Mark McLoughlin [Mon, 1 Dec 2008 13:57:46 +0000 (13:57 +0000)]
KVM: make kvm_unregister_irq_ack_notifier() safe
We never pass a NULL notifier pointer here, but we may well
pass a notifier struct which hasn't previously been
registered.
Guard against this by using hlist_del_init() which will
not do anything if the node hasn't been added to the list
and, when removing the node, will ensure that a subsequent
call to hlist_del_init() will be fine too.
Fixes an oops seen when an assigned device is freed before
and IRQ is assigned to it.
Signed-off-by: Mark McLoughlin <markmc@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Mark McLoughlin [Mon, 1 Dec 2008 13:57:45 +0000 (13:57 +0000)]
KVM: remove the IRQ ACK notifier assertions
We will obviously never pass a NULL struct kvm_irq_ack_notifier* to
this functions. They are always embedded in the assigned device
structure, so the assertion add nothing.
The irqchip_in_kernel() assertion is very out of place - clearly
this little abstraction needs to know nothing about the upper
layer details.
Signed-off-by: Mark McLoughlin <markmc@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
The s390 backend of kvm never calls kvm_vcpu_uninit. This causes
a memory leak of vcpu->run pages.
Lets call kvm_vcpu_uninit in kvm_arch_vcpu_destroy to free
the vcpu->run.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Acked-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
KVM: s390: Fix refcounting and allow module unload
Currently it is impossible to unload the kvm module on s390.
This patch fixes kvm_arch_destroy_vm to release all cpus.
This make it possible to unload the module.
In addition we stop messing with the module refcount in arch code.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Acked-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Marcelo Tosatti [Tue, 25 Nov 2008 14:58:07 +0000 (15:58 +0100)]
KVM: MMU: optimize set_spte for page sync
The write protect verification in set_spte is unnecessary for page sync.
Its guaranteed that, if the unsync spte was writable, the target page
does not have a write protected shadow (if it had, the spte would have
been write protected under mmu_lock by rmap_write_protect before).
Same reasoning applies to mark_page_dirty: the gfn has been marked as
dirty via the pagefault path.
The cost of hash table and memslot lookups are quite significant if the
workload is pagetable write intensive resulting in increased mmu_lock
contention.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Sheng Yang [Mon, 24 Nov 2008 06:32:57 +0000 (14:32 +0800)]
KVM: MSI to INTx translate
Now we use MSI as default one, and translate MSI to INTx when guest need
INTx rather than MSI. For legacy device, we provide support for non-sharing
host IRQ.
Provide a parameter msi2intx for this method. The value is true by default in
x86 architecture.
We can't guarantee this mode can work on every device, but for most of us
tested, it works. If your device encounter some trouble with this mode, you can
try set msi2intx modules parameter to 0. If the device is OK with msi2intx=0,
then please report it to KVM mailing list or me. We may prepare a blacklist for
the device that can't work in this mode.
Signed-off-by: Sheng Yang <sheng@linux.intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Sheng Yang [Mon, 24 Nov 2008 06:32:56 +0000 (14:32 +0800)]
KVM: Enable MSI for device assignment
We enable guest MSI and host MSI support in this patch. The userspace want to
enable MSI should set KVM_DEV_IRQ_ASSIGN_ENABLE_MSI in the assigned_irq's flag.
Function would return -ENOTTY if can't enable MSI, userspace shouldn't set MSI
Enable bit when KVM_ASSIGN_IRQ return -ENOTTY with
KVM_DEV_IRQ_ASSIGN_ENABLE_MSI.
Userspace can tell the support of MSI device from #ifdef KVM_CAP_DEVICE_MSI.
Signed-off-by: Sheng Yang <sheng@linux.intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Glauber Costa [Mon, 24 Nov 2008 17:45:23 +0000 (15:45 -0200)]
x86: KVM guest: sign kvmclock as paravirt
Currently, we only set the KVM paravirt signature in case
of CONFIG_KVM_GUEST. However, it is possible to have it turned
off, while CONFIG_KVM_CLOCK is turned on. This is also a paravirt
case, and should be shown accordingly.
Signed-off-by: Glauber Costa <glommer@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Avi Kivity [Sun, 23 Nov 2008 16:08:57 +0000 (18:08 +0200)]
KVM: VMX: Conditionally request interrupt window after injecting irq
If we're injecting an interrupt, and another one is pending, request
an interrupt window notification so we don't have excess latency on the
second interrupt.
This shouldn't happen in practice since an EOI will be issued, giving a second
chance to request an interrupt window, but...
