- Better machine check handling for HV KVM
- Ability to support guests with threads=2, 4 or 8 on POWER9
- Fix for a race that could cause delayed recognition of signals
- Fix for a bug where POWER9 guests could sleep with interrupts
pending.
At present, interrupts are hard-disabled fairly late in the guest
entry path, in the assembly code. Since we check for pending signals
for the vCPU(s) task(s) earlier in the guest entry path, it is
possible for a signal to be delivered before we enter the guest but
not be noticed until after we exit the guest for some other reason.
Similarly, it is possible for the scheduler to request a reschedule
while we are in the guest entry path, and we won't notice until after
we have run the guest, potentially for a whole timeslice.
Furthermore, with a radix guest on POWER9, we can take the interrupt
with the MMU on. In this case we end up leaving interrupts
hard-disabled after the guest exit, and they are likely to stay
hard-disabled until we exit to userspace or context-switch to
another process. This was masking the fact that we were also not
setting the RI (recoverable interrupt) bit in the MSR, meaning
that if we had taken an interrupt, it would have crashed the host
kernel with an unrecoverable interrupt message.
To close these races, we need to check for signals and reschedule
requests after hard-disabling interrupts, and then keep interrupts
hard-disabled until we enter the guest. If there is a signal or a
reschedule request from another CPU, it will send an IPI, which will
cause a guest exit.
This puts the interrupt disabling before we call kvmppc_start_thread()
for all the secondary threads of this core that are going to run vCPUs.
The reason for that is that once we have started the secondary threads
there is no easy way to back out without going through at least part
of the guest entry path. However, kvmppc_start_thread() includes some
code for radix guests which needs to call smp_call_function(), which
must be called with interrupts enabled. To solve this problem, this
patch moves that code into a separate function that is called earlier.
When the guest exit is caused by an external interrupt, a hypervisor
doorbell or a hypervisor maintenance interrupt, we now handle these
using the replay facility. __kvmppc_vcore_entry() now returns the
trap number that caused the exit on this thread, and instead of the
assembly code jumping to the handler entry, we return to C code with
interrupts still hard-disabled and set the irq_happened flag in the
PACA, so that when we do local_irq_enable() the appropriate handler
gets called.
With all this, we now have the interrupt soft-enable flag clear while
we are in the guest. This is useful because code in the real-mode
hypercall handlers that checks whether interrupts are enabled will
now see that they are disabled, which is correct, since interrupts
are hard-disabled in the real-mode code.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Since commit b009031f74 ("KVM: PPC: Book3S HV: Take out virtual
core piggybacking code", 2016-09-15), we only have at most one
vcore per subcore. Previously, the fact that there might be more
than one vcore per subcore meant that we had the notion of a
"master vcore", which was the vcore that controlled thread 0 of
the subcore. We also needed a list per subcore in the core_info
struct to record which vcores belonged to each subcore. Now that
there can only be one vcore in the subcore, we can replace the
list with a simple pointer and get rid of the notion of the
master vcore (and in fact treat every vcore as a master vcore).
We can also get rid of the subcore_vm[] field in the core_info
struct since it is never read.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
The macro insn_fetch marks the 'type' argument as having a specified
alignment. Type attributes can only be applied to structs, unions, or
enums, but insn_fetch is only ever invoked with integral types, so Clang
produces 19 -Wignored-attributes warnings for this source file.
Signed-off-by: Nick Desaulniers <nick.desaulniers@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
- vcpu request overhaul
- allow timer and PMU to have their interrupt number
selected from userspace
- workaround for Cavium erratum 30115
- handling of memory poisonning
- the usual crop of fixes and cleanups
-----BEGIN PGP SIGNATURE-----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=6hNK
-----END PGP SIGNATURE-----
Merge tag 'kvmarm-for-4.13' of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD
KVM/ARM updates for 4.13
- vcpu request overhaul
- allow timer and PMU to have their interrupt number
selected from userspace
- workaround for Cavium erratum 30115
- handling of memory poisonning
- the usual crop of fixes and cleanups
Conflicts:
arch/s390/include/asm/kvm_host.h
If the TSC deadline timer is programmed really close to the deadline or
even in the past, the computation in vmx_set_hv_timer will program the
absolute target tsc value to vmcs preemption timer field w/ delta == 0,
then plays a vmentry and an upcoming vmx preemption timer fire vmexit
dance, the lapic timer injection is delayed due to this duration. Actually
the lapic timer which is emulated by hrtimer can handle this correctly.
