_MSR_VP_INDEX to be set from the host side. For
compatibility it's initialized to KVM vcpu index. Also a few variables
are renamed to make clear distinction betweed this Hyper-V vp_index and
KVM vcpu_id (== APIC id). Besides, a new capability,
KVM_CAP_HYPERV_VP_INDEX, is added to allow the userspace to skip
attempting msr writes where unsupported, to avoid spamming error logs.
Signed-off-by: Roman Kagan <rkagan@virtuozzo.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Adds another flag bit (bit 2) to MSR_KVM_ASYNC_PF_EN. If bit 2 is 1,
async page faults are delivered to L1 as #PF vmexits; if bit 2 is 0,
kvm_can_do_async_pf returns 0 if in guest mode.
This is similar to what svm.c wanted to do all along, but it is only
enabled for Linux as L1 hypervisor. Foreign hypervisors must never
receive async page faults as vmexits, because they'd probably be very
confused about that.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Add an nested_apf field to vcpu->arch.exception to identify an async page
fault, and constructs the expected vm-exit information fields. Force a
nested VM exit from nested_vmx_check_exception() if the injected #PF is
async page fault.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
This patch adds the L1 guest async page fault #PF vmexit handler, such
by L1 similar to ordinary async page fault.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
[Passed insn parameters to kvm_mmu_page_fault().]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
This patch removes all arguments except the first in
kvm_x86_ops->queue_exception since they can extract the arguments from
vcpu->arch.exception themselves.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
There is a flaw in the Hyper-V SynIC implementation in KVM: when message
page or event flags page is enabled by setting the corresponding msr,
KVM zeroes it out. This is problematic because on migration the
corresponding MSRs are loaded on the destination, so the content of
those pages is lost.
This went unnoticed so far because the only user of those pages was
in-KVM hyperv synic timers, which could continue working despite that
zeroing.
Newer QEMU uses those pages for Hyper-V VMBus implementation, and
zeroing them breaks the migration.
Besides, in newer QEMU the content of those pages is fully managed by
QEMU, so zeroing them is undesirable even when writing the MSRs from the
guest side.
To support this new scheme, introduce a new capability,
KVM_CAP_HYPERV_SYNIC2, which, when enabled, makes sure that the synic
pages aren't zeroed out in KVM.
Signed-off-by: Roman Kagan <rkagan@virtuozzo.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
The backwards_tsc_observed global introduced in commit 16a9602 is never
reset to false. If a VM happens to be running while the host is suspended
(a common source of the TSC jumping backwards), master clock will never
be enabled again for any VM. In contrast, if no VM is running while the
host is suspended, master clock is unaffected. This is inconsistent and
unnecessarily strict. Let's track the backwards_tsc_observed variable
separately and let each VM start with a clean slate.
Real world impact: My Windows VMs get slower after my laptop undergoes a
suspend/resume cycle. The only way to get the perf back is unloading and
reloading the kvm module.
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Enable the Virtual VMLOAD VMSAVE feature. This is done by setting bit 1
at position B8h in the vmcb.
The processor must have nested paging enabled, be in 64-bit mode and
have support for the Virtual VMLOAD VMSAVE feature for the bit to be set
in the vmcb.
Signed-off-by: Janakarajan Natarajan <Janakarajan.Natarajan@amd.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Rename the lbr_ctl variable to better reflect the purpose of the field -
provide support for virtualization extensions.
Signed-off-by: Janakarajan Natarajan <Janakarajan.Natarajan@amd.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
The lbr_ctl variable in the vmcb control area is used to enable or
disable Last Branch Record (LBR) virtualization. However, this is to be
done using only bit 0 of the variable. To correct this and to prepare
for a new feature, change the current usage to work only on a particular
bit.
Signed-off-by: Janakarajan Natarajan <Janakarajan.Natarajan@amd.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
kvm_skip_emulated_instruction handles the singlestep debug exception
which is something we almost always want. This commit (specifically
the change in rdmsr_interception) makes the debug.flat KVM unit test
pass on AMD.
Two call sites still call skip_emulated_instruction directly:
* In svm_queue_exception where it's used only for moving the rip forward
* In task_switch_interception which is analogous to handle_task_switch
in VMX
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
kvm_vm_release() did not have slots_lock when calling
kvm_io_bus_unregister_dev() and this went unnoticed until 4a12f95177
("KVM: mark kvm->busses as rcu protected") added dynamic checks.