Xiantao Zhang [Fri, 21 Nov 2008 09:16:07 +0000 (17:16 +0800)]
KVM: ia64: Add handler for crashed vmm
Since vmm runs in an isolated address space and it is just a copy
of host's kvm-intel module, so once vmm crashes, we just crash all guests
running on it instead of crashing whole kernel.
Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Xiantao Zhang [Fri, 21 Nov 2008 12:58:11 +0000 (20:58 +0800)]
KVM: ia64: Define printk function for kvm-intel module
kvm-intel module is relocated to an isolated address space
with kernel, so it can't call host kernel's printk for debug
purpose. In the module, we implement the printk to output debug
info of vmm.
Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Eduardo Habkost [Mon, 17 Nov 2008 21:03:24 +0000 (19:03 -0200)]
x86: disable VMX on all CPUs on reboot
On emergency_restart, we may need to use an NMI to disable virtualization
on all CPUs. We do that using nmi_shootdown_cpus() if VMX is enabled.
Note: With this patch, we will run the NMI stuff only when the CPU where
emergency_restart() was called has VMX enabled. This should work on most
cases because KVM enables VMX on all CPUs, but we may miss the small
window where KVM is doing that. Also, I don't know if all code using
VMX out there always enable VMX on all CPUs like KVM does. We have two
other alternatives for that:
a) Have an API that all code that enables VMX on any CPU should use
to tell the kernel core that it is going to enable VMX on the CPUs.
b) Always call nmi_shootdown_cpus() if the CPU supports VMX. This is
a bit intrusive and more risky, as it would run nmi_shootdown_cpus()
on emergency_reboot() even on systems where virtualization is never
enabled.
Finding a proper point to hook the nmi_shootdown_cpus() call isn't
trivial, as the non-emergency machine_restart() (that doesn't need the
NMI tricks) uses machine_emergency_restart() directly.
The solution to make this work without adding a new function or argument
to machine_ops was setting a 'reboot_emergency' flag that tells if
native_machine_emergency_restart() needs to do the virt cleanup or not.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Eduardo Habkost [Mon, 17 Nov 2008 21:03:23 +0000 (19:03 -0200)]
kdump: forcibly disable VMX and SVM on machine_crash_shutdown()
We need to disable virtualization extensions on all CPUs before booting
the kdump kernel, otherwise the kdump kernel booting will fail, and
rebooting after the kdump kernel did its task may also fail.
We do it using cpu_emergency_vmxoff() and cpu_emergency_svm_disable(),
that should always work, because those functions check if the CPUs
support SVM or VMX before doing their tasks.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Eduardo Habkost [Mon, 17 Nov 2008 21:03:20 +0000 (19:03 -0200)]
KVM: SVM: move has_svm() code to asm/virtext.h
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>
Hollis Blanchard [Mon, 10 Nov 2008 20:57:36 +0000 (14:57 -0600)]
KVM: ppc: fix userspace mapping invalidation on context switch
We used to defer invalidating userspace TLB entries until jumping out of the
kernel. This was causing MMU weirdness most easily triggered by using a pipe in
the guest, e.g. "dmesg | tail". I believe the problem was that after the guest
kernel changed the PID (part of context switch), the old process's mappings
were still present, and so copy_to_user() on the "return to new process" path
ended up using stale mappings.
Testing with large pages (64K) exposed the problem, probably because with 4K
pages, pressure on the TLB faulted all process A's mappings out before the
guest kernel could insert any for process B.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Hollis Blanchard [Mon, 10 Nov 2008 20:57:35 +0000 (14:57 -0600)]
KVM: ppc: use prefetchable mappings for guest memory
Bare metal Linux on 440 can "overmap" RAM in the kernel linear map, so that it
can use large (256MB) mappings even if memory isn't a multiple of 256MB. To
prevent the hardware prefetcher from loading from an invalid physical address
through that mapping, it's marked Guarded.
However, KVM must ensure that all guest mappings are backed by real physical
RAM (since a deliberate access through a guarded mapping could still cause a
machine check). Accordingly, we don't need to make our mappings guarded, so
let's allow prefetching as the designers intended.
Curiously this patch didn't affect performance at all on the quick test I
tried, but it's clearly the right thing to do anyways and may improve other
workloads.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Sheng Yang [Tue, 11 Nov 2008 07:30:40 +0000 (15:30 +0800)]
KVM: Fix kernel allocated memory slot
Commit 7fd49de9773fdcb7b75e823b21c1c5dc1e218c14 "KVM: ensure that memslot
userspace addresses are page-aligned" broke kernel space allocated memory
slot, for the userspace_addr is invalid.
