and kvm_vcpu_kick() wakes up the target vcpu. We know which requests do
not need the wake up and use it to optimize the loop.
Thanks to that, this patch doesn't change the behavior of current users
(the all don't need the wake up) and only prepares for future where the
wake up is going to be needed.
I think that most requests do not need the wake up, so we would flip the
bit then.
Later on, kvm_make_request() will take care of kicking too, using this
bit to make the decision whether to kick or not.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com> Reviewed-by: Andrew Jones <drjones@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Some operations must ensure that the guest is not running with stale
data, but if the guest is halted, then the update can wait until another
event happens. kvm_make_all_requests() currently doesn't wake up, so we
can mark all requests used with it.
First 8 bits were arbitrarily reserved for request numbers.
Most uses of requests have the request type as a constant, so a compiler
will optimize the '&'.
An alternative would be to have an inline function that would return
whether the request needs a wake-up or not, but I like this one better
even though it might produce worse assembly.
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com> Reviewed-by: Andrew Jones <drjones@redhat.com> Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM: x86: always use kvm_make_request instead of set_bit
Reviewed-by: David Hildenbrand <david@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com> Reviewed-by: Andrew Jones <drjones@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM: add kvm_{test,clear}_request to replace {test,clear}_bit
Users were expected to use kvm_check_request() for testing and clearing,
but request have expanded their use since then and some users want to
only test or do a faster clear.
Make sure that requests are not directly accessed with bit operations.
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com> Reviewed-by: Andrew Jones <drjones@redhat.com> Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Jason J. Herne [Mon, 20 Mar 2017 13:57:42 +0000 (09:57 -0400)]
s390: kvm: Cpu model support for msa6, msa7 and msa8
msa6 and msa7 require no changes.
msa8 adds kma instruction and feature area.
Signed-off-by: Jason J. Herne <jjherne@linux.vnet.ibm.com> Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Jason J. Herne [Thu, 2 Feb 2017 13:25:19 +0000 (08:25 -0500)]
s390/cpacf: Introduce kma instruction
Provide a kma instruction definition for use by callers of __cpacf_query.
Signed-off-by: Jason J. Herne <jjherne@linux.vnet.ibm.com> Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Jason J. Herne [Tue, 21 Feb 2017 14:00:54 +0000 (09:00 -0500)]
s390/cpacf: query instructions use unique parameters for compatibility with KMA
The new KMA instruction requires unique parameters. Update __cpacf_query to
generate a compatible assembler instruction.
Signed-off-by: Jason J. Herne <jjherne@linux.vnet.ibm.com> Acked-by: Harald Freudenberger <freude@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
KVM: x86: remove irq disablement around KVM_SET_CLOCK/KVM_GET_CLOCK
The disablement of interrupts at KVM_SET_CLOCK/KVM_GET_CLOCK
attempts to disable software suspend from causing "non atomic behaviour" of
the operation:
Add a helper function to compute the kernel time and convert nanoseconds
back to CPU specific cycles. Note that these must not be called in preemptible
context, as that would mean the kernel could enter software suspend state,
which would cause non-atomic operation.
However, assume the kernel can enter software suspend at the following 2 points:
monotonic_to_bootbased() should be correct relative to a ktime_get_ts(&ts)
performed after point 1 (that is after resuming from software suspend),
hypothetical_ktime_get_ts()
Therefore it is also correct for the ktime_get_ts(&ts) before point 1,
which is
Note CLOCK_MONOTONIC does not count during suspension.
So remove the irq disablement, which causes the following warning on
-RT kernels:
With this reasoning, and the -RT bug that the irq disablement causes
(because spin_lock is now a sleeping lock), remove the IRQ protection as it
causes:
Guests that are heavy on futexes end up IPI'ing each other a lot. That
can lead to significant slowdowns and latency increase for those guests
when running within KVM.
If only a single guest is needed on a host, we have a lot of spare host
CPU time we can throw at the problem. Modern CPUs implement a feature
called "MWAIT" which allows guests to wake up sleeping remote CPUs without
an IPI - thus without an exit - at the expense of never going out of guest
context.
The decision whether this is something sensible to use should be up to the
VM admin, so to user space. We can however allow MWAIT execution on systems
that support it properly hardware wise.
This patch adds a CAP to user space and a KVM cpuid leaf to indicate
availability of native MWAIT execution. With that enabled, the worst a
guest can do is waste as many cycles as a "jmp ." would do, so it's not
a privilege problem.
We consciously do *not* expose the feature in our CPUID bitmap, as most
people will want to benefit from sleeping vCPUs to allow for over commit.
