Paul Mackerras [Sat, 27 Apr 2013 00:28:37 +0000 (00:28 +0000)]
KVM: PPC: Book3S: Add API for in-kernel XICS emulation
This adds the API for userspace to instantiate an XICS device in a VM
and connect VCPUs to it. The API consists of a new device type for
the KVM_CREATE_DEVICE ioctl, a new capability KVM_CAP_IRQ_XICS, which
functions similarly to KVM_CAP_IRQ_MPIC, and the KVM_IRQ_LINE ioctl,
which is used to assert and deassert interrupt inputs of the XICS.
The XICS device has one attribute group, KVM_DEV_XICS_GRP_SOURCES.
Each attribute within this group corresponds to the state of one
interrupt source. The attribute number is the same as the interrupt
source number.
This does not support irq routing or irqfd yet.
Signed-off-by: Paul Mackerras <paulus@samba.org> Acked-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
Jan Kiszka [Sun, 28 Apr 2013 07:24:41 +0000 (09:24 +0200)]
KVM: nVMX: Skip PF interception check when queuing during nested run
While a nested run is pending, vmx_queue_exception is only called to
requeue exceptions that were previously picked up via
vmx_cancel_injection. Therefore, we must not check for PF interception
by L1, possibly causing a bogus nested vmexit.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
KVM guests today use 8bit APIC ids allowing for 256 ID's. Reserving one
ID for Broadcast interrupts should leave 255 ID's. In case of KVM there
is no need for reserving another ID for IO-APIC so the hard max limit for
VCPUS can be increased from 254 to 255. (This was confirmed by Gleb Natapov
http://article.gmane.org/gmane.comp.emulators.kvm.devel/99713 )
Alex Williamson [Tue, 16 Apr 2013 19:49:18 +0000 (13:49 -0600)]
kvm: Allow build-time configuration of KVM device assignment
We hope to at some point deprecate KVM legacy device assignment in
favor of VFIO-based assignment. Towards that end, allow legacy
device assignment to be deconfigured.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Reviewed-by: Alexander Graf <agraf@suse.de> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Jan Kiszka [Sun, 28 Apr 2013 08:50:52 +0000 (10:50 +0200)]
KVM: x86: Rework request for immediate exit
The VMX implementation of enable_irq_window raised
KVM_REQ_IMMEDIATE_EXIT after we checked it in vcpu_enter_guest. This
caused infinite loops on vmentry. Fix it by letting enable_irq_window
signal the need for an immediate exit via its return value and drop
KVM_REQ_IMMEDIATE_EXIT.
This issue only affects nested VMX scenarios.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Paul Mackerras [Wed, 17 Apr 2013 20:32:26 +0000 (20:32 +0000)]
KVM: PPC: Book3S: Facilities to save/restore XICS presentation ctrler state
This adds the ability for userspace to save and restore the state
of the XICS interrupt presentation controllers (ICPs) via the
KVM_GET/SET_ONE_REG interface. Since there is one ICP per vcpu, we
simply define a new 64-bit register in the ONE_REG space for the ICP
state. The state includes the CPU priority setting, the pending IPI
priority, and the priority and source number of any pending external
interrupt.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
Paul Mackerras [Wed, 17 Apr 2013 20:32:04 +0000 (20:32 +0000)]
KVM: PPC: Book3S: Add support for ibm,int-on/off RTAS calls
This adds support for the ibm,int-on and ibm,int-off RTAS calls to the
in-kernel XICS emulation and corrects the handling of the saved
priority by the ibm,set-xive RTAS call. With this, ibm,int-off sets
the specified interrupt's priority in its saved_priority field and
sets the priority to 0xff (the least favoured value). ibm,int-on
restores the saved_priority to the priority field, and ibm,set-xive
sets both the priority and the saved_priority to the specified
priority value.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
Paul Mackerras [Wed, 17 Apr 2013 20:31:41 +0000 (20:31 +0000)]
KVM: PPC: Book3S HV: Improve real-mode handling of external interrupts
This streamlines our handling of external interrupts that come in
while we're in the guest. First, when waking up a hardware thread
that was napping, we split off the "napping due to H_CEDE" case
earlier, and use the code that handles an external interrupt (0x500)
in the guest to handle that too. Secondly, the code that handles
those external interrupts now checks if any other thread is exiting
to the host before bouncing an external interrupt to the guest, and
also checks that there is actually an external interrupt pending for
the guest before setting the LPCR MER bit (mediated external request).