Signed-off-by: Sheng Yang <sheng@linux.intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
KVM: ensure that memslot userspace addresses are page-aligned
Bad page translation and silent guest failure ensue if the userspace address is
not page-aligned. I hit this problem using large (host) pages with qemu,
because qemu currently has a hardcoded 4096-byte alignment for guest memory
allocations.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Nitin A Kamble [Wed, 5 Nov 2008 23:56:21 +0000 (15:56 -0800)]
KVM: Fix cpuid iteration on multiple leaves per eac
The code to traverse the cpuid data array list for counting type of leaves is
currently broken.
This patches fixes the 2 things in it.
1. Set the 1st counting entry's flag KVM_CPUID_FLAG_STATE_READ_NEXT. Without
it the code will never find a valid entry.
2. Also the stop condition in the for loop while looking for the next unflaged
entry is broken. It needs to stop when it find one matching entry;
and in the case of count of 1, it will be the same entry found in this
iteration.
Signed-Off-By: Nitin A Kamble <nitin.a.kamble@intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Nitin A Kamble [Wed, 5 Nov 2008 23:37:36 +0000 (15:37 -0800)]
KVM: Fix cpuid leaf 0xb loop termination
For cpuid leaf 0xb the bits 8-15 in ECX register define the end of counting
leaf. The previous code was using bits 0-7 for this purpose, which is
a bug.
Signed-off-by: Nitin A Kamble <nitin.a.kamble@intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
In kvmppc_deliver_interrupt is just one case left in the switch and it is a
rare one (less than 8%) when looking at the exit numbers. Therefore we can
at least drop the switch/case and if an if. I inserted an unlikely too, but
that's open for discussion.
In kvmppc_can_deliver_interrupt all frequent cases are in the default case.
I know compilers are smart but we can make it easier for them. By writing
down all options and removing the default case combined with the fact that
ithe values are constants 0..15 should allow the compiler to write an easy
jump table.
Modifying kvmppc_can_deliver_interrupt pointed me to the fact that gcc seems
to be unable to reduce priority_exception[x] to a build time constant.
Therefore I changed the usage of the translation arrays in the interrupt
delivery path completely. It is now using priority without translation to irq
on the full irq delivery path.
To be able to do that ivpr regs are stored by their priority now.
Additionally the decision made in kvmppc_can_deliver_interrupt is already
sufficient to get the value of interrupt_msr_mask[x]. Therefore we can replace
the 16x4byte array used here with a single 4byte variable (might still be one
miss, but the chance to find this in cache should be better than the right
entry of the whole array).
Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com> Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Currently we use an unnecessary if&switch to detect some cases.
To be honest we don't need the ligh_exits counter anyway, because we can
calculate it out of others. Sum_exits can also be calculated, so we can
remove that too.
MMIO, DCR and INTR can be counted on other places without these
additional control structures (The INTR case was never hit anyway).
The handling of BOOKE_INTERRUPT_EXTERNAL/BOOKE_INTERRUPT_DECREMENTER is
similar, but we can avoid the additional if when copying 3 lines of code.
I thought about a goto there to prevent duplicate lines, but rewriting three
lines should be better style than a goto cross switch/case statements (its
also not enough code to justify a new inline function).
Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com> Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
When changing some msr bits e.g. problem state we need to take special
care of that. We call the function in our mtmsr emulation (not needed for
wrtee[i]), but we don't call kvmppc_set_msr if we change msr via set_regs
ioctl.
It's a corner case we never hit so far, but I assume it should be
kvmppc_set_msr in our arch set regs function (I found it because it is also
a corner case when using pv support which would miss the update otherwise).
Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com> Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
KVM: ppc: create struct kvm_vcpu_44x and introduce container_of() accessor
This patch doesn't yet move all 44x-specific data into the new structure, but
is the first step down that path. In the future we may also want to create a
struct kvm_vcpu_booke.
Based on patch from Liu Yu <yu.liu@freescale.com>.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
KVM: ppc: refactor instruction emulation into generic and core-specific pieces
Cores provide 3 emulation hooks, implemented for example in the new
4xx_emulate.c:
kvmppc_core_emulate_op
kvmppc_core_emulate_mtspr
kvmppc_core_emulate_mfspr
Strictly speaking the last two aren't necessary, but provide for more
informative error reporting ("unknown SPR").
