The global_invalidates() function contains a check that is intended
to tell whether we are currently executing in the context of a hypercall
issued by the guest. The reason is that the optimization of using a
local TLB invalidate instruction is only valid in that context. The
check was testing local_paca->kvm_hstate.kvm_vcore, which gets set
when entering the guest but no longer gets cleared when exiting the
guest. To fix this, we use the kvm_vcpu field instead, which does
get cleared when exiting the guest, by the kvmppc_release_hwthread()
calls inside kvmppc_run_core().
The effect of having the check wrong was that when kvmppc_do_h_remove()
got called from htab_write() on the destination machine during a
migration, it cleared the current cpu's bit in kvm->arch.need_tlb_flush.
This meant that when the guest started running in the destination VM,
it may miss out on doing a complete TLB flush, and therefore may end
up using stale TLB entries from a previous guest that used the same
LPID value.
This should make migration more reliable.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Commit b005255e12 ("KVM: PPC: Book3S HV: Context-switch new POWER8
SPRs") added a definition of KVM_REG_PPC_WORT with the same register
number as the existing KVM_REG_PPC_VRSAVE (though in fact the
definitions are not identical because of the different register sizes.)
For clarity, this moves KVM_REG_PPC_WORT to the next unused number,
and also adds it to api.txt.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
We worked around some nasty KVM magic page hcall breakages:
1) NX bit not honored, so ignore NX when we detect it
2) LE guests swizzle hypercall instruction
Without these fixes in place, there's no way it would make sense to expose kvm
hypercalls to a guest. Chances are immensely high it would trip over and break.
So add a new CAP that gives user space a hint that we have workarounds for the
bugs above in place. It can use those as hint to disable PV hypercalls when
the guest CPU is anything POWER7 or higher and the host does not have fixes
in place.
Signed-off-by: Alexander Graf <agraf@suse.de>
When we reset the in-kernel MPIC controller, we forget to reset some hidden
state such as destmask and output. This state is usually set when the guest
writes to the IDR register for a specific IRQ line.
To make sure we stay in sync and don't forget hidden state, treat reset of
the IDR register as a simple write of the IDR register. That automatically
updates all the hidden state as well.
Reported-by: Paul Janzen <pcj@pauljanzen.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
There are LE Linux guests out there that don't handle hypercalls correctly.
Instead of interpreting the instruction stream from device tree as big endian
they assume it's a little endian instruction stream and fail.
When we see an illegal instruction from such a byte reversed instruction stream,
bail out graciously and just declare every hcall as error.
Signed-off-by: Alexander Graf <agraf@suse.de>
We get an array of instructions from the hypervisor via device tree that
we write into a buffer that gets executed whenever we want to make an
ePAPR compliant hypercall.
However, the hypervisor passes us these instructions in BE order which
we have to manually convert to LE when we want to run them in LE mode.
With this fixup in place, I can successfully run LE kernels with KVM
PV enabled on PR KVM.
Signed-off-by: Alexander Graf <agraf@suse.de>
Use make_dsisr instead of open coding it. This also have
the added benefit of handling alignment interrupt on additional
instructions.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Although it's optional, IBM POWER cpus always had DAR value set on
alignment interrupt. So don't try to compute these values.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Because old kernels enable the magic page and then choke on NXed trampoline
code we have to disable NX by default in KVM when we use the magic page.
However, since commit b18db0b8 we have successfully fixed that and can now
leave NX enabled, so tell the hypervisor about this.
Signed-off-by: Alexander Graf <agraf@suse.de>
Old guests try to use the magic page, but map their trampoline code inside
of an NX region.
Since we can't fix those old kernels, try to detect whether the guest is sane
or not. If not, just disable NX functionality in KVM so that old guests at
least work at all. For newer guests, add a bit that we can set to keep NX
functionality available.
Signed-off-by: Alexander Graf <agraf@suse.de>
On recent IBM Power CPUs, while the hashed page table is looked up using
the page size from the segmentation hardware (i.e. the SLB), it is
possible to have the HPT entry indicate a larger page size. Thus for
example it is possible to put a 16MB page in a 64kB segment, but since
the hash lookup is done using a 64kB page size, it may be necessary to
put multiple entries in the HPT for a single 16MB page. This
capability is called mixed page-size segment (MPSS). With MPSS,
there are two relevant page sizes: the base page size, which is the
size used in searching the HPT, and the actual page size, which is the
size indicated in the HPT entry. [ Note that the actual page size is
always >= base page size ].