This patch fixes it by firing the lapic timer and injecting a timer interrupt
immediately during the next vmentry if the TSC deadline timer is programmed
really close to the deadline or even in the past. This saves ~300 cycles on
the tsc_deadline_timer test of apic.flat.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Move the code to cancel the hv timer into the caller, just before
it starts the hrtimer. Check availability of the hv timer in
start_hv_timer.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
There are many cases in which the hv timer must be canceled. Split out
a new function to avoid duplication.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
- initial machine check forwarding
- migration support for the CMMA page hinting information
- cleanups
- fixes
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.14 (GNU/Linux)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=L7Ia
-----END PGP SIGNATURE-----
Merge tag 'kvm-s390-next-4.13-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into HEAD
KVM: s390: fixes and features for 4.13
- initial machine check forwarding
- migration support for the CMMA page hinting information
- cleanups
- fixes
The memory operand fetched for INVVPID is 128 bits. Bits 63:16 are
reserved and must be zero. Otherwise, the instruction fails with
VMfail(Invalid operand to INVEPT/INVVPID). If the INVVPID_TYPE is 0
(individual address invalidation), then bits 127:64 must be in
canonical form, or the instruction fails with VMfail(Invalid operand
to INVEPT/INVVPID).
Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
With vsie feature enabled, kvm can support nested guests (guest-3).
So inject machine check to the guest-2 if it happens when the nested
guest is running. And guest-2 will detect the machine check belongs
to guest-3 and reinject it into guest-3.
The host (guest-1) tries to inject the machine check to the picked
destination vcpu if it's a floating machine check.
Signed-off-by: QingFeng Hao <haoqf@linux.vnet.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
If the exit flag of SIE indicates that a machine check has happened
during guest's running and needs to be injected, inject it to the guest
accordingly.
But some machine checks, e.g. Channel Report Pending (CRW), refer to
host conditions only (the guest's channel devices are not managed by
the kernel directly) and are therefore not injected into the guest.
External Damage (ED) is also not reinjected into the guest because ETR
conditions are gone in Linux and STP conditions are not enabled in the
guest, and ED contains only these 8 ETR and STP conditions.
In general, instruction-processing damage, system recovery,
storage error, service-processor damage and channel subsystem damage
will be reinjected into the guest, and the remain (System damage,
timing-facility damage, warning, ED and CRW) will be handled on the host.
Signed-off-by: QingFeng Hao <haoqf@linux.vnet.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
This provides the basic plumbing for handling machine checks when
running guests
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.14 (GNU/Linux)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=CTzL
-----END PGP SIGNATURE-----
Merge tag 'nmiforkvm' of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into kernelorgnext
s390,kvm: provide plumbing for machines checks when running guests
This provides the basic plumbing for handling machine checks when
running guests
Toggle display total number of events by guest (debugfs only).
When switching to display of events by guest, field filters remain
active. I.e. the number of events per guest reported considers only
events matching the filters. Likewise with pid/guest filtering.
Note that when switching to display of events by guest, DebugfsProvider
remains to collect data for events as it did before, but the read()
method summarizes the values by pid.
Signed-off-by: Stefan Raspl <raspl@linux.vnet.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It might be handy to display the full history of event stats to compare
the current event distribution against any available historic data.
Since we have that available for debugfs, we offer a respective command
line option to display what's available.
Signed-off-by: Stefan Raspl <raspl@linux.vnet.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Fix an instance where print_all_gnames() is called without the mandatory
argument, resulting in a stack trace.
To reproduce, simply press 'g' in interactive mode.
Signed-off-by: Stefan Raspl <raspl@linux.vnet.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
enable_nmi_window is supposed to be a no-op if we know that we'll see
a VM exit by the time the NMI window opens. This commit adds two more
cases:
* We intercept stgi so we don't need to singlestep on GIF=0.
* We emulate nested vmexit so we don't need to singlestep when nested
VM exit is required.
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Singlestepping is enabled by setting the TF flag and care must be
taken to not let the guest see (and reuse at an inconvenient time)
the modified rflag value. One such case is event injection, as part
of which flags are pushed on the stack and restored later on iret.
This commit disables singlestepping when we're about to inject an
event and forces an immediate exit for us to re-evaluate the NMI
related state.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
These flags are used internally by SVM so it's cleaner to not leak
them to callers of svm_get_rflags. This is similar to how the TF
flag is handled on KVM_GUESTDBG_SINGLESTEP by kvm_get_rflags and
kvm_set_rflags.