Reported-by: "Gabriel L. Somlo" <gsomlo@gmail.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
[agraf: fix amd, change commit message] Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Kyle Huey [Mon, 20 Mar 2017 08:16:28 +0000 (01:16 -0700)]
KVM: x86: virtualize cpuid faulting
Hardware support for faulting on the cpuid instruction is not required to
emulate it, because cpuid triggers a VM exit anyways. KVM handles the relevant
MSRs (MSR_PLATFORM_INFO and MSR_MISC_FEATURES_ENABLE) and upon a
cpuid-induced VM exit checks the cpuid faulting state and the CPL.
kvm_require_cpl is even kind enough to inject the GP fault for us.
Signed-off-by: Kyle Huey <khuey@kylehuey.com> Reviewed-by: David Matlack <dmatlack@google.com>
[Return "1" from kvm_emulate_cpuid, it's not void. - Paolo] Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
vmm_exclusive=0 leads to KVM setting X86_CR4_VMXE always and calling
VMXON only when the vcpu is loaded. X86_CR4_VMXE is used as an
indication in cpu_emergency_vmxoff() (called on kdump) if VMXOFF has to be
called. This is obviously not the case if both are used independtly.
Calling VMXOFF without a previous VMXON will result in an exception.
In addition, X86_CR4_VMXE is used as a mean to test if VMX is already in
use by another VMM in hardware_enable(). So there can't really be
co-existance. If the other VMM is prepared for co-existance and does a
similar check, only one VMM can exist. If the other VMM is not prepared
and blindly sets/clears X86_CR4_VMXE, we will get inconsistencies with
X86_CR4_VMXE.
As we also had bug reports related to clearing of vmcs with vmm_exclusive=0
this seems to be pretty much untested. So let's better drop it.
While at it, directly move setting/clearing X86_CR4_VMXE into
kvm_cpu_vmxon/off.
Signed-off-by: David Hildenbrand <david@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Farhan Ali [Fri, 24 Feb 2017 21:12:56 +0000 (16:12 -0500)]
KVM: s390: Support keyless subset guest mode
If the KSS facility is available on the machine, we also make it
available for our KVM guests.
The KSS facility bypasses storage key management as long as the guest
does not issue a related instruction. When that happens, the control is
returned to the host, which has to turn off KSS for a guest vcpu
before retrying the instruction.
Signed-off-by: Corey S. McQuay <csmcquay@linux.vnet.ibm.com> Signed-off-by: Farhan Ali <alifm@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Farhan Ali [Thu, 23 Feb 2017 18:39:17 +0000 (13:39 -0500)]
s390/sclp: Detect KSS facility
Let's detect the keyless subset facility.
Signed-off-by: Farhan Ali <alifm@linux.vnet.ibm.com> Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com> Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Thomas Huth [Wed, 5 Apr 2017 13:58:51 +0000 (15:58 +0200)]
KVM: PPC: Book3S PR: Do not fail emulation with mtspr/mfspr for unknown SPRs
According to the PowerISA 2.07, mtspr and mfspr should not always
generate an illegal instruction exception when being used with an
undefined SPR, but rather treat the instruction as a NOP or inject a
privilege exception in some cases, too - depending on the SPR number.
Also turn the printk here into a ratelimited print statement, so that
the guest can not flood the dmesg log of the host by issueing lots of
illegal mtspr/mfspr instruction here.
Signed-off-by: Thomas Huth <thuth@redhat.com> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
KVM: PPC: VFIO: Add in-kernel acceleration for VFIO
This allows the host kernel to handle H_PUT_TCE, H_PUT_TCE_INDIRECT
and H_STUFF_TCE requests targeted an IOMMU TCE table used for VFIO
without passing them to user space which saves time on switching
to user space and back.
This adds H_PUT_TCE/H_PUT_TCE_INDIRECT/H_STUFF_TCE handlers to KVM.
KVM tries to handle a TCE request in the real mode, if failed
it passes the request to the virtual mode to complete the operation.
If it a virtual mode handler fails, the request is passed to
the user space; this is not expected to happen though.
To avoid dealing with page use counters (which is tricky in real mode),
this only accelerates SPAPR TCE IOMMU v2 clients which are required
to pre-register the userspace memory. The very first TCE request will
be handled in the VFIO SPAPR TCE driver anyway as the userspace view
of the TCE table (iommu_table::it_userspace) is not allocated till
the very first mapping happens and we cannot call vmalloc in real mode.
If we fail to update a hardware IOMMU table unexpected reason, we just
clear it and move on as there is nothing really we can do about it -
for example, if we hot plug a VFIO device to a guest, existing TCE tables
will be mirrored automatically to the hardware and there is no interface
to report to the guest about possible failures.