This also makes sure that we clear the "ceded" flag when we handle a
wakeup from cede in real mode, and fixes a potential infinite loop
in kvmppc_run_vcpu() which can occur if we ever end up with the ceded
flag set but MSR[EE] off.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
KVM: PPC: Book3S HV: Add support for real mode ICP in XICS emulation
This adds an implementation of the XICS hypercalls in real mode for HV
KVM, which allows us to avoid exiting the guest MMU context on all
threads for a variety of operations such as fetching a pending
interrupt, EOI of messages, IPIs, etc.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
KVM: PPC: Book3S HV: Speed up wakeups of CPUs on HV KVM
Currently, we wake up a CPU by sending a host IPI with
smp_send_reschedule() to thread 0 of that core, which will take all
threads out of the guest, and cause them to re-evaluate their
interrupt status on the way back in.
This adds a mechanism to differentiate real host IPIs from IPIs sent
by KVM for guest threads to poke each other, in order to target the
guest threads precisely when possible and avoid that global switch of
the core to host state.
We then use this new facility in the in-kernel XICS code.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
KVM: PPC: Book3S: Add kernel emulation for the XICS interrupt controller
This adds in-kernel emulation of the XICS (eXternal Interrupt
Controller Specification) interrupt controller specified by PAPR, for
both HV and PR KVM guests.
The XICS emulation supports up to 1048560 interrupt sources.
Interrupt source numbers below 16 are reserved; 0 is used to mean no
interrupt and 2 is used for IPIs. Internally these are represented in
blocks of 1024, called ICS (interrupt controller source) entities, but
that is not visible to userspace.
Each vcpu gets one ICP (interrupt controller presentation) entity,
used to store the per-vcpu state such as vcpu priority, pending
interrupt state, IPI request, etc.
This does not include any API or any way to connect vcpus to their
ICP state; that will be added in later patches.
This is based on an initial implementation by Michael Ellerman
<michael@ellerman.id.au> reworked by Benjamin Herrenschmidt and
Paul Mackerras.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
[agraf: fix typo, add dependency on !KVM_MPIC] Signed-off-by: Alexander Graf <agraf@suse.de>
Michael Ellerman [Wed, 17 Apr 2013 20:30:00 +0000 (20:30 +0000)]
KVM: PPC: Book3S: Add infrastructure to implement kernel-side RTAS calls
For pseries machine emulation, in order to move the interrupt
controller code to the kernel, we need to intercept some RTAS
calls in the kernel itself. This adds an infrastructure to allow
in-kernel handlers to be registered for RTAS services by name.
A new ioctl, KVM_PPC_RTAS_DEFINE_TOKEN, then allows userspace to
associate token values with those service names. Then, when the
guest requests an RTAS service with one of those token values, it
will be handled by the relevant in-kernel handler rather than being
passed up to userspace as at present.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
[agraf: fix warning] Signed-off-by: Alexander Graf <agraf@suse.de>
Scott Wood [Thu, 25 Apr 2013 14:11:24 +0000 (14:11 +0000)]
kvm/ppc/mpic: Eliminate mmio_mapped
We no longer need to keep track of this now that MPIC destruction
always happens either during VM destruction (after MMIO has been
destroyed) or during a failed creation (before the fd has been exposed
to userspace, and thus before the MMIO region could have been
registered).
Signed-off-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de>
Alexander Graf [Thu, 25 Apr 2013 09:50:03 +0000 (11:50 +0200)]
KVM: IA64: Carry non-ia64 changes into ia64
We changed a few things in non-ia64 code paths. This patch blindly applies
the changes to the ia64 code as well, hoping it proves useful in case anyone
revives the ia64 kvm code.
