* [PATCH 01/33] KVM: SVM: support writing 0 to K8 performance counter control registers
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 02/33] KVM: MMU: emulated cmpxchg8b should be atomic on i386 Avi Kivity
` (31 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Joerg Roedel
From: Joerg Roedel <joerg.roedel@amd.com>
This lets SVM ignore writes of the value 0 to the performance counter control
registers. Thus enabling them will still fail in the guest, but a write of 0
which keeps them disabled is accepted. This is required to boot Windows
Vista 64bit.
[avi: avoid fall-thru in switch statement]
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>
---
drivers/kvm/svm.c | 13 +++++++++++++
1 files changed, 13 insertions(+), 0 deletions(-)
diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c
index 9f8564a..839a75f 100644
--- a/drivers/kvm/svm.c
+++ b/drivers/kvm/svm.c
@@ -1155,7 +1155,20 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 data)
case MSR_IA32_SYSENTER_ESP:
svm->vmcb->save.sysenter_esp = data;
break;
+ case MSR_K7_EVNTSEL0:
+ case MSR_K7_EVNTSEL1:
+ case MSR_K7_EVNTSEL2:
+ case MSR_K7_EVNTSEL3:
+ /*
+ * only support writing 0 to the performance counters for now
+ * to make Windows happy. Should be replaced by a real
+ * performance counter emulation later.
+ */
+ if (data != 0)
+ goto unhandled;
+ break;
default:
+ unhandled:
return kvm_set_msr_common(vcpu, ecx, data);
}
return 0;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 02/33] KVM: MMU: emulated cmpxchg8b should be atomic on i386
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
2008-01-06 14:39 ` [PATCH 01/33] KVM: SVM: support writing 0 to K8 performance counter control registers Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 18:57 ` [kvm-devel] " Carlo Marcelo Arenas Belon
2008-01-06 14:39 ` [PATCH 03/33] KVM: Portability: Move kvm{pic,ioapic} accesors to x86 specific code Avi Kivity
` (30 subsequent siblings)
32 siblings, 1 reply; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Marcelo Tosatti
From: Marcelo Tosatti <marcelo@kvack.org>
Emulate cmpxchg8b atomically on i386. This is required to avoid a guest
pte walker from seeing a splitted write.
[avi: make it compile]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
drivers/kvm/x86.c | 26 ++++++++++++++++++++++++++
1 files changed, 26 insertions(+), 0 deletions(-)
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index 9db4e32..3b79684 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -25,6 +25,7 @@
#include <linux/vmalloc.h>
#include <linux/module.h>
#include <linux/mman.h>
+#include <linux/highmem.h>
#include <asm/uaccess.h>
#include <asm/msr.h>
@@ -1674,6 +1675,31 @@ static int emulator_cmpxchg_emulated(unsigned long addr,
reported = 1;
printk(KERN_WARNING "kvm: emulating exchange as write\n");
}
+#ifndef CONFIG_X86_64
+ /* guests cmpxchg8b have to be emulated atomically */
+ if (bytes == 8) {
+ gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr);
+ struct page *page;
+ char *addr;
+ u64 val;
+
+ if (gpa == UNMAPPED_GVA ||
+ (gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
+ goto emul_write;
+
+ if (((gpa + bytes - 1) & PAGE_MASK) != (gpa & PAGE_MASK))
+ goto emul_write;
+
+ val = *(u64 *)new;
+ page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT);
+ addr = kmap_atomic(page, KM_USER0);
+ set_64bit((u64 *)(addr + offset_in_page(gpa)), val);
+ kunmap_atomic(addr, KM_USER0);
+ kvm_release_page_dirty(page);
+ }
+emul_write:
+#endif
+
return emulator_write_emulated(addr, new, bytes, vcpu);
}
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* Re: [kvm-devel] [PATCH 02/33] KVM: MMU: emulated cmpxchg8b should be atomic on i386
2008-01-06 14:39 ` [PATCH 02/33] KVM: MMU: emulated cmpxchg8b should be atomic on i386 Avi Kivity
@ 2008-01-06 18:57 ` Carlo Marcelo Arenas Belon
2008-01-07 9:24 ` Avi Kivity
0 siblings, 1 reply; 36+ messages in thread
From: Carlo Marcelo Arenas Belon @ 2008-01-06 18:57 UTC (permalink / raw)
To: Avi Kivity; +Cc: linux-kernel, kvm-devel, Marcelo Tosatti
On Sun, Jan 06, 2008 at 04:39:08PM +0200, Avi Kivity wrote:
> From: Marcelo Tosatti <marcelo@kvack.org>
>
> Emulate cmpxchg8b atomically on i386. This is required to avoid a guest
> pte walker from seeing a splitted write.
>
> [avi: make it compile]
>
> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
> Signed-off-by: Avi Kivity <avi@qumranet.com>
> ---
This one doesn't look like the one in the current tree, and avoids the reuse
of addr by moving the call to get the gpa up instead of renaming the variable
as propossed in :
http://article.gmane.org/gmane.comp.emulators.kvm.devel/11511
IMHO renaming the parameter variable is probably better in the long run, as it
could result otherwise into scope related bugs like the one from kvm-59
> diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
> index 9db4e32..3b79684 100644
> --- a/drivers/kvm/x86.c
> +++ b/drivers/kvm/x86.c
> @@ -1674,6 +1675,31 @@ static int emulator_cmpxchg_emulated(unsigned long addr,
> reported = 1;
> printk(KERN_WARNING "kvm: emulating exchange as write\n");
> }
> +#ifndef CONFIG_X86_64
> + /* guests cmpxchg8b have to be emulated atomically */
> + if (bytes == 8) {
> + gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr);
> + struct page *page;
> + char *addr;
> + u64 val;
> +
> + if (gpa == UNMAPPED_GVA ||
> + (gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
> + goto emul_write;
> +
> + if (((gpa + bytes - 1) & PAGE_MASK) != (gpa & PAGE_MASK))
> + goto emul_write;
> +
> + val = *(u64 *)new;
> + page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT);
> + addr = kmap_atomic(page, KM_USER0);
> + set_64bit((u64 *)(addr + offset_in_page(gpa)), val);
> + kunmap_atomic(addr, KM_USER0);
> + kvm_release_page_dirty(page);
> + }
> +emul_write:
> +#endif
> +
> return emulator_write_emulated(addr, new, bytes, vcpu);
> }
>
> --
> 1.5.3.7
^ permalink raw reply [flat|nested] 36+ messages in thread* Re: [kvm-devel] [PATCH 02/33] KVM: MMU: emulated cmpxchg8b should be atomic on i386
2008-01-06 18:57 ` [kvm-devel] " Carlo Marcelo Arenas Belon
@ 2008-01-07 9:24 ` Avi Kivity
0 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-07 9:24 UTC (permalink / raw)
To: Carlo Marcelo Arenas Belon; +Cc: kvm-devel, Marcelo Tosatti, linux-kernel
Carlo Marcelo Arenas Belon wrote:
> On Sun, Jan 06, 2008 at 04:39:08PM +0200, Avi Kivity wrote:
>
>> From: Marcelo Tosatti <marcelo@kvack.org>
>>
>> Emulate cmpxchg8b atomically on i386. This is required to avoid a guest
>> pte walker from seeing a splitted write.