This adds new attribute - KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE - to
the VFIO KVM device. It takes a VFIO group fd and SPAPR TCE table fd
and associates a physical IOMMU table with the SPAPR TCE table (which
is a guest view of the hardware IOMMU table). The iommu_table object
is cached and referenced so we do not have to look up for it in real mode.
This does not implement the UNSET counterpart as there is no use for it -
once the acceleration is enabled, the existing userspace won't
disable it unless a VFIO container is destroyed; this adds necessary
cleanup to the KVM_DEV_VFIO_GROUP_DEL handler.
This advertises the new KVM_CAP_SPAPR_TCE_VFIO capability to the user
space.
This adds real mode version of WARN_ON_ONCE() as the generic version
causes problems with rcu_sched. Since we testing what vmalloc_to_phys()
returns in the code, this also adds a check for already existing
vmalloc_to_phys() call in kvmppc_rm_h_put_tce_indirect().
This finally makes use of vfio_external_user_iommu_id() which was
introduced quite some time ago and was considered for removal.
Tests show that this patch increases transmission speed from 220MB/s
to 750..1020MB/s on 10Gb network (Chelsea CXGB3 10Gb ethernet card).
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Acked-by: Alex Williamson <alex.williamson@redhat.com> Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
KVM: PPC: Use preregistered memory API to access TCE list
VFIO on sPAPR already implements guest memory pre-registration
when the entire guest RAM gets pinned. This can be used to translate
the physical address of a guest page containing the TCE list
from H_PUT_TCE_INDIRECT.
This makes use of the pre-registrered memory API to access TCE list
pages in order to avoid unnecessary locking on the KVM memory
reverse map as we know that all of guest memory is pinned and
we have a flat array mapping GPA to HPA which makes it simpler and
quicker to index into that array (even with looking up the
kernel page tables in vmalloc_to_phys) than it is to find the memslot,
lock the rmap entry, look up the user page tables, and unlock the rmap
entry. Note that the rmap pointer is initialized to NULL
where declared (not in this patch).
If a requested chunk of memory has not been preregistered, this will
fall back to non-preregistered case and lock rmap.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
The guest view TCE tables are per KVM anyway (not per VCPU) so pass kvm*
there. This will be used in the following patches where we will be
attaching VFIO containers to LIOBNs via ioctl() to KVM (rather than
to VCPU).
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
KVM: PPC: Enable IOMMU_API for KVM_BOOK3S_64 permanently
It does not make much sense to have KVM in book3s-64 and
not to have IOMMU bits for PCI pass through support as it costs little
and allows VFIO to function on book3s KVM.
Having IOMMU_API always enabled makes it unnecessary to have a lot of
"#ifdef IOMMU_API" in arch/powerpc/kvm/book3s_64_vio*. With those
ifdef's we could have only user space emulated devices accelerated
(but not VFIO) which do not seem to be very useful.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
KVM: PPC: Reserve KVM_CAP_SPAPR_TCE_VFIO capability number
This adds a capability number for in-kernel support for VFIO on
SPAPR platform.
The capability will tell the user space whether in-kernel handlers of
H_PUT_TCE can handle VFIO-targeted requests or not. If not, the user space
must not attempt allocating a TCE table in the host kernel via
the KVM_CREATE_SPAPR_TCE KVM ioctl because in that case TCE requests
will not be passed to the user space which is desired action in
the situation like that.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
KVM: PPC: Align the table size to system page size
At the moment the userspace can request a table smaller than a page size
and this value will be stored as kvmppc_spapr_tce_table::size.
However the actual allocated size will still be aligned to the system
page size as alloc_page() is used there.
This aligns the table size up to the system page size. It should not
change the existing behaviour but when in-kernel TCE acceleration patchset
reaches the upstream kernel, this will allow small TCE tables be
accelerated as well: PCI IODA iommu_table allocator already aligns
the size and, without this patch, an IOMMU group won't attach to LIOBN
due to the mismatching table size.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
KVM: PPC: Book3S PR: Preserve storage control bits
PR KVM page fault handler performs eaddr to pte translation for a guest,
however kvmppc_mmu_book3s_64_xlate() does not preserve WIMG bits
(storage control) in the kvmppc_pte struct. If PR KVM is running as
a second level guest under HV KVM, and PR KVM tries inserting HPT entry,
this fails in HV KVM if it already has this mapping.
This preserves WIMG bits between kvmppc_mmu_book3s_64_xlate() and
kvmppc_mmu_map_page().
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
At the moment kvmppc_mmu_map_page() returns -1 if
mmu_hash_ops.hpte_insert() fails for any reason so the page fault handler
resumes the guest and it faults on the same address again.
This adds distinction to kvmppc_mmu_map_page() to return -EIO if
mmu_hash_ops.hpte_insert() failed for a reason other than full pteg.