Alexander Graf [Tue, 16 Apr 2013 23:54:26 +0000 (01:54 +0200)]
KVM: PPC: MPIC: Restrict to e500 platforms
The code as is doesn't make any sense on non-e500 platforms. Restrict it
there, so that people don't get wrong ideas on what would actually work.
This patch should get reverted as soon as it's possible to either run e500
guests on non-e500 hosts or the MPIC emulation gains support for non-e500
modes.
Alexander Graf [Tue, 16 Apr 2013 22:37:57 +0000 (00:37 +0200)]
KVM: PPC: MPIC: Add support for KVM_IRQ_LINE
Now that all pieces are in place for reusing generic irq infrastructure,
we can copy x86's implementation of KVM_IRQ_LINE irq injection and simply
reuse it for PPC, as it will work there just as well.
Scott Wood [Fri, 12 Apr 2013 14:08:47 +0000 (14:08 +0000)]
kvm/ppc/mpic: add KVM_CAP_IRQ_MPIC
Enabling this capability connects the vcpu to the designated in-kernel
MPIC. Using explicit connections between vcpus and irqchips allows
for flexibility, but the main benefit at the moment is that it
simplifies the code -- KVM doesn't need vm-global state to remember
which MPIC object is associated with this vm, and it doesn't need to
care about ordering between irqchip creation and vcpu creation.
Signed-off-by: Scott Wood <scottwood@freescale.com>
[agraf: add stub functions for kvmppc_mpic_{dis,}connect_vcpu] Signed-off-by: Alexander Graf <agraf@suse.de>
Scott Wood [Fri, 12 Apr 2013 14:08:46 +0000 (14:08 +0000)]
kvm/ppc/mpic: in-kernel MPIC emulation
Hook the MPIC code up to the KVM interfaces, add locking, etc.
Signed-off-by: Scott Wood <scottwood@freescale.com>
[agraf: add stub function for kvmppc_mpic_set_epr, non-booke, 64bit] Signed-off-by: Alexander Graf <agraf@suse.de>
Scott Wood [Fri, 12 Apr 2013 14:08:43 +0000 (14:08 +0000)]
kvm/ppc/mpic: import hw/openpic.c from QEMU
This is QEMU's hw/openpic.c from commit abd8d4a4d6dfea7ddea72f095f993e1de941614e ("Update version for
1.4.0-rc0"), run through Lindent with no other changes to ease merging
future changes between Linux and QEMU. Remaining style issues
(including those introduced by Lindent) will be fixed in a later patch.
Signed-off-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de>
Scott Wood [Fri, 12 Apr 2013 14:08:42 +0000 (14:08 +0000)]
kvm: add device control API
Currently, devices that are emulated inside KVM are configured in a
hardcoded manner based on an assumption that any given architecture
only has one way to do it. If there's any need to access device state,
it is done through inflexible one-purpose-only IOCTLs (e.g.
KVM_GET/SET_LAPIC). Defining new IOCTLs for every little thing is
cumbersome and depletes a limited numberspace.
This API provides a mechanism to instantiate a device of a certain
type, returning an ID that can be used to set/get attributes of the
device. Attributes may include configuration parameters (e.g.
register base address), device state, operational commands, etc. It
is similar to the ONE_REG API, except that it acts on devices rather
than vcpus.
Both device types and individual attributes can be tested without having
to create the device or get/set the attribute, without the need for
separately managing enumerated capabilities.
Signed-off-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de>
Alexander Graf [Mon, 15 Apr 2013 21:23:21 +0000 (23:23 +0200)]
KVM: Move irq routing setup to irqchip.c
Setting up IRQ routes is nothing IOAPIC specific. Extract everything
that really is generic code into irqchip.c and only leave the ioapic
specific bits to irq_comm.c.
Signed-off-by: Alexander Graf <agraf@suse.de> Acked-by: Michael S. Tsirkin <mst@redhat.com>
Alexander Graf [Mon, 15 Apr 2013 21:04:10 +0000 (23:04 +0200)]
KVM: Extract generic irqchip logic into irqchip.c
The current irq_comm.c file contains pieces of code that are generic
across different irqchip implementations, as well as code that is
fully IOAPIC specific.
