An MSI interrupt should only be delivered to the lowest priority CPU
when it has RH=1, regardless of the delivery mode. Modified
kvm_is_dm_lowest_prio() to check for either irq->delivery_mode == APIC_DM_LOWPRI
or irq->msi_redir_hint.
Moved kvm_is_dm_lowest_prio() into lapic.h and renamed to
kvm_lowest_prio_delivery().
Changed a check in kvm_irq_delivery_to_apic_fast() from
irq->delivery_mode == APIC_DM_LOWPRI to kvm_is_dm_lowest_prio().
Signed-off-by: James Sullivan <sullivan.james.f@gmail.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Extended struct kvm_lapic_irq with bool msi_redir_hint, which will
be used to determine if the delivery of the MSI should target only
the lowest priority CPU in the logical group specified for delivery.
(In physical dest mode, the RH bit is not relevant). Initialized the value
of msi_redir_hint to true when RH=1 in kvm_set_msi_irq(), and initialized
to false in all other cases.
Added value of msi_redir_hint to a debug message dump of an IRQ in
apic_send_ipi().
Signed-off-by: James Sullivan <sullivan.james.f@gmail.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Change to u16 if they only contain data in the low 16 bits.
Change the level field to bool, since we assign 1 sometimes, but
just mask icr_low with APIC_INT_ASSERT in apic_send_ipi.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
x86 architecture defines differences between the reset and INIT sequences.
INIT does not initialize the FPU (including MMX, XMM, YMM, etc.), TSC, PMU,
MSRs (in general), MTRRs machine-check, APIC ID, APIC arbitration ID and BSP.
References (from Intel SDM):
"If the MP protocol has completed and a BSP is chosen, subsequent INITs (either
to a specific processor or system wide) do not cause the MP protocol to be
repeated." [8.4.2: MP Initialization Protocol Requirements and Restrictions]
[Table 9-1. IA-32 Processor States Following Power-up, Reset, or INIT]
"If the processor is reset by asserting the INIT# pin, the x87 FPU state is not
changed." [9.2: X87 FPU INITIALIZATION]
"The state of the local APIC following an INIT reset is the same as it is after
a power-up or hardware reset, except that the APIC ID and arbitration ID
registers are not affected." [10.4.7.3: Local APIC State After an INIT Reset
("Wait-for-SIPI" State)]
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1428924848-28212-1-git-send-email-namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Introducing KVM_CAP_DISABLE_QUIRKS for disabling x86 quirks that were previous
created in order to overcome QEMU issues. Those issue were mostly result of
invalid VM BIOS. Currently there are two quirks that can be disabled:
1. KVM_QUIRK_LINT0_REENABLED - LINT0 was enabled after boot
2. KVM_QUIRK_CD_NW_CLEARED - CD and NW are cleared after boot
These two issues are already resolved in recent releases of QEMU, and would
therefore be disabled by QEMU.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1428879221-29996-1-git-send-email-namit@cs.technion.ac.il>
[Report capability from KVM_CHECK_EXTENSION too. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Use __kvm_guest_{enter|exit} instead of kvm_guest_{enter|exit}
where interrupts are disabled.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The kvmclock spec says that the host will increment a version field to
an odd number, then update stuff, then increment it to an even number.
The host is buggy and doesn't do this, and the result is observable
when one vcpu reads another vcpu's kvmclock data.
There's no good way for a guest kernel to keep its vdso from reading
a different vcpu's kvmclock data, but we don't need to care about
changing VCPUs as long as we read a consistent data from kvmclock.
(VCPU can change outside of this loop too, so it doesn't matter if we
return a value not fit for this VCPU.)
Based on a patch by Radim Krčmář.
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Acked-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Pull fourth vfs update from Al Viro:
"d_inode() annotations from David Howells (sat in for-next since before
the beginning of merge window) + four assorted fixes"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
RCU pathwalk breakage when running into a symlink overmounting something
fix I_DIO_WAKEUP definition
direct-io: only inc/dec inode->i_dio_count for file systems
fs/9p: fix readdir()
VFS: assorted d_backing_inode() annotations
VFS: fs/inode.c helpers: d_inode() annotations
VFS: fs/cachefiles: d_backing_inode() annotations
VFS: fs library helpers: d_inode() annotations
VFS: assorted weird filesystems: d_inode() annotations
VFS: normal filesystems (and lustre): d_inode() annotations
VFS: security/: d_inode() annotations
VFS: security/: d_backing_inode() annotations
VFS: net/: d_inode() annotations
VFS: net/unix: d_backing_inode() annotations
VFS: kernel/: d_inode() annotations
VFS: audit: d_backing_inode() annotations
VFS: Fix up some ->d_inode accesses in the chelsio driver
VFS: Cachefiles should perform fs modifications on the top layer only
VFS: AF_UNIX sockets should call mknod on the top layer only
The host's decision to enable machine check exceptions should remain
in force during non-root mode. KVM was writing 0 to cr4 on VCPU reset
and passed a slightly-modified 0 to the vmcs.guest_cr4 value.
Tested: Built.
On earlier version, tested by injecting machine check
while a guest is spinning.
Before the change, if guest CR4.MCE==0, then the machine check is
escalated to Catastrophic Error (CATERR) and the machine dies.
If guest CR4.MCE==1, then the machine check causes VMEXIT and is
handled normally by host Linux. After the change, injecting a machine
check causes normal Linux machine check handling.
Signed-off-by: Ben Serebrin <serebrin@google.com>
Reviewed-by: Venkatesh Srinivas <venkateshs@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Soft mmu uses direct shadow page to fill guest large mapping with small
pages if huge mapping is disallowed on host. So zapping direct shadow
page works well both for soft mmu and hard mmu, it's just less widely
applicable.
Fix the comment to reflect this.
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Message-Id: <552C91BA.1010703@linux.intel.com>
[Fix comment wording further. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
As Andres pointed out:
| I don't understand the value of this check here. Are we looking for a
| broken memslot? Shouldn't this be a BUG_ON? Is this the place to care
| about these things? npages is capped to KVM_MEM_MAX_NR_PAGES, i.e.
| 2^31. A 64 bit overflow would be caused by a gigantic gfn_start which
| would be trouble in many other ways.
This patch drops the memslot overflow check to make the codes more simple.
Reviewed-by: Andres Lagar-Cavilla <andreslc@google.com>
Signed-off-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Message-Id: <1429064694-3072-1-git-send-email-wanpeng.li@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Sparse is reporting a "we previously assumed 'src' could be null" error.
This is true as far as the static analyzer can see, but in practice only
IPIs can set shorthand to self and they also set 'src', so it's ok.
Still, move the initialization of x2apic_ipi (and thus the NULL check for
src right before the first use.
While at it, initializing ret to "false" is somewhat confusing because of
the almost immediate assigned of "true" to the same variable. Thus,
initialize it to "true" and modify it in the only path that used to use
the value from "bool ret = false". There is no change in generated code
from this change.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
kvm_init_msr_list is currently called before hardware_setup. As a result,
vmx_mpx_supported always returns false when kvm_init_msr_list checks whether to
save MSR_IA32_BNDCFGS.
Move kvm_init_msr_list after vmx_hardware_setup is called to fix this issue.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1428864435-4732-1-git-send-email-namit@cs.technion.ac.il>
Cc: stable@vger.kernel.org # 3.15+
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Pull timer updates from Ingo Molnar:
"The main changes in this cycle were:
- clockevents state machine cleanups and enhancements (Viresh Kumar)
- clockevents broadcast notifier horror to state machine conversion
and related cleanups (Thomas Gleixner, Rafael J Wysocki)
- clocksource and timekeeping core updates (John Stultz)
- clocksource driver updates and fixes (Ben Dooks, Dmitry Osipenko,
Hans de Goede, Laurent Pinchart, Maxime Ripard, Xunlei Pang)
- y2038 fixes (Xunlei Pang, John Stultz)
- NMI-safe ktime_get_raw_fast() and general refactoring of the clock
code, in preparation to perf's per event clock ID support (Peter
Zijlstra)
- generic sched/clock fixes, optimizations and cleanups (Daniel
Thompson)
- clockevents cpu_down() race fix (Preeti U Murthy)"
* 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (94 commits)
timers/PM: Drop unnecessary braces from tick_freeze()
timers/PM: Fix up tick_unfreeze()
timekeeping: Get rid of stale comment
clockevents: Cleanup dead cpu explicitely
clockevents: Make tick handover explicit
clockevents: Remove broadcast oneshot control leftovers
sched/idle: Use explicit broadcast oneshot control function
ARM: Tegra: Use explicit broadcast oneshot control function
ARM: OMAP: Use explicit broadcast oneshot control function
intel_idle: Use explicit broadcast oneshot control function
ACPI/idle: Use explicit broadcast control function
ACPI/PAD: Use explicit broadcast oneshot control function
x86/amd/idle, clockevents: Use explicit broadcast oneshot control functions
clockevents: Provide explicit broadcast oneshot control functions
clockevents: Remove the broadcast control leftovers
ARM: OMAP: Use explicit broadcast control function
intel_idle: Use explicit broadcast control function
cpuidle: Use explicit broadcast control function
ACPI/processor: Use explicit broadcast control function
ACPI/PAD: Use explicit broadcast control function
...
ARM/ARM64: fixes for live migration, irqfd and ioeventfd support (enabling
vhost, too), page aging
s390: interrupt handling rework, allowing to inject all local interrupts
via new ioctl and to get/set the full local irq state for migration
and introspection. New ioctls to access memory by virtual address,
and to get/set the guest storage keys. SIMD support.
MIPS: FPU and MIPS SIMD Architecture (MSA) support. Includes some patches
from Ralf Baechle's MIPS tree.
x86: bugfixes (notably for pvclock, the others are small) and cleanups.
Another small latency improvement for the TSC deadline timer.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)
iQEcBAABAgAGBQJVJ9vmAAoJEL/70l94x66DoMEH/R3rh8IMf4jTiWRkcqohOMPX
k1+NaSY/lCKayaSgggJ2hcQenMbQoXEOdslvaA/H0oC+VfJGK+lmU6E63eMyyhjQ
Y+Px6L85NENIzDzaVu/TIWWuhil5PvIRr3VO8cvntExRoCjuekTUmNdOgCvN2ObW
wswN2qRdPIeEj2kkulbnye+9IV4G0Ne9bvsmUdOdfSSdi6ZcV43JcvrpOZT++mKj
RrKB+3gTMZYGJXMMLBwMkdl8mK1ozriD+q0mbomT04LUyGlPwYLl4pVRDBqyksD7
KsSSybaK2E4i5R80WEljgDMkNqrCgNfg6VZe4n9Y+CfAAOToNnkMJaFEi+yuqbs=
=yu2b
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM updates from Paolo Bonzini:
"First batch of KVM changes for 4.1
The most interesting bit here is irqfd/ioeventfd support for ARM and
ARM64.
Summary:
ARM/ARM64:
fixes for live migration, irqfd and ioeventfd support (enabling
vhost, too), page aging
s390:
interrupt handling rework, allowing to inject all local interrupts
via new ioctl and to get/set the full local irq state for migration
and introspection. New ioctls to access memory by virtual address,
and to get/set the guest storage keys. SIMD support.
MIPS:
FPU and MIPS SIMD Architecture (MSA) support. Includes some
patches from Ralf Baechle's MIPS tree.
x86:
bugfixes (notably for pvclock, the others are small) and cleanups.
Another small latency improvement for the TSC deadline timer"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (146 commits)
KVM: use slowpath for cross page cached accesses
kvm: mmu: lazy collapse small sptes into large sptes
KVM: x86: Clear CR2 on VCPU reset
KVM: x86: DR0-DR3 are not clear on reset
KVM: x86: BSP in MSR_IA32_APICBASE is writable
KVM: x86: simplify kvm_apic_map
KVM: x86: avoid logical_map when it is invalid
KVM: x86: fix mixed APIC mode broadcast
KVM: x86: use MDA for interrupt matching
kvm/ppc/mpic: drop unused IRQ_testbit
KVM: nVMX: remove unnecessary double caching of MAXPHYADDR
KVM: nVMX: checks for address bits beyond MAXPHYADDR on VM-entry
KVM: x86: cache maxphyaddr CPUID leaf in struct kvm_vcpu
KVM: vmx: pass error code with internal error #2
x86: vdso: fix pvclock races with task migration
KVM: remove kvm_read_hva and kvm_read_hva_atomic
KVM: x86: optimize delivery of TSC deadline timer interrupt
KVM: x86: extract blocking logic from __vcpu_run
kvm: x86: fix x86 eflags fixed bit
KVM: s390: migrate vcpu interrupt state
...
Dirty logging tracks sptes in 4k granularity, meaning that large sptes
have to be split. If live migration is successful, the guest in the
source machine will be destroyed and large sptes will be created in the
destination. However, the guest continues to run in the source machine
(for example if live migration fails), small sptes will remain around
and cause bad performance.
This patch introduce lazy collapsing of small sptes into large sptes.
The rmap will be scanned in ioctl context when dirty logging is stopped,
dropping those sptes which can be collapsed into a single large-page spte.
Later page faults will create the large-page sptes.
Reviewed-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Message-Id: <1428046825-6905-1-git-send-email-wanpeng.li@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
CR2 is not cleared as it should after reset. See Intel SDM table named "IA-32
Processor States Following Power-up, Reset, or INIT".
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1427933438-12782-5-git-send-email-namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
DR0-DR3 are not cleared as they should during reset and when they are set from
userspace. It appears to be caused by c77fb5fe6f ("KVM: x86: Allow the guest
to run with dirty debug registers").
Force their reload on these situations.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1427933438-12782-4-git-send-email-namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
After reset, the CPU can change the BSP, which will be used upon INIT. Reset
should return the BSP which QEMU asked for, and therefore handled accordingly.
To quote: "If the MP protocol has completed and a BSP is chosen, subsequent
INITs (either to a specific processor or system wide) do not cause the MP
protocol to be repeated."