At the moment only pSeries_lpar_hpte_insert() returns -2 if
plpar_pte_enter() failed with a code other than H_PTEG_FULL.
Other mmu_hash_ops.hpte_insert() instances can only fail with
-1 "full pteg".
With this change, if PR KVM fails to update HPT, it can signal
the userspace about this instead of returning to guest and having
the very same page fault over and over again.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Markus Elfring [Fri, 20 Jan 2017 15:30:18 +0000 (16:30 +0100)]
KVM: PPC: e500: Use kcalloc() in e500_mmu_host_init()
* A multiplication for the size determination of a memory allocation
indicated that an array data structure should be processed.
Thus use the corresponding function "kcalloc".
This issue was detected by using the Coccinelle software.
* Replace the specification of a data type by a pointer dereference
to make the corresponding size determination a bit safer according to
the Linux coding style convention.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Paul Mackerras [Thu, 23 Mar 2017 00:55:16 +0000 (11:55 +1100)]
KVM: PPC: Add MMIO emulation for remaining floating-point instructions
For completeness, this adds emulation of the lfiwax and lfiwzx
instructions. With this, all floating-point load and store instructions
as of Power ISA V2.07 are emulated.
Previously, most of these would cause an emulation failure exit to
userspace, though ldu and lwa got treated incorrectly as ld, and
stdu got treated incorrectly as std.
This also tidies up some of the formatting and updates the comment
listing instructions that still need to be implemented.
With this, all integer loads and stores that are defined in the Power
ISA v2.07 are emulated, except for those that are permitted to trap
when used on cache-inhibited or write-through mappings (and which do
in fact trap on POWER8), that is, lmw/stmw, lswi/stswi, lswx/stswx,
lq/stq, and l[bhwdq]arx/st[bhwdq]cx.
KVM: PPC: Add MMIO emulation for stdx (store doubleword indexed)
This adds missing stdx emulation for emulated MMIO accesses by KVM
guests. This allows the Mellanox mlx5_core driver from recent kernels
to work when MMIO emulation is enforced by userspace.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
Bin Lu [Tue, 21 Feb 2017 13:12:36 +0000 (21:12 +0800)]
KVM: PPC: Book3S: Add MMIO emulation for FP and VSX instructions
This patch provides the MMIO load/store emulation for instructions
of 'double & vector unsigned char & vector signed char & vector
unsigned short & vector signed short & vector unsigned int & vector
signed int & vector double '.
The instructions that this adds emulation for are:
Paul Mackerras [Wed, 22 Mar 2017 10:02:08 +0000 (21:02 +1100)]
KVM: PPC: Provide functions for queueing up FP/VEC/VSX unavailable interrupts
This provides functions that can be used for generating interrupts
indicating that a given functional unit (floating point, vector, or
VSX) is unavailable. These functions will be used in instruction
emulation code.
KVM: nVMX: fix AD condition when handling EPT violation
I have introduced this bug when applying and simplifying Paolo's patch
as we agreed on the list. The original was "x &= ~y; if (z) x |= y;".
Here is the story of a bad workflow:
A maintainer was already testing with the intended change, but it was
applied only to a testing repo on a different machine. When the time
to push tested patches to kvm/next came, he realized that this change
was missing and quickly added it to the maintenance repo, didn't test
again (because the change is trivial, right), and pushed the world to
fire.
Wanpeng Li [Tue, 11 Apr 2017 09:49:21 +0000 (02:49 -0700)]
x86/kvm: virt_xxx memory barriers instead of mandatory barriers
virt_xxx memory barriers are implemented trivially using the low-level
__smp_xxx macros, __smp_xxx is equal to a compiler barrier for strong
TSO memory model, however, mandatory barriers will unconditional add
memory barriers, this patch replaces the rmb() in kvm_steal_clock() by
virt_rmb().
Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim Krčmář <rkrcmar@redhat.com> Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
KVM: x86: fix maintaining of kvm_clock stability on guest CPU hotplug
VCPU TSC synchronization is perfromed in kvm_write_tsc() when the TSC
value being set is within 1 second from the expected, as obtained by
extrapolating of the TSC in already synchronized VCPUs.
This is naturally achieved on all VCPUs at VM start and resume;
however on VCPU hotplug it is not: the newly added VCPU is created
with TSC == 0 while others are well ahead.
To compensate for that, consider host-initiated kvm_write_tsc() with
TSC == 0 a special case requiring synchronization regardless of the
current TSC on other VCPUs.
KVM: x86: don't take kvm->irq_lock when creating IRQCHIP
I don't see any reason any more for this lock, seemed to be used to protect
removal of kvm->arch.vpic / kvm->arch.vioapic when already partially
inititalized, now access is properly protected using kvm->arch.irqchip_mode
and this shouldn't be necessary anymore.