Without this change, the flags may propagate from host VMCB to nested
VMCB or vice versa while singlestepping over a nested VM enter/exit,
and then get stuck in inappropriate places.
Example: NMI singlestepping is enabled while running L1 guest. The
instruction to step over is VMRUN and nested vmrun emulation stashes
rflags to hsave->save.rflags. Then if singlestepping is disabled
while still in L2, TF/RF will be cleared from the nested VMCB but the
next nested VM exit will restore them from hsave->save.rflags and
cause an unexpected DB exception.
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Nested hypervisor should not see singlestep VM exits if singlestepping
was enabled internally by KVM. Windows is particularly sensitive to this
and known to bluescreen on unexpected VM exits.
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Just moving the code to a new helper in preparation for following
commits.
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When a machine check happens in the guest, related mcck info (mcic,
external damage code, ...) is stored in the vcpu's lowcore on the host.
Then the machine check handler's low-level part is executed, followed
by the high-level part.
If the high-level part's execution is interrupted by a new machine check
happening on the same vcpu on the host, the mcck info in the lowcore is
overwritten with the new machine check's data.
If the high-level part's execution is scheduled to a different cpu,
the mcck info in the lowcore is uncertain.
Therefore, for both cases, the further reinjection to the guest will use
the wrong data.
Let's backup the mcck info in the lowcore to the sie page
for further reinjection, so that the right data will be used.
Add new member into struct sie_page to store related machine check's
info of mcic, failing storage address and external damage code.
Signed-off-by: QingFeng Hao <haoqf@linux.vnet.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Add the logic to check if the machine check happens when the guest is
running. If yes, set the exit reason -EINTR in the machine check's
interrupt handler. Refactor s390_do_machine_check to avoid panicing
the host for some kinds of machine checks which happen
when guest is running.
Reinject the instruction processing damage's machine checks including
Delayed Access Exception instead of damaging the host if it happens
in the guest because it could be caused by improper update on TLB entry
or other software case and impacts the guest only.
Signed-off-by: QingFeng Hao <haoqf@linux.vnet.ibm.com>
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
The call to kvm_put_kvm was removed from error handling in commit
506cfba9e7 ("KVM: don't use anon_inode_getfd() before possible
failures"), but it is _not_ a memory leak. Reuse Al's explanation
to avoid that someone else makes the same mistake.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Replaces "S_IRUGO | S_IWUSR" with 0644. The reason is that symbolic
permissions considered harmful:
https://lwn.net/Articles/696229/
Signed-off-by: Roman Storozhenko <romeusmeister@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This is really trivial; there is a dup (1 << 16) in the code
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Stefan Traby <stefan@hello-penguin.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Once we enable ARCH_SUPPORTS_MEMORY_FAILURE on arm64, notifications for
broken memory can call memory_failure() in mm/memory-failure.c to offline
pages of memory, possibly signalling user space processes and notifying all
the in-kernel users.
memory_failure() has two modes, early and late. Early is used by
machine-managers like Qemu to receive a notification when a memory error is
notified to the host. These can then be relayed to the guest before the
affected page is accessed. To enable this, the process must set
PR_MCE_KILL_EARLY in PR_MCE_KILL_SET using the prctl() syscall.
Once the early notification has been handled, nothing stops the
machine-manager or guest from accessing the affected page. If the
machine-manager does this the page will fail to be mapped and SIGBUS will
be sent. This patch adds the equivalent path for when the guest accesses
the page, sending SIGBUS to the machine-manager.
These two signals can be distinguished by the machine-manager using their
si_code: BUS_MCEERR_AO for 'action optional' early notifications, and
BUS_MCEERR_AR for 'action required' synchronous/late notifications.
Do as x86 does, and deliver the SIGBUS when we discover pfn ==
KVM_PFN_ERR_HWPOISON. Use the hugepage size as si_addr_lsb if this vma was
allocated as a hugepage. Transparent hugepages will be split by
memory_failure() before we see them here.
Cc: Punit Agrawal <punit.agrawal@arm.com>
Signed-off-by: James Morse <james.morse@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
For naturally aligned and sized data structures avoid superfluous
packed and aligned attributes.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
In some cases, userspace needs to get or set all ais states for example
migration. So we introduce a new group KVM_DEV_FLIC_AISM_ALL to provide
interfaces to get or set the adapter-interruption-suppression mode for
all ISCs. The corresponding documentation is updated.