>>
>> [avi: make it compile]
>>
>> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
>> Signed-off-by: Avi Kivity <avi@qumranet.com>
>> ---
>>
>
> This one doesn't look like the one in the current tree, and avoids the reuse
> of addr by moving the call to get the gpa up instead of renaming the variable
> as propossed in :
>
>
This is a cumulative patch that is the original patch plus all the fixes
rolled in. I'd to avoid churn for style issues at this point, maybe we
can address it later.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 36+ messages in thread
* [PATCH 03/33] KVM: Portability: Move kvm{pic,ioapic} accesors to x86 specific code
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
2008-01-06 14:39 ` [PATCH 01/33] KVM: SVM: support writing 0 to K8 performance counter control registers Avi Kivity
2008-01-06 14:39 ` [PATCH 02/33] KVM: MMU: emulated cmpxchg8b should be atomic on i386 Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 04/33] KVM: Portability: Introduce kvm_vcpu_arch Avi Kivity
` (29 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
drivers/kvm/irq.h | 18 ++++++++++++++++--
drivers/kvm/kvm.h | 18 +++---------------
drivers/kvm/x86.h | 1 -
3 files changed, 19 insertions(+), 18 deletions(-)
diff --git a/drivers/kvm/irq.h b/drivers/kvm/irq.h
index 730a87c..e88d939 100644
--- a/drivers/kvm/irq.h
+++ b/drivers/kvm/irq.h
@@ -26,6 +26,7 @@
#include <linux/hrtimer.h>
#include <asm/kvm.h>
#include "iodev.h"
+#include "x86.h"
struct kvm;
struct kvm_vcpu;
@@ -63,8 +64,6 @@ struct kvm_pic {
struct kvm_pic *kvm_create_pic(struct kvm *kvm);
void kvm_pic_set_irq(void *opaque, int irq, int level);
int kvm_pic_read_irq(struct kvm_pic *s);
-int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
-int kvm_cpu_has_interrupt(struct kvm_vcpu *v);
void kvm_pic_update_irq(struct kvm_pic *s);
#define IOAPIC_NUM_PINS KVM_IOAPIC_NUM_PINS
@@ -147,6 +146,21 @@ do { \
#define ASSERT(x) do { } while (0)
#endif
+static inline struct kvm_pic *pic_irqchip(struct kvm *kvm)
+{
+ return kvm->vpic;
+}
+
+static inline struct kvm_ioapic *ioapic_irqchip(struct kvm *kvm)
+{
+ return kvm->vioapic;
+}
+
+static inline int irqchip_in_kernel(struct kvm *kvm)
+{
+ return pic_irqchip(kvm) != NULL;
+}
+
void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu);
int kvm_apic_accept_pic_intr(struct kvm_vcpu *vcpu);
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index ceefb44..668a830 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -273,21 +273,6 @@ struct kvm {
struct kvm_vm_stat stat;
};
-static inline struct kvm_pic *pic_irqchip(struct kvm *kvm)
-{
- return kvm->vpic;
-}
-
-static inline struct kvm_ioapic *ioapic_irqchip(struct kvm *kvm)
-{
- return kvm->vioapic;
-}
-
-static inline int irqchip_in_kernel(struct kvm *kvm)
-{
- return pic_irqchip(kvm) != NULL;
-}
-
/* The guest did something we don't support. */
#define pr_unimpl(vcpu, fmt, ...) \
do { \
@@ -417,6 +402,9 @@ void kvm_free_physmem(struct kvm *kvm);
struct kvm *kvm_arch_create_vm(void);
void kvm_arch_destroy_vm(struct kvm *kvm);
+int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
+int kvm_cpu_has_interrupt(struct kvm_vcpu *v);
+
static inline void kvm_guest_enter(void)
{
account_system_vtime(current);
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index 78396d6..0fc7020 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -12,7 +12,6 @@
#define KVM_X86_H
#include "kvm.h"
-#include "irq.h"
#include <linux/types.h>
#include <linux/mm.h>
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 04/33] KVM: Portability: Introduce kvm_vcpu_arch
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (2 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 03/33] KVM: Portability: Move kvm{pic,ioapic} accesors to x86 specific code Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 05/33] KVM: Portability: Split mmu-related static inline functions to mmu.h Avi Kivity
` (28 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
Move all the architecture-specific fields in kvm_vcpu into a new struct
kvm_vcpu_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>
---
drivers/kvm/ioapic.c | 10 +-
drivers/kvm/kvm_main.c | 2 +-
drivers/kvm/lapic.c | 91 ++++----
drivers/kvm/mmu.c | 142 ++++++------
drivers/kvm/paging_tmpl.h | 16 +-
drivers/kvm/svm.c | 122 ++++++------
drivers/kvm/vmx.c | 212 +++++++++---------
drivers/kvm/x86.c | 522 ++++++++++++++++++++++----------------------
drivers/kvm/x86.h | 25 ++-
drivers/kvm/x86_emulate.c | 24 +-
10 files changed, 585 insertions(+), 581 deletions(-)
diff --git a/drivers/kvm/ioapic.c b/drivers/kvm/ioapic.c
index e7debfa..0491036 100644
--- a/drivers/kvm/ioapic.c
+++ b/drivers/kvm/ioapic.c
@@ -158,7 +158,7 @@ static u32 ioapic_get_delivery_bitmask(struct kvm_ioapic *ioapic, u8 dest,
if (dest_mode == 0) { /* Physical mode. */
if (dest == 0xFF) { /* Broadcast. */
for (i = 0; i < KVM_MAX_VCPUS; ++i)
- if (kvm->vcpus[i] && kvm->vcpus[i]->apic)
+ if (kvm->vcpus[i] && kvm->vcpus[i]->arch.apic)
mask |= 1 << i;
return mask;
}
@@ -166,8 +166,8 @@ static u32 ioapic_get_delivery_bitmask(struct kvm_ioapic *ioapic, u8 dest,
vcpu = kvm->vcpus[i];
if (!vcpu)
continue;
- if (kvm_apic_match_physical_addr(vcpu->apic, dest)) {
- if (vcpu->apic)
+ if (kvm_apic_match_physical_addr(vcpu->arch.apic, dest)) {
+ if (vcpu->arch.apic)
mask = 1 << i;
break;
}
@@ -177,8 +177,8 @@ static u32 ioapic_get_delivery_bitmask(struct kvm_ioapic *ioapic, u8 dest,
vcpu = kvm->vcpus[i];
if (!vcpu)
continue;
- if (vcpu->apic &&
- kvm_apic_match_logical_addr(vcpu->apic, dest))
+ if (vcpu->arch.apic &&
+ kvm_apic_match_logical_addr(vcpu->arch.apic, dest))
mask |= 1 << vcpu->vcpu_id;
}
ioapic_debug("mask %x\n", mask);