Signed-off-by: David Hildenbrand <david@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
It seemed like a nice idea to encapsulate access to kvm->arch.vpic. But
as the usage is already mixed, internal locks are taken outside of i8259.c
and grepping for "vpic" only is much easier, let's just get rid of
pic_irqchip().
Signed-off-by: David Hildenbrand <david@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
KVM: x86: check against irqchip_mode in ioapic_in_kernel()
KVM_IRQCHIP_KERNEL implies a fully inititalized ioapic, while
kvm->arch.vioapic might temporarily be set but invalidated again if e.g.
setting of default routing fails when setting KVM_CREATE_IRQCHIP.
Signed-off-by: David Hildenbrand <david@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
KVM: x86: check against irqchip_mode in pic_in_kernel()
Let's avoid checking against kvm->arch.vpic. We have kvm->arch.irqchip_mode
for that now.
KVM_IRQCHIP_KERNEL implies a fully inititalized pic, while kvm->arch.vpic
might temporarily be set but invalidated again if e.g. kvm_ioapic_init()
fails when setting KVM_CREATE_IRQCHIP. Although current users seem to be
fine, this avoids future bugs.
Signed-off-by: David Hildenbrand <david@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
KVM: x86: check against irqchip_mode in kvm_set_routing_entry()
Let's replace the checks for pic_in_kernel() and ioapic_in_kernel() by
checks against irqchip_mode.
Also make sure that creation of any route is only possible if we have
an lapic in kernel (irqchip_in_kernel()) or if we are currently
inititalizing the irqchip.
This is necessary to switch pic_in_kernel() and ioapic_in_kernel() to
irqchip_mode, too.
Signed-off-by: David Hildenbrand <david@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
KVM: s390: fix stale machine check data for guarded storage
When delivering a machine check the CPU state is "loaded", which
means that some registers are already in the host registers.
Before writing the register content into the machine check
save area, we must make sure that we save the content of the
registers into the data structures that are used for delivering
the machine check.
We already do the right thing for access, vector/floating point
registers, let's do the same for guarded storage.
Fixes: 4e0b1ab72b8a ("KVM: s390: gs support for kvm guests") Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
If the guest does not use the host register management, but it uses
the sdnx area, we must fill in a proper sdnxo value (address of sdnx
and the sdnxc).
Reported-by: David Hildenbrand <david@redhat.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Merge tag 'kvm-s390-next-4.12-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux
From: Christian Borntraeger <borntraeger@de.ibm.com>
KVM: s390: features for 4.12
1. guarded storage support for guests
This contains an s390 base Linux feature branch that is necessary
to implement the KVM part
2. Provide an interface to implement adapter interruption suppression
which is necessary for proper zPCI support
3. Use more defines instead of numbers
4. Provide logging for lazy enablement of runtime instrumentation
Jim Mattson [Wed, 5 Apr 2017 16:14:40 +0000 (09:14 -0700)]
kvm: nVMX: Disallow userspace-injected exceptions in guest mode
The userspace exception injection API and code path are entirely
unprepared for exceptions that might cause a VM-exit from L2 to L1, so
the best course of action may be to simply disallow this for now.
1. The API provides no mechanism for userspace to specify the new DR6
bits for a #DB exception or the new CR2 value for a #PF
exception. Presumably, userspace is expected to modify these registers
directly with KVM_SET_SREGS before the next KVM_RUN ioctl. However, in
the event that L1 intercepts the exception, these registers should not
be changed. Instead, the new values should be provided in the
exit_qualification field of vmcs12 (Intel SDM vol 3, section 27.1).
2. In the case of a userspace-injected #DB, inject_pending_event()
clears DR7.GD before calling vmx_queue_exception(). However, in the
event that L1 intercepts the exception, this is too early, because
DR7.GD should not be modified by a #DB that causes a VM-exit directly
(Intel SDM vol 3, section 27.1).
3. If the injected exception is a #PF, nested_vmx_check_exception()
doesn't properly check whether or not L1 is interested in the
associated error code (using the #PF error code mask and match fields
from vmcs12). It may either return 0 when it should call
nested_vmx_vmexit() or vice versa.
4. nested_vmx_check_exception() assumes that it is dealing with a
hardware-generated exception intercept from L2, with some of the
relevant details (the VM-exit interruption-information and the exit
qualification) live in vmcs02. For userspace-injected exceptions, this
is not the case.
5. prepare_vmcs12() assumes that when its exit_intr_info argument
specifies valid information with a valid error code that it can VMREAD
the VM-exit interruption error code from vmcs02. For
userspace-injected exceptions, this is not the case.
