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: Mark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: Marcelo 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This patch adds a tracepoint for the event that the guest
executed the INVLPGA instruction.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This patch adds a tracepoint for a nested #vmexit that gets
re-injected to the guest.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This patch adds a tracepoint for every #vmexit we get from a
nested guest.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
This patch adds a dedicated kvm tracepoint for a nested
vmrun.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo 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: Alexander Graf <agraf@suse.de>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo 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: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi 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: Zachary Amsden <zamsden@redhat.com>
Signed-off-by: Marcelo 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo 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: Alexander Graf <agraf@suse.de>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi 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: Marcelo Tosatti <mtosatti@redhat.com>
Acked-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch updates percpu related symbols in x86 such that percpu
symbols are unique and don't clash with local symbols. This serves
two purposes of decreasing the possibility of global percpu symbol
collision and allowing dropping per_cpu__ prefix from percpu symbols.
* arch/x86/kernel/cpu/common.c: rename local variable to avoid collision
* arch/x86/kvm/svm.c: s/svm_data/sd/ for local variables to avoid collision
* arch/x86/kernel/cpu/cpu_debug.c: s/cpu_arr/cpud_arr/
s/priv_arr/cpud_priv_arr/
s/cpu_priv_count/cpud_priv_count/
* arch/x86/kernel/cpu/intel_cacheinfo.c: s/cpuid4_info/ici_cpuid4_info/
s/cache_kobject/ici_cache_kobject/
s/index_kobject/ici_index_kobject/
* arch/x86/kernel/ds.c: s/cpu_context/cpu_ds_context/
Partly based on Rusty Russell's "alloc_percpu: rename percpu vars
which cause name clashes" patch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: (kvm) Avi Kivity <avi@redhat.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: x86@kernel.org
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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch removes one indentation level from nested_svm_intr and
makes the logic more readable.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch removes the usage of nested_svm_do from the vmrun emulation
path.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch removes the usage of nested_svm_do from the vmload and
vmsave emulation code paths.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Makes the code of this function more readable by removing on
indentation level for the core logic.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi 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: Joerg Roedel <joerg.roedel@amd.com>
Acked-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi 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: Joerg Roedel <joerg.roedel@amd.com>
Acked-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi 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: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This makes it more clear for which purpose these members in the vcpu_svm
exist.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Acked-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi 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: Joerg Roedel <joerg.roedel@amd.com>
Acked-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi 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: Joerg Roedel <joerg.roedel@amd.com>
Acked-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Only copy the necessary parts of the vmcb save area on vmrun and save
precious time.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Acked-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
It is more efficient to copy only the relevant parts of the vmcb back to
the nested vmcb when we emulate an vmexit.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Acked-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch makes the code easier to read when it comes to setting,
clearing and checking the status of the virtualized global
interrupt flag for the VCPU.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
If userspace knows that the kernel part supports 1GB pages it can enable
the corresponding cpuid bit so that guests actually use GB pages.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Use kvm_get_gdt() and kvm_read_ldt() to reduce inline assembly code.
Cc: Avi Kivity <avi@redhat.com>
Cc: kvm@vger.kernel.org
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Linux tries to disable the flush filter on all AMD K8 CPUs. Since KVM
does not handle the needed MSR, the injected #GP will panic the Linux
kernel. Ignore setting of the HWCR.FFDIS bit in this MSR to let Linux
boot with an AMD K8 family guest CPU.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This allows use of the powerful ftrace infrastructure.
See Documentation/trace/ for usage information.
[avi, stephen: various build fixes]
[sheng: fix control register breakage]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
While trying to get Hyper-V running, I realized that the interrupt injection
mechanisms that are in place right now are not 100% correct.
This patch makes nested SVM's interrupt injection behave more like on a
real machine.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
SVM adds another way to do INVLPG by ASID which Hyper-V makes use of,
so let's implement it!
For now we just do the same thing invlpg does, as asid switching
means we flush the mmu anyways. That might change one day though.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Hyper-V uses some MSRs, some of which are actually reserved for BIOS usage.
But let's be nice today and have it its way, because otherwise it fails
terribly.
[jaswinder: fix build for linux-next changes]
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The host never reads cr2 in process context, so are free to clobber it. The
vmx code does this, so we can safely remove the save/restore code.