Luckily, there should be no race at that point:
=============================
WARNING: suspicious RCU usage
4.12.0.kvm+ #0 Not tainted
-----------------------------
./include/linux/kvm_host.h:479 suspicious rcu_dereference_check() usage!
lockdep_rcu_suspicious+0xc5/0x100
kvm_io_bus_unregister_dev+0x173/0x190 [kvm]
kvm_free_pit+0x28/0x80 [kvm]
kvm_arch_sync_events+0x2d/0x30 [kvm]
kvm_put_kvm+0xa7/0x2a0 [kvm]
kvm_vm_release+0x21/0x30 [kvm]
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
vmx_complete_atomic_exit should call kvm_machine_check for any
VM-entry failure due to a machine-check event. Such an exit should be
recognized solely by its basic exit reason (i.e. the low 16 bits of
the VMCS exit reason field). None of the other VMCS exit information
fields contain valid information when the VM-exit is due to "VM-entry
failure due to machine-check event".
Signed-off-by: Jim Mattson <jmattson@google.com>
Reviewed-by: Xiao Guangrong <xiaoguangrong@tencent.com>
[Changed VM_EXIT_INTR_INFO condition to better describe its reason.]
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
kvm master clock usually has a different frequency than the kernel boot
clock. This is not a problem until the master clock is updated;
update uses the current kernel boot clock to compute new kvm clock,
which erases any kvm clock cycles that might have built up due to
frequency difference over a long period.
KVM_SET_CLOCK is one of places where we can safely update master clock
as the guest-visible clock is going to be shifted anyway.
The problem with current code is that it updates the kvm master clock
after updating the offset. If the master clock was enabled before
calling KVM_SET_CLOCK, then it might have built up a significant delta
from kernel boot clock.
In the worst case, the time set by userspace would be shifted by so much
that it couldn't have been set at any point during KVM_SET_CLOCK.
To fix this, move kvm_gen_update_masterclock() before computing
kvmclock_offset, which means that the master clock and kernel boot clock
will be sufficiently close together.
Another solution would be to replace get_kvmclock_ns() with
"ktime_get_boot_ns() + ka->kvmclock_offset", which is marginally more
accurate, but would break symmetry with KVM_GET_CLOCK.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Inconsistencies result from shadowing only accesses to the full
64-bits of a 64-bit VMCS field, but not shadowing accesses to the high
32-bits of the field. The "high" part of a 64-bit field should be
shadowed whenever the full 64-bit field is shadowed.
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Allow the L1 guest to specify the last page of addressable guest
physical memory for an L2 MSR permission bitmap. Also remove the
vmcs12_read_any() check that should never fail.
Fixes: 3af18d9c5f ("KVM: nVMX: Prepare for using hardware MSR bitmap")
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
According to the SDM, if the "use I/O bitmaps" VM-execution control is
1, bits 11:0 of each I/O-bitmap address must be 0. Neither address
should set any bits beyond the processor's physical-address width.
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The VMCS launch state is not set to "launched" unless the VMLAUNCH
actually succeeds. VMLAUNCH failure includes VM-exits with bit 31 set.
Note that this change does not address the general problem that a
failure to launch/resume vmcs02 (i.e. vmx->fail) is not handled
correctly.
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This exit ended up being reported, but the currently exposed data does not provide
much of a starting point for debugging. In the reported case, the vmexit was
an EPT misconfiguration (MMIO access). Let userspace report ethe exit qualification
and, if relevant, the GPA.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
- Better machine check handling for HV KVM
- Ability to support guests with threads=2, 4 or 8 on POWER9
- Fix for a race that could cause delayed recognition of signals
- Fix for a bug where POWER9 guests could sleep with interrupts pending.
ARM:
- VCPU request overhaul
- allow timer and PMU to have their interrupt number selected from userspace
- workaround for Cavium erratum 30115
- handling of memory poisonning
- the usual crop of fixes and cleanups
s390:
- initial machine check forwarding
- migration support for the CMMA page hinting information
- cleanups and fixes
x86:
- nested VMX bugfixes and improvements
- more reliable NMI window detection on AMD
- APIC timer optimizations
Generic:
- VCPU request overhaul + documentation of common code patterns
- kvm_stat improvements
There is a small conflict in arch/s390 due to an arch-wide field rename.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)
iQEcBAABAgAGBQJZW4XTAAoJEL/70l94x66DkhMH/izpk54KI17PtyQ9VYI2sYeZ
BWK6Kl886g3ij4pFi3pECqjDJzWaa3ai+vFfzzpJJ8OkCJT5Rv4LxC5ERltVVmR8
A3T1I/MRktSC0VJLv34daPC2z4Lco/6SPipUpPnL4bE2HATKed4vzoOjQ3tOeGTy
dwi7TFjKwoVDiM7kPPDRnTHqCe5G5n13sZ49dBe9WeJ7ttJauWqoxhlYosCGNPEj
g8ZX8+cvcAhVnz5uFL8roqZ8ygNEQq2mgkU18W8ZZKuiuwR0gdsG0gSBFNTdwIMK
NoreRKMrw0+oLXTIB8SZsoieU6Qi7w3xMAMabe8AJsvYtoersugbOmdxGCr1lsA=
=OD7H
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM updates from Paolo Bonzini:
"PPC:
- Better machine check handling for HV KVM
- Ability to support guests with threads=2, 4 or 8 on POWER9
- Fix for a race that could cause delayed recognition of signals
- Fix for a bug where POWER9 guests could sleep with interrupts pending.
