Commit Graph

80690 Commits

Author SHA1 Message Date
Amit Shah
9327fd1195 KVM: Make unloading of FPU state when putting vcpu arch-independent
Instead of having each architecture do it individually, we
do this in the arch-independent code (just x86 as of now).

[avi: add svm to the mix, which was added to mainline during the
 2.6.24-rc process]

Signed-off-by: Amit Shah <amit.shah@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:05 +02:00
Avi Kivity
4cee576493 KVM: MMU: Add some mmu statistics
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:04 +02:00
Avi Kivity
ba1389b7a0 KVM: Extend stats support for VM stats
This is in addition to the current virtual cpu statistics.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:04 +02:00
Avi Kivity
f2b5756bb3 KVM: Add instruction emulation statistics 2008-01-30 17:53:04 +02:00
Avi Kivity
f096ed8588 KVM: Add fpu_reload counter
Measure the number of times we switch the fpu state.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:04 +02:00
Avi Kivity
e1beb1d37c KVM: Replace 'light_exits' stat with 'host_state_reload'
This is a little more accurate (since it counts actual reloads, not potential
reloads), and reverses the sense of the statistic to measure a bad event like
most of the other stats (e.g. we want to minimize all counters).

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:04 +02:00
Zhang Xiantao
d19a9cd275 KVM: Portability: Add two hooks to handle kvm_create and destroy vm
Add two arch hooks to handle kvm_create_vm and kvm destroy_vm. Now, just
put io_bus init and destory in common.

Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:04 +02:00
Zhang Xiantao
a16b043cc9 KVM: Remove __init attributes for kvm_init_debug and kvm_init_msr_list
Since their callers are not declared with __init.

Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:04 +02:00
Joe Perches
56919c5c97 KVM: Remove ptr comparisons to 0
Fix sparse warnings "Using plain integer as NULL pointer"

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:03 +02:00
Zhang Xiantao
8b0067913d KVM: Portability: Make kvm_vcpu_ioctl_translate arch dependent
Move kvm_vcpu_ioctl_translate to arch, since mmu would be put under arch.

Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:03 +02:00
Avi Kivity
e08aa78ae5 KVM: VMX: Consolidate register usage in vmx_vcpu_run()
We pass vcpu, vmx->fail, and vmx->launched to assembly code, but all three
are fields within vmx.  Consolidate by only passing in vmx and offsets for
the rest.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:03 +02:00
Zhang Xiantao
018d00d2fe KVM: Portability: move KVM_CHECK_EXTENSION
Make KVM_CHECK_EXTENSION code into a function, all archs can define its
capability independently.

Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Acked-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:03 +02:00
Sheng Yang
a7e6c88a78 KVM: x86 emulator: modify 'lods', and 'stos' not to depend on CR2
The current 'lods' and 'stos' is depending on incoming CR2 rather than decode
memory address from registers.

Signed-off-by: Sheng Yang <sheng.yang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:03 +02:00
Zhang Xiantao
f8c16bbaa9 KVM: Portability: Move x86 specific code from kvm_init() to kvm_arch()
Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Acked-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:03 +02:00
Zhang Xiantao
cb498ea2ce KVM: Portability: Combine kvm_init and kvm_init_x86
Will be called once arch module registers itself.

Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Acked-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:02 +02:00
Zhang Xiantao
e9b11c1755 KVM: Portability: Add vcpu and hardware management arch hooks
Add the following hooks:

  void decache_vcpus_on_cpu(int cpu);
  int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu);
  void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu);
  void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu);
  void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu);
  void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu);
  struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id);
  void kvm_arch_vcpu_destory(struct kvm_vcpu *vcpu);
  int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu);
  void kvm_arch_hardware_enable(void *garbage);
  void kvm_arch_hardware_disable(void *garbage);
  int kvm_arch_hardware_setup(void);
  void kvm_arch_hardware_unsetup(void);
  void kvm_arch_check_processor_compat(void *rtn);

Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:02 +02:00
Zhang Xiantao
97896d04a1 KVM: Portability: Move kvm_x86_ops to x86.c
Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Acked-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:02 +02:00
Zhang Xiantao
d825ed0a97 KVM: Portability: Move some includes to x86.c
Move some includes to x86.c from kvm_main.c, since the related functions
have been moved to x86.c

Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Acked-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:02 +02:00
Izik Eidus
e0506bcba5 KVM: Change kvm_{read,write}_guest() to use copy_{from,to}_user()
This changes kvm_write_guest_page/kvm_read_guest_page to use
copy_to_user/read_from_user, as a result we get better speed
and better dirty bit tracking.

Signed-off-by: Izik Eidus <izike@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:02 +02:00
Izik Eidus
539cb6608c KVM: introduce gfn_to_hva()
Convert a guest frame number to the corresponding host virtual address.