Signed-off-by: Avi Kivity <avi@redhat.com>
The performance counter MSRs are different for AMD and Intel CPUs and they
are chosen mainly by the CPUID vendor string. This patch catches writes to
all addresses (regardless of VMX/SVM path) and handles them in the generic
MSR handler routine. Writing a 0 into the event select register is something
we perfectly emulate ;-), so don't print out a warning to dmesg in this
case.
This fixes booting a 64bit Windows guest with an AMD CPUID on an Intel host.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Instead of reloading the pdptrs on every entry and exit (vmcs writes on vmx,
guest memory access on svm) extract them on demand.
Signed-off-by: Avi Kivity <avi@redhat.com>
Since AMD does not support sysenter in 64bit mode, the VMCB fields storing
the MSRs are truncated to 32bit upon VMRUN/#VMEXIT. So store the values
in a separate 64bit storage to avoid truncation.
[andre: fix amd->amd migration]
Signed-off-by: Christoph Egger <christoph.egger@amd.com>
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Use standard msr-index.h's MSR declaration.
MSR_IA32_TSC is better than MSR_IA32_TIME_STAMP_COUNTER as it also solves
80 column issue.
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
If a migrated vcpu matches the asid_generation value of the target pcpu,
there will be no TLB flush via TLB_CONTROL_FLUSH_ALL_ASID.
The check for vcpu.cpu in pre_svm_run is meaningless since svm_vcpu_load
already updated it on schedule in.
Such vcpu will VMRUN with stale TLB entries.
Based on original patch from Joerg Roedel (http://patchwork.kernel.org/patch/10021/)
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Acked-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
If NMI is received during handling of another NMI it should be injected
immediately after IRET from previous NMI handler, but SVM intercept IRET
before instruction execution so we can't inject pending NMI at this
point and there is not way to request exit when NMI window opens. This
patch fix SVM code to open NMI window after IRET by single stepping over
IRET instruction.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Re-inject event instead. This is what Intel suggest. Also use correct
instruction length when re-injecting soft fault/interrupt.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch replaces drop_interrupt_shadow with the more
general set_interrupt_shadow, that can either drop or raise
it, depending on its parameter. It also adds ->get_interrupt_shadow()
for future use.
Signed-off-by: Glauber Costa <glommer@redhat.com>
CC: H. Peter Anvin <hpa@zytor.com>
CC: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
If a task switch caused by an event remove it from the event queue.
VMX already does that.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
On AMD CPUs sometimes the DB bit in the stack segment
descriptor is left as 1, although the whole segment has
been made unusable. Clear it here to pass an Intel VMX
entry check when cross vendor migrating.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Shadow_mt_mask is out of date, now it have only been used as a flag to indicate
if TDP enabled. Get rid of it and use tdp_enabled instead.
Also put memory type logical in kvm_x86_ops->get_mt_mask().
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
It just returns pending IRQ vector from the queue for VMX/SVM.
Get IRQ directly from the queue before migration and put it back
after.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Saves many exits to userspace in a case of IRQ chip in userspace.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Start to use interrupt/exception queues like VMX does.
This also fix the bug that if exit was caused by a guest
internal exception access to IDT the exception was not
reinjected.
Use EVENTINJ to inject interrupts. Use VINT only for detecting when IRQ
windows is open again. EVENTINJ ensures
the interrupt is injected immediately and not delayed.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Use the same callback to inject irq/nmi events no matter what irqchip is
in use. Only from VMX for now.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
At the vector level, kernel and userspace irqchip are fairly similar.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
If a task switch was initiated because off a task gate in IDT and IDT
was accessed because of an external even the instruction should not
be skipped.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
There is no need to skip instruction if the reason for a task switch
is a task gate in IDT and access to it is caused by an external even.
The problem is currently solved only for VMX since there is no reliable
way to skip an instruction in SVM. We should emulate it instead.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
kvm_vcpu_block() unhalts vpu on an interrupt/timer without checking
if interrupt window is actually opened.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The prioritized bit vector manipulation functions are useful in both vmx and
svm.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
KVM optimizes guest port 80 accesses by passthing them through to the host.
Some AMD machines die on port 80 writes, allowing the guest to hard-lock the
host.
Remove the port passthrough to avoid the problem.