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index cfb27ef..5bfa200 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -670,7 +670,7 @@ static int kvm_vcpu_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
if (vmf->pgoff == 0)
page = virt_to_page(vcpu->run);
else if (vmf->pgoff == KVM_PIO_PAGE_OFFSET)
- page = virt_to_page(vcpu->pio_data);
+ page = virt_to_page(vcpu->arch.pio_data);
else
return VM_FAULT_SIGBUS;
get_page(page);
diff --git a/drivers/kvm/lapic.c b/drivers/kvm/lapic.c
index 466c37f..5c9f467 100644
--- a/drivers/kvm/lapic.c
+++ b/drivers/kvm/lapic.c
@@ -58,6 +58,7 @@
#define VEC_POS(v) ((v) & (32 - 1))
#define REG_POS(v) (((v) >> 5) << 4)
+
static inline u32 apic_get_reg(struct kvm_lapic *apic, int reg_off)
{
return *((u32 *) (apic->regs + reg_off));
@@ -90,7 +91,7 @@ static inline void apic_clear_vector(int vec, void *bitmap)
static inline int apic_hw_enabled(struct kvm_lapic *apic)
{
- return (apic)->vcpu->apic_base & MSR_IA32_APICBASE_ENABLE;
+ return (apic)->vcpu->arch.apic_base & MSR_IA32_APICBASE_ENABLE;
}
static inline int apic_sw_enabled(struct kvm_lapic *apic)
@@ -174,7 +175,7 @@ static inline int apic_find_highest_irr(struct kvm_lapic *apic)
int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu)
{
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
int highest_irr;
if (!apic)
@@ -187,7 +188,7 @@ EXPORT_SYMBOL_GPL(kvm_lapic_find_highest_irr);
int kvm_apic_set_irq(struct kvm_vcpu *vcpu, u8 vec, u8 trig)
{
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
if (!apic_test_and_set_irr(vec, apic)) {
/* a new pending irq is set in IRR */
@@ -272,7 +273,7 @@ static int apic_match_dest(struct kvm_vcpu *vcpu, struct kvm_lapic *source,
int short_hand, int dest, int dest_mode)
{
int result = 0;
- struct kvm_lapic *target = vcpu->apic;
+ struct kvm_lapic *target = vcpu->arch.apic;
apic_debug("target %p, source %p, dest 0x%x, "
"dest_mode 0x%x, short_hand 0x%x",
@@ -339,10 +340,10 @@ static int __apic_accept_irq(struct kvm_lapic *apic, int delivery_mode,
} else
apic_clear_vector(vector, apic->regs + APIC_TMR);
- if (vcpu->mp_state == VCPU_MP_STATE_RUNNABLE)
+ if (vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE)
kvm_vcpu_kick(vcpu);
- else if (vcpu->mp_state == VCPU_MP_STATE_HALTED) {
- vcpu->mp_state = VCPU_MP_STATE_RUNNABLE;
+ else if (vcpu->arch.mp_state == VCPU_MP_STATE_HALTED) {
+ vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
if (waitqueue_active(&vcpu->wq))
wake_up_interruptible(&vcpu->wq);
}
@@ -363,11 +364,11 @@ static int __apic_accept_irq(struct kvm_lapic *apic, int delivery_mode,
case APIC_DM_INIT:
if (level) {
- if (vcpu->mp_state == VCPU_MP_STATE_RUNNABLE)
+ if (vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE)
printk(KERN_DEBUG
"INIT on a runnable vcpu %d\n",
vcpu->vcpu_id);
- vcpu->mp_state = VCPU_MP_STATE_INIT_RECEIVED;
+ vcpu->arch.mp_state = VCPU_MP_STATE_INIT_RECEIVED;
kvm_vcpu_kick(vcpu);
} else {
printk(KERN_DEBUG
@@ -380,9 +381,9 @@ static int __apic_accept_irq(struct kvm_lapic *apic, int delivery_mode,
case APIC_DM_STARTUP:
printk(KERN_DEBUG "SIPI to vcpu %d vector 0x%02x\n",
vcpu->vcpu_id, vector);
- if (vcpu->mp_state == VCPU_MP_STATE_INIT_RECEIVED) {
- vcpu->sipi_vector = vector;
- vcpu->mp_state = VCPU_MP_STATE_SIPI_RECEIVED;
+ if (vcpu->arch.mp_state == VCPU_MP_STATE_INIT_RECEIVED) {
+ vcpu->arch.sipi_vector = vector;
+ vcpu->arch.mp_state = VCPU_MP_STATE_SIPI_RECEIVED;
if (waitqueue_active(&vcpu->wq))
wake_up_interruptible(&vcpu->wq);
}
@@ -411,7 +412,7 @@ static struct kvm_lapic *kvm_apic_round_robin(struct kvm *kvm, u8 vector,
next = 0;
if (kvm->vcpus[next] == NULL || !test_bit(next, &bitmap))
continue;
- apic = kvm->vcpus[next]->apic;
+ apic = kvm->vcpus[next]->arch.apic;
if (apic && apic_enabled(apic))
break;
apic = NULL;
@@ -482,12 +483,12 @@ static void apic_send_ipi(struct kvm_lapic *apic)
if (!vcpu)
continue;
- if (vcpu->apic &&
+ if (vcpu->arch.apic &&
apic_match_dest(vcpu, apic, short_hand, dest, dest_mode)) {
if (delivery_mode == APIC_DM_LOWEST)
set_bit(vcpu->vcpu_id, &lpr_map);
else
- __apic_accept_irq(vcpu->apic, delivery_mode,
+ __apic_accept_irq(vcpu->arch.apic, delivery_mode,
vector, level, trig_mode);
}
}
@@ -495,7 +496,7 @@ static void apic_send_ipi(struct kvm_lapic *apic)
if (delivery_mode == APIC_DM_LOWEST) {
target = kvm_get_lowest_prio_vcpu(vcpu->kvm, vector, lpr_map);
if (target != NULL)
- __apic_accept_irq(target->apic, delivery_mode,
+ __apic_accept_irq(target->arch.apic, delivery_mode,
vector, level, trig_mode);
}
}
@@ -772,15 +773,15 @@ static int apic_mmio_range(struct kvm_io_device *this, gpa_t addr)
void kvm_free_lapic(struct kvm_vcpu *vcpu)
{
- if (!vcpu->apic)
+ if (!vcpu->arch.apic)
return;
- hrtimer_cancel(&vcpu->apic->timer.dev);
+ hrtimer_cancel(&vcpu->arch.apic->timer.dev);
- if (vcpu->apic->regs_page)
- __free_page(vcpu->apic->regs_page);
+ if (vcpu->arch.apic->regs_page)
+ __free_page(vcpu->arch.apic->regs_page);
- kfree(vcpu->apic);
+ kfree(vcpu->arch.apic);
}
/*
@@ -791,7 +792,7 @@ void kvm_free_lapic(struct kvm_vcpu *vcpu)
void kvm_lapic_set_tpr(struct kvm_vcpu *vcpu, unsigned long cr8)
{
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
if (!apic)
return;
@@ -800,7 +801,7 @@ void kvm_lapic_set_tpr(struct kvm_vcpu *vcpu, unsigned long cr8)
u64 kvm_lapic_get_cr8(struct kvm_vcpu *vcpu)
{
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
u64 tpr;
if (!apic)
@@ -813,29 +814,29 @@ EXPORT_SYMBOL_GPL(kvm_lapic_get_cr8);
void kvm_lapic_set_base(struct kvm_vcpu *vcpu, u64 value)
{
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
if (!apic) {
value |= MSR_IA32_APICBASE_BSP;
- vcpu->apic_base = value;
+ vcpu->arch.apic_base = value;
return;
}
if (apic->vcpu->vcpu_id)
value &= ~MSR_IA32_APICBASE_BSP;
- vcpu->apic_base = value;