Signed-off-by: Izik Eidus <izike@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:01 +02:00
Izik Eidus
f9d46eb0e4 KVM: add kvm_is_error_hva()
Check for the "error hva", an address outside the user address space that
signals a bad gfn.

Signed-off-by: Izik Eidus <izike@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:01 +02:00
Avi Kivity
1a6f4d7fbd KVM: Simplify CPU_TASKS_FROZEN cpu notifier handling
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:01 +02:00
Izik Eidus
906e608b05 KVM: x86 emulator: remove 8 bytes operands emulator for call near instruction
it is removed beacuse it isnt supported on a real host

Signed-off-by: Izik Eidus <izike@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:01 +02:00
Eddie Dong
e5edaa01c4 KVM: VMX: wbinvd exiting
Add wbinvd VM Exit support to prepare for pass-through
device cache emulation and also enhance real time
responsiveness.

Signed-off-by: Yaozu (Eddie) Dong <eddie.dong@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:01 +02:00
Eddie Dong
8a70cc3d0f KVM: VMX: Comment VMX primary/secondary exec ctl definitions
Add comments for secondary/primary Processor-Based VM-execution controls.

Signed-off-by: Yaozu (Eddie) Dong <eddie.dong@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:01 +02:00
Avi Kivity
9c8cba3761 KVM: Fix faults during injection of real-mode interrupts
If vmx fails to inject a real-mode interrupt while fetching the interrupt
redirection table, it fails to record this in the vectoring information
field.  So we detect this condition and do it ourselves.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:01 +02:00
Avi Kivity
1155f76a81 KVM: VMX: Read & store IDT_VECTORING_INFO_FIELD
We'll want to write to it in order to fix real-mode irq injection problems,
but it is a read-only field.  Storing it in a variable solves that issue.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:00 +02:00
Avi Kivity
9c5623e3e4 KVM: VMX: Use vmx to inject real-mode interrupts
Instead of injecting real-mode interrupts by writing the interrupt frame into
guest memory, abuse vmx by injecting a software interrupt.  We need to
pretend the software interrupt instruction had a length > 0, so we have to
adjust rip backward.

This lets us not to mess with writing guest memory, which is complex and also
sleeps.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:00 +02:00
Dor Laor
12264760e4 KVM: Add make_page_dirty() to kvm_clear_guest_page()
Every write access to guest pages should be tracked.

Signed-off-by: Dor Laor <dor.laor@qumranet.com>
Signed-off-by: Izik Eidus <izike@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:00 +02:00
Hollis Blanchard
b6c7a5dccf KVM: Portability: Move x86 vcpu ioctl handlers to x86.c
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:00 +02:00
Hollis Blanchard
d075206073 KVM: Portability: Move x86 FPU handling to x86.c
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:00 +02:00
Hollis Blanchard
8776e5194f KVM: Portability: Move x86 instruction emulation code to x86.c
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:53:00 +02:00
Hollis Blanchard
417bc3041f KVM: Portability: Make exported debugfs data architecture-specific
Signed-off-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:59 +02:00
Avi Kivity
1c73ef6650 KVM: x86 emulator: Hoist modrm and abs decoding into separate functions
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:59 +02:00
Uri Lublin
3b6fff198c KVM: Make mark_page_dirty() work for aliased pages too.
Recommended by Izik Eidus.

Signed-off-by: Uri Lublin <uril@qumranet.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:59 +02:00
Avi Kivity
9f1ef3f8f5 KVM: Simplify decode_register_operand() calling convention
Now that rex_prefix is part of the decode cache, there is no need to pass
it along.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:59 +02:00
Avi Kivity
33615aa956 KVM: x86 emulator: centralize decoding of one-byte register access insns
Instructions like 'inc reg' that have the register operand encoded
in the opcode are currently specially decoded.  Extend
decode_register_operand() to handle that case, indicated by having
DstReg or SrcReg without ModRM.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:59 +02:00
Avi Kivity
3c118e24af KVM: x86 emulator: Extract the common code of SrcReg and DstReg
Share the common parts of SrcReg and DstReg decoding.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:59 +02:00
Carsten Otte
de7d789acd KVM: Portability: Move pio emulation functions to x86.c
This patch moves implementation of the following functions from
kvm_main.c to x86.c:
free_pio_guest_pages, vcpu_find_pio_dev, pio_copy_data, complete_pio,
kernel_pio, pio_string_write, kvm_emulate_pio, kvm_emulate_pio_string

The function inject_gp, which was duplicated by yesterday's patch
series, is removed from kvm_main.c now because it is not needed anymore.

