New AMD processors (Family 0x10 models 8+) support the Pause
Filter Feature.  This feature creates a new field in the VMCB
called Pause Filter Count.  If Pause Filter Count is greater
than 0 and intercepting PAUSEs is enabled, the processor will
increment an internal counter when a PAUSE instruction occurs
instead of intercepting.  When the internal counter reaches the
Pause Filter Count value, a PAUSE intercept will occur.

This feature can be used to detect contended spinlocks,
especially when the lock holding VCPU is not scheduled.
Rescheduling another VCPU prevents the VCPU seeking the
lock from wasting its quantum by spinning idly.

Experimental results show that most spinlocks are held
for less than 1000 PAUSE cycles or more than a few
thousand.  Default the Pause Filter Counter to 3000 to
detect the contended spinlocks.

Processor support for this feature is indicated by a CPUID
bit.

On a 24 core system running 4 guests each with 16 VCPUs,
this patch improved overall performance of each guest's
32 job kernbench by approximately 3-5% when combined
with a scheduler algorithm thati caused the VCPU to
sleep for a brief period. Further performance improvement
may be possible with a more sophisticated yield algorithm.
Signed-off-by: NMark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

With all important informations now delivered through
tracepoints we can savely remove the nsvm_printk debugging
code for nested svm.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

This patch adds a tracepoint for the event that the guest
executed the SKINIT instruction. This information is
important because SKINIT is an SVM extenstion not yet
implemented by nested SVM and we may need this information
for debugging hypervisors that do not yet run on nested SVM.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

This patch adds a tracepoint for the event that the guest
executed the INVLPGA instruction.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

This patch adds a special tracepoint for the event that a
nested #vmexit is injected because kvm wants to inject an
interrupt into the guest.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

This patch adds a tracepoint for a nested #vmexit that gets
re-injected to the guest.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

This patch adds a tracepoint for every #vmexit we get from a
nested guest.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

This patch adds a dedicated kvm tracepoint for a nested
vmrun.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

The nested SVM code emulates a #vmexit caused by a request
to open the irq window right in the request function. This
is a bug because the request function runs with preemption
and interrupts disabled but the #vmexit emulation might
sleep. This can cause a schedule()-while-atomic bug and is
fixed with this patch.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

If event_inj is valid on a #vmexit the host CPU would write
the contents to exit_int_info, so the hypervisor knows that
the event wasn't injected.

We don't do this in nested SVM by now which is a bug and
fixed by this patch.
Signed-off-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

Much of so far vendor-specific code for setting up guest debug can
actually be handled by the generic code. This also fixes a minor deficit
in the SVM part /wrt processing KVM_GUESTDBG_ENABLE.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

Both VMX and SVM require per-cpu memory allocation, which is done at module
init time, for only online cpus.

Backend was not allocating enough structure for all possible CPUs, so
new CPUs coming online could not be hardware enabled.
Signed-off-by: NZachary Amsden <zamsden@redhat.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

Signed-off-by: NZachary Amsden <zamsden@redhat.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

This patch replaces them with native_read_tsc() which can
also be used in expressions and saves a variable on the
stack in this case.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

The exit_int_info field is only written by the hardware and
never read. So it does not need to be copied on a vmrun
emulation.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

This patch reorganizes the logic in svm_interrupt_allowed to
make it better to read. This is important because the logic
is a lot more complicated with Nested SVM.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

X86 CPUs need to have some magic happening to enable the virtualization
extensions on them. This magic can result in unpleasant results for
users, like blocking other VMMs from working (vmx) or using invalid TLB
entries (svm).

Currently KVM activates virtualization when the respective kernel module
is loaded. This blocks us from autoloading KVM modules without breaking
other VMMs.

To circumvent this problem at least a bit, this patch introduces on
demand activation of virtualization. This means, that instead
virtualization is enabled on creation of the first virtual machine
and disabled on destruction of the last one.

So using this, KVM can be easily autoloaded, while keeping other
hypervisors usable.
Signed-off-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

nested_svm_map unnecessarily takes mmap_sem around gfn_to_page, since
gfn_to_page / get_user_pages are responsible for it.
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>
Acked-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NAvi Kivity <avi@redhat.com>

They're just copies of vcpu->run, which is readily accessible.
Signed-off-by: NAvi Kivity <avi@redhat.com>

When running nested we need to touch the l1 guests
tsc_offset. Otherwise changes will be lost or a wrong value
be read.

Cc: stable@kernel.org
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

When svm_vcpu_load is called while the vcpu is running in
guest mode the tsc adjustment made there is lost on the next
emulated #vmexit. This causes the tsc running backwards in
the guest. This patch fixes the issue by also adjusting the
tsc_offset in the emulated hsave area so that it will not
get lost.

Cc: stable@kernel.org
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

It is no longer possible to reproduce the problem any more, so presumably
it has been fixed.
Signed-off-by: NAvi Kivity <avi@redhat.com>

Nested SVM is (in my experience) stable enough to be enabled by
default. So omit the requirement to pass a module parameter.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

Not checking for this flag breaks any nested hypervisor that does not
set VINTR. So fix it with this patch.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch removes one indentation level from nested_svm_intr and
makes the logic more readable.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This check is not necessary. We have to sync the vcpu->arch.cr2 always
back to the VMCB. This patch remove the is_nested check.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch moves the handling for special nested vmexits like #pf to a
separate function. This makes the kvm_override parameter obsolete and
makes the code more readable.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

If nested svm fails to load the msrpm the vmrun succeeds with the old
msrpm which is not correct. This patch changes the logic to roll back
to host mode in case the msrpm cannot be loaded.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This function is not longer required. So remove it.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch removes the usage of nested_svm_do from the vmrun emulation
path.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch removes the usage of nested_svm_do from the vmload and
vmsave emulation code paths.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch changes nested svm to call nested_svm_exit_handled_msr
directly and not through nested_svm_do.

[alex: fix oops due to nested kmap_atomics]
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This patch is the starting point of removing nested_svm_do from the
nested svm code. The nested_svm_do function basically maps two guest
physical pages to host virtual addresses and calls a passed function
on it. This function pointer code flow is hard to read and not the
best technical solution here.
As a side effect this patch indroduces the nested_svm_[un]map helper
functions.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

Makes the code of this function more readable by removing on
indentation level for the core logic.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

If this function returns true a nested vmexit is required. Move that
vmexit into the nested_svm_exit_handled function. This also simplifies
the handling of nested #pf intercepts in this function.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Acked-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NAvi Kivity <avi@redhat.com>

When caching guest intercepts there is no need anymore for the
nested_svm_exit_handled_real function. So move its code into
nested_svm_exit_handled.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Acked-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NAvi Kivity <avi@redhat.com>

When the nested intercepts are cached we don't need to call
get_user_pages and/or map the nested vmcb on every nested #vmexit to
check who will handle the intercept.
Further this patch aligns the emulated svm behavior better to real
hardware.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

This makes it more clear for which purpose these members in the vcpu_svm
exist.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Acked-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NAvi Kivity <avi@redhat.com>

The interrupt completion code must run after nested exits are handled
because not injected interrupts or exceptions may be handled by the l1
guest first.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Acked-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NAvi Kivity <avi@redhat.com>

The vmcb control area contains more then 800 bytes of reserved fields
which are unnecessarily copied. Fix this by introducing a copy
function which only copies the relevant part and saves time.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Acked-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NAvi Kivity <avi@redhat.com>