[Intel SDM 8.4.2: MP Initialization Protocol Requirements and Restrictions]
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1427933438-12782-3-git-send-email-namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
recalculate_apic_map() uses two passes over all VCPUs. This is a relic
from time when we selected a global mode in the first pass and set up
the optimized table in the second pass (to have a consistent mode).
Recent changes made mixed mode unoptimized and we can do it in one pass.
Format of logical MDA is a function of the mode, so we encode it in
apic_logical_id() and drop obsoleted variables from the struct.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Message-Id: <1423766494-26150-5-git-send-email-rkrcmar@redhat.com>
[Add lid_bits temporary in apic_logical_id. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We want to support mixed modes and the easiest solution is to avoid
optimizing those weird and unlikely scenarios.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Message-Id: <1423766494-26150-4-git-send-email-rkrcmar@redhat.com>
[Add comment above KVM_APIC_MODE_* defines. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Broadcast allowed only one global APIC mode, but mixed modes are
theoretically possible. x2APIC IPI doesn't mean 0xff as broadcast,
the rest does.
x2APIC broadcasts are accepted by xAPIC. If we take SDM to be logical,
even addreses beginning with 0xff should be accepted, but real hardware
disagrees. This patch aims for simple code by considering most of real
behavior as undefined.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Message-Id: <1423766494-26150-3-git-send-email-rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In mixed modes, we musn't deliver xAPIC IPIs like x2APIC and vice versa.
Instead of preserving the information in apic_send_ipi(), we regain it
by converting all destinations into correct MDA in the slow path.
This allows easier reasoning about subsequent matching.
Our kvm_apic_broadcast() had an interesting design decision: it didn't
consider IOxAPIC 0xff as broadcast in x2APIC mode ...
everything worked because IOxAPIC can't set that in physical mode and
logical mode considered it as a message for first 8 VCPUs.
This patch interprets IOxAPIC 0xff as x2APIC broadcast.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Message-Id: <1423766494-26150-2-git-send-email-rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
After speed-up of cpuid_maxphyaddr() it can be called frequently:
instead of heavyweight enumeration of CPUID entries it returns a cached
pre-computed value. It is also inlined now. So caching its result became
unnecessary and can be removed.
Signed-off-by: Eugene Korenevsky <ekorenevsky@gmail.com>
Message-Id: <20150329205644.GA1258@gnote>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
On each VM-entry CPU should check the following VMCS fields for zero bits
beyond physical address width:
- APIC-access address
- virtual-APIC address
- posted-interrupt descriptor address
This patch adds these checks required by Intel SDM.
Signed-off-by: Eugene Korenevsky <ekorenevsky@gmail.com>
Message-Id: <20150329205627.GA1244@gnote>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
cpuid_maxphyaddr(), which performs lot of memory accesses is called
extensively across KVM, especially in nVMX code.
This patch adds a cached value of maxphyaddr to vcpu.arch to reduce the
pressure onto CPU cache and simplify the code of cpuid_maxphyaddr()
callers. The cached value is initialized in kvm_arch_vcpu_init() and
reloaded every time CPUID is updated by usermode. It is obvious that
these reloads occur infrequently.
Signed-off-by: Eugene Korenevsky <ekorenevsky@gmail.com>
Message-Id: <20150329205612.GA1223@gnote>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Exposing the on-stack error code with internal error is cheap and
potentially useful.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Message-Id: <1428001865-32280-1-git-send-email-rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The newly-added tracepoint shows the following results on
the tscdeadline_latency test:
qemu-kvm-8387 [002] 6425.558974: kvm_vcpu_wakeup: poll time 10407 ns
qemu-kvm-8387 [002] 6425.558984: kvm_vcpu_wakeup: poll time 0 ns
qemu-kvm-8387 [002] 6425.561242: kvm_vcpu_wakeup: poll time 10477 ns
qemu-kvm-8387 [002] 6425.561251: kvm_vcpu_wakeup: poll time 0 ns
and so on. This is because we need to go through kvm_vcpu_block again
after the timer IRQ is injected. Avoid it by polling once before
entering kvm_vcpu_block.
On my machine (Xeon E5 Sandy Bridge) this removes about 500 cycles (7%)
from the latency of the TSC deadline timer.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Rename the old __vcpu_run to vcpu_run, and extract part of it to a new
function vcpu_block.
The next patch will add a new condition in vcpu_block, avoid extra
indentation.
Reviewed-by: David Matlack <dmatlack@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Guest can't be booted w/ ept=0, there is a message dumped as below:
If you're running a guest on an Intel machine without unrestricted mode
support, the failure can be most likely due to the guest entering an invalid
state for Intel VT. For example, the guest maybe running in big real mode
which is not supported on less recent Intel processors.
EAX=00000011 EBX=f000d2f6 ECX=00006cac EDX=000f8956
ESI=bffbdf62 EDI=00000000 EBP=00006c68 ESP=00006c68
EIP=0000d187 EFL=00000004 [-----P-] CPL=0 II=0 A20=1 SMM=0 HLT=0
ES =e000 000e0000 ffffffff 00809300 DPL=0 DS16 [-WA]
CS =f000 000f0000 ffffffff 00809b00 DPL=0 CS16 [-RA]
SS =0000 00000000 ffffffff 00809300 DPL=0 DS16 [-WA]
DS =0000 00000000 ffffffff 00809300 DPL=0 DS16 [-WA]
FS =0000 00000000 ffffffff 00809300 DPL=0 DS16 [-WA]
GS =0000 00000000 ffffffff 00809300 DPL=0 DS16 [-WA]
LDT=0000 00000000 0000ffff 00008200 DPL=0 LDT
TR =0000 00000000 0000ffff 00008b00 DPL=0 TSS32-busy
GDT= 000f6a80 00000037
IDT= 000f6abe 00000000
CR0=00000011 CR2=00000000 CR3=00000000 CR4=00000000
DR0=0000000000000000 DR1=0000000000000000 DR2=0000000000000000 DR3=0000000000000000
DR6=00000000ffff0ff0 DR7=0000000000000400
EFER=0000000000000000
Code=01 1e b8 6a 2e 0f 01 16 74 6a 0f 20 c0 66 83 c8 01 0f 22 c0 <66> ea 8f d1 0f 00 08 00 b8 10 00 00 00 8e d8 8e c0 8e d0 8e e0 8e e8 89 c8 ff e2 89 c1 b8X
X86 eflags bit 1 is fixed set, which means that 1 << 1 is set instead of 1,
this patch fix it.
Signed-off-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Message-Id: <1428473294-6633-1-git-send-email-wanpeng.li@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Use the normal return values for bool functions
Signed-off-by: Joe Perches <joe@perches.com>
Message-Id: <9f593eb2f43b456851cd73f7ed09654ca58fb570.1427759009.git.joe@perches.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A trivial code cleanup. This `if` is redundant.
Signed-off-by: Eugene Korenevsky <ekorenevsky@gmail.com>
Message-Id: <20150328222717.GA6508@gnote>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Some constants are redfined in emulate.c. Avoid it.
s/SELECTOR_RPL_MASK/SEGMENT_RPL_MASK
s/SELECTOR_TI_MASK/SEGMENT_TI_MASK
No functional change.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1427635984-8113-3-git-send-email-namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The eflags are redefined (using other defines) in emulate.c.
Use the definition from processor-flags.h as some mess already started.
No functional change.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1427635984-8113-2-git-send-email-namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If the source of BSF and BSR is zero, the destination register should not
change. That is how real hardware behaves. If we set the destination even with
the same value that we had before, we may clear bits [63:32] unnecassarily.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1427719163-5429-4-git-send-email-namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
POPA should assign the values to the registers as usual registers are assigned.
In other words, 32-bits register assignments should clear bits [63:32] of the
register.
Split the code of register assignments that will be used by future changes as
well.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1427719163-5429-3-git-send-email-namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
On legacy mode CMOV emulation should still clear bits [63:32] even if the
assignment is not done. The previous fix 140bad89fd ("KVM: x86: emulation of
dword cmov on long-mode should clear [63:32]") was incomplete.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Message-Id: <1427719163-5429-2-git-send-email-namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If data is read from PIC with invalid access size, the return data stays
uninitialized even though success is returned.
Fix this by always initializing the data.
Signed-off-by: Petr Matousek <pmatouse@redhat.com>
Reported-by: Nadav Amit <nadav.amit@gmail.com>
Message-Id: <20150311111609.GG8544@dhcp-25-225.brq.redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In preparation of adding another tkr field, rename this one to
tkr_mono. Also rename tk_read_base::base_mono to tk_read_base::base,
since the structure is not specific to CLOCK_MONOTONIC and the mono
name got added to the tk_read_base instance.
Lots of trivial churn.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: John Stultz <john.stultz@linaro.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20150319093400.344679419@infradead.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
If the guest CPU is supposed to support rdtscp and the host has rdtscp
enabled in the secondary execution controls, we can also expose this
feature to L1. Just extend nested_vmx_exit_handled to properly route
EXIT_REASON_RDTSCP.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
virt/kvm was never really a good include directory for anything else
than locally included headers.
With the move of iodev.h there is no need anymore to add this
directory the compiler's include path, so remove it from the x86 kvm
Makefile.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
iodev.h contains definitions for the kvm_io_bus framework. This is
needed both by the generic KVM code in virt/kvm as well as by
architecture specific code under arch/. Putting the header file in
virt/kvm and using local includes in the architecture part seems at
least dodgy to me, so let's move the file into include/kvm, so that a
more natural "#include <kvm/iodev.h>" can be used by all of the code.
This also solves a problem later when using struct kvm_io_device
in arm_vgic.h.
Fixing up the FSF address in the GPL header and a wrong include path
on the way.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
This is needed in e.g. ARM vGIC emulation, where the MMIO handling
depends on the VCPU that does the access.
Signed-off-by: Nikolay Nikolaev <n.nikolaev@virtualopensystems.com>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
An overhead from function call is not appropriate for its size and
frequency of execution.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
kvm_ioapic_update_eoi() wasn't called if directed EOI was enabled.
We need to do that for irq notifiers. (Like with edge interrupts.)
Fix it by skipping EOI broadcast only.
Bug: https://bugzilla.kernel.org/show_bug.cgi?id=82211
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Tested-by: Bandan Das <bsd@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
I hit this path on a AMD box and thought
someone was playing a April Fool's joke on me.
Signed-off-by: Bandan Das <bsd@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This patch fix the following sparse warnings:
for arch/x86/kvm/x86.c:
warning: symbol 'emulator_read_write' was not declared. Should it be static?
warning: symbol 'emulator_write_emulated' was not declared. Should it be static?
warning: symbol 'emulator_get_dr' was not declared. Should it be static?
warning: symbol 'emulator_set_dr' was not declared. Should it be static?
for arch/x86/kvm/pmu.c:
warning: symbol 'fixed_pmc_events' was not declared. Should it be static?
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This patch fix the following sparse warning:
for file arch/x86/kvm/x86.c:
warning: Using plain integer as NULL pointer
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
If EPT was enabled, unrestricted_guest was allowed in L1 regardless of
L0. L1 triple faulted when running L2 guest that required emulation.
Another side effect was 'WARN_ON_ONCE(vmx->nested.nested_run_pending)'
in L0's dmesg:
WARNING: CPU: 0 PID: 0 at arch/x86/kvm/vmx.c:9190 nested_vmx_vmexit+0x96e/0xb00 [kvm_intel] ()
Prevent this scenario by masking SECONDARY_EXEC_UNRESTRICTED_GUEST when
the host doesn't have it enabled.
Fixes: 78051e3b7e ("KVM: nVMX: Disable unrestricted mode if ept=0")
Cc: stable@vger.kernel.org
Tested-By: Kashyap Chamarthy <kchamart@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
While in L2, leave all #UD to L2 and do not try to emulate it. If L1 is
interested in doing this, it reports its interest via the exception
bitmap, and we never get into handle_exception of L0 anyway.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
For a very long time (since 2b3d2a20), the path handling a vmmcall
instruction of the guest on an Intel host only applied the patch but no
longer handled the hypercall. The reverse case, vmcall on AMD hosts, is
fine. As both em_vmcall and em_vmmcall actually have to do the same, we
can fix the issue by consolidating both into the same handler.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Another patch in my war on emulate_on_interception() use as a svm exit handler.
These were pulled out of a larger patch at the suggestion of Radim Krcmar, see
https://lkml.org/lkml/2015/2/25/559
Changes since v1:
* fixed typo introduced after test, retested
Signed-off-by: David Kaplan <david.kaplan@amd.com>
[separated out just cr_interception part from larger removal of
INTERCEPT_CR0_WRITE, forward ported, tested]
Signed-off-by: Joel Schopp <joel.schopp@amd.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
In commit 3af18d9c5f ("KVM: nVMX: Prepare for using hardware MSR bitmap"),
we are setting MSR_BITMAP in prepare_vmcs02 if we should use hardware. This
is not enough since the field will be modified by following vmx_set_efer.
Fix this by setting vmx_msr_bitmap_nested in vmx_set_msr_bitmap if vcpu is
in guest mode.
Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
If data is read from PIC with invalid access size, the return data stays
uninitialized even though success is returned.
Fix this by always initializing the data.
Signed-off-by: Petr Matousek <pmatouse@redhat.com>
Reported-by: Nadav Amit <nadav.amit@gmail.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
POWER supports irqfds but forgot to advertise them. Some userspace does
not check for the capability, but others check it---thus they work on
x86 and s390 but not POWER.
To avoid that other architectures in the future make the same mistake, let
common code handle KVM_CAP_IRQFD the same way as KVM_CAP_IRQFD_RESAMPLE.
Reported-and-tested-by: Greg Kurz <gkurz@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org
Fixes: 297e21053a
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
No need to re-decode WBINVD since we know what it is from the intercept.