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Acked-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:59 +02:00
Carsten Otte
bbd9b64e37 KVM: Portability: Move x86 emulation and mmio device hook to x86.c
This patch moves the following functions to from kvm_main.c to x86.c:
emulator_read/write_std, vcpu_find_pervcpu_dev, vcpu_find_mmio_dev,
emulator_read/write_emulated, emulator_write_phys,
emulator_write_emulated_onepage, emulator_cmpxchg_emulated,
get_setment_base, emulate_invlpg, emulate_clts, emulator_get/set_dr,
kvm_report_emulation_failure, emulate_instruction

The following data type is moved to x86.c:
struct x86_emulate_ops emulate_ops

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Acked-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:58 +02:00
Carsten Otte
15c4a6406f KVM: Portability: Move kvm_get/set_msr[_common] to x86.c
This patch moves the implementation of the functions of kvm_get/set_msr,
kvm_get/set_msr_common, and set_efer from kvm_main.c to x86.c. The
definition of EFER_RESERVED_BITS is moved too.

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Acked-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:58 +02:00
Anthony Liguori
aab61cc0d2 KVM: Fix gfn_to_page() acquiring mmap_sem twice
KVM's nopage handler calls gfn_to_page() which acquires the mmap_sem when
calling out to get_user_pages().  nopage handlers are already invoked with the
mmap_sem held though.  Introduce a __gfn_to_page() for use by the nopage
handler which requires the lock to already be held.

This was noticed by tglx.

Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:58 +02:00
Sheng Yang
f78e0e2ee4 KVM: VMX: Enable memory mapped TPR shadow (FlexPriority)
This patch based on CR8/TPR patch, and enable the TPR shadow (FlexPriority)
for 32bit Windows.  Since TPR is accessed very frequently by 32bit
Windows, especially SMP guest, with FlexPriority enabled, we saw significant
performance gain.

Signed-off-by: Sheng Yang <sheng.yang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:58 +02:00
Carsten Otte
a03490ed29 KVM: Portability: Move control register helper functions to x86.c
This patch moves the definitions of CR0_RESERVED_BITS,
CR4_RESERVED_BITS, and CR8_RESERVED_BITS along with the following
functions from kvm_main.c to x86.c:
set_cr0(), set_cr3(), set_cr4(), set_cr8(), get_cr8(), lmsw(),
load_pdptrs()
The static function wrapper inject_gp is duplicated in kvm_main.c and
x86.c for now, the version in kvm_main.c should disappear once the last
user of it is gone too.
The function load_pdptrs is no longer static, and now defined in x86.h
for the time being, until the last user of it is gone from kvm_main.c.

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:58 +02:00
Carsten Otte
6866b83ed7 KVM: Portability: move get/set_apic_base to x86.c
This patch moves the implementation of get_apic_base and set_apic_base
from kvm_main.c to x86.c

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:58 +02:00
Carsten Otte
5fb76f9be1 KVM: Portability: Move memory segmentation to x86.c
This patch moves the definition of segment_descriptor_64 for AMD64 and
EM64T from kvm_main.c to segment_descriptor.h. It also adds a proper
#ifndef...#define...#endif around that header file.
The implementation of segment_base is moved from kvm_main.c to x86.c.

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:57 +02:00
Carsten Otte
1fe779f8ec KVM: Portability: Split kvm_vm_ioctl v3
This patch splits kvm_vm_ioctl into archtecture independent parts, and
x86 specific parts which go to kvm_arch_vcpu_ioctl in x86.c.
The patch is unchanged since last submission.

Common ioctls for all architectures are:
KVM_CREATE_VCPU, KVM_GET_DIRTY_LOG, KVM_SET_USER_MEMORY_REGION

x86 specific ioctls are:
KVM_SET_MEMORY_REGION,
KVM_GET/SET_NR_MMU_PAGES, KVM_SET_MEMORY_ALIAS, KVM_CREATE_IRQCHIP,
KVM_CREATE_IRQ_LINE, KVM_GET/SET_IRQCHIP
KVM_SET_TSS_ADDR

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:57 +02:00
Avi Kivity
b733bfb524 KVM: MMU: Topup the mmu memory preallocation caches before emulating an insn
Emulation may cause a shadow pte to be instantiated, which requires
memory resources.  Make sure the caches are filled to avoid an oops.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:57 +02:00
Avi Kivity
3067714cf5 KVM: Move page fault processing to common code
The code that dispatches the page fault and emulates if we failed to map
is duplicated across vmx and svm.  Merge it to simplify further bugfixing.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:57 +02:00
Avi Kivity
c7e75a3db4 KVM: x86 emulator: don't depend on cr2 for mov abs emulation
The 'mov abs' instruction family (opcodes 0xa0 - 0xa3) still depends on cr2
provided by the page fault handler.  This is wrong for several reasons:

- if an instruction accessed misaligned data that crosses a page boundary,
  and if the fault happened on the second page, cr2 will point at the
  second page, not the data itself.

- if we're emulating in real mode, or due to a FlexPriority exit, there
  is no cr2 generated.

So, this change adds decoding for this instruction form and drops reliance
on cr2.

Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-01-30 17:52:57 +02:00