Cc: stable@kernel.org
Reported-by: Piotr Jaroszyński <p.jaroszynski@gmail.com>
Tested-by: Piotr Jaroszyński <p.jaroszynski@gmail.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
AMDs VMCB does not have an explicit unusable segment descriptor field,
so we emulate it by using "not present". This has to be setup before
the fixups, because this field is used there.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
In the segment descriptor _cache_ the accessed bit is always set
(although it can be cleared in the descriptor itself). Since Intel
checks for this condition on a VMENTRY, set this bit in the AMD path
to enable cross vendor migration.
Cc: stable@kernel.org
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
Acked-By: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Looks like neither the direction nor the rep prefix are used anymore.
Drop related evaluations from SVM's and VMX's I/O exit handlers.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
AMD K10 CPUs implement the FFXSR feature that gets enabled using
EFER. Let's check if the virtual CPU description includes that
CPUID feature bit and allow enabling it then.
This is required for Windows Server 2008 in Hyper-V mode.
v2 adds CPUID capability exposure
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
VMware ESX checks if the microcode level is correct when using a barcelona
CPU, in order to see if it actually can use SVM. Let's tell it we're on the
safe side...
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Add the remaining bits to make use of debug registers also for guest
debugging, thus enabling the use of hardware breakpoints and
watchpoints.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
So far KVM only had basic x86 debug register support, once introduced to
realize guest debugging that way. The guest itself was not able to use
those registers.
This patch now adds (almost) full support for guest self-debugging via
hardware registers. It refactors the code, moving generic parts out of
SVM (VMX was already cleaned up by the KVM_SET_GUEST_DEBUG patches), and
it ensures that the registers are properly switched between host and
guest.
This patch also prepares debug register usage by the host. The latter
will (once wired-up by the following patch) allow for hardware
breakpoints/watchpoints in guest code. If this is enabled, the guest
will only see faked debug registers without functionality, but with
content reflecting the guest's modifications.
Tested on Intel only, but SVM /should/ work as well, but who knows...
Known limitations: Trapping on tss switch won't work - most probably on
Intel.
Credits also go to Joerg Roedel - I used his once posted debugging
series as platform for this patch.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This rips out the support for KVM_DEBUG_GUEST and introduces a new IOCTL
instead: KVM_SET_GUEST_DEBUG. The IOCTL payload consists of a generic
part, controlling the "main switch" and the single-step feature. The
arch specific part adds an x86 interface for intercepting both types of
debug exceptions separately and re-injecting them when the host was not
interested. Moveover, the foundation for guest debugging via debug
registers is layed.
To signal breakpoint events properly back to userland, an arch-specific
data block is now returned along KVM_EXIT_DEBUG. For x86, the arch block
contains the PC, the debug exception, and relevant debug registers to
tell debug events properly apart.
The availability of this new interface is signaled by
KVM_CAP_SET_GUEST_DEBUG. Empty stubs for not yet supported archs are
provided.
Note that both SVM and VTX are supported, but only the latter was tested
yet. Based on the experience with all those VTX corner case, I would be
fairly surprised if SVM will work out of the box.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Normally setting the SVME bit in EFER is not allowed, as we did
not support SVM. Not since we do, we should also allow enabling
SVM mode.
v2 comes as last patch, so we don't enable half-ready code
v4 introduces a module option to enable SVM
v6 warns that nesting is enabled
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
KVM tries to read the VM_CR MSR to find out if SVM was disabled by
the BIOS. So implement read support for this MSR to make nested
SVM running.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
This adds the #VMEXIT intercept, so we return to the level 1 guest
when something happens in the level 2 guest that should return to
the level 1 guest.
v2 implements HIF handling and cleans up exception interception
v3 adds support for V_INTR_MASKING_MASK
v4 uses the host page hsave
v5 removes IOPM merging code
v6 moves mmu code out of the atomic section
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch implements VMRUN. VMRUN enters a virtual CPU and runs that
in the same context as the normal guest CPU would run.
So basically it is implemented the same way, a normal CPU would do it.