Signed-off-by: David Kaplan <David.Kaplan@amd.com>
[extracted from larger unlrelated patch, forward ported, tested,style cleanup]
Signed-off-by: Joel Schopp <joel.schopp@amd.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Currently kvm_emulate() skips the instruction but kvm_emulate_* sometimes
don't. The end reult is the caller ends up doing the skip themselves.
Let's make them consistant.
Signed-off-by: Joel Schopp <joel.schopp@amd.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
kvm_kvfree() provides exactly the same functionality as the
new common kvfree() function - so let's simply replace the
kvm function with the common function.
Signed-off-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This patch fixes the bug discussed in
https://www.mail-archive.com/kvm@vger.kernel.org/msg109813.html
This patch uses a new field named irr_delivered to record the
delivery status of edge-triggered interrupts, and clears the
delivered interrupts in kvm_get_ioapic. So it has the same effect
of commit 0bc830b05c
("KVM: ioapic: clear IRR for edge-triggered interrupts at delivery")
while avoids the bug of Windows guests.
Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
KVM has nice wrappers to access the register values, clean up a few places
that should use them but currently do not.
Signed-off-by: David Kaplan <david.kaplan@amd.com>
[forward port and testing]
Signed-off-by: Joel Schopp <joel.schopp@amd.com>
Acked-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
In commit b4eef9b36d, we started to use hwapic_isr_update() != NULL
instead of kvm_apic_vid_enabled(vcpu->kvm). This didn't work because
SVM had it defined and "apicv" path in apic_{set,clear}_isr() does not
change apic->isr_count, because it should always be 1. The initial
value of apic->isr_count was based on kvm_apic_vid_enabled(vcpu->kvm),
which is always 0 for SVM, so KVM could have injected interrupts when it
shouldn't.
Fix it by implicitly setting SVM's hwapic_isr_update to NULL and make the
initial isr_count depend on hwapic_isr_update() for good measure.
Fixes: b4eef9b36d ("kvm: x86: vmx: NULL out hwapic_isr_update() in case of !enable_apicv")
Reported-and-tested-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This has been broken for a long time: it broke first in 2.6.35, then was
almost fixed in 2.6.36 but this one-liner slipped through the cracks.
The bug shows up as an infinite loop in Windows 7 (and newer) boot on
32-bit hosts without EPT.
Windows uses CMPXCHG8B to write to page tables, which causes a
page fault if running without EPT; the emulator is then called from
kvm_mmu_page_fault. The loop then happens if the higher 4 bytes are
not 0; the common case for this is that the NX bit (bit 63) is 1.
Fixes: 6550e1f165
Fixes: 16518d5ada
Cc: stable@vger.kernel.org # 2.6.35+
Reported-by: Erik Rull <erik.rull@rdsoftware.de>
Tested-by: Erik Rull <erik.rull@rdsoftware.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
'apic' is not defined if !CONFIG_X86_64 && !CONFIG_X86_LOCAL_APIC.
Posted interrupt makes no sense without CONFIG_SMP, and
CONFIG_X86_LOCAL_APIC will be set with it.
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Pull x86 perf updates from Ingo Molnar:
"This series tightens up RDPMC permissions: currently even highly
sandboxed x86 execution environments (such as seccomp) have permission
to execute RDPMC, which may leak various perf events / PMU state such
as timing information and other CPU execution details.
This 'all is allowed' RDPMC mode is still preserved as the
(non-default) /sys/devices/cpu/rdpmc=2 setting. The new default is
that RDPMC access is only allowed if a perf event is mmap-ed (which is
needed to correctly interpret RDPMC counter values in any case).
As a side effect of these changes CR4 handling is cleaned up in the
x86 code and a shadow copy of the CR4 value is added.
The extra CR4 manipulation adds ~ <50ns to the context switch cost
between rdpmc-capable and rdpmc-non-capable mms"
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86: Add /sys/devices/cpu/rdpmc=2 to allow rdpmc for all tasks
perf/x86: Only allow rdpmc if a perf_event is mapped
perf: Pass the event to arch_perf_update_userpage()
perf: Add pmu callbacks to track event mapping and unmapping
x86: Add a comment clarifying LDT context switching
x86: Store a per-cpu shadow copy of CR4
x86: Clean up cr4 manipulation
Common: Optional support for adding a small amount of polling on each HLT
instruction executed in the guest (or equivalent for other architectures).
This can improve latency up to 50% on some scenarios (e.g. O_DSYNC writes
or TCP_RR netperf tests). This also has to be enabled manually for now,
but the plan is to auto-tune this in the future.
ARM/ARM64: the highlights are support for GICv3 emulation and dirty page
tracking
s390: several optimizations and bugfixes. Also a first: a feature
exposed by KVM (UUID and long guest name in /proc/sysinfo) before
it is available in IBM's hypervisor! :)
MIPS: Bugfixes.
x86: Support for PML (page modification logging, a new feature in
Broadwell Xeons that speeds up dirty page tracking), nested virtualization
improvements (nested APICv---a nice optimization), usual round of emulation
fixes. There is also a new option to reduce latency of the TSC deadline
timer in the guest; this needs to be tuned manually.
Some commits are common between this pull and Catalin's; I see you
have already included his tree.
ARM has other conflicts where functions are added in the same place
by 3.19-rc and 3.20 patches. These are not large though, and entirely
within KVM.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)
iQEcBAABAgAGBQJU28rkAAoJEL/70l94x66DXqQH/1TDOfJIjW7P2kb0Sw7Fy1wi
cEX1KO/VFxAqc8R0E/0Wb55CXyPjQJM6xBXuFr5cUDaIjQ8ULSktL4pEwXyyv/s5
DBDkN65mriry2w5VuEaRLVcuX9Wy+tqLQXWNkEySfyb4uhZChWWHvKEcgw5SqCyg
NlpeHurYESIoNyov3jWqvBjr4OmaQENyv7t2c6q5ErIgG02V+iCux5QGbphM2IC9
LFtPKxoqhfeB2xFxTOIt8HJiXrZNwflsTejIlCl/NSEiDVLLxxHCxK2tWK/tUXMn
JfLD9ytXBWtNMwInvtFm4fPmDouv2VDyR0xnK2db+/axsJZnbxqjGu1um4Dqbak=
=7gdx
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM update from Paolo Bonzini:
"Fairly small update, but there are some interesting new features.
Common:
Optional support for adding a small amount of polling on each HLT
instruction executed in the guest (or equivalent for other
architectures). This can improve latency up to 50% on some
scenarios (e.g. O_DSYNC writes or TCP_RR netperf tests). This
also has to be enabled manually for now, but the plan is to
auto-tune this in the future.
ARM/ARM64:
The highlights are support for GICv3 emulation and dirty page
tracking
s390:
Several optimizations and bugfixes. Also a first: a feature
exposed by KVM (UUID and long guest name in /proc/sysinfo) before
it is available in IBM's hypervisor! :)
MIPS:
Bugfixes.
x86:
Support for PML (page modification logging, a new feature in
Broadwell Xeons that speeds up dirty page tracking), nested
virtualization improvements (nested APICv---a nice optimization),
usual round of emulation fixes.
There is also a new option to reduce latency of the TSC deadline
timer in the guest; this needs to be tuned manually.
Some commits are common between this pull and Catalin's; I see you
have already included his tree.
Powerpc:
Nothing yet.
The KVM/PPC changes will come in through the PPC maintainers,
because I haven't received them yet and I might end up being
offline for some part of next week"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (130 commits)
KVM: ia64: drop kvm.h from installed user headers
KVM: x86: fix build with !CONFIG_SMP
KVM: x86: emulate: correct page fault error code for NoWrite instructions
KVM: Disable compat ioctl for s390
KVM: s390: add cpu model support
KVM: s390: use facilities and cpu_id per KVM
KVM: s390/CPACF: Choose crypto control block format
s390/kernel: Update /proc/sysinfo file with Extended Name and UUID
KVM: s390: reenable LPP facility
KVM: s390: floating irqs: fix user triggerable endless loop
kvm: add halt_poll_ns module parameter
kvm: remove KVM_MMIO_SIZE
KVM: MIPS: Don't leak FPU/DSP to guest
KVM: MIPS: Disable HTW while in guest
KVM: nVMX: Enable nested posted interrupt processing
KVM: nVMX: Enable nested virtual interrupt delivery
KVM: nVMX: Enable nested apic register virtualization
KVM: nVMX: Make nested control MSRs per-cpu
KVM: nVMX: Enable nested virtualize x2apic mode
KVM: nVMX: Prepare for using hardware MSR bitmap
...
<asm/apic.h> isn't included directly and without CONFIG_SMP, an option
that automagically pulls it can't be enabled.
Reported-by: Jim Davis <jim.epost@gmail.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Pull RCU updates from Ingo Molnar:
"The main RCU changes in this cycle are:
- Documentation updates.
- Miscellaneous fixes.
- Preemptible-RCU fixes, including fixing an old bug in the
interaction of RCU priority boosting and CPU hotplug.
- SRCU updates.
- RCU CPU stall-warning updates.
- RCU torture-test updates"
* 'core-rcu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (54 commits)
rcu: Initialize tiny RCU stall-warning timeouts at boot
rcu: Fix RCU CPU stall detection in tiny implementation
rcu: Add GP-kthread-starvation checks to CPU stall warnings
rcu: Make cond_resched_rcu_qs() apply to normal RCU flavors
rcu: Optionally run grace-period kthreads at real-time priority
ksoftirqd: Use new cond_resched_rcu_qs() function
ksoftirqd: Enable IRQs and call cond_resched() before poking RCU
rcutorture: Add more diagnostics in rcu_barrier() test failure case
torture: Flag console.log file to prevent holdovers from earlier runs
torture: Add "-enable-kvm -soundhw pcspk" to qemu command line
rcutorture: Handle different mpstat versions
rcutorture: Check from beginning to end of grace period
rcu: Remove redundant rcu_batches_completed() declaration
rcutorture: Drop rcu_torture_completed() and friends
rcu: Provide rcu_batches_completed_sched() for TINY_RCU
rcutorture: Use unsigned for Reader Batch computations
rcutorture: Make build-output parsing correctly flag RCU's warnings
rcu: Make _batches_completed() functions return unsigned long
rcutorture: Issue warnings on close calls due to Reader Batch blows
documentation: Fix smp typo in memory-barriers.txt
...
NoWrite instructions (e.g. cmp or test) never set the "write access"
bit in the error code, even if one of the operands is treated as a
destination.
Fixes: c205fb7d7d
Cc: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch introduces a new module parameter for the KVM module; when it
is present, KVM attempts a bit of polling on every HLT before scheduling
itself out via kvm_vcpu_block.
This parameter helps a lot for latency-bound workloads---in particular
I tested it with O_DSYNC writes with a battery-backed disk in the host.
In this case, writes are fast (because the data doesn't have to go all
the way to the platters) but they cannot be merged by either the host or
the guest. KVM's performance here is usually around 30% of bare metal,
or 50% if you use cache=directsync or cache=writethrough (these
parameters avoid that the guest sends pointless flush requests, and
at the same time they are not slow because of the battery-backed cache).
The bad performance happens because on every halt the host CPU decides
to halt itself too. When the interrupt comes, the vCPU thread is then
migrated to a new physical CPU, and in general the latency is horrible
because the vCPU thread has to be scheduled back in.
With this patch performance reaches 60-65% of bare metal and, more
important, 99% of what you get if you use idle=poll in the guest. This
means that the tunable gets rid of this particular bottleneck, and more
work can be done to improve performance in the kernel or QEMU.
Of course there is some price to pay; every time an otherwise idle vCPUs
is interrupted by an interrupt, it will poll unnecessarily and thus
impose a little load on the host. The above results were obtained with
a mostly random value of the parameter (500000), and the load was around
1.5-2.5% CPU usage on one of the host's core for each idle guest vCPU.
The patch also adds a new stat, /sys/kernel/debug/kvm/halt_successful_poll,
that can be used to tune the parameter. It counts how many HLT
instructions received an interrupt during the polling period; each
successful poll avoids that Linux schedules the VCPU thread out and back
in, and may also avoid a likely trip to C1 and back for the physical CPU.
While the VM is idle, a Linux 4 VCPU VM halts around 10 times per second.
Of these halts, almost all are failed polls. During the benchmark,
instead, basically all halts end within the polling period, except a more
or less constant stream of 50 per second coming from vCPUs that are not
running the benchmark. The wasted time is thus very low. Things may
be slightly different for Windows VMs, which have a ~10 ms timer tick.
The effect is also visible on Marcelo's recently-introduced latency
test for the TSC deadline timer. Though of course a non-RT kernel has
awful latency bounds, the latency of the timer is around 8000-10000 clock
cycles compared to 20000-120000 without setting halt_poll_ns. For the TSC
deadline timer, thus, the effect is both a smaller average latency and
a smaller variance.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Context switches and TLB flushes can change individual bits of CR4.
CR4 reads take several cycles, so store a shadow copy of CR4 in a
per-cpu variable.
To avoid wasting a cache line, I added the CR4 shadow to
cpu_tlbstate, which is already touched in switch_mm. The heaviest
users of the cr4 shadow will be switch_mm and __switch_to_xtra, and
__switch_to_xtra is called shortly after switch_mm during context
switch, so the cacheline is likely to be hot.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Vince Weaver <vince@deater.net>
Cc: "hillf.zj" <hillf.zj@alibaba-inc.com>
Cc: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/3a54dd3353fffbf84804398e00dfdc5b7c1afd7d.1414190806.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
CR4 manipulation was split, seemingly at random, between direct
(write_cr4) and using a helper (set/clear_in_cr4). Unfortunately,
the set_in_cr4 and clear_in_cr4 helpers also poke at the boot code,
which only a small subset of users actually wanted.