ARM:
- VCPU request overhaul
- allow timer and PMU to have their interrupt number selected from userspace
- workaround for Cavium erratum 30115
- handling of memory poisonning
- the usual crop of fixes and cleanups
s390:
- initial machine check forwarding
- migration support for the CMMA page hinting information
- cleanups and fixes
x86:
- nested VMX bugfixes and improvements
- more reliable NMI window detection on AMD
- APIC timer optimizations
Generic:
- VCPU request overhaul + documentation of common code patterns
- kvm_stat improvements"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (124 commits)
Update my email address
kvm: vmx: allow host to access guest MSR_IA32_BNDCFGS
x86: kvm: mmu: use ept a/d in vmcs02 iff used in vmcs12
kvm: x86: mmu: allow A/D bits to be disabled in an mmu
x86: kvm: mmu: make spte mmio mask more explicit
x86: kvm: mmu: dead code thanks to access tracking
KVM: PPC: Book3S: Fix typo in XICS-on-XIVE state saving code
KVM: PPC: Book3S HV: Close race with testing for signals on guest entry
KVM: PPC: Book3S HV: Simplify dynamic micro-threading code
KVM: x86: remove ignored type attribute
KVM: LAPIC: Fix lapic timer injection delay
KVM: lapic: reorganize restart_apic_timer
KVM: lapic: reorganize start_hv_timer
kvm: nVMX: Check memory operand to INVVPID
KVM: s390: Inject machine check into the nested guest
KVM: s390: Inject machine check into the guest
tools/kvm_stat: add new interactive command 'b'
tools/kvm_stat: add new command line switch '-i'
tools/kvm_stat: fix error on interactive command 'g'
KVM: SVM: suppress unnecessary NMI singlestep on GIF=0 and nested exit
...
It's easier for host applications, such as QEMU, if they can always
access guest MSR_IA32_BNDCFGS in VMCS, even though MPX is disabled in
guest cpuid.
Cc: stable@vger.kernel.org
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Pull x86 mm updates from Ingo Molnar:
"The main changes in this cycle were:
- Continued work to add support for 5-level paging provided by future
Intel CPUs. In particular we switch the x86 GUP code to the generic
implementation. (Kirill A. Shutemov)
- Continued work to add PCID CPU support to native kernels as well.
In this round most of the focus is on reworking/refreshing the TLB
flush infrastructure for the upcoming PCID changes. (Andy
Lutomirski)"
* 'x86-mm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (34 commits)
x86/mm: Delete a big outdated comment about TLB flushing
x86/mm: Don't reenter flush_tlb_func_common()
x86/KASLR: Fix detection 32/64 bit bootloaders for 5-level paging
x86/ftrace: Exclude functions in head64.c from function-tracing
x86/mmap, ASLR: Do not treat unlimited-stack tasks as legacy mmap
x86/mm: Remove reset_lazy_tlbstate()
x86/ldt: Simplify the LDT switching logic
x86/boot/64: Put __startup_64() into .head.text
x86/mm: Add support for 5-level paging for KASLR
x86/mm: Make kernel_physical_mapping_init() support 5-level paging
x86/mm: Add sync_global_pgds() for configuration with 5-level paging
x86/boot/64: Add support of additional page table level during early boot
x86/boot/64: Rename init_level4_pgt and early_level4_pgt
x86/boot/64: Rewrite startup_64() in C
x86/boot/compressed: Enable 5-level paging during decompression stage
x86/boot/efi: Define __KERNEL32_CS GDT on 64-bit configurations
x86/boot/efi: Fix __KERNEL_CS definition of GDT entry on 64-bit configurations
x86/boot/efi: Cleanup initialization of GDT entries
x86/asm: Fix comment in return_from_SYSCALL_64()
x86/mm/gup: Switch GUP to the generic get_user_page_fast() implementation
...
