Some userspace (e.g. QEMU 1.1) munge the d and g bits of segment
descriptors, causing us not to recognize them as unusable segments
with emulate_invalid_guest_state=1. Relax the check by testing for
segment not present (a non-present segment cannot be usable).
Signed-off-by: Avi Kivity <avi@redhat.com>
The operand size for these instructions is 8 bytes in long mode, even without
a REX prefix. Set it explicitly.
Triggered while booting Linux with emulate_invalid_guest_state=1.
Signed-off-by: Avi Kivity <avi@redhat.com>
Opcode 0F A2.
Used by Linux during the mode change trampoline while in a state that is
not virtualizable on vmx without unrestricted_guest, so we need to emulate
it is emulate_invalid_guest_state=1.
Signed-off-by: Avi Kivity <avi@redhat.com>
Instead of getting an exact leaf, follow the spec and fall back to the last
main leaf instead. This lets us easily emulate the cpuid instruction in the
emulator.
Signed-off-by: Avi Kivity <avi@redhat.com>
Introduce kvm_cpuid() to perform the leaf limit check and calculate
register values, and let kvm_emulate_cpuid() just handle reading and
writing the registers from/to the vcpu. This allows us to reuse
kvm_cpuid() in a context where directly reading and writing registers
is not desired.
Signed-off-by: Avi Kivity <avi@redhat.com>
In protected mode, the CPL is defined as the lower two bits of CS, as set by
the last far jump. But during the transition to protected mode, there is no
last far jump, so we need to return zero (the inherited real mode CPL).
Fix by reading CPL from the cache during the transition. This isn't 100%
correct since we don't set the CPL cache on a far jump, but since protected
mode transition will always jump to a segment with RPL=0, it will always
work.
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently the MMU's ->new_cr3() callback does nothing when guest paging
is disabled or when two-dimentional paging (e.g. EPT on Intel) is active.
This means that an emulated write to cr3 can be lost; kvm_set_cr3() will
write vcpu-arch.cr3, but the GUEST_CR3 field in the VMCS will retain its
old value and this is what the guest sees.
This bug did not have any effect until now because:
- with unrestricted guest, or with svm, we never emulate a mov cr3 instruction
- without unrestricted guest, and with paging enabled, we also never emulate a
mov cr3 instruction
- without unrestricted guest, but with paging disabled, the guest's cr3 is
ignored until the guest enables paging; at this point the value from arch.cr3
is loaded correctly my the mov cr0 instruction which turns on paging
However, the patchset that enables big real mode causes us to emulate mov cr3
instructions in protected mode sometimes (when guest state is not virtualizable
by vmx); this mov cr3 is effectively ignored and will crash the guest.
The fix is to make nonpaging_new_cr3() call mmu_free_roots() to force a cr3
reload. This is awkward because now all the new_cr3 callbacks to the same
thing, and because mmu_free_roots() is somewhat of an overkill; but fixing
that is more complicated and will be done after this minimal fix.
Observed in the Window XP 32-bit installer while bringing up secondary vcpus.
Signed-off-by: Avi Kivity <avi@redhat.com>
Implementation of PV EOI using shared memory.
This reduces the number of exits an interrupt
causes as much as by half.
The idea is simple: there's a bit, per APIC, in guest memory,
that tells the guest that it does not need EOI.
We set it before injecting an interrupt and clear
before injecting a nested one. Guest tests it using
a test and clear operation - this is necessary
so that host can detect interrupt nesting -
and if set, it can skip the EOI MSR.
There's a new MSR to set the address of said register
in guest memory. Otherwise not much changed:
- Guest EOI is not required
- Register is tested & ISR is automatically cleared on exit
For testing results see description of previous patch
'kvm_para: guest side for eoi avoidance'.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Each time we need to cancel injection we invoke same code
(cancel_injection callback). Move it towards the end of function using
the familiar goto on error pattern.
Will make it easier to do more cleanups for PV EOI.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Commit eb0dc6d0368072236dcd086d7fdc17fd3c4574d4 introduced apic
attention bitmask but kvm still syncs lapic unconditionally.
As that commit suggested and in anticipation of adding more attention
bits, only sync lapic if(apic_attention).
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
We perform ISR lookups twice: during interrupt
injection and on EOI. Typical workloads only have
a single bit set there. So we can avoid ISR scans by
1. counting bits as we set/clear them in ISR
2. on set, caching the injected vector number
3. on clear, invalidating the cache
The real purpose of this is enabling PV EOI
which needs to quickly validate the vector.