- apic->base_address = apic->vcpu->apic_base &
+ vcpu->arch.apic_base = value;
+ apic->base_address = apic->vcpu->arch.apic_base &
MSR_IA32_APICBASE_BASE;
/* with FSB delivery interrupt, we can restart APIC functionality */
apic_debug("apic base msr is 0x%016" PRIx64 ", and base address is "
- "0x%lx.\n", apic->vcpu->apic_base, apic->base_address);
+ "0x%lx.\n", apic->vcpu->arch.apic_base, apic->base_address);
}
u64 kvm_lapic_get_base(struct kvm_vcpu *vcpu)
{
- return vcpu->apic_base;
+ return vcpu->arch.apic_base;
}
EXPORT_SYMBOL_GPL(kvm_lapic_get_base);
@@ -847,7 +848,7 @@ void kvm_lapic_reset(struct kvm_vcpu *vcpu)
apic_debug("%s\n", __FUNCTION__);
ASSERT(vcpu);
- apic = vcpu->apic;
+ apic = vcpu->arch.apic;
ASSERT(apic != NULL);
/* Stop the timer in case it's a reset to an active apic */
@@ -878,19 +879,19 @@ void kvm_lapic_reset(struct kvm_vcpu *vcpu)
update_divide_count(apic);
atomic_set(&apic->timer.pending, 0);
if (vcpu->vcpu_id == 0)
- vcpu->apic_base |= MSR_IA32_APICBASE_BSP;
+ vcpu->arch.apic_base |= MSR_IA32_APICBASE_BSP;
apic_update_ppr(apic);
apic_debug(KERN_INFO "%s: vcpu=%p, id=%d, base_msr="
"0x%016" PRIx64 ", base_address=0x%0lx.\n", __FUNCTION__,
vcpu, kvm_apic_id(apic),
- vcpu->apic_base, apic->base_address);
+ vcpu->arch.apic_base, apic->base_address);
}
EXPORT_SYMBOL_GPL(kvm_lapic_reset);
int kvm_lapic_enabled(struct kvm_vcpu *vcpu)
{
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
int ret = 0;
if (!apic)
@@ -915,7 +916,7 @@ static int __apic_timer_fn(struct kvm_lapic *apic)
atomic_inc(&apic->timer.pending);
if (waitqueue_active(q)) {
- apic->vcpu->mp_state = VCPU_MP_STATE_RUNNABLE;
+ apic->vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
wake_up_interruptible(q);
}
if (apic_lvtt_period(apic)) {
@@ -961,7 +962,7 @@ int kvm_create_lapic(struct kvm_vcpu *vcpu)
if (!apic)
goto nomem;
- vcpu->apic = apic;
+ vcpu->arch.apic = apic;
apic->regs_page = alloc_page(GFP_KERNEL);
if (apic->regs_page == NULL) {
@@ -976,7 +977,7 @@ int kvm_create_lapic(struct kvm_vcpu *vcpu)
hrtimer_init(&apic->timer.dev, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
apic->timer.dev.function = apic_timer_fn;
apic->base_address = APIC_DEFAULT_PHYS_BASE;
- vcpu->apic_base = APIC_DEFAULT_PHYS_BASE;
+ vcpu->arch.apic_base = APIC_DEFAULT_PHYS_BASE;
kvm_lapic_reset(vcpu);
apic->dev.read = apic_mmio_read;
@@ -994,7 +995,7 @@ EXPORT_SYMBOL_GPL(kvm_create_lapic);
int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu)
{
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
int highest_irr;
if (!apic || !apic_enabled(apic))
@@ -1010,11 +1011,11 @@ int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu)
int kvm_apic_accept_pic_intr(struct kvm_vcpu *vcpu)
{
- u32 lvt0 = apic_get_reg(vcpu->apic, APIC_LVT0);
+ u32 lvt0 = apic_get_reg(vcpu->arch.apic, APIC_LVT0);
int r = 0;
if (vcpu->vcpu_id == 0) {
- if (!apic_hw_enabled(vcpu->apic))
+ if (!apic_hw_enabled(vcpu->arch.apic))
r = 1;
if ((lvt0 & APIC_LVT_MASKED) == 0 &&
GET_APIC_DELIVERY_MODE(lvt0) == APIC_MODE_EXTINT)
@@ -1025,7 +1026,7 @@ int kvm_apic_accept_pic_intr(struct kvm_vcpu *vcpu)
void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu)
{
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
if (apic && apic_lvt_enabled(apic, APIC_LVTT) &&
atomic_read(&apic->timer.pending) > 0) {
@@ -1036,7 +1037,7 @@ void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu)
void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec)
{
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
if (apic && apic_lvt_vector(apic, APIC_LVTT) == vec)
apic->timer.last_update = ktime_add_ns(
@@ -1047,7 +1048,7 @@ void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec)
int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu)
{
int vector = kvm_apic_has_interrupt(vcpu);
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
if (vector == -1)
return -1;
@@ -1060,9 +1061,9 @@ int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu)
void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu)
{
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
- apic->base_address = vcpu->apic_base &
+ apic->base_address = vcpu->arch.apic_base &
MSR_IA32_APICBASE_BASE;
apic_set_reg(apic, APIC_LVR, APIC_VERSION);
apic_update_ppr(apic);
@@ -1073,7 +1074,7 @@ void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu)
void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu)
{
- struct kvm_lapic *apic = vcpu->apic;
+ struct kvm_lapic *apic = vcpu->arch.apic;
struct hrtimer *timer;
if (!apic)
diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c
index 92ac0d1..da1dedb 100644
--- a/drivers/kvm/mmu.c
+++ b/drivers/kvm/mmu.c
@@ -180,7 +180,7 @@ EXPORT_SYMBOL_GPL(kvm_mmu_set_nonpresent_ptes);
static int is_write_protection(struct kvm_vcpu *vcpu)
{
- return vcpu->cr0 & X86_CR0_WP;
+ return vcpu->arch.cr0 & X86_CR0_WP;
}
static int is_cpuid_PSE36(void)
@@ -190,7 +190,7 @@ static int is_cpuid_PSE36(void)
static int is_nx(struct kvm_vcpu *vcpu)
{
- return vcpu->shadow_efer & EFER_NX;
+ return vcpu->arch.shadow_efer & EFER_NX;
}
static int is_present_pte(unsigned long pte)
@@ -292,18 +292,18 @@ static int mmu_topup_memory_caches(struct kvm_vcpu *vcpu)
int r;
kvm_mmu_free_some_pages(vcpu);
- r = mmu_topup_memory_cache(&vcpu->mmu_pte_chain_cache,
+ r = mmu_topup_memory_cache(&vcpu->arch.mmu_pte_chain_cache,
pte_chain_cache, 4);
if (r)
goto out;
- r = mmu_topup_memory_cache(&vcpu->mmu_rmap_desc_cache,
+ r = mmu_topup_memory_cache(&vcpu->arch.mmu_rmap_desc_cache,
rmap_desc_cache, 1);
if (r)
goto out;
- r = mmu_topup_memory_cache_page(&vcpu->mmu_page_cache, 8);
+ r = mmu_topup_memory_cache_page(&vcpu->arch.mmu_page_cache, 8);
if (r)
goto out;
- r = mmu_topup_memory_cache(&vcpu->mmu_page_header_cache,