EPT A/D was enabled in the vmcs02 EPTP regardless of the vmcs12's EPTP
value. The problem is that enabling A/D changes the behavior of L2's
x86 page table walks as seen by L1. With A/D enabled, x86 page table
walks are always treated as EPT writes.
Commit ae1e2d1082 ("kvm: nVMX: support EPT accessed/dirty bits",
2017-03-30) tried to work around this problem by clearing the write
bit in the exit qualification for EPT violations triggered by page
walks. However, that fixup introduced the opposite bug: page-table walks
that actually set x86 A/D bits were *missing* the write bit in the exit
qualification.
This patch fixes the problem by disabling EPT A/D in the shadow MMU
when EPT A/D is disabled in vmcs12's EPTP.
Signed-off-by: Peter Feiner <pfeiner@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Adds the plumbing to disable A/D bits in the MMU based on a new role
bit, ad_disabled. When A/D is disabled, the MMU operates as though A/D
aren't available (i.e., using access tracking faults instead).
To avoid SP -> kvm_mmu_page.role.ad_disabled lookups all over the
place, A/D disablement is now stored in the SPTE. This state is stored
in the SPTE by tweaking the use of SPTE_SPECIAL_MASK for access
tracking. Rather than just setting SPTE_SPECIAL_MASK when an
access-tracking SPTE is non-present, we now always set
SPTE_SPECIAL_MASK for access-tracking SPTEs.
Signed-off-by: Peter Feiner <pfeiner@google.com>
[Use role.ad_disabled even for direct (non-shadow) EPT page tables. Add
documentation and a few MMU_WARN_ONs. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Specify both a mask (i.e., bits to consider) and a value (i.e.,
pattern of bits that indicates a special PTE) for mmio SPTEs. On
Intel, this lets us pack even more information into the
(SPTE_SPECIAL_MASK | EPT_VMX_RWX_MASK) mask we use for access
tracking liberating all (SPTE_SPECIAL_MASK | (non-misconfigured-RWX))
values.
Signed-off-by: Peter Feiner <pfeiner@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The MMU always has hardware A bits or access tracking support, thus
it's unnecessary to handle the scenario where we have neither.
Signed-off-by: Peter Feiner <pfeiner@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The macro insn_fetch marks the 'type' argument as having a specified
alignment. Type attributes can only be applied to structs, unions, or
enums, but insn_fetch is only ever invoked with integral types, so Clang
produces 19 -Wignored-attributes warnings for this source file.
Signed-off-by: Nick Desaulniers <nick.desaulniers@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
- vcpu request overhaul
- allow timer and PMU to have their interrupt number
selected from userspace
- workaround for Cavium erratum 30115
- handling of memory poisonning
- the usual crop of fixes and cleanups
-----BEGIN PGP SIGNATURE-----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=6hNK
-----END PGP SIGNATURE-----
Merge tag 'kvmarm-for-4.13' of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD
KVM/ARM updates for 4.13
- vcpu request overhaul
- allow timer and PMU to have their interrupt number
selected from userspace
- workaround for Cavium erratum 30115
- handling of memory poisonning
- the usual crop of fixes and cleanups
Conflicts:
arch/s390/include/asm/kvm_host.h
In preparation for an objtool rewrite which will have broader checks,
whitelist functions and files which cause problems because they do
unusual things with the stack.
These whitelists serve as a TODO list for which functions and files
don't yet have undwarf unwinder coverage. Eventually most of the
whitelists can be removed in favor of manual CFI hint annotations or
objtool improvements.
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: live-patching@vger.kernel.org
Link: http://lkml.kernel.org/r/7f934a5d707a574bda33ea282e9478e627fb1829.1498659915.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
If the TSC deadline timer is programmed really close to the deadline or
even in the past, the computation in vmx_set_hv_timer will program the
absolute target tsc value to vmcs preemption timer field w/ delta == 0,
then plays a vmentry and an upcoming vmx preemption timer fire vmexit
dance, the lapic timer injection is delayed due to this duration. Actually
the lapic timer which is emulated by hrtimer can handle this correctly.
This patch fixes it by firing the lapic timer and injecting a timer interrupt
immediately during the next vmentry if the TSC deadline timer is programmed
really close to the deadline or even in the past. This saves ~300 cycles on
the tsc_deadline_timer test of apic.flat.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Move the code to cancel the hv timer into the caller, just before
it starts the hrtimer. Check availability of the hv timer in
start_hv_timer.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
There are many cases in which the hv timer must be canceled. Split out
a new function to avoid duplication.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