Signed-off-by: Jim Mattson <jmattson@google.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
KVM: x86: fix user triggerable warning in kvm_apic_accept_events()
If we already entered/are about to enter SMM, don't allow switching to
INIT/SIPI_RECEIVED, otherwise the next call to kvm_apic_accept_events()
will report a warning.
Same applies if we are already in MP state INIT_RECEIVED and SMM is
requested to be turned on. Refuse to set the VCPU events in this case.
Fixes: cd7764fe9f73 ("KVM: x86: latch INITs while in system management mode") Cc: stable@vger.kernel.org # 4.2+ Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: David Hildenbrand <david@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Paolo Bonzini [Fri, 31 Mar 2017 11:53:23 +0000 (13:53 +0200)]
kvm: make KVM_COALESCED_MMIO_PAGE_OFFSET public
Its value has never changed; we might as well make it part of the ABI instead
of using the return value of KVM_CHECK_EXTENSION(KVM_CAP_COALESCED_MMIO).
Because PPC does not always make MMIO available, the code has to be made
dependent on CONFIG_KVM_MMIO rather than KVM_COALESCED_MMIO_PAGE_OFFSET.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Paolo Bonzini [Fri, 31 Mar 2017 11:53:22 +0000 (13:53 +0200)]
kvm: make KVM_CAP_COALESCED_MMIO architecture agnostic
Remove code from architecture files that can be moved to virt/kvm, since there
is already common code for coalesced MMIO.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: David Hildenbrand <david@redhat.com>
[Removed a pointless 'break' after 'return'.] Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Paolo Bonzini [Thu, 30 Mar 2017 09:55:30 +0000 (11:55 +0200)]
kvm: nVMX: support EPT accessed/dirty bits
Now use bit 6 of EPTP to optionally enable A/D bits for EPTP. Another
thing to change is that, when EPT accessed and dirty bits are not in use,
VMX treats accesses to guest paging structures as data reads. When they
are in use (bit 6 of EPTP is set), they are treated as writes and the
corresponding EPT dirty bit is set. The MMU didn't know this detail,
so this patch adds it.
We also have to fix up the exit qualification. It may be wrong because
KVM sets bit 6 but the guest might not.
L1 emulates EPT A/D bits using write permissions, so in principle it may
be possible for EPT A/D bits to be used by L1 even though not available
in hardware. The problem is that guest page-table walks will be treated
as reads rather than writes, so they would not cause an EPT violation.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[Fixed typo in walk_addr_generic() comment and changed bit clear +
conditional-set pattern in handle_ept_violation() to conditional-clear] Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Paolo Bonzini [Thu, 30 Mar 2017 09:55:29 +0000 (11:55 +0200)]
kvm: x86: MMU support for EPT accessed/dirty bits
This prepares the MMU paging code for EPT accessed and dirty bits,
which can be enabled optionally at runtime. Code that updates the
accessed and dirty bits will need a pointer to the struct kvm_mmu.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Paolo Bonzini [Thu, 30 Mar 2017 09:55:28 +0000 (11:55 +0200)]
KVM: VMX: remove bogus check for invalid EPT violation
handle_ept_violation is checking for "guest-linear-address invalid" +
"not a paging-structure walk". However, _all_ EPT violations without
a valid guest linear address are paging structure walks, because those
EPT violations happen when loading the guest PDPTEs.
Therefore, the check can never be true, and even if it were, KVM doesn't
care about the guest linear address; it only uses the guest *physical*
address VMCS field. So, remove the check altogether.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Jim Mattson <jmattson@google.com> Reviewed-by: David Hildenbrand <david@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Paolo Bonzini [Wed, 8 Mar 2017 09:49:19 +0000 (10:49 +0100)]
KVM: nVMX: we support 1GB EPT pages
Large pages at the PDPE level can be emulated by the MMU, so the bit
can be set unconditionally in the EPT capabilities MSR. The same is
true of 2MB EPT pages, though all Intel processors with EPT in practice
support those.
Reviewed-by: David Hildenbrand <david@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Mon, 27 Mar 2017 12:30:40 +0000 (14:30 +0200)]
KVM: x86: drop legacy device assignment
Legacy device assignment has been deprecated since 4.2 (released
1.5 years ago). VFIO is better and everyone should have switched to it.
If they haven't, this should convince them. :)
Reviewed-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Mon, 27 Mar 2017 12:37:28 +0000 (14:37 +0200)]
KVM: VMX: require virtual NMI support
Virtual NMIs are only missing in Prescott and Yonah chips. Both are obsolete
for virtualization usage---Yonah is 32-bit only even---so drop vNMI emulation.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Borislav Petkov [Sun, 26 Mar 2017 21:51:24 +0000 (23:51 +0200)]
kvm/svm: Setup MCG_CAP on AMD properly
MCG_CAP[63:9] bits are reserved on AMD. However, on an AMD guest, this
MSR returns 0x100010a. More specifically, bit 24 is set, which is simply
wrong. That bit is MCG_SER_P and is present only on Intel. Thus, clean
up the reserved bits in order not to confuse guests.