+ r = mmu_topup_memory_cache(&vcpu->arch.mmu_page_header_cache,
mmu_page_header_cache, 4);
out:
return r;
@@ -311,10 +311,10 @@ out:
static void mmu_free_memory_caches(struct kvm_vcpu *vcpu)
{
- mmu_free_memory_cache(&vcpu->mmu_pte_chain_cache);
- mmu_free_memory_cache(&vcpu->mmu_rmap_desc_cache);
- mmu_free_memory_cache_page(&vcpu->mmu_page_cache);
- mmu_free_memory_cache(&vcpu->mmu_page_header_cache);
+ mmu_free_memory_cache(&vcpu->arch.mmu_pte_chain_cache);
+ mmu_free_memory_cache(&vcpu->arch.mmu_rmap_desc_cache);
+ mmu_free_memory_cache_page(&vcpu->arch.mmu_page_cache);
+ mmu_free_memory_cache(&vcpu->arch.mmu_page_header_cache);
}
static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc,
@@ -330,7 +330,7 @@ static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc,
static struct kvm_pte_chain *mmu_alloc_pte_chain(struct kvm_vcpu *vcpu)
{
- return mmu_memory_cache_alloc(&vcpu->mmu_pte_chain_cache,
+ return mmu_memory_cache_alloc(&vcpu->arch.mmu_pte_chain_cache,
sizeof(struct kvm_pte_chain));
}
@@ -341,7 +341,7 @@ static void mmu_free_pte_chain(struct kvm_pte_chain *pc)
static struct kvm_rmap_desc *mmu_alloc_rmap_desc(struct kvm_vcpu *vcpu)
{
- return mmu_memory_cache_alloc(&vcpu->mmu_rmap_desc_cache,
+ return mmu_memory_cache_alloc(&vcpu->arch.mmu_rmap_desc_cache,
sizeof(struct kvm_rmap_desc));
}
@@ -568,9 +568,9 @@ static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu,
if (!vcpu->kvm->n_free_mmu_pages)
return NULL;
- sp = mmu_memory_cache_alloc(&vcpu->mmu_page_header_cache, sizeof *sp);
- sp->spt = mmu_memory_cache_alloc(&vcpu->mmu_page_cache, PAGE_SIZE);
- sp->gfns = mmu_memory_cache_alloc(&vcpu->mmu_page_cache, PAGE_SIZE);
+ sp = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_header_cache, sizeof *sp);
+ sp->spt = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
+ sp->gfns = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
set_page_private(virt_to_page(sp->spt), (unsigned long)sp);
list_add(&sp->link, &vcpu->kvm->active_mmu_pages);
ASSERT(is_empty_shadow_page(sp->spt));
@@ -692,11 +692,11 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
struct hlist_node *node;
role.word = 0;
- role.glevels = vcpu->mmu.root_level;
+ role.glevels = vcpu->arch.mmu.root_level;
role.level = level;
role.metaphysical = metaphysical;
role.access = access;
- if (vcpu->mmu.root_level <= PT32_ROOT_LEVEL) {
+ if (vcpu->arch.mmu.root_level <= PT32_ROOT_LEVEL) {
quadrant = gaddr >> (PAGE_SHIFT + (PT64_PT_BITS * level));
quadrant &= (1 << ((PT32_PT_BITS - PT64_PT_BITS) * level)) - 1;
role.quadrant = quadrant;
@@ -718,7 +718,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
sp->gfn = gfn;
sp->role = role;
hlist_add_head(&sp->hash_link, bucket);
- vcpu->mmu.prefetch_page(vcpu, sp);
+ vcpu->arch.mmu.prefetch_page(vcpu, sp);
if (!metaphysical)
rmap_write_protect(vcpu->kvm, gfn);
if (new_page)
@@ -768,7 +768,7 @@ static void kvm_mmu_reset_last_pte_updated(struct kvm *kvm)
for (i = 0; i < KVM_MAX_VCPUS; ++i)
if (kvm->vcpus[i])
- kvm->vcpus[i]->last_pte_updated = NULL;
+ kvm->vcpus[i]->arch.last_pte_updated = NULL;
}
static void kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp)
@@ -875,7 +875,7 @@ static void page_header_update_slot(struct kvm *kvm, void *pte, gfn_t gfn)
struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva)
{
- gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, gva);
+ gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
if (gpa == UNMAPPED_GVA)
return NULL;
@@ -962,7 +962,7 @@ unshadowed:
else
kvm_release_page_clean(page);
if (!ptwrite || !*ptwrite)
- vcpu->last_pte_updated = shadow_pte;
+ vcpu->arch.last_pte_updated = shadow_pte;
}
static void nonpaging_new_cr3(struct kvm_vcpu *vcpu)
@@ -972,7 +972,7 @@ static void nonpaging_new_cr3(struct kvm_vcpu *vcpu)
static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
{
int level = PT32E_ROOT_LEVEL;
- hpa_t table_addr = vcpu->mmu.root_hpa;
+ hpa_t table_addr = vcpu->arch.mmu.root_hpa;
int pt_write = 0;
for (; ; level--) {
@@ -1024,29 +1024,29 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu)
int i;
struct kvm_mmu_page *sp;
- if (!VALID_PAGE(vcpu->mmu.root_hpa))
+ if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
return;
#ifdef CONFIG_X86_64
- if (vcpu->mmu.shadow_root_level == PT64_ROOT_LEVEL) {
- hpa_t root = vcpu->mmu.root_hpa;
+ if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
+ hpa_t root = vcpu->arch.mmu.root_hpa;
sp = page_header(root);
--sp->root_count;
- vcpu->mmu.root_hpa = INVALID_PAGE;
+ vcpu->arch.mmu.root_hpa = INVALID_PAGE;
return;
}
#endif
for (i = 0; i < 4; ++i) {
- hpa_t root = vcpu->mmu.pae_root[i];
+ hpa_t root = vcpu->arch.mmu.pae_root[i];
if (root) {
root &= PT64_BASE_ADDR_MASK;
sp = page_header(root);
--sp->root_count;
}
- vcpu->mmu.pae_root[i] = INVALID_PAGE;
+ vcpu->arch.mmu.pae_root[i] = INVALID_PAGE;
}
- vcpu->mmu.root_hpa = INVALID_PAGE;
+ vcpu->arch.mmu.root_hpa = INVALID_PAGE;
}
static void mmu_alloc_roots(struct kvm_vcpu *vcpu)
@@ -1055,41 +1055,41 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu)
gfn_t root_gfn;
struct kvm_mmu_page *sp;
- root_gfn = vcpu->cr3 >> PAGE_SHIFT;
+ root_gfn = vcpu->arch.cr3 >> PAGE_SHIFT;
#ifdef CONFIG_X86_64
- if (vcpu->mmu.shadow_root_level == PT64_ROOT_LEVEL) {
- hpa_t root = vcpu->mmu.root_hpa;
+ if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
+ hpa_t root = vcpu->arch.mmu.root_hpa;
ASSERT(!VALID_PAGE(root));
sp = kvm_mmu_get_page(vcpu, root_gfn, 0,
PT64_ROOT_LEVEL, 0, ACC_ALL, NULL, NULL);
root = __pa(sp->spt);
++sp->root_count;
- vcpu->mmu.root_hpa = root;
+ vcpu->arch.mmu.root_hpa = root;
return;
}
#endif
for (i = 0; i < 4; ++i) {
- hpa_t root = vcpu->mmu.pae_root[i];
+ hpa_t root = vcpu->arch.mmu.pae_root[i];
ASSERT(!VALID_PAGE(root));