This patch replaces all cr4 access in functions that don't leave cr4
exactly the way they found it with new helpers cr4_set_bits,
cr4_clear_bits, and cr4_set_bits_and_update_boot.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Vince Weaver <vince@deater.net>
Cc: "hillf.zj" <hillf.zj@alibaba-inc.com>
Cc: Valdis Kletnieks <Valdis.Kletnieks@vt.edu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/495a10bdc9e67016b8fd3945700d46cfd5c12c2f.1414190806.git.luto@amacapital.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
If vcpu has a interrupt in vmx non-root mode, injecting that interrupt
requires a vmexit. With posted interrupt processing, the vmexit
is not needed, and interrupts are fully taken care of by hardware.
In nested vmx, this feature avoids much more vmexits than non-nested vmx.
When L1 asks L0 to deliver L1's posted interrupt vector, and the target
VCPU is in non-root mode, we use a physical ipi to deliver POSTED_INTR_NV
to the target vCPU. Using POSTED_INTR_NV avoids unexpected interrupts
if a concurrent vmexit happens and L1's vector is different with L0's.
The IPI triggers posted interrupt processing in the target physical CPU.
In case the target vCPU was not in guest mode, complete the posted
interrupt delivery on the next entry to L2.
Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
With virtual interrupt delivery, the hardware lets KVM use a more
efficient mechanism for interrupt injection. This is an important feature
for nested VMX, because it reduces vmexits substantially and they are
much more expensive with nested virtualization. This is especially
important for throughput-bound scenarios.
Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We can reduce apic register virtualization cost with this feature,
it is also a requirement for virtual interrupt delivery and posted
interrupt processing.
Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
To enable nested apicv support, we need per-cpu vmx
control MSRs:
1. If in-kernel irqchip is enabled, we can enable nested
posted interrupt, we should set posted intr bit in
the nested_vmx_pinbased_ctls_high.
2. If in-kernel irqchip is disabled, we can not enable
nested posted interrupt, the posted intr bit
in the nested_vmx_pinbased_ctls_high will be cleared.
Since there would be different settings about in-kernel
irqchip between VMs, different nested control MSRs
are needed.
Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When L2 is using x2apic, we can use virtualize x2apic mode to
gain higher performance, especially in apicv case.
This patch also introduces nested_vmx_check_apicv_controls
for the nested apicv patches.
Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Currently, if L1 enables MSR_BITMAP, we will emulate this feature, all
of L2's msr access is intercepted by L0. Features like "virtualize
x2apic mode" require that the MSR bitmap is enabled, or the hardware
will exit and for example not virtualize the x2apic MSRs. In order to
let L1 use these features, we need to build a merged bitmap that only
not cause a VMEXIT if 1) L1 requires that 2) the bit is not required by
the processor for APIC virtualization.
For now the guests are still run with MSR bitmap disabled, but this
patch already introduces nested_vmx_merge_msr_bitmap for future use.
Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Revert 7c6a98dfa1, given
that testing PIR is not necessary anymore.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
With APICv, LAPIC timer interrupt is always delivered via IRR:
apic_find_highest_irr syncs PIR to IRR.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A function pointer was not NULLed, causing kvm_vcpu_reload_apic_access_page to
go down the wrong path and OOPS when doing put_page(NULL).
This did not happen on old processors, only when setting the module option
explicitly.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We forgot to re-check LAPIC after splitting the loop in commit
173beedc16 (KVM: x86: Software disabled APIC should still deliver
NMIs, 2014-11-02).
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Fixes: 173beedc16
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We cannot hit the bug now, but future patches will expose this path.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The majority of this patch turns
result = 0; if (CODE) result = 1; return result;
into
return CODE;
because we return bool now.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch adds PML support in VMX. A new module parameter 'enable_pml' is added
to allow user to enable/disable it manually.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Reviewed-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch adds new kvm_x86_ops dirty logging hooks to enable/disable dirty
logging for particular memory slot, and to flush potentially logged dirty GPAs
before reporting slot->dirty_bitmap to userspace.
kvm x86 common code calls these hooks when they are available so PML logic can
be hidden to VMX specific. SVM won't be impacted as these hooks remain NULL
there.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Reviewed-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch changes the second parameter of kvm_mmu_slot_remove_write_access from
'slot id' to 'struct kvm_memory_slot *' to align with kvm_x86_ops dirty logging
hooks, which will be introduced in further patch.
Better way is to change second parameter of kvm_arch_commit_memory_region from
'struct kvm_userspace_memory_region *' to 'struct kvm_memory_slot * new', but it
requires changes on other non-x86 ARCH too, so avoid it now.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Reviewed-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch avoids unnecessary dirty GPA logging to PML buffer in EPT violation
path by setting D-bit manually prior to the occurrence of the write from guest.
We only set D-bit manually in set_spte, and leave fast_page_fault path
unchanged, as fast_page_fault is very unlikely to happen in case of PML.
For the hva <-> pa change case, the spte is updated to either read-only (host
pte is read-only) or be dropped (host pte is writeable), and both cases will be
handled by above changes, therefore no change is necessary.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Reviewed-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch adds new mmu layer functions to clear/set D-bit for memory slot, and
to write protect superpages for memory slot.
In case of PML, CPU logs the dirty GPA automatically to PML buffer when CPU
updates D-bit from 0 to 1, therefore we don't have to write protect 4K pages,
instead, we only need to clear D-bit in order to log that GPA.
For superpages, we still write protect it and let page fault code to handle
dirty page logging, as we still need to split superpage to 4K pages in PML.
As PML is always enabled during guest's lifetime, to eliminate unnecessary PML
GPA logging, we set D-bit manually for the slot with dirty logging disabled.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Reviewed-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We don't have to write protect guest memory for dirty logging if architecture
supports hardware dirty logging, such as PML on VMX, so rename it to be more
generic.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Reviewed-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When assigning devices to large memory guests (>=128GB guest
memory in the failure case) the functions to create the
IOMMU page-tables for the whole guest might run for a very
long time. On non-preemptible kernels this might cause
Soft-Lockup warnings. Fix these by adding a cond_resched()
to the mapping and unmapping loops.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
On long-mode, when far call that changes cs.l takes place, the stack size is
determined by the new mode. For instance, if we go from 32-bit mode to 64-bit
mode, the stack-size if 64. KVM uses the old stack size.
Fix it.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If we got a wraparound of 32-bit operand, and the limit is 0xffffffff, read and
writes should be successful. It just needs to be done in two segments.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Unnecassary define was left after commit 7d882ffa81 ("KVM: x86: Revert
NoBigReal patch in the emulator").
Commit 39f062ff51 ("KVM: x86: Generate #UD when memory operand is required")
was missing undef.
Fix it.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
ARPL and MOVSXD are encoded the same and their execution depends on the
execution mode. The operand sizes of each instruction are different.
Currently, ARPL is detected too late, after the decoding was already done, and
therefore may result in spurious exception (instead of failed emulation).
Introduce a group to the emulator to handle instructions according to execution
mode (32/64 bits). Note: in order not to make changes that may affect
performance, the new ModeDual can only be applied to instructions with ModRM.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The IRET instruction should clear NMI masking, but the current implementation
does not do so.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Indeed, Intel SDM specifically states that for the RET instruction "In 64-bit
mode, the default operation size of this instruction is the stack-address size,
i.e. 64 bits."
However, experiments show this is not the case. Here is for example objdump of
small 64-bit asm:
4004f1: ca 14 00 lret $0x14
4004f4: 48 cb lretq
4004f6: 48 ca 14 00 lretq $0x14
Therefore, remove the Stack flag from far-ret instructions.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Intel SDM says for CMPXCHG: "To simplify the interface to the processor’s bus,
the destination operand receives a write cycle without regard to the result of
the comparison.". This means the destination page should be dirtied.
Fix it to by writing back the original value if cmpxchg failed.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
1. Generic
- sparse warning (make function static)
- optimize locking
- bugfixes for interrupt injection
- fix MVPG addressing modes
2. hrtimer/wakeup fun
A recent change can cause KVM hangs if adjtime is used in the host.
The hrtimer might wake up too early or too late. Too early is fatal
as vcpu_block will see that the wakeup condition is not met and
sleep again. This CPU might never wake up again.
This series addresses this problem. adjclock slowing down the host
clock will result in too late wakeups. This will require more work.
In addition to that we also change the hrtimer from REALTIME to
MONOTONIC to avoid similar problems with timedatectl set-time.
3. sigp rework
We will move all "slow" sigps to QEMU (protected with a capability that
can be enabled) to avoid several races between concurrent SIGP orders.
4. Optimize the shadow page table
Provide an interface to announce the maximum guest size. The kernel
will use that to make the pagetable 2,3,4 (or theoretically) 5 levels.
5. Provide an interface to set the guest TOD
We now use two vm attributes instead of two oneregs, as oneregs are
vcpu ioctl and we don't want to call them from other threads.
6. Protected key functions
The real HMC allows to enable/disable protected key CPACF functions.
Lets provide an implementation + an interface for QEMU to activate
this the protected key instructions.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.14 (GNU/Linux)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=zLjR
-----END PGP SIGNATURE-----
Merge tag 'kvm-s390-next-20150122' of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into kvm-next
KVM: s390: fixes and features for kvm/next (3.20)
1. Generic
- sparse warning (make function static)
- optimize locking
- bugfixes for interrupt injection
- fix MVPG addressing modes
2. hrtimer/wakeup fun
A recent change can cause KVM hangs if adjtime is used in the host.
The hrtimer might wake up too early or too late. Too early is fatal
as vcpu_block will see that the wakeup condition is not met and
sleep again. This CPU might never wake up again.
This series addresses this problem. adjclock slowing down the host
clock will result in too late wakeups. This will require more work.
In addition to that we also change the hrtimer from REALTIME to
MONOTONIC to avoid similar problems with timedatectl set-time.
3. sigp rework
We will move all "slow" sigps to QEMU (protected with a capability that
can be enabled) to avoid several races between concurrent SIGP orders.
4. Optimize the shadow page table
Provide an interface to announce the maximum guest size. The kernel
will use that to make the pagetable 2,3,4 (or theoretically) 5 levels.
5. Provide an interface to set the guest TOD
We now use two vm attributes instead of two oneregs, as oneregs are
vcpu ioctl and we don't want to call them from other threads.
6. Protected key functions
The real HMC allows to enable/disable protected key CPACF functions.
Lets provide an implementation + an interface for QEMU to activate
this the protected key instructions.
SYSENTER emulation is broken in several ways:
1. It misses the case of 16-bit code segments completely (CVE-2015-0239).
2. MSR_IA32_SYSENTER_CS is checked in 64-bit mode incorrectly (bits 0 and 1 can
still be set without causing #GP).
3. MSR_IA32_SYSENTER_EIP and MSR_IA32_SYSENTER_ESP are not masked in
legacy-mode.
4. There is some unneeded code.
Fix it.
Cc: stable@vger.linux.org
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
STR and SLDT with rip-relative operand can cause a host kernel oops.
Mark them as DstMem as well.
Cc: stable@vger.linux.org
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The return value of kvm_arch_vcpu_postcreate is not checked in its
caller. This is okay, because only x86 provides vcpu_postcreate right
now and it could only fail if vcpu_load failed. But that is not
possible during KVM_CREATE_VCPU (kvm_arch_vcpu_load is void, too), so
just get rid of the unchecked return value.
Signed-off-by: Dominik Dingel <dingel@linux.vnet.ibm.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Pull RCU updates from Paul E. McKenney:
- Documentation updates.
- Miscellaneous fixes.
- Preemptible-RCU fixes, including fixing an old bug in the
interaction of RCU priority boosting and CPU hotplug.
- SRCU updates.
- RCU CPU stall-warning updates.
- RCU torture-test updates.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
SuSE's 2.6.16 kernel fails to boot if the delta between tsc_timestamp
and rdtsc is larger than a given threshold:
* If we get more than the below threshold into the future, we rerequest
* the real time from the host again which has only little offset then
* that we need to adjust using the TSC.
*
* For now that threshold is 1/5th of a jiffie. That should be good
* enough accuracy for completely broken systems, but also give us swing
* to not call out to the host all the time.
*/
#define PVCLOCK_DELTA_MAX ((1000000000ULL / HZ) / 5)
Disable masterclock support (which increases said delta) in case the
boot vcpu does not use MSR_KVM_SYSTEM_TIME_NEW.
Upstreams kernels which support pvclock vsyscalls (and therefore make
use of PVCLOCK_STABLE_BIT) use MSR_KVM_SYSTEM_TIME_NEW.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
arch/x86/kvm/x86.c:495:5: sparse: symbol 'kvm_read_nested_guest_page' was not declared. Should it be static?
arch/x86/kvm/x86.c:646:5: sparse: symbol '__kvm_set_xcr' was not declared. Should it be static?
arch/x86/kvm/x86.c:1183:15: sparse: symbol 'max_tsc_khz' was not declared. Should it be static?
arch/x86/kvm/x86.c:1237:6: sparse: symbol 'kvm_track_tsc_matching' was not declared. Should it be static?
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
No TLB flush is needed when there's no valid rmap in memory slot.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Removes some functions that are not used anywhere:
cpu_has_vmx_eptp_writeback() cpu_has_vmx_eptp_uncacheable()
This was partially found by using a static code analysis program called cppcheck.
Signed-off-by: Rickard Strandqvist <rickard_strandqvist@spectrumdigital.se>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We now have a generic function that does most of the work of
kvm_vm_ioctl_get_dirty_log, now use it.
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Mario Smarduch <m.smarduch@samsung.com>
When emulating an instruction that reads the destination memory operand (i.e.,
instructions without the Mov flag in the emulator), the operand is first read.
If a page-fault is detected in this phase, the error-code which would be
delivered to the VM does not indicate that the access that caused the exception
is a write one. This does not conform with real hardware, and may cause the VM
to enter the page-fault handler twice for no reason (once for read, once for
write).
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When software changes D bit (either from 1 to 0, or 0 to 1), the
corresponding TLB entity in the hardware won't be updated immediately. We
should flush it to guarantee the consistence of D bit between TLB and
MMU page table in memory. This is especially important when clearing
the D bit, since it may cause false negatives in reporting dirtiness.
