Removing direct calls to KVM.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Avoid using ctxt->vcpu; we can do everything with ->get_cr() and ->set_cr().

A side effect is that we no longer activate the fpu on emulated CLTS; but that
should be very rare.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Avoid use of ctxt->vcpu.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Requires ctxt->vcpu, which is to be abolished.  Replace with open calls
to get_msr().
Signed-off-by: NAvi Kivity <avi@redhat.com>

Replacing direct calls to realmode_lgdt(), realmode_lidt().
Signed-off-by: NAvi Kivity <avi@redhat.com>

Unneeded for register access.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Making the emulator caller agnostic.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Making the emulator caller agnostic.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Making the emulator caller agnostic.

[Takuya Yoshikawa: fix typo leading to LDT failures]
Signed-off-by: NTakuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: NAvi Kivity <avi@redhat.com>

Making the emulator caller agnostic.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Making the emulator caller agnostic.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Clean up lines longer than 80 columns.  No code changes.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Since segments need to be handled slightly differently when fetching
instructions, we add a __linearize helper that accepts a new 'fetch' boolean.

[avi: fix oops caused by wrong segmented_address initialization order]
Signed-off-by: NNelson Elhage <nelhage@ksplice.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

The last_guest_tsc is used in vcpu_load to adjust the
tsc_offset since tsc-scaling is merged. So the
last_guest_tsc needs to be updated in vcpu_put instead of
the the last_host_tsc. This is fixed with this patch.
Reported-by: NJan Kiszka <jan.kiszka@web.de>
Tested-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch fixes a bug in the nested-svm path when
decode-assists is available on the machine. After a
selective-cr0 intercept is detected the rip is advanced
unconditionally. This causes the l1-guest to continue
running with an l2-rip.
This bug was with the sel_cr0 unit-test on decode-assists
capable hardware.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

Currently, setting a large (i.e. negative) base address for %cs does not work on
a 64-bit host. The "JOS" teaching operating system, used by MIT and other
universities, relies on such segments while bootstrapping its way to full
virtual memory management.
Signed-off-by: NNelson Elhage <nelhage@ksplice.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

Just remove useless function define kvm_inject_pit_timer_irqs() from
file arch/x86/kvm/i8254.h

Signed-off-by:Duan Jiong<djduanjiong@gmail.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

Just remove useless function define kvm_pic_clear_isr_ack() and
pit_has_pending_timer()

Signed-off-by: Duan Jiong<djduanjiong@gmail.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch avoids gcc issuing the following warning when KVM_MAX_VCPUS=1:
warning: array subscript is above array bounds

kvm_for_each_vcpu currently checks to see if the index for the vcpu is
valid /after/ loading it. We don't run into problems because the address
is still inside the enclosing struct kvm and we never deference or write
to it, so this isn't a security issue.

The warning occurs when KVM_MAX_VCPUS=1 because the increment portion of
the loop will *always* cause the loop to load an invalid location since
++idx will always be > 0.

This patch moves the load so that the check occurs before the load and
we don't run into the compiler warning.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NJeff Mahoney <jeffm@suse.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

When doing a soft int, we need to bump eip before pushing it to
the stack.  Otherwise we'll do the int a second time.

[apw@canonical.com: merged eip update as per Jan's recommendation.]
Signed-off-by: NSerge E. Hallyn <serge.hallyn@ubuntu.com>
Signed-off-by: NAndy Whitcroft <apw@canonical.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

em_push() is a simple wrapper of emulate_push().  So this patch replaces
emulate_push() with em_push() and removes the unnecessary former.

In addition, the unused ops arguments are removed from emulate_pusha()
and emulate_grp45().
Signed-off-by: NTakuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: NAvi Kivity <avi@redhat.com>

PUSH emulation stores the value by calling writeback() after setting
the dst operand appropriately in emulate_push().

This writeback() using dst is not needed at all because we know the
target is the stack.  So this patch makes emulate_push() call, newly
introduced, segmented_write() directly.

By this, many inlined writeback()'s are removed.
Signed-off-by: NTakuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This stops "CMP r/m, reg" to write back the data into memory.
Pointed out by Avi.

The writeback suppression now covers CMP, CMPS, SCAS.
Signed-off-by: NTakuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: NAvi Kivity <avi@redhat.com>

In case certain allocations fail, vmx_create_vcpu may return 0 as error
instead of a negative value encoded via ERR_PTR. This causes a NULL
pointer dereferencing later on in kvm_vm_ioctl_vcpu_create.
Reported-by: NSasha Levin <levinsasha928@gmail.com>
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

Currently we sync registers back and forth before/after exiting
to userspace for IO, but during IO device model shouldn't need to
read/write the registers, so we can as well skip those sync points. The
only exaception is broken vmware backdor interface. The new code sync
registers content during IO only if registers are read from/written to
by userspace in the middle of the IO operation and this almost never
happens in practise.
Signed-off-by: NGleb Natapov <gleb@redhat.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

Signed-off-by: NAvi Kivity <avi@redhat.com>

For reuse later.
Signed-off-by: NAvi Kivity <avi@redhat.com>

So it can call emulate_gp() without forward declarations.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Needed for segment read/write checks.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Preparing to add segment checks.
Signed-off-by: NAvi Kivity <avi@redhat.com>

It's going to get more complicated soon.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Will help later adding proper segment checks.
Signed-off-by: NAvi Kivity <avi@redhat.com>

When the emulation of vmload or vmsave fails because the
guest passed an unsupported physical address it gets an #GP
with rip pointing to the instruction after vmsave/vmload.
This is a bug and fixed by this patch.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch implements two new vm-ioctls to get and set the
virtual_tsc_khz if the machine supports tsc-scaling. Setting
the tsc-frequency is only possible before userspace creates
any vcpu.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

With TSC scaling in SVM the tsc-offset needs to be
calculated differently. This patch propagates this
calculation into the architecture specific modules so that
this complexity can be handled there.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch implements a call-back into the architecture code
to allow the propagation of changes to the virtual tsc_khz
of the vcpu.
On SVM it updates the tsc_ratio variable, on VMX it does
nothing.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

The calculation of the tsc_delta value to ensure a
forward-going tsc for the guest is a function of the
host-tsc. This works as long as the guests tsc_khz is equal
to the hosts tsc_khz. With tsc-scaling hardware support this
is not longer true and the tsc_delta needs to be calculated
using guest_tsc values.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch changes the kvm_guest_time_update function to use
TSC frequency the guest actually has for updating its clock.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch enhances the kvm_amd module with functions to
support the TSC_RATE_MSR which can be used to set a given
tsc frequency for the guest vcpu.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

VMMCALL requires EFER.SVME to be enabled in the host, not in the guest, which
is what check_svme() checks.
Signed-off-by: NAvi Kivity <avi@redhat.com>