- if (vcpu->mmu.root_level == PT32E_ROOT_LEVEL) {
- if (!is_present_pte(vcpu->pdptrs[i])) {
- vcpu->mmu.pae_root[i] = 0;
+ if (vcpu->arch.mmu.root_level == PT32E_ROOT_LEVEL) {
+ if (!is_present_pte(vcpu->arch.pdptrs[i])) {
+ vcpu->arch.mmu.pae_root[i] = 0;
continue;
}
- root_gfn = vcpu->pdptrs[i] >> PAGE_SHIFT;
- } else if (vcpu->mmu.root_level == 0)
+ root_gfn = vcpu->arch.pdptrs[i] >> PAGE_SHIFT;
+ } else if (vcpu->arch.mmu.root_level == 0)
root_gfn = 0;
sp = kvm_mmu_get_page(vcpu, root_gfn, i << 30,
PT32_ROOT_LEVEL, !is_paging(vcpu),
ACC_ALL, NULL, NULL);
root = __pa(sp->spt);
++sp->root_count;
- vcpu->mmu.pae_root[i] = root | PT_PRESENT_MASK;
+ vcpu->arch.mmu.pae_root[i] = root | PT_PRESENT_MASK;
}
- vcpu->mmu.root_hpa = __pa(vcpu->mmu.pae_root);
+ vcpu->arch.mmu.root_hpa = __pa(vcpu->arch.mmu.pae_root);
}
static gpa_t nonpaging_gva_to_gpa(struct kvm_vcpu *vcpu, gva_t vaddr)
@@ -1109,7 +1109,7 @@ static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva,
return r;
ASSERT(vcpu);
- ASSERT(VALID_PAGE(vcpu->mmu.root_hpa));
+ ASSERT(VALID_PAGE(vcpu->arch.mmu.root_hpa));
gfn = gva >> PAGE_SHIFT;
@@ -1124,7 +1124,7 @@ static void nonpaging_free(struct kvm_vcpu *vcpu)
static int nonpaging_init_context(struct kvm_vcpu *vcpu)
{
- struct kvm_mmu *context = &vcpu->mmu;
+ struct kvm_mmu *context = &vcpu->arch.mmu;
context->new_cr3 = nonpaging_new_cr3;
context->page_fault = nonpaging_page_fault;
@@ -1171,7 +1171,7 @@ static void paging_free(struct kvm_vcpu *vcpu)
static int paging64_init_context_common(struct kvm_vcpu *vcpu, int level)
{
- struct kvm_mmu *context = &vcpu->mmu;
+ struct kvm_mmu *context = &vcpu->arch.mmu;
ASSERT(is_pae(vcpu));
context->new_cr3 = paging_new_cr3;
@@ -1192,7 +1192,7 @@ static int paging64_init_context(struct kvm_vcpu *vcpu)
static int paging32_init_context(struct kvm_vcpu *vcpu)
{
- struct kvm_mmu *context = &vcpu->mmu;
+ struct kvm_mmu *context = &vcpu->arch.mmu;
context->new_cr3 = paging_new_cr3;
context->page_fault = paging32_page_fault;
@@ -1213,7 +1213,7 @@ static int paging32E_init_context(struct kvm_vcpu *vcpu)
static int init_kvm_mmu(struct kvm_vcpu *vcpu)
{
ASSERT(vcpu);
- ASSERT(!VALID_PAGE(vcpu->mmu.root_hpa));
+ ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
if (!is_paging(vcpu))
return nonpaging_init_context(vcpu);
@@ -1228,9 +1228,9 @@ static int init_kvm_mmu(struct kvm_vcpu *vcpu)
static void destroy_kvm_mmu(struct kvm_vcpu *vcpu)
{
ASSERT(vcpu);
- if (VALID_PAGE(vcpu->mmu.root_hpa)) {
- vcpu->mmu.free(vcpu);
- vcpu->mmu.root_hpa = INVALID_PAGE;
+ if (VALID_PAGE(vcpu->arch.mmu.root_hpa)) {
+ vcpu->arch.mmu.free(vcpu);
+ vcpu->arch.mmu.root_hpa = INVALID_PAGE;
}
}
@@ -1250,7 +1250,7 @@ int kvm_mmu_load(struct kvm_vcpu *vcpu)
if (r)
goto out;
mmu_alloc_roots(vcpu);
- kvm_x86_ops->set_cr3(vcpu, vcpu->mmu.root_hpa);
+ kvm_x86_ops->set_cr3(vcpu, vcpu->arch.mmu.root_hpa);
kvm_mmu_flush_tlb(vcpu);
out:
mutex_unlock(&vcpu->kvm->lock);
@@ -1323,7 +1323,7 @@ static void mmu_pte_write_flush_tlb(struct kvm_vcpu *vcpu, u64 old, u64 new)
static bool last_updated_pte_accessed(struct kvm_vcpu *vcpu)
{
- u64 *spte = vcpu->last_pte_updated;
+ u64 *spte = vcpu->arch.last_pte_updated;
return !!(spte && (*spte & PT_ACCESSED_MASK));
}
@@ -1350,15 +1350,15 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
pgprintk("%s: gpa %llx bytes %d\n", __FUNCTION__, gpa, bytes);
++vcpu->kvm->stat.mmu_pte_write;
kvm_mmu_audit(vcpu, "pre pte write");
- if (gfn == vcpu->last_pt_write_gfn
+ if (gfn == vcpu->arch.last_pt_write_gfn
&& !last_updated_pte_accessed(vcpu)) {
- ++vcpu->last_pt_write_count;
- if (vcpu->last_pt_write_count >= 3)
+ ++vcpu->arch.last_pt_write_count;
+ if (vcpu->arch.last_pt_write_count >= 3)
flooded = 1;
} else {
- vcpu->last_pt_write_gfn = gfn;
- vcpu->last_pt_write_count = 1;
- vcpu->last_pte_updated = NULL;
+ vcpu->arch.last_pt_write_gfn = gfn;
+ vcpu->arch.last_pt_write_count = 1;
+ vcpu->arch.last_pte_updated = NULL;
}
index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
bucket = &vcpu->kvm->mmu_page_hash[index];
@@ -1420,7 +1420,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
{
- gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, gva);
+ gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
return kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT);
}
@@ -1443,7 +1443,7 @@ int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code)
enum emulation_result er;
mutex_lock(&vcpu->kvm->lock);
- r = vcpu->mmu.page_fault(vcpu, cr2, error_code);
+ r = vcpu->arch.mmu.page_fault(vcpu, cr2, error_code);
if (r < 0)
goto out;
@@ -1486,7 +1486,7 @@ static void free_mmu_pages(struct kvm_vcpu *vcpu)
struct kvm_mmu_page, link);
kvm_mmu_zap_page(vcpu->kvm, sp);
}
- free_page((unsigned long)vcpu->mmu.pae_root);
+ free_page((unsigned long)vcpu->arch.mmu.pae_root);
}
static int alloc_mmu_pages(struct kvm_vcpu *vcpu)
@@ -1508,9 +1508,9 @@ static int alloc_mmu_pages(struct kvm_vcpu *vcpu)
page = alloc_page(GFP_KERNEL | __GFP_DMA32);
if (!page)
goto error_1;
- vcpu->mmu.pae_root = page_address(page);
+ vcpu->arch.mmu.pae_root = page_address(page);
for (i = 0; i < 4; ++i)
- vcpu->mmu.pae_root[i] = INVALID_PAGE;
+ vcpu->arch.mmu.pae_root[i] = INVALID_PAGE;
return 0;
@@ -1522,7 +1522,7 @@ error_1:
int kvm_mmu_create(struct kvm_vcpu *vcpu)
{
ASSERT(vcpu);
- ASSERT(!VALID_PAGE(vcpu->mmu.root_hpa));
+ ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
return alloc_mmu_pages(vcpu);
}
@@ -1530,7 +1530,7 @@ int kvm_mmu_create(struct kvm_vcpu *vcpu)
int kvm_mmu_setup(struct kvm_vcpu *vcpu)
{
ASSERT(vcpu);
- ASSERT(!VALID_PAGE(vcpu->mmu.root_hpa));
+ ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa));
return init_kvm_mmu(vcpu);
}
@@ -1659,11 +1659,11 @@ static void audit_mappings_page(struct kvm_vcpu *vcpu, u64 page_pte,