Split the generic bits out into irqchip.c.
Signed-off-by: Alexander Graf <agraf@suse.de> Acked-by: Michael S. Tsirkin <mst@redhat.com>
Alexander Graf [Mon, 15 Apr 2013 19:12:53 +0000 (21:12 +0200)]
KVM: Move irq routing to generic code
The IRQ routing set ioctl lives in the hacky device assignment code inside
of KVM today. This is definitely the wrong place for it. Move it to the much
more natural kvm_main.c.
Signed-off-by: Alexander Graf <agraf@suse.de> Acked-by: Michael S. Tsirkin <mst@redhat.com>
Alexander Graf [Mon, 15 Apr 2013 08:49:31 +0000 (10:49 +0200)]
KVM: Drop __KVM_HAVE_IOAPIC condition on irq routing
We have a capability enquire system that allows user space to ask kvm
whether a feature is available.
The point behind this system is that we can have different kernel
configurations with different capabilities and user space can adjust
accordingly.
Because features can always be non existent, we can drop any #ifdefs
on CAP defines that could be used generically, like the irq routing
bits. These can be easily reused for non-IOAPIC systems as well.
Signed-off-by: Alexander Graf <agraf@suse.de> Acked-by: Michael S. Tsirkin <mst@redhat.com>
Alexander Graf [Wed, 17 Apr 2013 11:29:30 +0000 (13:29 +0200)]
KVM: Introduce CONFIG_HAVE_KVM_IRQ_ROUTING
Quite a bit of code in KVM has been conditionalized on availability of
IOAPIC emulation. However, most of it is generically applicable to
platforms that don't have an IOPIC, but a different type of irq chip.
Make code that only relies on IRQ routing, not an APIC itself, on
CONFIG_HAVE_KVM_IRQ_ROUTING, so that we can reuse it later.
Signed-off-by: Alexander Graf <agraf@suse.de> Acked-by: Michael S. Tsirkin <mst@redhat.com>
Alexander Graf [Mon, 15 Apr 2013 08:42:33 +0000 (10:42 +0200)]
KVM: Add KVM_IRQCHIP_NUM_PINS in addition to KVM_IOAPIC_NUM_PINS
The concept of routing interrupt lines to an irqchip is nothing
that is IOAPIC specific. Every irqchip has a maximum number of pins
that can be linked to irq lines.
So let's add a new define that allows us to reuse generic code for
non-IOAPIC platforms.
Signed-off-by: Alexander Graf <agraf@suse.de> Acked-by: Michael S. Tsirkin <mst@redhat.com>
Paul Mackerras [Thu, 18 Apr 2013 19:51:04 +0000 (19:51 +0000)]
KVM: PPC: Book3S HV: Report VPA and DTL modifications in dirty map
At present, the KVM_GET_DIRTY_LOG ioctl doesn't report modifications
done by the host to the virtual processor areas (VPAs) and dispatch
trace logs (DTLs) registered by the guest. This is because those
modifications are done either in real mode or in the host kernel
context, and in neither case does the access go through the guest's
HPT, and thus no change (C) bit gets set in the guest's HPT.
However, the changes done by the host do need to be tracked so that
the modified pages get transferred when doing live migration. In
order to track these modifications, this adds a dirty flag to the
struct representing the VPA/DTL areas, and arranges to set the flag
when the VPA/DTL gets modified by the host. Then, when we are
collecting the dirty log, we also check the dirty flags for the
VPA and DTL for each vcpu and set the relevant bit in the dirty log
if necessary. Doing this also means we now need to keep track of
the guest physical address of the VPA/DTL areas.
