Commit 69acc0d3ba ("KVM: PPC: Resolve real-mode handlers through
function exports") resulted in vcpu->arch.trampoline_lowmem and
vcpu->arch.trampoline_enter ending up with kernel virtual addresses
rather than physical addresses.  This is OK on 64-bit Book3S machines,
which ignore the top 4 bits of the effective address in real mode,
but on 32-bit Book3S machines, accessing these addresses in real mode
causes machine check interrupts, as the hardware uses the whole
effective address as the physical address in real mode.

This fixes the problem by using __pa() to convert these addresses
to physical addresses.
Signed-off-by: NPaul Mackerras <paulus@samba.org>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Suggested by Ingo and Avi.

Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NTakuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

The current name does not explain the meaning well.  So give it a better
name "retry_walk" to show that we are trying the walk again.

This was suggested by Ingo Molnar.

Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NTakuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

Avoid two step jump to the error handling part.  This eliminates the use
of the variables present and rsvd_fault.

We also use the const type qualifier to show that write/user/fetch_fault
do not change in the function.

Both of these were suggested by Ingo Molnar.

Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NTakuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

This reverts commit bee931d31e588b8eb86b7edee32fac2d16930cd7.

TLB flush should be done lazily during guest entry, in
kvm_mmu_load().
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

Only look in the 4 entries that could possibly contain the
entry we're looking for.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Dynamically assign host PIDs to guest PIDs, splitting each guest PID into
multiple host (shadow) PIDs based on kernel/user and MSR[IS/DS].  Use
both PID0 and PID1 so that the shadow PIDs for the right mode can be
selected, that correspond both to guest TID = zero and guest TID = guest
PID.

This allows us to significantly reduce the frequency of needing to
invalidate the entire TLB.  When the guest mode or PID changes, we just
update the host PID0/PID1.  And since the allocation of shadow PIDs is
global, multiple guests can share the TLB without conflict.

Note that KVM does not yet support the guest setting PID1 or PID2 to
a value other than zero.  This will need to be fixed for nested KVM
to work.  Until then, we enforce the requirement for guest PID1/PID2
to stay zero by failing the emulation if the guest tries to set them
to something else.
Signed-off-by: NLiu Yu <yu.liu@freescale.com>
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Instead of a fully separate set of TLB entries, keep just the
pfn and dirty status.
Signed-off-by: NLiu Yu <yu.liu@freescale.com>
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

This is a shared page used for paravirtualization.  It is always present
in the guest kernel's effective address space at the address indicated
by the hypercall that enables it.

The physical address specified by the hypercall is not used, as
e500 does not have real mode.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

This allows large pages to be used on guest mappings backed by things like
/dev/mem, resulting in a significant speedup when guest memory
is mapped this way (it's useful for directly-assigned MMIO, too).

This is not a substitute for hugetlbfs integration, but is useful for
configurations where devices are directly assigned on chips without an
IOMMU -- in these cases, we need guest physical and true physical to
match, and be contiguous, so static reservation and mapping via /dev/mem
is the most straightforward way to set things up.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

This is in line with what other architectures do, and will allow us to
map things other than ordinary, unreserved kernel pages -- such as
dedicated devices, or large contiguous reserved regions.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

This avoids races.  It also means that we use the shadow TLB way,
rather than the hardware hint -- if this is a problem, we could do
a tlbsx before inserting a TLB0 entry.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Since TLB1 loading doesn't check the shadow TLB before allocating another
entry, you can get duplicates.

Once shadow PIDs are enabled in a later patch, we won't need to
invalidate the TLB on every switch, so this optimization won't be
needed anyway.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

This is done lazily.  The SPE save will be done only if the guest has
used SPE since the last preemption or heavyweight exit.  Restore will be
done only on demand, when enabling MSR_SPE in the shadow MSR, in response
to an SPE fault or mtmsr emulation.

For SPEFSCR, Linux already switches it on context switch (non-lazily), so
the only remaining bit is to save it between qemu and the guest.
Signed-off-by: NLiu Yu <yu.liu@freescale.com>
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Keep the guest MSR and the guest-mode true MSR separate, rather than
modifying the guest MSR on each guest entry to produce a true MSR.

Any bits which should be modified based on guest MSR must be explicitly
propagated from vcpu->arch.shared->msr to vcpu->arch.shadow_msr in
kvmppc_set_msr().

While we're modifying the guest entry code, reorder a few instructions
to bury some load latencies.
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Previously, these macros hardcoded THREAD_EVR0 as the base of the save
area, relative to the base register passed.  This base offset is now
passed as a separate macro parameter, allowing reuse with other SPE
save areas, such as used by KVM.
Acked-by: NKumar Gala <galak@kernel.crashing.org>
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

giveup_spe() saves the SPE state which is protected by MSR[SPE].
However, modifying SPEFSCR does not trap when MSR[SPE]=0.
And since SPEFSCR is already saved/restored in _switch(),
not all the callers want to save SPEFSCR again.
Thus, saving SPEFSCR should not belong to giveup_spe().

This patch moves SPEFSCR saving to flush_spe_to_thread(),
and cleans up the caller that needs to save SPEFSCR accordingly.
Signed-off-by: NLiu Yu <yu.liu@freescale.com>
Acked-by: NKumar Gala <galak@kernel.crashing.org>
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Up until now, Book3S KVM had variables stored in the kernel that a kernel module
or the kvm code in the kernel could read from to figure out where some real mode
helper functions are located.