Signed-off-by: Borislav Petkov <bp@suse.de> Cc: Joerg Roedel <joro@8bytes.org> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim Krčmář <rkrcmar@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Jim Mattson [Wed, 15 Mar 2017 14:56:11 +0000 (07:56 -0700)]
kvm: vmx: Don't use INVVPID when EPT is enabled
According to the Intel SDM, volume 3, section 28.3.2: Creating and
Using Cached Translation Information, "No linear mappings are used
while EPT is in use." INVEPT will invalidate both the guest-physical
mappings and the combined mappings in the TLBs and paging-structure
caches, so an INVVPID is superfluous.
Signed-off-by: Jim Mattson <jmattson@google.com> Reviewed-by: David Hildenbrand <david@redhat.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Yi Min Zhao [Fri, 10 Mar 2017 08:29:38 +0000 (09:29 +0100)]
KVM: s390: introduce AIS capability
Introduce a cap to enable AIS facility bit, and add documentation
for this capability.
Signed-off-by: Yi Min Zhao <zyimin@linux.vnet.ibm.com> Signed-off-by: Fei Li <sherrylf@linux.vnet.ibm.com> Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Merge tag 'kvm_mips_4.12_1' of git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/kvm-mips
From: James Hogan <james.hogan@imgtec.com>
KVM: MIPS: VZ support, Octeon III, and TLBR
Add basic support for the MIPS Virtualization Module (generally known as
MIPS VZ) in KVM. We primarily support the ImgTec P5600, P6600, I6400,
and Cavium Octeon III cores so far. Support is included for the
following VZ / guest hardware features:
- MIPS32 and MIPS64, r5 (VZ requires r5 or later) and r6
- TLBs with GuestID (IMG cores) or Root ASID Dealias (Octeon III)
- Shared physical root/guest TLB (IMG cores)
- FPU / MSA
- Cop0 timer (up to 1GHz for now due to soft timer limit)
- Segmentation control (EVA)
- Hardware page table walker (HTW) both for root and guest TLB
Also included is a proper implementation of the TLBR instruction for the
trap & emulate MIPS KVM implementation.
Preliminary MIPS architecture changes are applied directly with Ralf's
ack.
Yi Min Zhao [Mon, 20 Feb 2017 02:15:01 +0000 (10:15 +0800)]
KVM: s390: introduce adapter interrupt inject function
Inject adapter interrupts on a specified adapter which allows to
retrieve the adapter flags, e.g. if the adapter is subject to AIS
facility or not. And add documentation for this interface.
For adapters subject to AIS, handle the airq injection suppression
for a given ISC according to the interruption mode:
- before injection, if NO-Interruptions Mode, just return 0 and
suppress, otherwise, allow the injection.
- after injection, if SINGLE-Interruption Mode, change it to
NO-Interruptions Mode to suppress the following interrupts.
Besides, add tracepoint for suppressed airq and AIS mode transitions.
Signed-off-by: Yi Min Zhao <zyimin@linux.vnet.ibm.com> Signed-off-by: Fei Li <sherrylf@linux.vnet.ibm.com> Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Fei Li [Fri, 17 Feb 2017 09:06:26 +0000 (17:06 +0800)]
KVM: s390: introduce ais mode modify function
Provide an interface for userspace to modify AIS
(adapter-interruption-suppression) mode state, and add documentation
for the interface. Allowed target modes are ALL-Interruptions mode
and SINGLE-Interruption mode.
We introduce the 'simm' and 'nimm' fields in kvm_s390_float_interrupt
to store interruption modes for each ISC. Each bit in 'simm' and
'nimm' targets to one ISC, and collaboratively indicate three modes:
ALL-Interruptions, SINGLE-Interruption and NO-Interruptions. This
interface can initiate most transitions between the states; transition
from SINGLE-Interruption to NO-Interruptions via adapter interrupt
injection will be introduced in a following patch. The meaningful
combinations are as follows:
interruption mode | simm bit | nimm bit
------------------|----------|----------
ALL | 0 | 0
SINGLE | 1 | 0
NO | 1 | 1
Besides, add tracepoint to track AIS mode transitions.