So as not to lose track of modifications to a VPA/DTL area when it gets
unregistered, or when a new area gets registered in its place, we need
to transfer the dirty state to the rmap chain. This adds code to
kvmppc_unpin_guest_page() to do that if the area was dirty. To simplify
that code, we now require that all VPA, DTL and SLB shadow buffer areas
fit within a single host page. Guests already comply with this
requirement because pHyp requires that these areas not cross a 4k
boundary.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
Paul Mackerras [Thu, 18 Apr 2013 19:50:24 +0000 (19:50 +0000)]
KVM: PPC: Book3S HV: Make HPT reading code notice R/C bit changes
At present, the code that determines whether a HPT entry has changed,
and thus needs to be sent to userspace when it is copying the HPT,
doesn't consider a hardware update to the reference and change bits
(R and C) in the HPT entries to constitute a change that needs to
be sent to userspace. This adds code to check for changes in R and C
when we are scanning the HPT to find changed entries, and adds code
to set the changed flag for the HPTE when we update the R and C bits
in the guest view of the HPTE.
Since we now need to set the HPTE changed flag in book3s_64_mmu_hv.c
as well as book3s_hv_rm_mmu.c, we move the note_hpte_modification()
function into kvm_book3s_64.h.
Current Linux guest kernels don't use the hardware updates of R and C
in the HPT, so this change won't affect them. Linux (or other) kernels
might in future want to use the R and C bits and have them correctly
transferred across when a guest is migrated, so it is better to correct
this deficiency.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
Mihai Caraman [Thu, 11 Apr 2013 00:03:08 +0000 (00:03 +0000)]
KVM: PPC: e500: Expose MMU registers via ONE_REG
MMU registers were exposed to user-space using sregs interface. Add them
to ONE_REG interface using kvmppc_get_one_reg/kvmppc_set_one_reg delegation
mechanism.
Signed-off-by: Mihai Caraman <mihai.caraman@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de>
Refactor Book3E ONE_REG ioctl implementation to use kvmppc_get_one_reg/
kvmppc_set_one_reg delegation interface introduced by Book3S. This is
necessary for MMU SPRs which are platform specifics.
Get rid of useless case braces in the process.
Signed-off-by: Mihai Caraman <mihai.caraman@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de>
KVM: extend EMULATE_EXIT_USER to support different exit reasons
Currently the instruction emulator code returns EMULATE_EXIT_USER
and common code initializes the "run->exit_reason = .." and
"vcpu->arch.hcall_needed = .." with one fixed reason.
But there can be different reasons when emulator need to exit
to user space. To support that the "run->exit_reason = .."
and "vcpu->arch.hcall_needed = .." initialization is moved a
level up to emulator.
Signed-off-by: Bharat Bhushan <bharat.bhushan@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de>
Instruction emulation return EMULATE_DO_PAPR when it requires
exit to userspace on book3s. Similar return is required
for booke. EMULATE_DO_PAPR reads out to be confusing so it is
renamed to EMULATE_EXIT_USER.
Signed-off-by: Bharat Bhushan <bharat.bhushan@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de>
This patch defines the interface parameter for KVM_SET_GUEST_DEBUG
ioctl support. Follow up patches will use this for setting up
hardware breakpoints, watchpoints and software breakpoints.
Also kvm_arch_vcpu_ioctl_set_guest_debug() is brought one level below.
This is because I am not sure what is required for book3s. So this ioctl
behaviour will not change for book3s.
Signed-off-by: Bharat Bhushan <bharat.bhushan@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de>
If we load the complete EFER MSR on entry or exit, EFER.LMA (and LME)
loading is skipped. Their consistency is already checked now before
starting the transition.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Jan Kiszka [Sat, 20 Apr 2013 08:52:36 +0000 (10:52 +0200)]
KVM: nVMX: Validate EFER values for VM_ENTRY/EXIT_LOAD_IA32_EFER
As we may emulate the loading of EFER on VM-entry and VM-exit, implement
the checks that VMX performs on the guest and host values on vmlaunch/
vmresume. Factor out kvm_valid_efer for this purpose which checks for
set reserved bits.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Jan Kiszka [Sun, 14 Apr 2013 19:04:26 +0000 (21:04 +0200)]
KVM: nVMX: Fix conditions for NMI injection
The logic for checking if interrupts can be injected has to be applied
also on NMIs. The difference is that if NMI interception is on these
events are consumed and blocked by the VM exit.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Andrew Honig [Thu, 18 Apr 2013 16:38:14 +0000 (09:38 -0700)]
KVM: x86: Fix memory leak in vmx.c
If userspace creates and destroys multiple VMs within the same process
we leak 20k of memory in the userspace process context per VM. This
patch frees the memory in kvm_arch_destroy_vm. If the process exits
without closing the VM file descriptor or the file descriptor has been
shared with another process then we don't free the memory.