But non PV guests also benefit: with this patch,
and without interrupt nesting, apic_find_highest_isr
will always return immediately without scanning ISR.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
The following commit did not care about the error handling path:
commit c1a7b32a14
KVM: Avoid wasting pages for small lpage_info arrays
If memory allocation fails, vfree() will be called with the address
returned by kzalloc(). This patch fixes this issue.
Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: Avi Kivity <avi@redhat.com>
EPT Dirty bit use bit 9 as Intel SDM definition, to avoid conflict, change
PT_FIRST_AVAIL_BITS_SHIFT to 10.
Signed-off-by: Xudong Hao <xudong.hao@intel.com>
Signed-off-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Size is not needed to return one from pre-allocated objects.
Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
I see this in 3.5-rc1:
arch/x86/kvm/mmu.c: In function ‘kvm_test_age_rmapp’:
arch/x86/kvm/mmu.c:1271: warning: ‘iter.desc’ may be used uninitialized in this function
The line in question was introduced by commit
1e3f42f03c
static int kvm_test_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
unsigned long data)
{
- u64 *spte;
+ u64 *sptep;
+ struct rmap_iterator iter; <- line 1271
int young = 0;
/*
The reason I think is that the compiler assumes that
the rmap value could be 0, so
static u64 *rmap_get_first(unsigned long rmap, struct rmap_iterator
*iter)
{
if (!rmap)
return NULL;
if (!(rmap & 1)) {
iter->desc = NULL;
return (u64 *)rmap;
}
iter->desc = (struct pte_list_desc *)(rmap & ~1ul);
iter->pos = 0;
return iter->desc->sptes[iter->pos];
}
will not initialize iter.desc, but the compiler isn't
smart enough to see that
for (sptep = rmap_get_first(*rmapp, &iter); sptep;
sptep = rmap_get_next(&iter)) {
will immediately exit in this case.
I checked by adding
if (!*rmapp)
goto out;
on top which is clearly equivalent but disables the warning.
This patch uses uninitialized_var to disable the warning without
increasing code size.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Introduces a couple of print functions, which are essentially wrappers
around standard printk functions, with a KVM: prefix.
Functions introduced or modified are:
- kvm_err(fmt, ...)
- kvm_info(fmt, ...)
- kvm_debug(fmt, ...)
- kvm_pr_unimpl(fmt, ...)
- pr_unimpl(vcpu, fmt, ...) -> vcpu_unimpl(vcpu, fmt, ...)
Signed-off-by: Christoffer Dall <c.dall@virtualopensystems.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
For example migration between Westmere and Nehelem hosts, caught in big real mode.
The code that fixes the segments for real mode guest was moved from enter_rmode
to vmx_set_segments. enter_rmode calls vmx_set_segments for each segment.
Signed-off-by: Orit Wasserman <owasserm@rehdat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
mmu_shrink() needlessly iterates over all VMs even though it will not
attempt to free mmu pages from more than one on them. Fix that and also
check used mmu pages count outside of VM lock to skip inactive VMs faster.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Add kernel parameter to control A/D bits support, it's on by default.
Signed-off-by: Haitao Shan <haitao.shan@intel.com>
Signed-off-by: Xudong Hao <xudong.hao@intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
lpage_info is created for each large level even when the memory slot is
not for RAM. This means that when we add one slot for a PCI device, we
end up allocating at least KVM_NR_PAGE_SIZES - 1 pages by vmalloc().
To make things worse, there is an increasing number of devices which
would result in more pages being wasted this way.
This patch mitigates this problem by using kvm_kvzalloc().
Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: Avi Kivity <avi@redhat.com>
The huge page size is 4M on non-PAE host, but 2M page size is used in
transparent_hugepage_adjust(), so the page we get after adjust the
mapping level is not the head page, the BUG_ON() will be triggered
Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Pull KVM changes from Avi Kivity:
"Changes include additional instruction emulation, page-crossing MMIO,
faster dirty logging, preventing the watchdog from killing a stopped
guest, module autoload, a new MSI ABI, and some minor optimizations
and fixes. Outside x86 we have a small s390 and a very large ppc
update.
Regarding the new (for kvm) rebaseless workflow, some of the patches
that were merged before we switch trees had to be rebased, while
others are true pulls. In either case the signoffs should be correct
now."