We use "ibm,segment-page-sizes" device tree node to advertise
the MPSS support to PAPR guest. The penc encoding indicates whether
we support a specific combination of base page size and actual
page size in the same segment. We also use the penc value in the
LP encoding of HPTE entry.
This patch exposes MPSS support to KVM guest by advertising the
feature via "ibm,segment-page-sizes". It also adds the necessary changes
to decode the base page size and the actual page size correctly from the
HPTE entry.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Today when KVM tries to reserve memory for the hash page table it
allocates from the normal page allocator first. If that fails it
falls back to CMA's reserved region. One of the side effects of
this is that we could end up exhausting the page allocator and
get linux into OOM conditions while we still have plenty of space
available in CMA.
This patch addresses this issue by first trying hash page table
allocation from CMA's reserved region before falling back to the normal
page allocator. So if we run out of memory, we really are out of memory.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
POWER8 introduces transactional memory which brings along a number of new
registers and MSR bits.
Implementing all of those is a pretty big headache, so for now let's at least
emulate enough to make Linux's context switching code happy.
Signed-off-by: Alexander Graf <agraf@suse.de>
POWER8 introduces a new facility called the "Event Based Branch" facility.
It contains of a few registers that indicate where a guest should branch to
when a defined event occurs and it's in PR mode.
We don't want to really enable EBB as it will create a big mess with !PR guest
mode while hardware is in PR and we don't really emulate the PMU anyway.
So instead, let's just leave it at emulation of all its registers.
Signed-off-by: Alexander Graf <agraf@suse.de>
POWER8 implements a new register called TAR. This register has to be
enabled in FSCR and then from KVM's point of view is mere storage.
This patch enables the guest to use TAR.
Signed-off-by: Alexander Graf <agraf@suse.de>
POWER8 introduced a new interrupt type called "Facility unavailable interrupt"
which contains its status message in a new register called FSCR.
Handle these exits and try to emulate instructions for unhandled facilities.
Follow-on patches enable KVM to expose specific facilities into the guest.
Signed-off-by: Alexander Graf <agraf@suse.de>
In parallel to the Processor ID Register (PIR) threaded POWER8 also adds a
Thread ID Register (TIR). Since PR KVM doesn't emulate more than one thread
per core, we can just always expose 0 here.
Signed-off-by: Alexander Graf <agraf@suse.de>
When we expose a POWER8 CPU into the guest, it will start accessing PMU SPRs
that we don't emulate. Just ignore accesses to them.
Signed-off-by: Alexander Graf <agraf@suse.de>
With the previous patches applied, we can now successfully use PR KVM on
little endian hosts which means we can now allow users to select it.
However, HV KVM still needs some work, so let's keep the kconfig conflict
on that one.
Signed-off-by: Alexander Graf <agraf@suse.de>
When the host CPU we're running on doesn't support dcbz32 itself, but the
guest wants to have dcbz only clear 32 bytes of data, we loop through every
executable mapped page to search for dcbz instructions and patch them with
a special privileged instruction that we emulate as dcbz32.
The only guests that want to see dcbz act as 32byte are book3s_32 guests, so
we don't have to worry about little endian instruction ordering. So let's
just always search for big endian dcbz instructions, also when we're on a
little endian host.
Signed-off-by: Alexander Graf <agraf@suse.de>
The shared (magic) page is a data structure that contains often used
supervisor privileged SPRs accessible via memory to the user to reduce
the number of exits we have to take to read/write them.
When we actually share this structure with the guest we have to maintain
it in guest endianness, because some of the patch tricks only work with
native endian load/store operations.
Since we only share the structure with either host or guest in little
endian on book3s_64 pr mode, we don't have to worry about booke or book3s hv.
For booke, the shared struct stays big endian. For book3s_64 hv we maintain
the struct in host native endian, since it never gets shared with the guest.
For book3s_64 pr we introduce a variable that tells us which endianness the
shared struct is in and route every access to it through helper inline
functions that evaluate this variable.
Signed-off-by: Alexander Graf <agraf@suse.de>
We expose a blob of hypercall instructions to user space that it gives to
the guest via device tree again. That blob should contain a stream of
instructions necessary to do a hypercall in big endian, as it just gets
passed into the guest and old guests use them straight away.
Signed-off-by: Alexander Graf <agraf@suse.de>
When the guest does an RTAS hypercall it keeps all RTAS variables inside a
big endian data structure.
To make sure we don't have to bother about endianness inside the actual RTAS
handlers, let's just convert the whole structure to host endian before we
call our RTAS handlers and back to big endian when we return to the guest.