Co-Authored-By: Yi Min Zhao <zyimin@linux.vnet.ibm.com> Signed-off-by: Yi Min Zhao <zyimin@linux.vnet.ibm.com> Signed-off-by: Fei Li <sherrylf@linux.vnet.ibm.com> Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Fei Li [Thu, 19 Jan 2017 16:02:26 +0000 (17:02 +0100)]
KVM: s390: interface for suppressible I/O adapters
In order to properly implement adapter-interruption suppression, we
need a way for userspace to specify which adapters are subject to
suppression. Let's convert the existing (and unused) 'pad' field into
a 'flags' field and define a flag value for suppressible adapters.
Besides, add documentation for the interface.
Signed-off-by: Fei Li <sherrylf@linux.vnet.ibm.com> Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
powerpc/vfio_spapr_tce: Add reference counting to iommu_table
So far iommu_table obejcts were only used in virtual mode and had
a single owner. We are going to change this by implementing in-kernel
acceleration of DMA mapping requests. The proposed acceleration
will handle requests in real mode and KVM will keep references to tables.
This adds a kref to iommu_table and defines new helpers to update it.
This replaces iommu_free_table() with iommu_tce_table_put() and makes
iommu_free_table() static. iommu_tce_table_get() is not used in this patch
but it will be in the following patch.
Since this touches prototypes, this also removes @node_name parameter as
it has never been really useful on powernv and carrying it for
the pseries platform code to iommu_free_table() seems to be quite
useless as well.
This should cause no behavioral change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Acked-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
At the moment iommu_table can be disposed by either calling
iommu_table_free() directly or it_ops::free(); the only implementation
of free() is in IODA2 - pnv_ioda2_table_free() - and it calls
iommu_table_free() anyway.
As we are going to have reference counting on tables, we need an unified
way of disposing tables.
This moves it_ops::free() call into iommu_free_table() and makes use
of the latter. The free() callback now handles only platform-specific
data.
As from now on the iommu_free_table() calls it_ops->free(), we need
to have it_ops initialized before calling iommu_free_table() so this
moves this initialization in pnv_pci_ioda2_create_table().
This should cause no behavioral change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Acked-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/powernv/iommu: Add real mode version of iommu_table_ops::exchange()
In real mode, TCE tables are invalidated using special
cache-inhibited store instructions which are not available in
virtual mode
This defines and implements exchange_rm() callback. This does not
define set_rm/clear_rm/flush_rm callbacks as there is no user for those -
exchange/exchange_rm are only to be used by KVM for VFIO.
The exchange_rm callback is defined for IODA1/IODA2 powernv platforms.
This replaces list_for_each_entry_rcu with its lockless version as
from now on pnv_pci_ioda2_tce_invalidate() can be called in
the real mode too.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Stefan Raspl [Fri, 10 Mar 2017 12:40:15 +0000 (13:40 +0100)]
tools/kvm_stat: add interactive command 'r'
Provide an interactive command to reset the tracepoint statistics.
Requires some extra work for debugfs, as the counters cannot be reset.
On the up side, this offers us the opportunity to have debugfs values
reset on startup and whenever a filter is modified, becoming consistent
with the tracepoint provider. As a bonus, 'kvmstat -dt' will now provide
useful output, instead of mixing values in totally different orders of
magnitude.
Furthermore, we avoid unnecessary resets when any of the filters is
"changed" interactively to the previous value.
Signed-off-by: Stefan Raspl <raspl@linux.vnet.ibm.com> Acked-by: Janosch Frank <frankja@linux.vnet.ibm.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Stefan Raspl [Fri, 10 Mar 2017 12:40:13 +0000 (13:40 +0100)]
tools/kvm_stat: add option '--guest'
Add a new option '-g'/'--guest' to select a particular process by providing
the QEMU guest name.
Notes:
- The logic to figure out the pid corresponding to the guest name might look
scary, but works pretty reliably in practice; in the unlikely event that it
returns add'l flukes, it will bail out and hint at using '-p' instead, no
harm done.
- Mixing '-g' and '-p' is possible, and the final instance specified on the
command line is the significant one. This is consistent with current
behavior for '-p' which, if specified multiple times, also regards the final
instance as the significant one.
Signed-off-by: Stefan Raspl <raspl@linux.vnet.ibm.com> Reviewed-by: Janosch Frank <frankja@linux.vnet.ibm.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Stefan Raspl [Fri, 10 Mar 2017 12:40:12 +0000 (13:40 +0100)]
tools/kvm_stat: remove regex filter on empty input
Behavior on empty/0 input for regex and pid filtering was inconsistent, as
the former would keep the current filter, while the latter would (naturally)
remove any pid filtering.
Make things consistent by falling back to the default filter on empty input
for the regex filter dialogue.
Signed-off-by: Stefan Raspl <raspl@linux.vnet.ibm.com> Reviewed-by: Marc Hartmayer <mhartmay@linux.vnet.ibm.com> Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
projects / mirror_ubuntu-artful-kernel.git / log