printk(KERN_ERR "audit: (%s) nontrapping pte"
" in nonleaf level: levels %d gva %lx"
" level %d pte %llx\n", audit_msg,
- vcpu->mmu.root_level, va, level, ent);
+ vcpu->arch.mmu.root_level, va, level, ent);
audit_mappings_page(vcpu, ent, va, level - 1);
} else {
- gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, va);
+ gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, va);
struct page *page = gpa_to_page(vcpu, gpa);
hpa_t hpa = page_to_phys(page);
@@ -1671,7 +1671,7 @@ static void audit_mappings_page(struct kvm_vcpu *vcpu, u64 page_pte,
&& (ent & PT64_BASE_ADDR_MASK) != hpa)
printk(KERN_ERR "xx audit error: (%s) levels %d"
" gva %lx gpa %llx hpa %llx ent %llx %d\n",
- audit_msg, vcpu->mmu.root_level,
+ audit_msg, vcpu->arch.mmu.root_level,
va, gpa, hpa, ent,
is_shadow_present_pte(ent));
else if (ent == shadow_notrap_nonpresent_pte
@@ -1688,13 +1688,13 @@ static void audit_mappings(struct kvm_vcpu *vcpu)
{
unsigned i;
- if (vcpu->mmu.root_level == 4)
- audit_mappings_page(vcpu, vcpu->mmu.root_hpa, 0, 4);
+ if (vcpu->arch.mmu.root_level == 4)
+ audit_mappings_page(vcpu, vcpu->arch.mmu.root_hpa, 0, 4);
else
for (i = 0; i < 4; ++i)
- if (vcpu->mmu.pae_root[i] & PT_PRESENT_MASK)
+ if (vcpu->arch.mmu.pae_root[i] & PT_PRESENT_MASK)
audit_mappings_page(vcpu,
- vcpu->mmu.pae_root[i],
+ vcpu->arch.mmu.pae_root[i],
i << 30,
2);
}
diff --git a/drivers/kvm/paging_tmpl.h b/drivers/kvm/paging_tmpl.h
index fb19596..56b88f7 100644
--- a/drivers/kvm/paging_tmpl.h
+++ b/drivers/kvm/paging_tmpl.h
@@ -129,11 +129,11 @@ static int FNAME(walk_addr)(struct guest_walker *walker,
pgprintk("%s: addr %lx\n", __FUNCTION__, addr);
walk:
- walker->level = vcpu->mmu.root_level;
- pte = vcpu->cr3;
+ walker->level = vcpu->arch.mmu.root_level;
+ pte = vcpu->arch.cr3;
#if PTTYPE == 64
if (!is_long_mode(vcpu)) {
- pte = vcpu->pdptrs[(addr >> 30) & 3];
+ pte = vcpu->arch.pdptrs[(addr >> 30) & 3];
if (!is_present_pte(pte))
goto not_present;
--walker->level;
@@ -275,10 +275,10 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
if (!is_present_pte(walker->ptes[walker->level - 1]))
return NULL;
- shadow_addr = vcpu->mmu.root_hpa;
- level = vcpu->mmu.shadow_root_level;
+ shadow_addr = vcpu->arch.mmu.root_hpa;
+ level = vcpu->arch.mmu.shadow_root_level;
if (level == PT32E_ROOT_LEVEL) {
- shadow_addr = vcpu->mmu.pae_root[(addr >> 30) & 3];
+ shadow_addr = vcpu->arch.mmu.pae_root[(addr >> 30) & 3];
shadow_addr &= PT64_BASE_ADDR_MASK;
--level;
}
@@ -380,7 +380,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
if (!r) {
pgprintk("%s: guest page fault\n", __FUNCTION__);
inject_page_fault(vcpu, addr, walker.error_code);
- vcpu->last_pt_write_count = 0; /* reset fork detector */
+ vcpu->arch.last_pt_write_count = 0; /* reset fork detector */
return 0;
}
@@ -390,7 +390,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
shadow_pte, *shadow_pte, write_pt);
if (!write_pt)
- vcpu->last_pt_write_count = 0; /* reset fork detector */
+ vcpu->arch.last_pt_write_count = 0; /* reset fork detector */
/*
* mmio: emulate if accessible, otherwise its a guest fault.
diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c
index 839a75f..6c100b1 100644
--- a/drivers/kvm/svm.c
+++ b/drivers/kvm/svm.c
@@ -99,20 +99,20 @@ static inline u32 svm_has(u32 feat)
static inline u8 pop_irq(struct kvm_vcpu *vcpu)
{
- int word_index = __ffs(vcpu->irq_summary);
- int bit_index = __ffs(vcpu->irq_pending[word_index]);
+ int word_index = __ffs(vcpu->arch.irq_summary);
+ int bit_index = __ffs(vcpu->arch.irq_pending[word_index]);
int irq = word_index * BITS_PER_LONG + bit_index;
- clear_bit(bit_index, &vcpu->irq_pending[word_index]);
- if (!vcpu->irq_pending[word_index])
- clear_bit(word_index, &vcpu->irq_summary);
+ clear_bit(bit_index, &vcpu->arch.irq_pending[word_index]);
+ if (!vcpu->arch.irq_pending[word_index])
+ clear_bit(word_index, &vcpu->arch.irq_summary);
return irq;
}
static inline void push_irq(struct kvm_vcpu *vcpu, u8 irq)
{
- set_bit(irq, vcpu->irq_pending);
- set_bit(irq / BITS_PER_LONG, &vcpu->irq_summary);
+ set_bit(irq, vcpu->arch.irq_pending);
+ set_bit(irq / BITS_PER_LONG, &vcpu->arch.irq_summary);
}
static inline void clgi(void)
@@ -185,7 +185,7 @@ static void svm_set_efer(struct kvm_vcpu *vcpu, u64 efer)
efer &= ~EFER_LME;
to_svm(vcpu)->vmcb->save.efer = efer | MSR_EFER_SVME_MASK;
- vcpu->shadow_efer = efer;
+ vcpu->arch.shadow_efer = efer;
}
static void svm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr,
@@ -227,10 +227,10 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
svm->vmcb->save.rip,
svm->next_rip);
- vcpu->rip = svm->vmcb->save.rip = svm->next_rip;
+ vcpu->arch.rip = svm->vmcb->save.rip = svm->next_rip;
svm->vmcb->control.int_state &= ~SVM_INTERRUPT_SHADOW_MASK;
- vcpu->interrupt_window_open = 1;
+ vcpu->arch.interrupt_window_open = 1;
}
static int has_svm(void)
@@ -559,8 +559,8 @@ static int svm_vcpu_reset(struct kvm_vcpu *vcpu)
if (vcpu->vcpu_id != 0) {
svm->vmcb->save.rip = 0;
- svm->vmcb->save.cs.base = svm->vcpu.sipi_vector << 12;
- svm->vmcb->save.cs.selector = svm->vcpu.sipi_vector << 8;
+ svm->vmcb->save.cs.base = svm->vcpu.arch.sipi_vector << 12;
+ svm->vmcb->save.cs.selector = svm->vcpu.arch.sipi_vector << 8;
}
return 0;
@@ -597,9 +597,9 @@ static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id)
fx_init(&svm->vcpu);
svm->vcpu.fpu_active = 1;
- svm->vcpu.apic_base = 0xfee00000 | MSR_IA32_APICBASE_ENABLE;
+ svm->vcpu.arch.apic_base = 0xfee00000 | MSR_IA32_APICBASE_ENABLE;
if (svm->vcpu.vcpu_id == 0)
- svm->vcpu.apic_base |= MSR_IA32_APICBASE_BSP;
+ svm->vcpu.arch.apic_base |= MSR_IA32_APICBASE_BSP;
return &svm->vcpu;
@@ -633,7 +633,7 @@ static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
* increasing TSC.