Long term I'd like to have instruction decoding autogenerated from tables of
opcodes, and that way we could aggregate universal, Book E, and core-specific
instructions more easily and without redundant switch statements.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
ppc: Create disassemble.h to extract instruction fields
This is used in a couple places in KVM, but isn't KVM-specific.
However, this patch doesn't modify other in-kernel emulation code:
- xmon uses a direct copy of ppc_opc.c from binutils
- emulate_instruction() doesn't need it because it can use a series
of mask tests.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Acked-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Avi Kivity <avi@redhat.com>
KVM: ppc: Refactor powerpc.c to relocate 440-specific code
This introduces a set of core-provided hooks. For 440, some of these are
implemented by booke.c, with the rest in (the new) 44x.c.
Note that these hooks are link-time, not run-time. Since it is not possible to
build a single kernel for both e500 and 440 (for example), using function
pointers would only add overhead.
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Izik Eidus [Fri, 3 Oct 2008 14:40:32 +0000 (17:40 +0300)]
KVM: MMU: Fix aliased gfns treated as unaliased
Some areas of kvm x86 mmu are using gfn offset inside a slot without
unaliasing the gfn first. This patch makes sure that the gfn will be
unaliased and add gfn_to_memslot_unaliased() to save the calculating
of the gfn unaliasing in case we have it unaliased already.
Sheng Yang [Fri, 31 Oct 2008 04:37:41 +0000 (12:37 +0800)]
KVM: Enable Function Level Reset for assigned device
Ideally, every assigned device should in a clear condition before and after
assignment, so that the former state of device won't affect later work.
Some devices provide a mechanism named Function Level Reset, which is
defined in PCI/PCI-e document. We should execute it before and after device
assignment.
(But sadly, the feature is new, and most device on the market now don't
support it. We are considering using D0/D3hot transmit to emulate it later,
but not that elegant and reliable as FLR itself.)
[Update: Reminded by Xiantao, execute FLR after we ensure that the device can
be assigned to the guest.]
Signed-off-by: Sheng Yang <sheng@linux.intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Xiantao Zhang [Thu, 23 Oct 2008 07:03:38 +0000 (15:03 +0800)]
KVM: ia64: Remove lock held by halted vcpu
Remove the lock protection for kvm halt logic, otherwise,
once other vcpus want to acquire the lock, and they have to
wait all vcpus are waken up from halt.
Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Xiantao Zhang [Thu, 23 Oct 2008 06:56:44 +0000 (14:56 +0800)]
KVM: ia64: Re-organize data sturure of guests' data area
1. Increase the size of data area to 64M
2. Support more vcpus and memory, 128 vcpus and 256G memory are supported
for guests.
3. Add the boundary check for memory and vcpu allocation.
With this patch, kvm guest's data area looks as follow:
*
* +----------------------+ ------- KVM_VM_DATA_SIZE
* | vcpu[n]'s data | | ___________________KVM_STK_OFFSET
* | | | / |
* | .......... | | /vcpu's struct&stack |
* | .......... | | /---------------------|---- 0
* | vcpu[5]'s data | | / vpd |
* | vcpu[4]'s data | |/-----------------------|
* | vcpu[3]'s data | / vtlb |
* | vcpu[2]'s data | /|------------------------|
* | vcpu[1]'s data |/ | vhpt |
* | vcpu[0]'s data |____________________________|
* +----------------------+ |
* | memory dirty log | |
* +----------------------+ |
* | vm's data struct | |
* +----------------------+ |
* | | |
* | | |
* | | |
* | | |
* | | |
* | | |
* | | |
* | vm's p2m table | |
* | | |
* | | |
* | | | |
* vm's data->| | | |
* +----------------------+ ------- 0
* To support large memory, needs to increase the size of p2m.
* To support more vcpus, needs to ensure it has enough space to
* hold vcpus' data.
*/
Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
KVM: VMX: Handle mmio emulation when guest state is invalid
If emulate_invalid_guest_state is enabled, the emulator is called
when guest state is invalid. Until now, we reported an mmio failure
when emulate_instruction() returned EMULATE_DO_MMIO. This patch adds
the case where emulate_instruction() failed and an MMIO emulation
is needed.