Signed-off-by: Alexander Graf <agraf@suse.de>
The HTAB on PPC is always in big endian. When we access it via hypercalls
on behalf of the guest and we're running on a little endian host, we need
to make sure we swap the bits accordingly.
Signed-off-by: Alexander Graf <agraf@suse.de>
The default MSR when user space does not define anything should be identical
on little and big endian hosts, so remove MSR_LE from it.
Signed-off-by: Alexander Graf <agraf@suse.de>
The "shadow SLB" in the PACA is shared with the hypervisor, so it has to
be big endian. We access the shadow SLB during world switch, so let's make
sure we access it in big endian even when we're on a little endian host.
Signed-off-by: Alexander Graf <agraf@suse.de>
The HTAB is always big endian. We access the guest's HTAB using
copy_from/to_user, but don't yet take care of the fact that we might
be running on an LE host.
Wrap all accesses to the guest HTAB with big endian accessors.
Signed-off-by: Alexander Graf <agraf@suse.de>
The HTAB is always big endian. We access the guest's HTAB using
copy_from/to_user, but don't yet take care of the fact that we might
be running on an LE host.
Wrap all accesses to the guest HTAB with big endian accessors.
Signed-off-by: Alexander Graf <agraf@suse.de>
Commit 9308ab8e2d made C/R HTAB updates go byte-wise into the target HTAB.
However, it didn't update the guest's copy of the HTAB, but instead the
host local copy of it.
Write to the guest's HTAB instead.
Signed-off-by: Alexander Graf <agraf@suse.de>
CC: Paul Mackerras <paulus@samba.org>
Acked-by: Paul Mackerras <paulus@samba.org>
This patch make sure we inherit the LE bit correctly in different case
so that we can run Little Endian distro in PR mode
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
The dcbtls instruction is able to lock data inside the L1 cache.
We don't want to give the guest actual access to hardware cache locks,
as that could influence other VMs on the same system. But we can tell
the guest that its locking attempt failed.
By implementing the instruction we at least don't give the guest a
program exception which it definitely does not expect.
Signed-off-by: Alexander Graf <agraf@suse.de>
The L1 instruction cache control register contains bits that indicate
that we're still handling a request. Mask those out when we set the SPR
so that a read doesn't assume we're still doing something.
Signed-off-by: Alexander Graf <agraf@suse.de>
The DR7 masking which is done on task switch emulation should be in hex format
(clearing the local breakpoints enable bits 0,2,4 and 6).
Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
I noticed on some of my systems that page fault tracing doesn't
work:
cd /sys/kernel/debug/tracing
echo 1 > events/exceptions/enable
cat trace;
# nothing shows up
I eventually traced it down to CONFIG_KVM_GUEST. At least in a
KVM VM, enabling that option breaks page fault tracing, and
disabling fixes it. I tried on some old kernels and this does
not appear to be a regression: it never worked.
There are two page-fault entry functions today. One when tracing
is on and another when it is off. The KVM code calls do_page_fault()
directly instead of calling the traced version:
> dotraplinkage void __kprobes
> do_async_page_fault(struct pt_regs *regs, unsigned long
> error_code)
> {
> enum ctx_state prev_state;
>
> switch (kvm_read_and_reset_pf_reason()) {
> default:
> do_page_fault(regs, error_code);
> break;
> case KVM_PV_REASON_PAGE_NOT_PRESENT:
I'm also having problems with the page fault tracing on bare
metal (same symptom of no trace output). I'm unsure if it's
related.
Steven had an alternative to this which has zero overhead when
tracing is off where this includes the standard noops even when
tracing is disabled. I'm unconvinced that the extra complexity
of his apporach:
http://lkml.kernel.org/r/20140508194508.561ed220@gandalf.local.home
is worth it, expecially considering that the KVM code is already
making page fault entry slower here. This solution is
dirt-simple.
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: x86@kernel.org
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Gleb Natapov <gleb@redhat.com>
Cc: kvm@vger.kernel.org
Cc: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: "H. Peter Anvin" <hpa@zytor.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
CS.RPL is not equal to the CPL in the few instructions between
setting CR0.PE and reloading CS. And CS.DPL is also not equal
to the CPL for conforming code segments.
However, SS.DPL *is* always equal to the CPL except for the weird
case of SYSRET on AMD processors, which sets SS.DPL=SS.RPL from the
value in the STAR MSR, but force CPL=3 (Intel instead forces
SS.DPL=SS.RPL=CPL=3).