It's still possible for a user space process to leak memory if the last
process to close the fd for the VM is not the process that created it.
However, this is an unexpected case that's only caused by a user space
process that's misbehaving.
Signed-off-by: Andrew Honig <ahonig@google.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Abel Gordon [Thu, 18 Apr 2013 11:39:55 +0000 (14:39 +0300)]
KVM: nVMX: Enable and disable shadow vmcs functionality
Once L1 loads VMCS12 we enable shadow-vmcs capability and copy all the VMCS12
shadowed fields to the shadow vmcs. When we release the VMCS12, we also
disable shadow-vmcs capability.
Signed-off-by: Abel Gordon <abelg@il.ibm.com> Reviewed-by: Orit Wasserman <owasserm@redhat.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Abel Gordon [Thu, 18 Apr 2013 11:37:25 +0000 (14:37 +0300)]
KVM: nVMX: Allocate shadow vmcs
Allocate a shadow vmcs used by the processor to shadow part of the fields
stored in the software defined VMCS12 (let L1 access fields without causing
exits). Note we keep a shadow vmcs only for the current vmcs12. Once a vmcs12
becomes non-current, its shadow vmcs is released.
Signed-off-by: Abel Gordon <abelg@il.ibm.com> Reviewed-by: Orit Wasserman <owasserm@redhat.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Abel Gordon [Thu, 18 Apr 2013 11:36:55 +0000 (14:36 +0300)]
KVM: nVMX: Fix VMXON emulation
handle_vmon doesn't check if L1 is already in root mode (VMXON
was previously called). This patch adds this missing check and calls
nested_vmx_failValid if VMX is already ON.
We need this check because L0 will allocate the shadow vmcs when L1
executes VMXON and we want to avoid host leaks (due to shadow vmcs
allocation) if L1 executes VMXON repeatedly.
Signed-off-by: Abel Gordon <abelg@il.ibm.com> Reviewed-by: Orit Wasserman <owasserm@redhat.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Abel Gordon [Thu, 18 Apr 2013 11:36:25 +0000 (14:36 +0300)]
KVM: nVMX: Refactor handle_vmwrite
Refactor existent code so we re-use vmcs12_write_any to copy fields from the
shadow vmcs specified by the link pointer (used by the processor,
implementation-specific) to the VMCS12 software format used by L0 to hold
the fields in L1 memory address space.
Signed-off-by: Abel Gordon <abelg@il.ibm.com> Reviewed-by: Orit Wasserman <owasserm@redhat.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Abel Gordon [Thu, 18 Apr 2013 11:35:55 +0000 (14:35 +0300)]
KVM: nVMX: Introduce vmread and vmwrite bitmaps
Prepare vmread and vmwrite bitmaps according to a pre-specified list of fields.
These lists are intended to specifiy most frequent accessed fields so we can
minimize the number of fields that are copied from/to the software controlled
VMCS12 format to/from to processor-specific shadow vmcs. The lists were built
measuring the VMCS fields access rate after L2 Ubuntu 12.04 booted when it was
running on top of L1 KVM, also Ubuntu 12.04. Note that during boot there were
additional fields which were frequently modified but they were not added to
these lists because after boot these fields were not longer accessed by L1.
Signed-off-by: Abel Gordon <abelg@il.ibm.com> Reviewed-by: Orit Wasserman <owasserm@redhat.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Abel Gordon [Thu, 18 Apr 2013 11:35:25 +0000 (14:35 +0300)]
KVM: nVMX: Detect shadow-vmcs capability
Add logic required to detect if shadow-vmcs is supported by the
processor. Introduce a new kernel module parameter to specify if L0 should use
shadow vmcs (or not) to run L1.