Sanity test was done on my machine with Intel processor.
Signed-off-by: Kai Huang <kai.huang@linux.intel.com>
[Check A bit too. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Adds a function kvm_vcpu_set_pending_timer instead of calling
kvm_make_request in lapic.c.
Signed-off-by: Nicholas Krause <xerofoify@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When access to descriptor in LDT/GDT wraparound outside long-mode, the address
of the descriptor should be truncated to 32-bit. Citing Intel SDM 2.1.1.1
"Global and Local Descriptor Tables in IA-32e Mode": "GDTR and LDTR registers
are expanded to 64-bits wide in both IA-32e sub-modes (64-bit mode and
compatibility mode)."
So in other cases, we need to truncate. Creating new function to return a
pointer to descriptor table to avoid too much code duplication.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
[Wrap 64-bit check with #ifdef CONFIG_X86_64, to avoid a "right shift count
>= width of type" warning and consequent undefined behavior. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When segment is loaded, the segment access bit is set unconditionally. In
fact, it should be set conditionally, based on whether the segment had the
accessed bit set before. In addition, it can improve performance.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
According to Intel SDM: "If the ESP register is used as a base register for
addressing a destination operand in memory, the POP instruction computes the
effective address of the operand after it increments the ESP register."
The current emulation does not behave so. The fix required to waste another
of the precious instruction flags and to check the flag in decode_modrm.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Currently, if em_call_far fails it returns success instead of the resulting
error-code. Fix it.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The KVM emulator does not emulate JMP and CALL that target a call gate or a
task gate. This patch does not try to implement these scenario as they are
presumably rare; yet it returns X86EMUL_UNHANDLEABLE error in such cases
instead of generating an exception.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Since the operand size of fnstcw and fnstsw is updated during the execution,
the emulation may cause spurious exceptions as it reads the memory beforehand.
Marking these instructions as Mov (since the previous value is ignored) and
DstMem16 to simplify the setting of operand size.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Although pop sreg updates RSP according to the operand size, only 2 bytes are
read. The current behavior may result in incorrect #GP or #PF exceptions.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Because ASSERT is just a printk, these would oops right away.
The assertion thus hardly adds anything.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The initialization function in mmu.c can always use walk_mmu, which
is known to be vcpu->arch.mmu. Only init_kvm_nested_mmu is used to
initialize vcpu->arch.nested_mmu.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add tracepoint to wait_lapic_expire.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
[Remind reader if early or late. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
For the hrtimer which emulates the tscdeadline timer in the guest,
add an option to advance expiration, and busy spin on VM-entry waiting
for the actual expiration time to elapse.
This allows achieving low latencies in cyclictest (or any scenario
which requires strict timing regarding timer expiration).
Reduces average cyclictest latency from 12us to 8us
on Core i5 desktop.
Note: this option requires tuning to find the appropriate value
for a particular hardware/guest combination. One method is to measure the
average delay between apic_timer_fn and VM-entry.
Another method is to start with 1000ns, and increase the value
in say 500ns increments until avg cyclictest numbers stop decreasing.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
kvm_x86_ops->test_posted_interrupt() returns true/false depending
whether 'vector' is set.
Next patch makes use of this interface.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In most cases calling hwapic_isr_update(), we always check if
kvm_apic_vid_enabled() == 1, but actually,
kvm_apic_vid_enabled()
-> kvm_x86_ops->vm_has_apicv()
-> vmx_vm_has_apicv() or '0' in svm case
-> return enable_apicv && irqchip_in_kernel(kvm)
So its a little cost to recall vmx_vm_has_apicv() inside
hwapic_isr_update(), here just NULL out hwapic_isr_update() in
case of !enable_apicv inside hardware_setup() then make all
related stuffs follow this. Note we don't check this under that
condition of irqchip_in_kernel() since we should make sure
definitely any caller don't work without in-kernel irqchip.
Signed-off-by: Tiejun Chen <tiejun.chen@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Remove FIXME comments about needing fault addresses to be returned. These
are propaagated from walk_addr_generic to gva_to_gpa and from there to
ops->read_std and ops->write_std.
Signed-off-by: Nicholas Krause <xerofoify@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When generating #PF VM-exit, check equality:
(PFEC & PFEC_MASK) == PFEC_MATCH
If there is equality, the 14 bit of exception bitmap is used to take decision
about generating #PF VM-exit. If there is inequality, inverted 14 bit is used.
Signed-off-by: Eugene Korenevsky <ekorenevsky@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch improve checks required by Intel Software Developer Manual.
- SMM MSRs are not allowed.
- microcode MSRs are not allowed.
- check x2apic MSRs only when LAPIC is in x2apic mode.
- MSR switch areas must be aligned to 16 bytes.
- address of first and last byte in MSR switch areas should not set any bits
beyond the processor's physical-address width.
Also it adds warning messages on failures during MSR switch. These messages
are useful for people who debug their VMMs in nVMX.
Signed-off-by: Eugene Korenevsky <ekorenevsky@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Several hypervisors need MSR auto load/restore feature.
We read MSRs from VM-entry MSR load area which specified by L1,
and load them via kvm_set_msr in the nested entry.
When nested exit occurs, we get MSRs via kvm_get_msr, writing
them to L1`s MSR store area. After this, we read MSRs from VM-exit
MSR load area, and load them via kvm_set_msr.
Signed-off-by: Wincy Van <fanwenyi0529@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
SRCU is not necessary to be compiled by default in all cases. For tinification
efforts not compiling SRCU unless necessary is desirable.
The current patch tries to make compiling SRCU optional by introducing a new
Kconfig option CONFIG_SRCU which is selected when any of the components making
use of SRCU are selected.
If we do not select CONFIG_SRCU, srcu.o will not be compiled at all.
text data bss dec hex filename
2007 0 0 2007 7d7 kernel/rcu/srcu.o
Size of arch/powerpc/boot/zImage changes from
text data bss dec hex filename
831552 64180 23944 919676 e087c arch/powerpc/boot/zImage : before
829504 64180 23952 917636 e0084 arch/powerpc/boot/zImage : after
so the savings are about ~2000 bytes.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
CC: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
CC: Josh Triplett <josh@joshtriplett.org>
CC: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
[ paulmck: resolve conflict due to removal of arch/ia64/kvm/Kconfig. ]
Since most virtual machines raise this message once, it is a bit annoying.
Make it KERN_DEBUG severity.
Cc: stable@vger.kernel.org
Fixes: 7a2e8aaf0f
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The commit 34a1cd60d1, "x86: vmx: move some vmx setting from
vmx_init() to hardware_setup()", tried to refactor some codes
specific to vmx hardware setting into hardware_setup(), but some
msr writing should depend on our previous setting condition like
enable_apicv, enable_ept and so on.
Reported-by: Jamie Heilman <jamie@audible.transient.net>
Tested-by: Jamie Heilman <jamie@audible.transient.net>
Signed-off-by: Tiejun Chen <tiejun.chen@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
- spring cleaning: removed support for IA64, and for hardware-assisted
virtualization on the PPC970
- ARM, PPC, s390 all had only small fixes
For x86:
- small performance improvements (though only on weird guests)
- usual round of hardware-compliancy fixes from Nadav
- APICv fixes
- XSAVES support for hosts and guests. XSAVES hosts were broken because
the (non-KVM) XSAVES patches inadvertently changed the KVM userspace
ABI whenever XSAVES was enabled; hence, this part is going to stable.
Guest support is just a matter of exposing the feature and CPUID leaves
support.
Right now KVM is broken for PPC BookE in your tree (doesn't compile).
I'll reply to the pull request with a patch, please apply it either
before the pull request or in the merge commit, in order to preserve
bisectability somewhat.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)
iQEcBAABAgAGBQJUkpg+AAoJEL/70l94x66DUmoH/jzXYkptSW9NGgm79KqxGJlD
lzLnLBkitVvx++Mz5YBhdJEhKKLUlCtifFT1zPJQ/pthQhIRSaaAwZyNGgUs5w5x
yMGKHiPQFyZRbmQtZhCInW0BftJoYHHciO3nUfHCZnp34My9MP2D55W7/z+fYFfQ
DuqBSE9ThyZJtZ4zh8NRA9fCOeuqwVYRyoBs820Wbsh4cpIBoIK63Dg7k+CLE+ZV
MZa/mRL6bAfsn9W5bnOUAgHJ3SPznnWbO3/g0aV+roL/5pffblprJx9lKNR08xUM
6hDFLop2gDehDJesDkY/o8Ckp1hEouvfsVpSShry4vcgtn0hgh2O5/6Orbmj6vE=
=Zwq1
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM update from Paolo Bonzini:
"3.19 changes for KVM:
- spring cleaning: removed support for IA64, and for hardware-
assisted virtualization on the PPC970
- ARM, PPC, s390 all had only small fixes
For x86:
- small performance improvements (though only on weird guests)
- usual round of hardware-compliancy fixes from Nadav
- APICv fixes
- XSAVES support for hosts and guests. XSAVES hosts were broken
because the (non-KVM) XSAVES patches inadvertently changed the KVM
userspace ABI whenever XSAVES was enabled; hence, this part is
going to stable. Guest support is just a matter of exposing the
feature and CPUID leaves support"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (179 commits)
KVM: move APIC types to arch/x86/
KVM: PPC: Book3S: Enable in-kernel XICS emulation by default
KVM: PPC: Book3S HV: Improve H_CONFER implementation
KVM: PPC: Book3S HV: Fix endianness of instruction obtained from HEIR register
KVM: PPC: Book3S HV: Remove code for PPC970 processors
KVM: PPC: Book3S HV: Tracepoints for KVM HV guest interactions
KVM: PPC: Book3S HV: Simplify locking around stolen time calculations
arch: powerpc: kvm: book3s_paired_singles.c: Remove unused function
arch: powerpc: kvm: book3s_pr.c: Remove unused function
arch: powerpc: kvm: book3s.c: Remove some unused functions
arch: powerpc: kvm: book3s_32_mmu.c: Remove unused function
KVM: PPC: Book3S HV: Check wait conditions before sleeping in kvmppc_vcore_blocked
KVM: PPC: Book3S HV: ptes are big endian
KVM: PPC: Book3S HV: Fix inaccuracies in ICP emulation for H_IPI
KVM: PPC: Book3S HV: Fix KSM memory corruption
KVM: PPC: Book3S HV: Fix an issue where guest is paused on receiving HMI
KVM: PPC: Book3S HV: Fix computation of tlbie operand
KVM: PPC: Book3S HV: Add missing HPTE unlock
KVM: PPC: BookE: Improve irq inject tracepoint
arm/arm64: KVM: Require in-kernel vgic for the arch timers
...
problems, clarifies VCPU init, and fixes a regression concerning the
VGIC init flow.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJUjsVhAAoJEEtpOizt6ddy5rIH/1V/YVwhprC55YqdHelU9Qu2
Muzsx+7F71NxC7xgMGFqPD1YrPR+hxvoPhy+ADOBlvcqlolrkDnV9I+8e3geaYNc
nZ/yEnoGTtbAggiS1smx7usBv34Z88Sd5txNjmj1cmHBy+VOWlyidWMkGBTsfBRe
mVc61BDUfyC47udgRHXhwS80sbHLJHElmADisFOVmQNBYwwiHiTdx0hMBMnHcC3Y
/3T0tKxHdeTISnmA+J+n7TcChtTIM4xqC6kwf3rw3b7XX8gdtTKylDHX2GLAg646
RdebAG2twmGpIc6SxXZbo38f3oY9OFo1Le5xZGa6iUjD56VDw/e4wg4iA2juo0Y=
=J2Ut
-----END PGP SIGNATURE-----
Merge tag 'kvm-arm-for-3.19-take2' of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD
Second round of changes for KVM for arm/arm64 for v3.19; fixes reboot
problems, clarifies VCPU init, and fixes a regression concerning the
VGIC init flow.
Conflicts:
arch/ia64/kvm/kvm-ia64.c [deleted in HEAD and modified in kvmarm]
commit d50eaa1803 ("KVM: x86: Perform limit checks when assigning EIP")
mistakenly used zero as cpl on em_ret_far. Use the actual one.
Fixes: d50eaa1803
Cc: stable@vger.kernel.org
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If L0 has disabled EPT, don't advertise unrestricted
mode at all since it depends on EPT to run real mode code.
Fixes: 92fbc7b195
Cc: stable@vger.kernel.org
Reviewed-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Bandan Das <bsd@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
to the trace_seq code. It also removed the return values to the
trace_seq_*() functions and use trace_seq_has_overflowed() to see if
the buffer filled up or not. This is similar to work being done to the
seq_file code as well in another tree.
Some of the other goodies include:
o Added some "!" (NOT) logic to the tracing filter.
o Fixed the frame pointer logic to the x86_64 mcount trampolines
o Added the logic for dynamic trampolines on !CONFIG_PREEMPT systems.
That is, the ftrace trampoline can be dynamically allocated
and be called directly by functions that only have a single hook
to them.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJUhbLGAAoJEEjnJuOKh9ldRV4H/3NcLbgGB2iu96la1zdYE6pG
Q7cDJMxXK80YIIL70h9G0IItcD4t62LMb72lfBnMGRj3msgFb3AgISW57EuI0Pxk
xk24wuIPoTG2S7v9sc3SboNFwO8qbtIjxD2OBmqIUrGo2sZIiGjyj3gX7mCY3uzL
WB2bUOSFz/22OgaANinR5EELHA3pZZCf54Vz1K9ndmtK0xp0j1a7xJShD6TrMdYv
mZ3zH5ViIhW4A3mdcMceh6fy2JLQAiEKF0uPTvcMMz7NlVul0mxyL/+10P7AE/3R
Ehw4fzmm4NDshPDtBOkKH0LsppgXzuItFuQUTpact3JlqTg++bV6onSsrkt1hlY=
=Z7Cm
-----END PGP SIGNATURE-----
Merge tag 'trace-3.19' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing updates from Steven Rostedt:
"There was a lot of clean ups and minor fixes. One of those clean ups
was to the trace_seq code. It also removed the return values to the
trace_seq_*() functions and use trace_seq_has_overflowed() to see if
the buffer filled up or not. This is similar to work being done to
the seq_file code as well in another tree.