Signed-off-by: Yi Min Zhao <zyimin@linux.vnet.ibm.com>
Reviewed-by: Halil Pasic <pasic@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
ifetch
While currently only used to fetch the original instruction on failure
for getting the instruction length code, we should make the page table
walking code future proof.
Suggested-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
* Add the struct used in the ioctls to get and set CMMA attributes.
* Add the two functions needed to get and set the CMMA attributes for
guest pages.
* Add the two ioctls that use the aforementioned functions.
Signed-off-by: Claudio Imbrenda <imbrenda@linux.vnet.ibm.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
* Add a migration state bitmap to keep track of which pages have dirty
CMMA information.
* Disable CMMA by default, so we can track if it's used or not. Enable
it on first use like we do for storage keys (unless we are doing a
migration).
* Creates a VM attribute to enter and leave migration mode.
* In migration mode, CMMA is disabled in the SIE block, so ESSA is
always interpreted and emulated in software.
* Free the migration state on VM destroy.
Signed-off-by: Claudio Imbrenda <imbrenda@linux.vnet.ibm.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Now that userspace can set the virtual SMT mode by enabling the
KVM_CAP_PPC_SMT capability, it is useful for userspace to be able
to query the set of possible virtual SMT modes. This provides a
new capability, KVM_CAP_PPC_SMT_POSSIBLE, to provide this
information. The return value is a bitmap of possible modes, with
bit N set if virtual SMT mode 2^N is available. That is, 1 indicates
SMT1 is available, 2 indicates that SMT2 is available, 3 indicates
that both SMT1 and SMT2 are available, and so on.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Enhance KVM to cause a guest exit with KVM_EXIT_NMI
exit reason upon a machine check exception (MCE) in
the guest address space if the KVM_CAP_PPC_FWNMI
capability is enabled (instead of delivering a 0x200
interrupt to guest). This enables QEMU to build error
log and deliver machine check exception to guest via
guest registered machine check handler.
This approach simplifies the delivery of machine
check exception to guest OS compared to the earlier
approach of KVM directly invoking 0x200 guest interrupt
vector.
This design/approach is based on the feedback for the
QEMU patches to handle machine check exception. Details
of earlier approach of handling machine check exception
in QEMU and related discussions can be found at:
https://lists.nongnu.org/archive/html/qemu-devel/2014-11/msg00813.html
Note:
This patch now directly invokes machine_check_print_event_info()
from kvmppc_handle_exit_hv() to print the event to host console
at the time of guest exit before the exception is passed on to the
guest. Hence, the host-side handling which was performed earlier
via machine_check_fwnmi is removed.
The reasons for this approach is (i) it is not possible
to distinguish whether the exception occurred in the
guest or the host from the pt_regs passed on the
machine_check_exception(). Hence machine_check_exception()
calls panic, instead of passing on the exception to
the guest, if the machine check exception is not
recoverable. (ii) the approach introduced in this
patch gives opportunity to the host kernel to perform
actions in virtual mode before passing on the exception
to the guest. This approach does not require complex
tweaks to machine_check_fwnmi and friends.
Signed-off-by: Aravinda Prasad <aravinda@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
It will be used in arch/powerpc/kvm/book3s_hv.c KVM module.
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
This introduces a new KVM capability to control how KVM behaves
on machine check exception (MCE) in HV KVM guests.
If this capability has not been enabled, KVM redirects machine check
exceptions to guest's 0x200 vector, if the address in error belongs to
the guest. With this capability enabled, KVM will cause a guest exit
with the exit reason indicating an NMI.
The new capability is required to avoid problems if a new kernel/KVM
is used with an old QEMU, running a guest that doesn't issue
"ibm,nmi-register". As old QEMU does not understand the NMI exit
type, it treats it as a fatal error. However, the guest could have
handled the machine check error if the exception was delivered to
guest's 0x200 interrupt vector instead of NMI exit in case of old
QEMU.
[paulus@ozlabs.org - Reworded the commit message to be clearer,
enable only on HV KVM.]