Signed-off-by: Abel Gordon <abelg@il.ibm.com> Reviewed-by: Orit Wasserman <owasserm@redhat.com> Signed-off-by: Gleb Natapov <gleb@redhat.com>
Paolo Bonzini [Mon, 15 Apr 2013 13:00:27 +0000 (15:00 +0200)]
KVM: nVMX: check vmcs12 for valid activity state
KVM does not use the activity state VMCS field, and does not support
it in nested VMX either (the corresponding bits in the misc VMX feature
MSR are zero). Fail entry if the activity state is set to anything but
"active".
Since the value will always be the same for L1 and L2, we do not need
to read and write the corresponding VMCS field on L1/L2 transitions,
either.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Gleb Natapov <gleb@redhat.com> Reviewed-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Yang Zhang [Thu, 11 Apr 2013 11:25:13 +0000 (19:25 +0800)]
KVM: Call common update function when ioapic entry changed.
Both TMR and EOI exit bitmap need to be updated when ioapic changed
or vcpu's id/ldr/dfr changed. So use common function instead eoi exit
bitmap specific function.
Yang Zhang [Thu, 11 Apr 2013 11:25:11 +0000 (19:25 +0800)]
KVM: VMX: Register a new IPI for posted interrupt
Posted Interrupt feature requires a special IPI to deliver posted interrupt
to guest. And it should has a high priority so the interrupt will not be
blocked by others.
Normally, the posted interrupt will be consumed by vcpu if target vcpu is
running and transparent to OS. But in some cases, the interrupt will arrive
when target vcpu is scheduled out. And host will see it. So we need to
register a dump handler to handle it.
Yang Zhang [Thu, 11 Apr 2013 11:25:10 +0000 (19:25 +0800)]
KVM: VMX: Enable acknowledge interupt on vmexit
The "acknowledge interrupt on exit" feature controls processor behavior
for external interrupt acknowledgement. When this control is set, the
processor acknowledges the interrupt controller to acquire the
interrupt vector on VM exit.
After enabling this feature, an interrupt which arrived when target cpu is
running in vmx non-root mode will be handled by vmx handler instead of handler
in idt. Currently, vmx handler only fakes an interrupt stack and jump to idt
table to let real handler to handle it. Further, we will recognize the interrupt
and only delivery the interrupt which not belong to current vcpu through idt table.
The interrupt which belonged to current vcpu will be handled inside vmx handler.
This will reduce the interrupt handle cost of KVM.
Also, interrupt enable logic is changed if this feature is turnning on:
Before this patch, hypervior call local_irq_enable() to enable it directly.
Now IF bit is set on interrupt stack frame, and will be enabled on a return from
interrupt handler if exterrupt interrupt exists. If no external interrupt, still
call local_irq_enable() to enable it.
Yang Zhang [Thu, 11 Apr 2013 11:21:41 +0000 (19:21 +0800)]
KVM: Use eoi to track RTC interrupt delivery status
Current interrupt coalescing logci which only used by RTC has conflict
with Posted Interrupt.
This patch introduces a new mechinism to use eoi to track interrupt:
When delivering an interrupt to vcpu, the pending_eoi set to number of
vcpu that received the interrupt. And decrease it when each vcpu writing
eoi. No subsequent RTC interrupt can deliver to vcpu until all vcpus
write eoi.
Yang Zhang [Thu, 11 Apr 2013 11:21:36 +0000 (19:21 +0800)]
KVM: Introduce struct rtc_status
rtc_status is used to track RTC interrupt delivery status. The pending_eoi
will be increased by vcpu who received RTC interrupt and will be decreased
when EOI to this interrupt.
Also, we use dest_map to record the destination vcpu to avoid the case that
vcpu who didn't get the RTC interupt, but issued EOI with same vector of RTC
and descreased pending_eoi by mistake.