So this patch:
- modifies SVM to update the CPL from SS.DPL rather than CS.RPL;
the above case with SYSRET is not broken further, and the way
to fix it would be to pass the CPL to userspace and back
- modifies VMX to always return the CPL from SS.DPL (except
forcing it to 0 if we are emulating real mode via vm86 mode;
in vm86 mode all DPLs have to be 3, but real mode does allow
privileged instructions). It also removes the CPL cache,
which becomes a duplicate of the SS access rights cache.
This fixes doing KVM_IOCTL_SET_SREGS exactly after setting
CR0.PE=1 but before CS has been reloaded.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Table 7-1 of the SDM mentions a check that the code segment's
DPL must match the selector's RPL. This was not done by KVM,
fix it.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
During task switch, all of CS.DPL, CS.RPL, SS.DPL must match (in addition
to all the other requirements) and will be the new CPL. So far this
worked by carefully setting the CS selector and flag before doing the
task switch; setting CS.selector will already change the CPL.
However, this will not work once we get the CPL from SS.DPL, because
then you will have to set the full segment descriptor cache to change
the CPL. ctxt->ops->cpl(ctxt) will then return the old CPL during the
task switch, and the check that SS.DPL == CPL will fail.
Temporarily assume that the CPL comes from CS.RPL during task switch
to a protected-mode task. This is the same approach used in QEMU's
emulation code, which (until version 2.0) manually tracks the CPL.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch splits the SIE state guest prefix at offset 4
into a prefix bit field. Additionally it provides the
access functions:
- kvm_s390_get_prefix()
- kvm_s390_set_prefix()
to access the prefix per vcpu.
Signed-off-by: Michael Mueller <mimu@linux.vnet.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
The patch adds functionality to retrieve the IBC configuration
by means of function sclp_get_ibc().
Signed-off-by: Michael Mueller <mimu@linux.vnet.ibm.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
If the sigp interpretation facility is installed, most SIGP EXTERNAL CALL
operations will be interpreted instead of intercepted. A partial execution
interception will occurr at the sending cpu only if the target cpu is in the
wait state ("W" bit in the cpuflags set). Instruction interception will only
happen in error cases (e.g. cpu addr invalid).
As a sending cpu might set the external call interrupt pending flags at the
target cpu at every point in time, we can't handle this kind of interrupt using
our kvm interrupt injection mechanism. The injection will be done automatically
by the SIE when preparing the start of the target cpu.
Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
CC: Thomas Huth <thuth@linux.vnet.ibm.com>
[Adopt external call injection to check for sigp interpretion]
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
The current trace definition doesn't work very well with the perf tool.
Perf shows a "insn_to_mnemonic not found" message. Let's handle the
decoding completely in a parseable format.
Signed-off-by: Alexander Yarygin <yarygin@linux.vnet.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
This patch adds a new decoder of SIE intercepted instructions.
The decoder implemented as a macro and potentially can be used in
both kernelspace and userspace.
Note that this simplified instruction decoder is only intended to be
used with the subset of instructions that may cause a SIE intercept.
Signed-off-by: Alexander Yarygin <yarygin@linux.vnet.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Use the symbolic translation tables from sie.h for decoding diag, sigp
and sie exit codes.
Signed-off-by: Alexander Yarygin <yarygin@linux.vnet.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
This patch defines tables of reasons for exiting from SIE mode
in a new sie.h header file. Tables contain SIE intercepted codes,
intercepted instructions and program interruptions codes.
Signed-off-by: Alexander Yarygin <yarygin@linux.vnet.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Use the new helper function kvm_arch_fault_in_page() for faulting-in
the guest pages and only inject addressing errors when we've really
hit a bad address (and return other error codes to userspace instead).
Signed-off-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Rework the function kvm_arch_fault_in_sync() to become a proper helper
function for faulting-in a guest page. Now it takes the guest address as
a parameter and does not ignore the possible error code from gmap_fault()
anymore (which could cause undetected error conditions before).
Signed-off-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
If the new PSW for program interrupts is invalid, the VM ends up
in an endless loop of specification exceptions. Since there is not
much left we can do in this case, we should better drop to userspace
instead so that the crash can be reported to the user.
Signed-off-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
As a program status word is also invalid (and thus generates an
specification exception) if the instruction address is not even,
we should test this in is_valid_psw(), too. This patch also exports
the function so that it becomes available for other parts of the
S390 KVM code as well.
Signed-off-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