Signed-off-by: Aravinda Prasad <aravinda@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
On a POWER9 system, it is possible for an interrupt to become pending
for a VCPU when that VCPU is about to cede (execute a H_CEDE hypercall)
and has already disabled interrupts, or in the H_CEDE processing up
to the point where the XIVE context is pulled from the hardware. In
such a case, the H_CEDE should not sleep, but should return immediately
to the guest. However, the conditions tested in kvmppc_vcpu_woken()
don't include the condition that a XIVE interrupt is pending, so the
VCPU could sleep until the next decrementer interrupt.
To fix this, we add a new xive_interrupt_pending() helper which looks
in the XIVE context that was pulled from the hardware to see if the
priority of any pending interrupt is higher (numerically lower than)
the CPU priority. If so then kvmppc_vcpu_woken() will return true.
If the XIVE context has never been used, then both the pipr and the
cppr fields will be zero and the test will indicate that no interrupt
is pending.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
On POWER9, we no longer have the restriction that we had on POWER8
where all threads in a core have to be in the same partition, so
the CPU threads are now independent. However, we still want to be
able to run guests with a virtual SMT topology, if only to allow
migration of guests from POWER8 systems to POWER9.
A guest that has a virtual SMT mode greater than 1 will expect to
be able to use the doorbell facility; it will expect the msgsndp
and msgclrp instructions to work appropriately and to be able to read
sensible values from the TIR (thread identification register) and
DPDES (directed privileged doorbell exception status) special-purpose
registers. However, since each CPU thread is a separate sub-processor
in POWER9, these instructions and registers can only be used within
a single CPU thread.
In order for these instructions to appear to act correctly according
to the guest's virtual SMT mode, we have to trap and emulate them.
We cause them to trap by clearing the HFSCR_MSGP bit in the HFSCR
register. The emulation is triggered by the hypervisor facility
unavailable interrupt that occurs when the guest uses them.
To cause a doorbell interrupt to occur within the guest, we set the
DPDES register to 1. If the guest has interrupts enabled, the CPU
will generate a doorbell interrupt and clear the DPDES register in
hardware. The DPDES hardware register for the guest is saved in the
vcpu->arch.vcore->dpdes field. Since this gets written by the guest
exit code, other VCPUs wishing to cause a doorbell interrupt don't
write that field directly, but instead set a vcpu->arch.doorbell_request
flag. This is consumed and set to 0 by the guest entry code, which
then sets DPDES to 1.
Emulating reads of the DPDES register is somewhat involved, because
it requires reading the doorbell pending interrupt status of all of the
VCPU threads in the virtual core, and if any of those VCPUs are
running, their doorbell status is only up-to-date in the hardware
DPDES registers of the CPUs where they are running. In order to get
a reasonable approximation of the current doorbell status, we send
those CPUs an IPI, causing an exit from the guest which will update
the vcpu->arch.vcore->dpdes field. We then use that value in
constructing the emulated DPDES register value.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
This allows userspace to set the desired virtual SMT (simultaneous
multithreading) mode for a VM, that is, the number of VCPUs that
get assigned to each virtual core. Previously, the virtual SMT mode
was fixed to the number of threads per subcore, and if userspace
wanted to have fewer vcpus per vcore, then it would achieve that by
using a sparse CPU numbering. This had the disadvantage that the
vcpu numbers can get quite large, particularly for SMT1 guests on
a POWER8 with 8 threads per core. With this patch, userspace can
set its desired virtual SMT mode and then use contiguous vcpu
numbering.
On POWER8, where the threading mode is "strict", the virtual SMT mode
must be less than or equal to the number of threads per subcore. On
POWER9, which implements a "loose" threading mode, the virtual SMT
mode can be any power of 2 between 1 and 8, even though there is
effectively one thread per subcore, since the threads are independent
and can all be in different partitions.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
This adds code to allow us to use a different value for the HFSCR
(Hypervisor Facilities Status and Control Register) when running the
guest from that which applies in the host. The reason for doing this
is to allow us to trap the msgsndp instruction and related operations
in future so that they can be virtualized. We also save the value of
HFSCR when a hypervisor facility unavailable interrupt occurs, because
the high byte of HFSCR indicates which facility the guest attempted to
access.
We save and restore the host value on guest entry/exit because some
bits of it affect host userspace execution.