Some of the other goodies include:
- Added some "!" (NOT) logic to the tracing filter.
- Fixed the frame pointer logic to the x86_64 mcount trampolines
- Added the logic for dynamic trampolines on !CONFIG_PREEMPT systems.
That is, the ftrace trampoline can be dynamically allocated and be
called directly by functions that only have a single hook to them"
* tag 'trace-3.19' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (55 commits)
tracing: Truncated output is better than nothing
tracing: Add additional marks to signal very large time deltas
Documentation: describe trace_buf_size parameter more accurately
tracing: Allow NOT to filter AND and OR clauses
tracing: Add NOT to filtering logic
ftrace/fgraph/x86: Have prepare_ftrace_return() take ip as first parameter
ftrace/x86: Get rid of ftrace_caller_setup
ftrace/x86: Have save_mcount_regs macro also save stack frames if needed
ftrace/x86: Add macro MCOUNT_REG_SIZE for amount of stack used to save mcount regs
ftrace/x86: Simplify save_mcount_regs on getting RIP
ftrace/x86: Have save_mcount_regs store RIP in %rdi for first parameter
ftrace/x86: Rename MCOUNT_SAVE_FRAME and add more detailed comments
ftrace/x86: Move MCOUNT_SAVE_FRAME out of header file
ftrace/x86: Have static tracing also use ftrace_caller_setup
ftrace/x86: Have static function tracing always test for function graph
kprobes: Add IPMODIFY flag to kprobe_ftrace_ops
ftrace, kprobes: Support IPMODIFY flag to find IP modify conflict
kprobes/ftrace: Recover original IP if pre_handler doesn't change it
tracing/trivial: Fix typos and make an int into a bool
tracing: Deletion of an unnecessary check before iput()
...
Intel SDM table 6-2 ("Priority Among Simultaneous Exceptions and Interrupts")
shows that faults from decoding the next instruction got higher priority than
general protection. Moving the protected-mode check before the CPL check to
avoid wrong exception on vm86 mode.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The pushf instruction does not push eflags.VM, so emulation should not do so as
well. Although eflags.RF should not be pushed as well, it is already cleared
by the time pushf is executed.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The macro GP already sets the flag Prefix. Remove the redundant flag for
0f_38_f0 and 0f_38_f1 opcodes.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We reused host EBX and ECX, but KVM might not support all features;
emulated XSAVE size should be smaller.
EBX depends on unknown XCR0, so we default to ECX.
SDM CPUID (EAX = 0DH, ECX = 0):
EBX Bits 31-00: Maximum size (bytes, from the beginning of the
XSAVE/XRSTOR save area) required by enabled features in XCR0. May
be different than ECX if some features at the end of the XSAVE save
area are not enabled.
ECX Bit 31-00: Maximum size (bytes, from the beginning of the
XSAVE/XRSTOR save area) of the XSAVE/XRSTOR save area required by
all supported features in the processor, i.e all the valid bit
fields in XCR0.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Tested-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add nested virtualization support for xsaves.
Signed-off-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add logic to get/set the XSS model-specific register.
Signed-off-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Initialize the XSS exit bitmap. It is zero so there should be no XSAVES
or XRSTORS exits.
Signed-off-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
- EAX=0Dh, ECX=1: output registers EBX/ECX/EDX are reserved.
- EAX=0Dh, ECX>1: output register ECX bit 0 is clear for all the CPUID
leaves we support, because variable "supported" comes from XCR0 and not
XSS. Bits above 0 are reserved, so ECX is overall zero. Output register
EDX is reserved.
Source: Intel Architecture Instruction Set Extensions Programming
Reference, ref. number 319433-022
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Tested-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This is the size of the XSAVES area. This starts providing guest support
for XSAVES (with no support yet for supervisor states, i.e. XSS == 0
always in guests for now).
Wanpeng Li suggested testing XSAVEC as well as XSAVES, since in practice
no real processor exists that only has one of them, and there is no
other way for userspace programs to compute the area of the XSAVEC
save area. CPUID(EAX=0xd,ECX=1).EBX provides an upper bound.
Suggested-by: Radim Krčmář <rkrcmar@redhat.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Tested-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Expose the XSAVES feature to the guest if the kvm_x86_ops say it is
available.
Signed-off-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Userspace is expecting non-compacted format for KVM_GET_XSAVE, but
struct xsave_struct might be using the compacted format. Convert
in order to preserve userspace ABI.
Likewise, userspace is passing non-compacted format for KVM_SET_XSAVE
but the kernel will pass it to XRSTORS, and we need to convert back.
Fixes: f31a9f7c71
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: stable@vger.kernel.org
Cc: H. Peter Anvin <hpa@linux.intel.com>
Tested-by: Nadav Amit <namit@cs.technion.ac.il>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
While fixing an x2apic bug,
17d68b7 KVM: x86: fix guest-initiated crash with x2apic (CVE-2013-6376)
we've made only one cluster available. This means that the amount of
logically addressible x2APICs was reduced to 16 and VCPUs kept
overwriting themselves in that region, so even the first cluster wasn't
set up correctly.
This patch extends x2APIC support back to the logical_map's limit, and
keeps the CVE fixed as messages for non-present APICs are dropped.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
They can't be violated now, but play it safe for the future.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
x2apic allows destinations > 0xff and we don't want them delivered to
lower APICs. They are correctly handled by doing nothing.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Physical mode can't address more than one APIC, but lowest-prio is
allowed, so we just reuse our paths.
SDM 10.6.2.1 Physical Destination:
Also, for any non-broadcast IPI or I/O subsystem initiated interrupt
with lowest priority delivery mode, software must ensure that APICs
defined in the interrupt address are present and enabled to receive
interrupts.
We could warn on top of that.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
False from kvm_irq_delivery_to_apic_fast() means that we don't handle it
in the fast path, but we still return false in cases that were perfectly
handled, fix that.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
0x830 MSR is 0x300 xAPIC MMIO, which is MSR_ICR.
Signed-off-by: Radim KrÄmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
x2APIC has no registers for DFR and ICR2 (see Intel SDM 10.12.1.2 "x2APIC
Register Address Space"). KVM needs to cause #GP on such accesses.
Fix it (DFR and ICR2 on read, ICR2 on write, DFR already handled on writes).
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Certain x86 instructions that use modrm operands only allow memory operand
(i.e., mod012), and cause a #UD exception otherwise. KVM ignores this fact.
Currently, the instructions that are such and are emulated by KVM are MOVBE,
MOVNTPS, MOVNTPD and MOVNTI. MOVBE is the most blunt example, since it may be
emulated by the host regardless of MMIO.
The fix introduces a new group for handling such instructions, marking mod3 as
illegal instruction.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This reverts commit 85c8555ff0 ("KVM: check for !is_zero_pfn() in
kvm_is_mmio_pfn()") and renames the function to kvm_is_reserved_pfn.
The problem being addressed by the patch above was that some ARM code
based the memory mapping attributes of a pfn on the return value of
kvm_is_mmio_pfn(), whose name indeed suggests that such pfns should
be mapped as device memory.
However, kvm_is_mmio_pfn() doesn't do quite what it says on the tin,
and the existing non-ARM users were already using it in a way which
suggests that its name should probably have been 'kvm_is_reserved_pfn'
from the beginning, e.g., whether or not to call get_page/put_page on
it etc. This means that returning false for the zero page is a mistake
and the patch above should be reverted.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This reverts commit 85c8555ff0 ("KVM: check for !is_zero_pfn() in
kvm_is_mmio_pfn()") and renames the function to kvm_is_reserved_pfn.
The problem being addressed by the patch above was that some ARM code
based the memory mapping attributes of a pfn on the return value of
kvm_is_mmio_pfn(), whose name indeed suggests that such pfns should
be mapped as device memory.
However, kvm_is_mmio_pfn() doesn't do quite what it says on the tin,
and the existing non-ARM users were already using it in a way which
suggests that its name should probably have been 'kvm_is_reserved_pfn'
from the beginning, e.g., whether or not to call get_page/put_page on
it etc. This means that returning false for the zero page is a mistake
and the patch above should be reverted.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
cs.base is declared as a __u64 variable and vector is a u32 so this
causes a static checker warning. The user indeed can set "sipi_vector"
to any u32 value in kvm_vcpu_ioctl_x86_set_vcpu_events(), but the
value should really have 8-bit precision only.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Create a new header, and hide the device assignment functions there.
Move struct kvm_assigned_dev_kernel to assigned-dev.c by modifying
arch/x86/kvm/iommu.c to take a PCI device struct.
Based on a patch by Radim Krcmar <rkrcmark@redhat.com>.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This feature is not supported inside KVM guests yet, because we do not emulate
MSR_IA32_XSS. Mask it out.
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Now that ia64 is gone, we can hide deprecated device assignment in x86.
Notable changes:
- kvm_vm_ioctl_assigned_device() was moved to x86/kvm_arch_vm_ioctl()
The easy parts were removed from generic kvm code, remaining
- kvm_iommu_(un)map_pages() would require new code to be moved
- struct kvm_assigned_dev_kernel depends on struct kvm_irq_ack_notifier
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
ia64 does not need them anymore. Ack notifiers become x86-specific
too.
Suggested-by: Gleb Natapov <gleb@kernel.org>
Reviewed-by: Radim Krcmar <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
To allow for the restructiong of the trace_seq code, we need users
of it to use the helper functions instead of accessing the internals
of the trace_seq structure itself.
Link: http://lkml.kernel.org/r/20141104160221.585025609@goodmis.org
Tested-by: Jiri Kosina <jkosina@suse.cz>
Acked-by: Jiri Kosina <jkosina@suse.cz>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: Mark Rustad <mark.d.rustad@intel.com>
Reviewed-by: Petr Mladek <pmladek@suse.cz>
Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Remove FIXME comments about needing fault addresses to be returned. These
are propaagated from walk_addr_generic to gva_to_gpa and from there to
ops->read_std and ops->write_std.
Signed-off-by: Nicholas Krause <xerofoify@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The check on the higher limit of the segment, and the check on the
maximum accessible size, is the same for both expand-up and
expand-down segments. Only the computation of "lim" varies.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
register_address has been a duplicate of address_mask ever since the
ancestor of __linearize was born in 90de84f50b (KVM: x86 emulator:
preserve an operand's segment identity, 2010-11-17).
However, we can put it to a better use by including the call to reg_read
in register_address. Similarly, the call to reg_rmw can be moved to
register_address_increment.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In __linearize there is check of the condition whether to check if masking of
the linear address is needed. It occurs immediately after switch that
evaluates the same condition. Merge them.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When SS is used using a non-canonical address, an #SS exception is generated on
real hardware. KVM emulator causes a #GP instead. Fix it to behave as real x86
CPU.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If branch (e.g., jmp, ret) causes limit violations, since the target IP >
limit, the #GP exception occurs before the branch. In other words, the RIP
pushed on the stack should be that of the branch and not that of the target.
To do so, we can call __linearize, with new EIP, which also saves us the code
which performs the canonical address checks. On the case of assigning an EIP >=
2^32 (when switching cs.l), we also safe, as __linearize will check the new EIP
does not exceed the limit and would trigger #GP(0) otherwise.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When segment is accessed, real hardware does not perform any privilege level
checks. In contrast, KVM emulator does. This causes some discrepencies from
real hardware. For instance, reading from readable code segment may fail due to
incorrect segment checks. In addition, it introduces unnecassary overhead.
To reference Intel SDM 5.5 ("Privilege Levels"): "Privilege levels are checked
when the segment selector of a segment descriptor is loaded into a segment
register." The SDM never mentions privilege level checks during memory access,
except for loading far pointers in section 5.10 ("Pointer Validation"). Those
are actually segment selector loads and are emulated in the similarily (i.e.,
regardless to __linearize checks).
This behavior was also checked using sysexit. A data-segment whose DPL=0 was
loaded, and after sysexit (CPL=3) it is still accessible.
Therefore, all the privilege level checks in __linearize are removed.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When performing segmented-read/write in the emulator for stack operations, it
ignores the stack size, and uses the ad_bytes as indication for the pointer
size. As a result, a wrong address may be accessed.
To fix this behavior, we can remove the masking of address in __linearize and
perform it beforehand. It is already done for the operands (so currently it is
inefficiently done twice). It is missing in two cases:
1. When using rip_relative
2. On fetch_bit_operand that changes the address.
This patch masks the address on these two occassions, and removes the masking
from __linearize.
Note that it does not mask EIP during fetch. In protected/legacy mode code
fetch when RIP >= 2^32 should result in #GP and not wrap-around. Since we make
limit checks within __linearize, this is the expected behavior.
Partial revert of commit 518547b32a (KVM: x86: Emulator does not
calculate address correctly, 2014-09-30).
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit 10e38fc7cab6 ("KVM: x86: Emulator flag for instruction that only support
16-bit addresses in real mode") introduced NoBigReal for instructions such as
MONITOR. Apparetnly, the Intel SDM description that led to this patch is
misleading. Since no instruction is using NoBigReal, it is safe to remove it,
we fully understand what the SDM means.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
MMIO_MAX_GEN is the same as MMIO_GEN_MASK. Use only one.
Signed-off-by: Tiejun Chen <tiejun.chen@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Instead, just use PFERR_{FETCH, PRESENT, WRITE}_MASK
inside handle_ept_violation() for slightly better code.
Signed-off-by: Tiejun Chen <tiejun.chen@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
apic_find_highest_irr assumes irr_pending is set if any vector in APIC_IRR is
set. If this assumption is broken and apicv is disabled, the injection of
interrupts may be deferred until another interrupt is delivered to the guest.