Fix up trivial conflicts in Documentation/feature-removal-schedule.txt
arch/powerpc/kvm/book3s_segment.S and arch/x86/include/asm/kvm_para.h.
I suspect the kvm_para.h resolution ends up doing the "do I have cpuid"
check effectively twice (it was done differently in two different
commits), but better safe than sorry ;)
* 'next' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (125 commits)
KVM: make asm-generic/kvm_para.h have an ifdef __KERNEL__ block
KVM: s390: onereg for timer related registers
KVM: s390: epoch difference and TOD programmable field
KVM: s390: KVM_GET/SET_ONEREG for s390
KVM: s390: add capability indicating COW support
KVM: Fix mmu_reload() clash with nested vmx event injection
KVM: MMU: Don't use RCU for lockless shadow walking
KVM: VMX: Optimize %ds, %es reload
KVM: VMX: Fix %ds/%es clobber
KVM: x86 emulator: convert bsf/bsr instructions to emulate_2op_SrcV_nobyte()
KVM: VMX: unlike vmcs on fail path
KVM: PPC: Emulator: clean up SPR reads and writes
KVM: PPC: Emulator: clean up instruction parsing
kvm/powerpc: Add new ioctl to retreive server MMU infos
kvm/book3s: Make kernel emulated H_PUT_TCE available for "PR" KVM
KVM: PPC: bookehv: Fix r8/r13 storing in level exception handler
KVM: PPC: Book3S: Enable IRQs during exit handling
KVM: PPC: Fix PR KVM on POWER7 bare metal
KVM: PPC: Fix stbux emulation
KVM: PPC: bookehv: Use lwz/stw instead of PPC_LL/PPC_STL for 32-bit fields
...
Currently the inject_pending_event() call during guest entry happens after
kvm_mmu_reload(). This is for historical reasons - we used to
inject_pending_event() in atomic context, while kvm_mmu_reload() needs task
context.
A problem is that nested vmx can cause the mmu context to be reset, if event
injection is intercepted and causes a #VMEXIT instead (the #VMEXIT resets
CR0/CR3/CR4). If this happens, we end up with invalid root_hpa, and since
kvm_mmu_reload() has already run, no one will fix it and we end up entering
the guest this way.
Fix by reordering event injection to be before kvm_mmu_reload(). Use
->cancel_injection() to undo if kvm_mmu_reload() fails.
https://bugzilla.kernel.org/show_bug.cgi?id=42980
Reported-by: Luke-Jr <luke-jr+linuxbugs@utopios.org>
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Using RCU for lockless shadow walking can increase the amount of memory
in use by the system, since RCU grace periods are unpredictable. We also
have an unconditional write to a shared variable (reader_counter), which
isn't good for scaling.
Replace that with a scheme similar to x86's get_user_pages_fast(): disable
interrupts during lockless shadow walk to force the freer
(kvm_mmu_commit_zap_page()) to wait for the TLB flush IPI to find the
processor with interrupts enabled.
We also add a new vcpu->mode, READING_SHADOW_PAGE_TABLES, to prevent
kvm_flush_remote_tlbs() from avoiding the IPI.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
On x86_64, we can defer %ds and %es reload to the heavyweight context switch,
since nothing in the lightweight paths uses the host %ds or %es (they are
ignored by the processor). Furthermore we can avoid the load if the segments
are null, by letting the hardware load the null segments for us. This is the
expected case.
On i386, we could avoid the reload entirely, since the entry.S paths take care
of reload, except for the SYSEXIT path which leaves %ds and %es set to __USER_DS.
So we set them to the same values as well.
Saves about 70 cycles out of 1600 (around 4%; noisy measurements).
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
The vmx exit code unconditionally restores %ds and %es to __USER_DS. This
can override the user's values, since %ds and %es are not saved and restored
in x86_64 syscalls. In practice, this isn't dangerous since nobody uses
segment registers in long mode, least of all programs that use KVM.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
The instruction emulation for bsrw is broken in KVM because
the code always uses bsr with 32 or 64 bit operand size for
emulation. Fix that by using emulate_2op_SrcV_nobyte() macro
to use guest operand size for emulation.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Although ModRM byte is fetched for group decoding, it is soon pushed
back to make decode_modrm() fetch it later again.
Now that ModRM flag can be found in the top level opcode tables, fetch
ModRM byte before group decoding to make the code simpler.
Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: Avi Kivity <avi@redhat.com>
Needed for the following patch which simplifies ModRM fetching code.
Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: Avi Kivity <avi@redhat.com>
cpuid eax should return the max leaf so that
guests can find out the valid range.
This matches Xen et al.
Update documentation to match.
Tested with -cpu host.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
If vcpu executes hlt instruction while async PF is waiting to be delivered
vcpu can block and deliver async PF only after another even wakes it
up. This happens because kvm_check_async_pf_completion() will remove
completion event from vcpu->async_pf.done before entering kvm_vcpu_block()
and this will make kvm_arch_vcpu_runnable() return false. The solution
is to make vcpu runnable when processing completion.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
We can't run PIT IRQ injection work in the interrupt context of the host
timer. This would allow the user to influence the handler complexity by
asking for a broadcast to a large number of VCPUs. Therefore, this work
was pushed into workqueue context in 9d244caf2e. However, this prevents
prioritizing the PIT injection over other task as workqueues share
kernel threads.
This replaces the workqueue with a kthread worker and gives that thread
a name in the format "kvm-pit/<owner-process-pid>". That allows to
identify and adjust the kthread priority according to the VM process
parameters.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
The patch introduces a bitmap that will hold reasons apic should be
checked during vmexit. This is in a preparation for vp eoi patch
that will add one more check on vmexit. With the bitmap we can do
if(apic_attention) to check everything simultaneously which will
add zero overhead on the fast path.
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently, MSI messages can only be injected to in-kernel irqchips by
defining a corresponding IRQ route for each message. This is not only
unhandy if the MSI messages are generated "on the fly" by user space,
IRQ routes are a limited resource that user space has to manage
carefully.
By providing a direct injection path, we can both avoid using up limited
resources and simplify the necessary steps for user land.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
This continues the theme started with vm_brk() and vm_munmap():
vm_mmap() does the same thing as do_mmap(), but additionally does the
required VM locking.
This uninlines (and rewrites it to be clearer) do_mmap(), which sadly
duplicates it in mm/mmap.c and mm/nommu.c. But that way we don't have
to export our internal do_mmap_pgoff() function.
Some day we hopefully don't have to export do_mmap() either, if all
modular users can become the simpler vm_mmap() instead. We're actually
very close to that already, with the notable exception of the (broken)
use in i810, and a couple of stragglers in binfmt_elf.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Like the vm_brk() function, this is the same as "do_munmap()", except it
does the VM locking for the caller.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
MMIO that are split across a page boundary are currently broken - the
code does not expect to be aborted by the exit to userspace for the
first MMIO fragment.
This patch fixes the problem by generalizing the current code for handling
16-byte MMIOs to handle a number of "fragments", and changes the MMIO
code to create those fragments.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Merge reason: development work has dependency on kvm patches merged
upstream.
Conflicts:
Documentation/feature-removal-schedule.txt
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
kvm_set_shared_msr() may not be called in preemptible context,
but vmx_set_msr() does so:
BUG: using smp_processor_id() in preemptible [00000000] code: qemu-kvm/22713
caller is kvm_set_shared_msr+0x32/0xa0 [kvm]
Pid: 22713, comm: qemu-kvm Not tainted 3.4.0-rc3+ #39
Call Trace:
[<ffffffff8131fa82>] debug_smp_processor_id+0xe2/0x100
[<ffffffffa0328ae2>] kvm_set_shared_msr+0x32/0xa0 [kvm]
[<ffffffffa03a103b>] vmx_set_msr+0x28b/0x2d0 [kvm_intel]
...
Making kvm_set_shared_msr() work in preemptible is cleaner, but
it's used in the fast path. Making two variants is overkill, so
this patch just disables preemption around the call.
Reported-by: Dave Jones <davej@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Its much cleaner to use PT_PAGE_TABLE_LEVEL than its numeric value.
Signed-off-by: Davidlohr Bueso <dave@gnu.org>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Intel spec says that TMR needs to be set/cleared
when IRR is set, but kvm also clears it on EOI.
I did some tests on a real (AMD based) system,
and I see same TMR values both before
and after EOI, so I think it's a minor bug in kvm.
This patch fixes TMR to be set/cleared on IRR set
only as per spec.
And now that we don't clear TMR, we can save
an atomic read of TMR on EOI that's not propagated
to ioapic, by checking whether ioapic needs
a specific vector first and calculating
the mode afterwards.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