We only do all this on POWER9, not on POWER8, because we are not
intending to virtualize any of the facilities controlled by HFSCR on
POWER8. In particular, the HFSCR bit that controls execution of
msgsndp and related operations does not exist on POWER8. The HFSCR
doesn't exist at all on POWER7.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
It is possible, through a narrow race condition, for a VCPU to exit
the guest with a H_CEDE hypercall while it has a doorbell interrupt
pending. In this case, the H_CEDE should return immediately, but in
fact it puts the VCPU to sleep until some other interrupt becomes
pending or a prod is received (via another VCPU doing H_PROD).
This fixes it by checking the DPDES (Directed Privileged Doorbell
Exception Status) bit for the thread along with the other interrupt
pending bits.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
This allows userspace (e.g. QEMU) to enable large decrementer mode for
the guest when running on a POWER9 host, by setting the LPCR_LD bit in
the guest LPCR value. With this, the guest exit code saves 64 bits of
the guest DEC value on exit. Other places that use the guest DEC
value check the LPCR_LD bit in the guest LPCR value, and if it is set,
omit the 32-bit sign extension that would otherwise be done.
This doesn't change the DEC emulation used by PR KVM because PR KVM
is not supported on POWER9 yet.
This is partly based on an earlier patch by Oliver O'Halloran.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
POWER9 DD1 has an erratum where writing to the TBU40 register, which
is used to apply an offset to the timebase, can cause the timebase to
lose counts. This results in the timebase on some CPUs getting out of
sync with other CPUs, which then results in misbehaviour of the
timekeeping code.
To work around the problem, we make KVM ignore the timebase offset for
all guests on POWER9 DD1 machines. This means that live migration
cannot be supported on POWER9 DD1 machines.
Cc: stable@vger.kernel.org # v4.10+
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
At present, HV KVM on POWER8 and POWER9 machines loses any instruction
or data breakpoint set in the host whenever a guest is run.
Instruction breakpoints are currently only used by xmon, but ptrace
and the perf_event subsystem can set data breakpoints as well as xmon.
To fix this, we save the host values of the debug registers (CIABR,
DAWR and DAWRX) before entering the guest and restore them on exit.
To provide space to save them in the stack frame, we expand the stack
frame allocated by kvmppc_hv_entry() from 112 to 144 bytes.
Fixes: b005255e12 ("KVM: PPC: Book3S HV: Context-switch new POWER8 SPRs", 2014-01-08)
Cc: stable@vger.kernel.org # v3.14+
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
When reading the cntpct_el0 in guest with VHE (Virtual Host Extension)
enabled in host, the "Unsupported guest sys_reg access" error reported.
The reason is cnthctl_el2.EL1PCTEN is not enabled, which is expected
to be done in kvm_timer_init_vhe(). The problem is kvm_timer_init_vhe
is called by cpu_init_hyp_mode, and which is called when VHE is disabled.
This patch remove the incorrect call to kvm_timer_init_vhe() from
cpu_init_hyp_mode(), and calls kvm_timer_init_vhe() to enable
cnthctl_el2.EL1PCTEN in cpu_hyp_reinit().
Fixes: 488f94d721 ("KVM: arm64: Access CNTHCTL_EL2 bit fields correctly on VHE systems")
Cc: stable@vger.kernel.org
Signed-off-by: Hu Huajun <huhuajun@huawei.com>
Reviewed-by: Christoffer Dall <cdall@linaro.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Almost all of the arm64 KVM code uses the sysreg mnemonics for AArch64
register descriptions. Move the last straggler over.
To match what we do for SYS_ICH_AP*R*_EL2, the SYS_ICC_AP*R*_EL1
mnemonics are expanded in <asm/sysreg.h>.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: kvmarm@lists.cs.columbia.edu
Acked-by: Christoffer Dall <cdall@linaro.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
Per ARM DDI 0487B.a, the registers are named ICC_IGRPEN*_EL1 rather than
ICC_GRPEN*_EL1. Correct our mnemonics and comments to match, before we
add more GICv3 register definitions.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: kvmarm@lists.cs.columbia.edu
Acked-by: Christoffer Dall <cdall@linaro.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
A write-to-read-only GICv3 access should UNDEF at EL1. But since
we're in complete paranoia-land with broken CPUs, let's assume the
worse and gracefully handle the case.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Christoffer Dall <cdall@linaro.org>
A read-from-write-only GICv3 access should UNDEF at EL1. But since
we're in complete paranoia-land with broken CPUs, let's assume the
worse and gracefully handle the case.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Reviewed-by: Christoffer Dall <cdall@linaro.org>
Signed-off-by: Christoffer Dall <cdall@linaro.org>