Ultimately, if no other interrupt should be injected to that vCPU, the pending
interrupt may be lost.
commit 56cc2406d6 ("KVM: nVMX: fix "acknowledge interrupt on exit" when APICv
is in use") changed the behavior of apic_clear_irr so irr_pending is cleared
after setting APIC_IRR vector. After this commit, if apic_set_irr and
apic_clear_irr run simultaneously, a race may occur, resulting in APIC_IRR
vector set, and irr_pending cleared. In the following example, assume a single
vector is set in IRR prior to calling apic_clear_irr:
apic_set_irr apic_clear_irr
------------ --------------
apic->irr_pending = true;
apic_clear_vector(...);
vec = apic_search_irr(apic);
// => vec == -1
apic_set_vector(...);
apic->irr_pending = (vec != -1);
// => apic->irr_pending == false
Nonetheless, it appears the race might even occur prior to this commit:
apic_set_irr apic_clear_irr
------------ --------------
apic->irr_pending = true;
apic->irr_pending = false;
apic_clear_vector(...);
if (apic_search_irr(apic) != -1)
apic->irr_pending = true;
// => apic->irr_pending == false
apic_set_vector(...);
Fixing this issue by:
1. Restoring the previous behavior of apic_clear_irr: clear irr_pending, call
apic_clear_vector, and then if APIC_IRR is non-zero, set irr_pending.
2. On apic_set_irr: first call apic_set_vector, then set irr_pending.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Logical destination mode can be used to send NMI IPIs even when all
APICs are software disabled, so if all APICs are software disabled we
should still look at the DFRs.
So the DFRs should all be the same, even if some or all APICs are
software disabled. However, the SDM does not say this, so tweak
the logic as follows:
- if one APIC is enabled and has LDR != 0, use that one to build the map.
This picks the right DFR in case an OS is only setting it for the
software-enabled APICs, or in case an OS is using logical addressing
on some APICs while leaving the rest in reset state (using LDR was
suggested by Radim).
- if all APICs are disabled, pick a random one to build the map.
We use the last one with LDR != 0 for simplicity.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Currently, the APIC logical map does not consider VCPUs whose local-apic is
software-disabled. However, NMIs, INIT, etc. should still be delivered to such
VCPUs. Therefore, the APIC mode should first be determined, and then the map,
considering all VCPUs should be constructed.
To address this issue, first find the APIC mode, and only then construct the
logical map.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When running the tsc_adjust kvm-unit-test on an AMD processor with the
IA32_TSC_ADJUST feature enabled, the WARN_ON in svm_adjust_tsc_offset can be
triggered. This WARN_ON checks for a negative adjustment in case __scale_tsc
is called; however it may trigger unnecessary warnings.
This patch moves the WARN_ON to trigger only if __scale_tsc will actually be
called from svm_adjust_tsc_offset. In addition make adj in kvm_set_msr_common
s64 since this can have signed values.
Signed-off-by: Chris J Arges <chris.j.arges@canonical.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
There's nothing to switch if the host and guest values are the same.
I am unable to find evidence that this makes any difference
whatsoever.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
[I could see a difference on Nehalem. From 5 runs:
userspace exit, guest!=host 12200 11772 12130 12164 12327
userspace exit, guest=host 11983 11780 11920 11919 12040
lightweight exit, guest!=host 3214 3220 3238 3218 3337
lightweight exit, guest=host 3178 3193 3193 3187 3220
This passes the t-test with 99% confidence for userspace exit,
98.5% confidence for lightweight exit. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
At least on Sandy Bridge, letting the CPU switch IA32_EFER is much
faster than switching it manually.
I benchmarked this using the vmexit kvm-unit-test (single run, but
GOAL multiplied by 5 to do more iterations):
Test Before After Change
cpuid 2000 1932 -3.40%
vmcall 1914 1817 -5.07%
mov_from_cr8 13 13 0.00%
mov_to_cr8 19 19 0.00%
inl_from_pmtimer 19164 10619 -44.59%
inl_from_qemu 15662 10302 -34.22%
inl_from_kernel 3916 3802 -2.91%
outl_to_kernel 2230 2194 -1.61%
mov_dr 172 176 2.33%
ipi (skipped) (skipped)
ipi+halt (skipped) (skipped)
ple-round-robin 13 13 0.00%
wr_tsc_adjust_msr 1920 1845 -3.91%
rd_tsc_adjust_msr 1892 1814 -4.12%
mmio-no-eventfd:pci-mem 16394 11165 -31.90%
mmio-wildcard-eventfd:pci-mem 4607 4645 0.82%
mmio-datamatch-eventfd:pci-mem 4601 4610 0.20%
portio-no-eventfd:pci-io 11507 7942 -30.98%
portio-wildcard-eventfd:pci-io 2239 2225 -0.63%
portio-datamatch-eventfd:pci-io 2250 2234 -0.71%
I haven't explicitly computed the significance of these numbers,
but this isn't subtle.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
[The results were reproducible on all of Nehalem, Sandy Bridge and
Ivy Bridge. The slowness of manual switching is because writing
to EFER with WRMSR triggers a TLB flush, even if the only bit you're
touching is SCE (so the page table format is not affected). Doing
the write as part of vmentry/vmexit, instead, does not flush the TLB,
probably because all processors that have EPT also have VPID. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
PCIDs are only supported in 64-bit mode. No need to clear bit 63
of CR3 unless the host is 64-bit.
Reported by Fengguang Wu's autobuilder.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The new trace event records:
* the id of vcpu being updated
* the pvclock_vcpu_time_info struct being written to guest memory
This is useful for debugging pvclock bugs, such as the bug fixed by
"[PATCH] kvm: x86: Fix kvm clock versioning.".
Signed-off-by: David Matlack <dmatlack@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
kvm updates the version number for the guest paravirt clock structure by
incrementing the version of its private copy. It does not read the guest
version, so will write version = 2 in the first update for every new VM,
including after restoring a saved state. If guest state is saved during
reading the clock, it could read and accept struct fields and guest TSC
from two different updates. This changes the code to increment the guest
version and write it back.
Signed-off-by: Owen Hofmann <osh@google.com>
Reviewed-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit 3b32004a66 ("KVM: x86: movnti minimum op size of 32-bit is not kept")
did not fully fix the minimum operand size of MONTI emulation. Still, MOVNTI
may be mistakenly performed using 16-bit opsize.
This patch add No16 flag to mark an instruction does not support 16-bits
operand size.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When the guest writes to the TSC, the masterclock TSC copy must be
updated as well along with the TSC_OFFSET update, otherwise a negative
tsc_timestamp is calculated at kvm_guest_time_update.
Once "if (!vcpus_matched && ka->use_master_clock)" is simplified to
"if (ka->use_master_clock)", the corresponding "if (!ka->use_master_clock)"
becomes redundant, so remove the do_request boolean and collapse
everything into a single condition.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Now that KVM injects #UD on "unhandlable" error, it makes better sense to
return such error on sysenter instead of directly injecting #UD to the guest.
This allows to track more easily the unhandlable cases the emulator does not
support.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
APIC base relocation is unsupported by KVM. If anyone uses it, the least should
be to report a warning in the hypervisor.
Note that KVM-unit-tests uses this feature for some reason, so running the
tests triggers the warning.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit 7fe864dc94 (KVM: x86: Mark VEX-prefix instructions emulation as
unimplemented, 2014-06-02) marked VEX instructions as such in protected
mode. VEX-prefix instructions are not supported relevant on real-mode
and VM86, but should cause #UD instead of being decoded as LES/LDS.
Fix this behaviour to be consistent with real hardware.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
[Check for mod == 3, rather than 2 or 3. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Task-switch emulation checks the privilege level prior to performing the
task-switch. This check is incorrect in the case of task-gates, in which the
tss.dpl is ignored, and can cause superfluous exceptions. Moreover this check
is unnecassary, since the CPU checks the privilege levels prior to exiting.
Intel SDM 25.4.2 says "If CALL or JMP accesses a TSS descriptor directly
outside IA-32e mode, privilege levels are checked on the TSS descriptor" prior
to exiting. AMD 15.14.1 says "The intercept is checked before the task switch
takes place but after the incoming TSS and task gate (if one was involved) have
been checked for correctness."
This patch removes the CPL checks for CALL and JMP.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When emulating LTR/LDTR/LGDT/LIDT, #GP should be injected if the base is
non-canonical. Otherwise, VM-entry will fail.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
LGDT and LIDT emulation logic is almost identical. Merge the logic into a
single point to avoid redundancy. This will be used by the next patch that
will ensure the bases of the loaded GDTR and IDTR are canonical.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If the emulation ends in fault, eflags should not be updated. However, several
instruction emulations (actually all the fastops) currently update eflags, if
the fault was detected afterwards (e.g., #PF during writeback).
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Although Intel SDM mentions bit 63 is reserved, MOV to CR3 can have bit 63 set.
As Intel SDM states in section 4.10.4 "Invalidation of TLBs and
Paging-Structure Caches": " MOV to CR3. ... If CR4.PCIDE = 1 and bit 63 of the
instruction’s source operand is 0 ..."
In other words, bit 63 is not reserved. KVM emulator currently consider bit 63
as reserved. Fix it.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
According to Intel SDM push of segment selectors is done in the following
manner: "if the operand size is 32-bits, either a zero-extended value is pushed
on the stack or the segment selector is written on the stack using a 16-bit
move. For the last case, all recent Core and Atom processors perform a 16-bit
move, leaving the upper portion of the stack location unmodified."
This patch modifies the behavior to match the core behavior.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
CMPS and SCAS instructions are evaluated in the wrong order. For reference (of
CMPS), see http://www.fermimn.gov.it/linux/quarta/x86/cmps.htm : "Note that the
direction of subtraction for CMPS is [SI] - [DI] or [ESI] - [EDI]. The left
operand (SI or ESI) is the source and the right operand (DI or EDI) is the
destination. This is the reverse of the usual Intel convention in which the
left operand is the destination and the right operand is the source."
Introducing em_cmp_r for this matter that performs comparison in reverse order
using fastop infrastructure to avoid a wrapper function.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
SYSCALL emulation currently clears in 64-bit mode eflags according to
MSR_SYSCALL_MASK. However, on bare-metal eflags[1] which is fixed to one
cannot be cleared, even if MSR_SYSCALL_MASK masks the bit. This wrong behavior
may result in failed VM-entry, as VT disallows entry with eflags[1] cleared.
This patch sets the bit after masking eflags on syscall.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In x86, you can only MOV-sreg to memory with either 16-bits or 64-bits size.
In contrast, KVM may write to 32-bits memory on MOV-sreg. This patch fixes KVM
behavior, and sets the destination operand size to two, if the destination is
memory.
When destination is registers, and the operand size is 32-bits, the high
16-bits in modern CPUs is filled with zero. This is handled correctly.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
x86 debug registers hold a linear address. Therefore, breakpoints detection
should consider CS.base, and check whether instruction linear address equals
(CS.base + RIP). This patch introduces a function to evaluate RIP linear
address and uses it for breakpoints detection.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
DR6[0:3] (previous breakpoint indications) are cleared when #DB is injected
during handle_exception, just as real hardware does. Similarily, handle_dr
should clear DR6[0:3].
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Real-mode exceptions do not deliver error code. As can be seen in Intel SDM
volume 2, real-mode exceptions do not have parentheses, which indicate
error-code. To avoid significant changes of the code, the error code is
"removed" during exception queueing.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In one occassion, decode_modrm uses the rm field after it is extended with
REX.B to determine the addressing mode. Doing so causes it not to read the
offset for rip-relative addressing with REX.B=1.
This patch moves the fetch where we already mask REX.B away instead.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A bug was reported as follows: when running Windows 7 32-bit guests on qemu-kvm,
sometimes the guests run into blue screen during reboot. The problem was that a
guest's RVI was not cleared when it rebooted. This patch has fixed the problem.
Signed-off-by: Wei Wang <wei.w.wang@intel.com>
Signed-off-by: Yang Zhang <yang.z.zhang@intel.com>
Tested-by: Rongrong Liu <rongrongx.liu@intel.com>, Da Chun <ngugc@qq.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Return a negative error code instead, and WARN() when we should be covering
the entire 2-bit space of vmcs_field_type's return value. For increased
robustness, add a BUILD_BUG_ON checking the range of vmcs_field_to_offset.
Suggested-by: Tiejun Chen <tiejun.chen@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Instead of vmx_init(), actually it would make reasonable sense to do
anything specific to vmx hardware setting in vmx_x86_ops->hardware_setup().
Signed-off-by: Tiejun Chen <tiejun.chen@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Just move this pair of functions down to make sure later we can
add something dependent on others.
Signed-off-by: Tiejun Chen <tiejun.chen@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The emulator could reuse an op->type from a previous instruction for some
immediate values. If it mistakenly considers the operands as memory
operands, it will performs a memory read and overwrite op->val.
Consider for instance the ROR instruction - src2 (the number of times)
would be read from memory instead of being used as immediate.
Mark every immediate operand as such to avoid this problem.
Cc: stable@vger.kernel.org
Fixes: c44b4c6ab8
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We mirror a subset of these registers in separate variables.
Using them directly should be faster.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
APIC-write VM exits are "trap-like": they save CS:RIP values for the
instruction after the write, and more importantly, the handler will
already see the new value in the virtual-APIC page. This means that
apic_reg_write cannot use kvm_apic_get_reg to omit timer cancelation
when mode changes.
timer_mode_mask shouldn't be changing as it depends on cpuid.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
APIC-write VM exits are "trap-like": they save CS:RIP values for the
instruction after the write, and more importantly, the handler will
already see the new value in the virtual-APIC page.