We also prepare all intercepts that get OR'ed with the original
intercepts, as we do not allow a level 2 guest to be intercepted less
than the first level guest.
v2 implements the following improvements:
- fixes the CPL check
- does not allocate iopm when not used
- remembers the host's IF in the HIF bit in the hflags
v3:
- make use of the new permission checking
- add support for V_INTR_MASKING_MASK
v4:
- use host page backed hsave
v5:
- remove IOPM merging code
v6:
- save cr4 so PAE l1 guests work
v7:
- return 0 on vmrun so we check the MSRs too
- fix MSR check to use the correct variable
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
This implements the VMLOAD and VMSAVE instructions, that usually surround
the VMRUN instructions. Both instructions load / restore the same elements,
so we only need to implement them once.
v2 fixes CPL checking and replaces memcpy by assignments
v3 makes use of the new permission checking
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Implement the hsave MSR, that gives the VCPU a GPA to save the
old guest state in.
v2 allows userspace to save/restore hsave
v4 dummys out the hsave MSR, so we use a host page
v6 remembers the guest's hsave and exports the MSR
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch implements the GIF flag and the clgi and stgi instructions that
set this flag. Only if the flag is set (default), interrupts can be received by
the CPU.
To keep the information about that somewhere, this patch adds a new hidden
flags vector. that is used to store information that does not go into the
vmcb, but is SVM specific.
I tried to write some code to make -no-kvm-irqchip work too, but the first
level guest won't even boot with that atm, so I ditched it.
v2 moves the hflags to x86 generic code
v3 makes use of the new permission helper
v6 only enables interrupt_window if GIF=1
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
These are helpers for the nested SVM implementation.
- nsvm_printk implements a debug printk variant
- nested_svm_do calls a handler that can accesses gpa-based memory
v3 makes use of the new permission checker
v6 changes:
- streamline nsvm_debug()
- remove printk(KERN_ERR)
- SVME check before CPL check
- give GP error code
- use new EFER constant
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
MSR_EFER_SVME_MASK, MSR_VM_CR and MSR_VM_HSAVE_PA are set in KVM
specific headers. Linux does have nice header files to collect
EFER bits and MSR IDs, so IMHO we should put them there.
While at it, I also changed the naming scheme to match that
of the other defines.
(introduced in v6)
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
The current VINTR intercept setters don't look clean to me. To make
the code easier to read and enable the possibilty to trap on a VINTR
set, this uses a helper function to set the VINTR intercept.
v2 uses two distinct functions for setting and clearing the bit
Acked-by: Joerg Roedel <joro@8bytes.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Simplify LAPIC TMCCT calculation by using hrtimer provided
function to query remaining time until expiration.
Fixes host hang with nested ESX.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Use a trick to keep the printk()s on has_svm() working as before. gcc
will take care of not generating code for the 'msg' stuff when the
function is called with a NULL msg argument.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
If we call the emulator we shouldn't call skip_emulated_instruction()
in the first place, since the emulator already computes the next rip
for us. Thus we move ->skip_emulated_instruction() out of
kvm_emulate_pio() and into handle_io() (and the svm equivalent). We
also replaced "return 0" by "break" in the "do_io:" case because now
the shadow register state needs to be committed. Otherwise eip will never
be updated.
Signed-off-by: Guillaume Thouvenin <guillaume.thouvenin@ext.bull.net>
Signed-off-by: Avi Kivity <avi@redhat.com>
The busy flag of the TR selector is not set by the hardware. This breaks
migration from amd hosts to intel hosts.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The hardware does not set the 'g' bit of the cs selector and this breaks
migration from amd hosts to intel hosts. Set this bit if the segment
limit is beyond 1 MB.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The effective memory type of EPT is the mixture of MSR_IA32_CR_PAT and memory
type field of EPT entry.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
With pages out of sync invlpg needs to be trapped. For now simply nuke
the entry.
Untested on AMD.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
As suggested by Avi, introduce accessors to read/write guest registers.
This simplifies the ->cache_regs/->decache_regs interface, and improves
register caching which is important for VMX, where the cost of
vmcs_read/vmcs_write is significant.
[avi: fix warnings]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Accesses to CR4 are intercepted even with Nested Paging enabled. But the code
does not check if the guest wants to do a global TLB flush. So this flush gets
lost. This patch adds the check and the flush to svm_set_cr4.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch introduces a guest TLB flush on every NPF exit in KVM. This fixes
random segfaults and #UD exceptions in the guest seen under some workloads
(e.g. long running compile workloads or tbench). A kernbench run with and
without that fix showed that it has a slowdown lower than 0.5%
Cc: stable@kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@qumranet.com>
When an event (such as an interrupt) is injected, and the stack is
shadowed (and therefore write protected), the guest will exit. The
current code will see that the stack is shadowed and emulate a few
instructions, each time postponing the injection. Eventually the
injection may succeed, but at that time the guest may be unwilling
to accept the interrupt (for example, the TPR may have changed).