Signed-off-by: Guillaume Thouvenin <guillaume.thouvenin@ext.bull.net> Signed-off-by: Avi Kivity <avi@redhat.com>
KVM: allow emulator to adjust rip for emulated pio instructions
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>
Amit Shah [Mon, 27 Oct 2008 09:04:17 +0000 (09:04 +0000)]
KVM: SVM: Set the 'g' bit of the cs selector for cross-vendor migration
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>
Jan Kiszka [Mon, 20 Oct 2008 08:20:03 +0000 (10:20 +0200)]
KVM: x86: Optimize NMI watchdog delivery
As suggested by Avi, this patch introduces a counter of VCPUs that have
LVT0 set to NMI mode. Only if the counter > 0, we push the PIT ticks via
all LAPIC LVT0 lines to enable NMI watchdog support.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Acked-by: Sheng Yang <sheng@linux.intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Jan Kiszka [Mon, 20 Oct 2008 08:20:02 +0000 (10:20 +0200)]
KVM: x86: Fix and refactor NMI watchdog emulation
This patch refactors the NMI watchdog delivery patch, consolidating
tests and providing a proper API for delivering watchdog events.
An included micro-optimization is to check only for apic_hw_enabled in
kvm_apic_local_deliver (the test for LVT mask is covering the
soft-disabled case already).
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Acked-by: Sheng Yang <sheng@linux.intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Sheng Yang [Thu, 16 Oct 2008 09:30:58 +0000 (17:30 +0800)]
KVM: VMX: Move private memory slot position
PCI device assignment would map guest MMIO spaces as separate slot, so it is
possible that the device has more than 2 MMIO spaces and overwrite current
private memslot.
The patch move private memory slot to the top of userspace visible memory slots.
Signed-off-by: Sheng Yang <sheng@linux.intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Sheng Yang [Thu, 9 Oct 2008 08:01:55 +0000 (16:01 +0800)]
KVM: VMX: Add PAT support for EPT
GUEST_PAT support is a new feature introduced by Intel Core i7 architecture.
With this, cpu would save/load guest and host PAT automatically, for EPT memory
type in guest depends on MSR_IA32_CR_PAT.
Also add save/restore for MSR_IA32_CR_PAT.
Signed-off-by: Sheng Yang <sheng@linux.intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Jan Kiszka [Fri, 26 Sep 2008 07:30:57 +0000 (09:30 +0200)]
KVM: VMX: work around lacking VNMI support
Older VMX supporting CPUs do not provide the "Virtual NMI" feature for
tracking the NMI-blocked state after injecting such events. For now
KVM is unable to inject NMIs on those CPUs.
Derived from Sheng Yang's suggestion to use the IRQ window notification
for detecting the end of NMI handlers, this patch implements virtual
NMI support without impact on the host's ability to receive real NMIs.
The downside is that the given approach requires some heuristics that
can cause NMI nesting in vary rare corner cases.
The approach works as follows:
- inject NMI and set a software-based NMI-blocked flag
- arm the IRQ window start notification whenever an NMI window is
requested
- if the guest exits due to an opening IRQ window, clear the emulated
NMI-blocked flag
- if the guest net execution time with NMI-blocked but without an IRQ
window exceeds 1 second, force NMI-blocked reset and inject anyway
This approach covers most practical scenarios:
- succeeding NMIs are seperated by at least one open IRQ window
- the guest may spin with IRQs disabled (e.g. due to a bug), but
leaving the NMI handler takes much less time than one second
- the guest does not rely on strict ordering or timing of NMIs
(would be problematic in virtualized environments anyway)
Successfully tested with the 'nmi n' monitor command, the kgdbts
testsuite on smp guests (additional patches required to add debug
register support to kvm) + the kernel's nmi_watchdog=1, and a Siemens-
specific board emulation (+ guest) that comes with its own NMI
watchdog mechanism.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Jan Kiszka [Fri, 26 Sep 2008 07:30:56 +0000 (09:30 +0200)]
KVM: VMX: Provide support for user space injected NMIs
This patch adds the required bits to the VMX side for user space
injected NMIs. As with the preexisting in-kernel irqchip support, the
CPU must provide the "virtual NMI" feature for proper tracking of the
NMI blocking state.
Based on the original patch by Sheng Yang.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Sheng Yang <sheng.yang@intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Jan Kiszka [Fri, 26 Sep 2008 07:30:52 +0000 (09:30 +0200)]
KVM: x86: Enable NMI Watchdog via in-kernel PIT source
LINT0 of the LAPIC can be used to route PIT events as NMI watchdog ticks
into the guest. This patch aligns the in-kernel irqchip emulation with
the user space irqchip with already supports this feature. The trick is
to route PIT interrupts to all LAPIC's LVT0 lines.
Rebased and slightly polished patch originally posted by Sheng Yang.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Sheng Yang <sheng.yang@intel.com> Signed-off-by: Avi Kivity <avi@redhat.com>
projects / linux-beck.git / log