This caused a bug if you used KVM_SET_IRQCHIP to set the SW-enabled bit
in the SPIV register. The chain of events is as follows:
* When the irqchip is added to the destination VM, the apic_sw_disabled
static key is incremented (1)
* When the KVM_SET_IRQCHIP ioctl is invoked, it is decremented (0)
* When the guest disables the bit in the SPIV register, e.g. as part of
shutdown, apic_set_spiv does not notice the change and the static key is
_not_ incremented.
* When the guest is destroyed, the static key is decremented (-1),
resulting in this trace:
WARNING: at kernel/jump_label.c:81 __static_key_slow_dec+0xa6/0xb0()
jump label: negative count!
[<ffffffff816bf898>] dump_stack+0x19/0x1b
[<ffffffff8107c6f1>] warn_slowpath_common+0x61/0x80
[<ffffffff8107c76c>] warn_slowpath_fmt+0x5c/0x80
[<ffffffff811931e6>] __static_key_slow_dec+0xa6/0xb0
[<ffffffff81193226>] static_key_slow_dec_deferred+0x16/0x20
[<ffffffffa0637698>] kvm_free_lapic+0x88/0xa0 [kvm]
[<ffffffffa061c63e>] kvm_arch_vcpu_uninit+0x2e/0xe0 [kvm]
[<ffffffffa05ff301>] kvm_vcpu_uninit+0x21/0x40 [kvm]
[<ffffffffa067cec7>] vmx_free_vcpu+0x47/0x70 [kvm_intel]
[<ffffffffa061bc50>] kvm_arch_vcpu_free+0x50/0x60 [kvm]
[<ffffffffa061ca22>] kvm_arch_destroy_vm+0x102/0x260 [kvm]
[<ffffffff810b68fd>] ? synchronize_srcu+0x1d/0x20
[<ffffffffa06030d1>] kvm_put_kvm+0xe1/0x1c0 [kvm]
[<ffffffffa06036f8>] kvm_vcpu_release+0x18/0x20 [kvm]
[<ffffffff81215c62>] __fput+0x102/0x310
[<ffffffff81215f4e>] ____fput+0xe/0x10
[<ffffffff810ab664>] task_work_run+0xb4/0xe0
[<ffffffff81083944>] do_exit+0x304/0xc60
[<ffffffff816c8dfc>] ? _raw_spin_unlock_irq+0x2c/0x50
[<ffffffff810fd22d>] ? trace_hardirqs_on_caller+0xfd/0x1c0
[<ffffffff8108432c>] do_group_exit+0x4c/0xc0
[<ffffffff810843b4>] SyS_exit_group+0x14/0x20
[<ffffffff816d33a9>] system_call_fastpath+0x16/0x1b
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The check in kvm_set_lapic_tscdeadline_msr() was trying to prevent a
situation where we lose a pending deadline timer in a MSR write.
Losing it is fine, because it effectively occurs before the timer fired,
so we should be able to cancel or postpone it.
Another problem comes from interaction with QEMU, or other userspace
that can set deadline MSR without a good reason, when timer is already
pending: one guest's deadline request results in more than one
interrupt because one is injected immediately on MSR write from
userspace and one through hrtimer later.
The solution is to remove the injection when replacing a pending timer
and to improve the usual QEMU path, we inject without a hrtimer when the
deadline has already passed.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Reported-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Make the code reusable.
If the timer was already pending, we shouldn't be waiting in a queue,
so wake_up can be skipped, simplifying the path.
There is no 'reinject' case => the comment is removed.
Current race behaves correctly.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If DR4/5 is accessed when it is unavailable (since CR4.DE is set), then #UD
should be generated even if CPL>0. This is according to Intel SDM Table 6-2:
"Priority Among Simultaneous Exceptions and Interrupts".
Note, that this may happen on the first DR access, even if the host does not
sets debug breakpoints. Obviously, it occurs when the host debugs the guest.
This patch moves the DR4/5 checks from __kvm_set_dr/_kvm_get_dr to handle_dr.
The emulator already checks DR4/5 availability in check_dr_read. Nested
virutalization related calls to kvm_set_dr/kvm_get_dr would not like to inject
exceptions to the guest.
As for SVM, the patch follows the previous logic as much as possible. Anyhow,
it appears the DR interception code might be buggy - even if the DR access
may cause an exception, the instruction is skipped.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When read access is performed using a readable code segment, the "conforming"
and "non-conforming" checks should not be done. As a result, read using
non-conforming readable code segment fails.
This is according to Intel SDM 5.6.1 ("Accessing Data in Code Segments").
The fix is not to perform the "non-conforming" checks if the access is not a
fetch; the relevant checks are already done when loading the segment.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
DR7.LE should be cleared during task-switch. This feature is poorly documented.
For reference, see:
http://pdos.csail.mit.edu/6.828/2005/readings/i386/s12_02.htm
SDM [17.2.4]:
This feature is not supported in the P6 family processors, later IA-32
processors, and Intel 64 processors.
AMD [2:13.1.1.4]:
This bit is ignored by implementations of the AMD64 architecture.
Intel's formulation could mean that it isn't even zeroed, but current
hardware indeed does not behave like that.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In long-mode, when the address size is 4 bytes, the linear address is not
truncated as the emulator mistakenly does. Instead, the offset within the
segment (the ea field) should be truncated according to the address size.
As Intel SDM says: "In 64-bit mode, the effective address components are added
and the effective address is truncated ... before adding the full 64-bit
segment base."
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Intel SDM 17.2.4 (Debug Control Register (DR7)) says: "The processor clears the
GD flag upon entering to the debug exception handler." This sentence may be
misunderstood as if it happens only on #DB due to debug-register protection,
but it happens regardless to the cause of the #DB.
Fix the behavior to match both real hardware and Bochs.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM does not deliver x2APIC broadcast messages with physical mode. Intel SDM
(10.12.9 ICR Operation in x2APIC Mode) states: "A destination ID value of
FFFF_FFFFH is used for broadcast of interrupts in both logical destination and
physical destination modes."
In addition, the local-apic enables cluster mode broadcast. As Intel SDM
10.6.2.2 says: "Broadcast to all local APICs is achieved by setting all
destination bits to one." This patch enables cluster mode broadcast.
The fix tries to combine broadcast in different modes through a unified code.
One rare case occurs when the source of IPI has its APIC disabled. In such
case, the source can still issue IPIs, but since the source is not obliged to
have the same LAPIC mode as the enabled ones, we cannot rely on it.
Since it is a rare case, it is unoptimized and done on the slow-path.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Reviewed-by: Wanpeng Li <wanpeng.li@linux.intel.com>
[As per Radim's review, use unsigned int for X2APIC_BROADCAST, return bool from
kvm_apic_broadcast. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
CR4.TSD is guest-owned; don't trap writes to it in VMX guests. This
avoids a VM exit on context switches into or out of a PR_TSC_SIGSEGV
task.
I think that this fixes an unintentional side-effect of:
4c38609ac5 KVM: VMX: Make guest cr4 mask more conservative
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If the operand size is not 64-bit, then the sysexit instruction should assign
ECX to RSP and EDX to RIP. The current code assigns the full 64-bits.
Fix it by masking.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In 64-bit, stack operations default to 64-bits, but can be overriden (to
16-bit) using opsize override prefix. In contrast, near-branches are always
64-bit. This patch distinguish between the different behaviors.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Breaking grp45 to the relevant functions to speed up the emulation and simplify
the code. In addition, it is necassary the next patch will distinguish between
far and near branches according to the flags.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Replace the current canonical address check with the new function which is
identical.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In order to access the shadow VMCS, we need to load it. At this point,
vmx->loaded_vmcs->vmcs and the actually loaded one start to differ. If
we now get preempted by Linux, vmx_vcpu_put and, on return, the
vmx_vcpu_load will work against the wrong vmcs. That can cause
copy_shadow_to_vmcs12 to corrupt the vmcs12 state.
Fix the issue by disabling preemption during the copy operation.
copy_vmcs12_to_shadow is safe from this issue as it is executed by
vmx_vcpu_run when preemption is already disabled before vmentry.
This bug is exposed by running Jailhouse within KVM on CPUs with
shadow VMCS support. Jailhouse never expects an interrupt pending
vmexit, but the bug can cause it if, after copy_shadow_to_vmcs12
is preempted, the active VMCS happens to have the virtual interrupt
pending flag set in the CPU-based execution controls.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit d1442d85cc ("KVM: x86: Handle errors when RIP is set during far
jumps") introduced a bug that caused the fix to be incomplete. Due to
incorrect evaluation, far jump to segment with L bit cleared (i.e., 32-bit
segment) and RIP with any of the high bits set (i.e, RIP[63:32] != 0) set may
not trigger #GP. As we know, this imposes a security problem.
In addition, the condition for two warnings was incorrect.
Fixes: d1442d85cc
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
[Add #ifdef CONFIG_X86_64 to avoid complaints of undefined behavior. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In order to access the shadow VMCS, we need to load it. At this point,
vmx->loaded_vmcs->vmcs and the actually loaded one start to differ. If
we now get preempted by Linux, vmx_vcpu_put and, on return, the
vmx_vcpu_load will work against the wrong vmcs. That can cause
copy_shadow_to_vmcs12 to corrupt the vmcs12 state.
Fix the issue by disabling preemption during the copy operation.
copy_vmcs12_to_shadow is safe from this issue as it is executed by
vmx_vcpu_run when preemption is already disabled before vmentry.
This bug is exposed by running Jailhouse within KVM on CPUs with
shadow VMCS support. Jailhouse never expects an interrupt pending
vmexit, but the bug can cause it if, after copy_shadow_to_vmcs12
is preempted, the active VMCS happens to have the virtual interrupt
pending flag set in the CPU-based execution controls.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit d1442d85cc ("KVM: x86: Handle errors when RIP is set during far
jumps") introduced a bug that caused the fix to be incomplete. Due to
incorrect evaluation, far jump to segment with L bit cleared (i.e., 32-bit
segment) and RIP with any of the high bits set (i.e, RIP[63:32] != 0) set may
not trigger #GP. As we know, this imposes a security problem.
In addition, the condition for two warnings was incorrect.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
[Add #ifdef CONFIG_X86_64 to avoid complaints of undefined behavior. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Emulation of code that is 14 bytes to the segment limit or closer
(e.g. RIP = 0xFFFFFFF2 after reset) is broken because we try to read as
many as 15 bytes from the beginning of the instruction, and __linearize
fails when the passed (address, size) pair reaches out of the segment.
To fix this, let __linearize return the maximum accessible size (clamped
to 2^32-1) for usage in __do_insn_fetch_bytes, and avoid the limit check
by passing zero for the desired size.
For expand-down segments, __linearize is performing a redundant check.
(u32)(addr.ea + size - 1) <= lim can only happen if addr.ea is close
to 4GB; in this case, addr.ea + size - 1 will also fail the check against
the upper bound of the segment (which is provided by the D/B bit).
After eliminating the redundant check, it is simple to compute
the *max_size for expand-down segments too.
Now that the limit check is done in __do_insn_fetch_bytes, we want
to inject a general protection fault there if size < op_size (like
__linearize would have done), instead of just aborting.
This fixes booting Tiano Core from emulated flash with EPT disabled.
Cc: stable@vger.kernel.org
Fixes: 719d5a9b24
Reported-by: Borislav Petkov <bp@suse.de>
Tested-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The error code for #GP and #SS is zero when the segment is used to
access an operand or an instruction. It is only non-zero when
a segment register is being loaded; for limit checks this means
cases such as:
* for #GP, when RIP is beyond the limit on a far call (before the first
instruction is executed). We do not implement this check, but it
would be in em_jmp_far/em_call_far.
* for #SS, if the new stack overflows during an inter-privilege-level
call to a non-conforming code segment. We do not implement stack
switching at all.
So use an error code of zero.
Reviewed-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Even after the recent fix, the assertion on paging_tmpl.h is triggered.
Apparently, the assertion wants to check that the PAE is always set on
long-mode, but does it in incorrect way. Note that the assertion is not
enabled unless the code is debugged by defining MMU_DEBUG.
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The decode phase of the x86 emulator assumes that every instruction with the
ModRM flag, and which can be used with RIP-relative addressing, has either
SrcMem or DstMem. This is not the case for several instructions - prefetch,
hint-nop and clflush.
Adding SrcMem|NoAccess for prefetch and hint-nop and SrcMem for clflush.
This fixes CVE-2014-8480.
Fixes: 41061cdb98
Cc: stable@vger.kernel.org
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Currently, all group15 instructions are decoded as clflush (e.g., mfence,
xsave). In addition, the clflush instruction requires no prefix (66/f2/f3)
would exist. If prefix exists it may encode a different instruction (e.g.,
clflushopt).
Creating a group for clflush, and different group for each prefix.
This has been the case forever, but the next patch needs the cflush group
in order to fix a bug introduced in 3.17.
Fixes: 41061cdb98
Cc: stable@vger.kernel.org
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A failure to decode the instruction can cause a NULL pointer access.
This is fixed simply by moving the "done" label as close as possible
to the return.
This fixes CVE-2014-8481.
Reported-by: Andy Lutomirski <luto@amacapital.net>
Cc: stable@vger.kernel.org
Fixes: 41061cdb98
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Once an instruction crosses a page boundary, the size read from the second page
disregards the common case that part of the operand resides on the first page.
As a result, fetch of long insturctions may fail, and thereby cause the
decoding to fail as well.
Cc: stable@vger.kernel.org
Fixes: 5cfc7e0f5e
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM_EXIT_UNKNOWN is a kvm bug, we don't really know whether it was
triggered by a priveledged application. Let's not kill the guest: WARN
and inject #UD instead.
Cc: stable@vger.kernel.org
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
On systems with invvpid instruction support (corresponding bit in
IA32_VMX_EPT_VPID_CAP MSR is set) guest invocation of invvpid
causes vm exit, which is currently not handled and results in
propagation of unknown exit to userspace.
Fix this by installing an invvpid vm exit handler.
This is CVE-2014-3646.
Cc: stable@vger.kernel.org
Signed-off-by: Petr Matousek <pmatouse@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