This occurs every once in a while during a Windows 2008 boot.
Fix by unshadowing the fault address if the fault was due to an event
injection.
Signed-off-by: Avi Kivity <avi@qumranet.com>
If NPT is enabled after loading both KVM modules on AMD and it should be
disabled, both KVM modules must be reloaded. If only the architecture module is
reloaded the behavior is undefined. With this patch it is possible to disable
NPT only by reloading the kvm_amd module.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
On suspend the svm_hardware_disable function is called which frees all svm_data
variables. On resume they are not re-allocated. This patch removes the
deallocation of svm_data from the hardware_disable function to the
hardware_unsetup function which is not called on suspend.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
KVM turns off hardware virtualization extensions during reboot, in order
to disassociate the memory used by the virtualization extensions from the
processor, and in order to have the system in a consistent state.
Unfortunately virtual machines may still be running while this goes on,
and once virtualization extensions are turned off, any virtulization
instruction will #UD on execution.
Fix by adding an exception handler to virtualization instructions; if we get
an exception during reboot, we simply spin waiting for the reset to complete.
If it's a true exception, BUG() so we can have our stack trace.
Signed-off-by: Avi Kivity <avi@qumranet.com>
Attached is a patch that fixes a guest crash when booting older Linux kernels.
The problem stems from the fact that we are currently emulating
MSR_K7_EVNTSEL[0-3], but not emulating MSR_K7_PERFCTR[0-3]. Because of this,
setup_k7_watchdog() in the Linux kernel receives a GPF when it attempts to
write into MSR_K7_PERFCTR, which causes an OOPs.
The patch fixes it by just "fake" emulating the appropriate MSRs, throwing
away the data in the process. This causes the NMI watchdog to not actually
work, but it's not such a big deal in a virtualized environment.
When we get a write to one of these counters, we printk_ratelimit() a warning.
I decided to print it out for all writes, even if the data is 0; it doesn't
seem to make sense to me to special case when data == 0.
Tested by myself on a RHEL-4 guest, and Joerg Roedel on a Windows XP 64-bit
guest.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
To distinguish between real page faults and nested page faults they should be
traced as different events. This is implemented by this patch.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch adds the missing kvmtrace markers to the svm
module of kvm.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
With an exit handler for INTR intercepts its possible to account them using
kvmtrace.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
With an exit handler for NMI intercepts its possible to account them using
kvmtrace.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Migrate the PIT timer to the physical CPU which vcpu0 is scheduled on,
similarly to what is done for the LAPIC timers, otherwise PIT interrupts
will be delayed until an unrelated event causes an exit.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
The function get_tdp_level() provided the number of tdp level for EPT and
NPT rather than the NPT specific macro.
Signed-off-by: Sheng Yang <sheng.yang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
There is not selective cr0 intercept bug. The code in the comment sets the
CR0.PG bit. But KVM sets the CR4.PG bit for SVM always to implement the paged
real mode. So the 'mov %eax,%cr0' instruction does not change the CR0.PG bit.
Selective CR0 intercepts only occur when a bit is actually changed. So its the
right behavior that there is no intercept on this instruction.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
With the usage of the V_TPR field this comment is now obsolete.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch disables the intercept of CR8 writes if the TPR is not masking
interrupts. This reduces the total number CR8 intercepts to below 1 percent of
what we have without this patch using Windows 64 bit guests.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
If the CR8 write intercept is disabled the V_TPR field of the VMCB needs to be
synced with the TPR field in the local apic.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch adds syncing of the lapic.tpr field to the V_TPR field of the VMCB.
With this change we can safely remove the CR8 read intercept.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Shutdown interception clears the vmcb, leaving the asid at zero (which is
illegal. so force a new asid on vmcb initialization.
Signed-off-by: Avi Kivity <avi@qumranet.com>
When KVM uses NPT there is no reason to intercept task switches. This patch
removes the intercept for it in that case.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
To properly forward a MCE occured while the guest is running to the host, we
have to intercept this exception and call the host handler by hand. This is
implemented by this patch.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch aligns the host version of the CR4.MCE bit with the CR4 active in
the guest. This is necessary to get MCE exceptions when the guest is running.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
The svm_set_cr4 function is indented with spaces. This patch replaces
them with tabs.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This emulates the x86 hardware task switch mechanism in software, as it is
unsupported by either vmx or svm. It allows operating systems which use it,
like freedos, to run as kvm guests.