This is all unnecessary. The high bits of the EA get ignore in real mode, so we
can just use the pointer as is. Also, it's a lot easier on relocations when we
use the normal way of resolving the address to a function, instead of jumping
through hoops.

This patch fixes compilation with CONFIG_RELOCATABLE=y.
Signed-off-by: NAlexander Graf <agraf@suse.de>

When http://www.spinics.net/lists/kvm-ppc/msg02664.html
was applied to produce commit b51e7aa7ed6d8d134d02df78300ab0f91cfff4d2,
the removal of the conversion in add_exit_timing was left out.
Signed-off-by: NStuart Yoder <stuart.yoder@freescale.com>
Signed-off-by: NScott Wood <scottwood@freescale.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

kvm_set_cr0() and kvm_set_cr4(), and possible other functions,
assume that kvm_mmu_reset_context() flushes the guest TLB.  However,
it does not.

Fix by flushing the tlb (and syncing the new root as well).
Signed-off-by: NAvi Kivity <avi@redhat.com>

When CR0.WP=0, we sometimes map user pages as kernel pages (to allow
the kernel to write to them).  Unfortunately this also allows the kernel
to fetch from these pages, even if CR4.SMEP is set.

Adjust for this by also setting NX on the spte in these circumstances.
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch exposes ERMS feature to KVM guests.

The REP MOVSB/STOSB instruction can enhance fast strings attempts to
move as much of the data with larger size load/stores as possible.
Signed-off-by: NYang, Wei <wei.y.yang@intel.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch exposes RDWRGSFS bit to KVM guests.
Signed-off-by: NYang, Wei <wei.y.yang@intel.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch adds RDWRGSFS support when setting CR4.
Signed-off-by: NYang, Wei <wei.y.yang@intel.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch removes RDWRGSFS bit from CR4_RESERVED_BITS.
Signed-off-by: NYang, Wei <wei.y.yang@intel.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch exposes DRNG feature to KVM guests.

The RDRAND instruction can provide software with sequences of
random numbers generated from white noise.
Signed-off-by: NYang, Wei <wei.y.yang@intel.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

commit 123108f1c1aafd51d6a5c79cc04d7999dd88a930 tried to fix KVMs
XSAVE valid feature scanning, but it was wrong. It was not considering
the sparse nature of this bitfield, instead reading values from
uninitialized members of the entries array.
This patch now separates subleaf indicies from KVM's array indicies
and fills the entry before querying it's value.
This fixes AVX support in KVM guests.
Signed-off-by: NAndre Przywara <andre.przywara@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

The documented behavior did not match the implemented one (which also
never changed).
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

KVM_MAX_MSIX_PER_DEV implies that up to that many MSI-X entries can be
requested. But the kernel so far rejected already the upper limit.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

KVM has an ioctl to define which signal mask should be used while running
inside VCPU_RUN. At least for big endian systems, this mask is different
on 32-bit and 64-bit systems (though the size is identical).

Add a compat wrapper that converts the mask to whatever the kernel accepts,
allowing 32-bit kvm user space to set signal masks.

This patch fixes qemu with --enable-io-thread on ppc64 hosts when running
32-bit user land.
Signed-off-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NAvi Kivity <avi@redhat.com>

Neither host_irq nor the guest_msi struct are used anymore today.
Tag the former, drop the latter to avoid confusion.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch adds instruction fetch checking when walking guest page table,
to implement SMEP when emulating instead of executing natively.
Signed-off-by: NYang, Wei <wei.y.yang@intel.com>
Signed-off-by: NShan, Haitao <haitao.shan@intel.com>
Signed-off-by: NLi, Xin <xin.li@intel.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch masks CPUID leaf 7 ebx against host capability word9.
Signed-off-by: NYang, Wei <wei.y.yang@intel.com>
Signed-off-by: NShan, Haitao <haitao.shan@intel.com>
Signed-off-by: NLi, Xin <xin.li@intel.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch adds SMEP handling when setting CR4.
Signed-off-by: NYang, Wei <wei.y.yang@intel.com>
Signed-off-by: NShan, Haitao <haitao.shan@intel.com>
Signed-off-by: NLi, Xin <xin.li@intel.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch removes SMEP bit from CR4_RESERVED_BITS.
Signed-off-by: NYang, Wei <wei.y.yang@intel.com>
Signed-off-by: NShan, Haitao <haitao.shan@intel.com>
Signed-off-by: NLi, Xin <xin.li@intel.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

The nested VMX feature is supposed to fully emulate VMX for the guest. This
(theoretically) not only allows it to run its own guests, but also also
to further emulate VMX for its own guests, and allow arbitrarily deep nesting.

This patch fixes a bug (discovered by Kevin Tian) in handling a VMLAUNCH
by L2, which prevented deeper nesting.

Deeper nesting now works (I only actually tested L3), but is currently
*absurdly* slow, to the point of being unusable.
Signed-off-by: NNadav Har'El <nyh@il.ibm.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

This saves a lot of pointless casts x86_emulate_ctxt and decode_cache.
Signed-off-by: NAvi Kivity <avi@redhat.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

The name eip conflicts with a field of the same name in x86_emulate_ctxt,
which we plan to fold decode_cache into.

The name _eip is unfortunate, but what's really needed is a refactoring
here, not a better name.
Signed-off-by: NAvi Kivity <avi@redhat.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

a is unused now on CONFIG_X86_32.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>