Signed-off-by: Izik Eidus <izike@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Fixes sparse warning as well.
arch/x86/kvm/svm.c:69:15: warning: symbol 'iopm_base' was not declared. Should it be static?
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch implements the Last Branch Record Virtualization (LBRV) feature of
the AMD Barcelona and Phenom processors into the kvm-amd module. It will only
be enabled if the guest enables last branch recording in the DEBUG_CTL MSR. So
there is no increased world switch overhead when the guest doesn't use these
MSRs.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Markus Rechberger <markus.rechberger@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch changes the kvm-amd module to allocate the SVM MSR permission map
per VCPU instead of a global map for all VCPUs. With this we have more
flexibility allowing specific guests to access virtualized MSRs. This is
required for LBR virtualization.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Markus Rechberger <markus.rechberger@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Change the parameter of the init_vmcb() function in the kvm-amd module from
struct vmcb to struct vcpu_svm.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Markus Rechberger <markus.rechberger@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch contains the SVM architecture dependent changes for KVM to enable
support for the Nested Paging feature of AMD Barcelona and Phenom processors.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
The generic x86 code has to know if the specific implementation uses Nested
Paging. In the generic code Nested Paging is called Two Dimensional Paging
(TDP) to avoid confusion with (future) TDP implementations of other vendors.
This patch exports the availability of TDP to the generic x86 code.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
To disable the use of the Nested Paging feature even if it is available in
hardware this patch adds a module parameter. Nested Paging can be disabled by
passing npt=0 to the kvm_amd module.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Let SVM detect if the Nested Paging feature is available on the hardware.
Disable it to keep this patch series bisectable.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
By moving the SVM feature detection from the each_cpu code to the hardware
setup code it runs only once. As an additional advance the feature check is now
available earlier in the module setup process.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This patch aligns the bits the guest can set in the EFER register with the
features in the host processor. Currently it lets EFER.NX disabled if the
processor does not support it and enables EFER.LME and EFER.LMA only for KVM on
64 bit hosts.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
While installing Windows XP 64 bit wants to access the DEBUGCTL and the last
branch record (LBR) MSRs. Don't allowing this in KVM causes the installation to
crash. This patch allow the access to these MSRs and fixes the issue.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Markus Rechberger <markus.rechberger@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
Explicitly enable the NM intercept in svm_set_cr0 if we enable TS in the guest
copy of CR0 for lazy FPU switching. This fixes guest SMP with Linux under SVM.
Without that patch Linux deadlocks or panics right after trying to boot the
other CPUs.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Markus Rechberger <markus.rechberger@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
If the guest writes to cr0 and leaves the TS flag at 0 while vcpu->fpu_active
is also 0, the TS flag in the guest's cr0 gets lost. This leads to corrupt FPU
state an causes Windows Vista 64bit to crash very soon after boot. This patch
fixes this bug.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Markus Rechberger <markus.rechberger@amd.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
When executing a test program called "crashme", we found the KVM guest cannot
survive more than ten seconds, then encounterd kernel panic. The basic concept
of "crashme" is generating random assembly code and trying to execute it.
After some fixes on emulator insn validity judgment, we found it's hard to
get the current emulator handle the invalid instructions correctly, for the
#UD trap for hypercall patching caused troubles. The problem is, if the opcode
itself was OK, but combination of opcode and modrm_reg was invalid, and one
operand of the opcode was memory (SrcMem or DstMem), the emulator will fetch
the memory operand first rather than checking the validity, and may encounter
an error there. For example, ".byte 0xfe, 0x34, 0xcd" has this problem.
In the patch, we simply check that if the invalid opcode wasn't vmcall/vmmcall,
then return from emulate_instruction() and inject a #UD to guest. With the
patch, the guest had been running for more than 12 hours.
Signed-off-by: Feng (Eric) Liu <eric.e.liu@intel.com>
Signed-off-by: Sheng Yang <sheng.yang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
This paves the way for multiple architecture support. Note that while
ioapic.c could potentially be shared with ia64, it is also moved.
Signed-off-by: Avi Kivity <avi@qumranet.com>