*/
rdtscll(tsc_this);
- delta = vcpu->host_tsc - tsc_this;
+ delta = vcpu->arch.host_tsc - tsc_this;
svm->vmcb->control.tsc_offset += delta;
vcpu->cpu = cpu;
kvm_migrate_apic_timer(vcpu);
@@ -652,7 +652,7 @@ static void svm_vcpu_put(struct kvm_vcpu *vcpu)
for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++)
wrmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]);
- rdtscll(vcpu->host_tsc);
+ rdtscll(vcpu->arch.host_tsc);
}
static void svm_vcpu_decache(struct kvm_vcpu *vcpu)
@@ -663,17 +663,17 @@ static void svm_cache_regs(struct kvm_vcpu *vcpu)
{
struct vcpu_svm *svm = to_svm(vcpu);
- vcpu->regs[VCPU_REGS_RAX] = svm->vmcb->save.rax;
- vcpu->regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp;
- vcpu->rip = svm->vmcb->save.rip;
+ vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax;
+ vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp;
+ vcpu->arch.rip = svm->vmcb->save.rip;
}
static void svm_decache_regs(struct kvm_vcpu *vcpu)
{
struct vcpu_svm *svm = to_svm(vcpu);
- svm->vmcb->save.rax = vcpu->regs[VCPU_REGS_RAX];
- svm->vmcb->save.rsp = vcpu->regs[VCPU_REGS_RSP];
- svm->vmcb->save.rip = vcpu->rip;
+ svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX];
+ svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP];
+ svm->vmcb->save.rip = vcpu->arch.rip;
}
static unsigned long svm_get_rflags(struct kvm_vcpu *vcpu)
@@ -771,24 +771,24 @@ static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
struct vcpu_svm *svm = to_svm(vcpu);
#ifdef CONFIG_X86_64
- if (vcpu->shadow_efer & EFER_LME) {
+ if (vcpu->arch.shadow_efer & EFER_LME) {
if (!is_paging(vcpu) && (cr0 & X86_CR0_PG)) {
- vcpu->shadow_efer |= EFER_LMA;
+ vcpu->arch.shadow_efer |= EFER_LMA;
svm->vmcb->save.efer |= EFER_LMA | EFER_LME;
}
if (is_paging(vcpu) && !(cr0 & X86_CR0_PG)) {
- vcpu->shadow_efer &= ~EFER_LMA;
+ vcpu->arch.shadow_efer &= ~EFER_LMA;
svm->vmcb->save.efer &= ~(EFER_LMA | EFER_LME);
}
}
#endif
- if ((vcpu->cr0 & X86_CR0_TS) && !(cr0 & X86_CR0_TS)) {
+ if ((vcpu->arch.cr0 & X86_CR0_TS) && !(cr0 & X86_CR0_TS)) {
svm->vmcb->control.intercept_exceptions &= ~(1 << NM_VECTOR);
vcpu->fpu_active = 1;
}
- vcpu->cr0 = cr0;
+ vcpu->arch.cr0 = cr0;
cr0 |= X86_CR0_PG | X86_CR0_WP;
cr0 &= ~(X86_CR0_CD | X86_CR0_NW);
svm->vmcb->save.cr0 = cr0;
@@ -796,7 +796,7 @@ static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
static void svm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
{
- vcpu->cr4 = cr4;
+ vcpu->arch.cr4 = cr4;
to_svm(vcpu)->vmcb->save.cr4 = cr4 | X86_CR4_PAE;
}
@@ -901,7 +901,7 @@ static void svm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long value,
svm->db_regs[dr] = value;
return;
case 4 ... 5:
- if (vcpu->cr4 & X86_CR4_DE) {
+ if (vcpu->arch.cr4 & X86_CR4_DE) {
*exception = UD_VECTOR;
return;
}
@@ -950,7 +950,7 @@ static int ud_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
static int nm_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
{
svm->vmcb->control.intercept_exceptions &= ~(1 << NM_VECTOR);
- if (!(svm->vcpu.cr0 & X86_CR0_TS))
+ if (!(svm->vcpu.arch.cr0 & X86_CR0_TS))
svm->vmcb->save.cr0 &= ~X86_CR0_TS;
svm->vcpu.fpu_active = 1;
@@ -1103,14 +1103,14 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 *data)
static int rdmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
{
- u32 ecx = svm->vcpu.regs[VCPU_REGS_RCX];
+ u32 ecx = svm->vcpu.arch.regs[VCPU_REGS_RCX];
u64 data;
if (svm_get_msr(&svm->vcpu, ecx, &data))
kvm_inject_gp(&svm->vcpu, 0);
else {
svm->vmcb->save.rax = data & 0xffffffff;
- svm->vcpu.regs[VCPU_REGS_RDX] = data >> 32;
+ svm->vcpu.arch.regs[VCPU_REGS_RDX] = data >> 32;
svm->next_rip = svm->vmcb->save.rip + 2;
skip_emulated_instruction(&svm->vcpu);
}
@@ -1176,9 +1176,9 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 data)
static int wrmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
{
- u32 ecx = svm->vcpu.regs[VCPU_REGS_RCX];
+ u32 ecx = svm->vcpu.arch.regs[VCPU_REGS_RCX];
u64 data = (svm->vmcb->save.rax & -1u)
- | ((u64)(svm->vcpu.regs[VCPU_REGS_RDX] & -1u) << 32);
+ | ((u64)(svm->vcpu.arch.regs[VCPU_REGS_RDX] & -1u) << 32);
svm->next_rip = svm->vmcb->save.rip + 2;
if (svm_set_msr(&svm->vcpu, ecx, data))
kvm_inject_gp(&svm->vcpu, 0);
@@ -1205,7 +1205,7 @@ static int interrupt_window_interception(struct vcpu_svm *svm,
* possible
*/
if (kvm_run->request_interrupt_window &&
- !svm->vcpu.irq_summary) {
+ !svm->vcpu.arch.irq_summary) {
++svm->vcpu.stat.irq_window_exits;
kvm_run->exit_reason = KVM_EXIT_IRQ_WINDOW_OPEN;
return 0;
@@ -1382,20 +1382,20 @@ static void kvm_reput_irq(struct vcpu_svm *svm)
push_irq(&svm->vcpu, control->int_vector);
}
- svm->vcpu.interrupt_window_open =
+ svm->vcpu.arch.interrupt_window_open =
!(control->int_state & SVM_INTERRUPT_SHADOW_MASK);
}
static void svm_do_inject_vector(struct vcpu_svm *svm)
{
struct kvm_vcpu *vcpu = &svm->vcpu;
- int word_index = __ffs(vcpu->irq_summary);
- int bit_index = __ffs(vcpu->irq_pending[word_index]);
+ int word_index = __ffs(vcpu->arch.irq_summary);
+ int bit_index = __ffs(vcpu->arch.irq_pending[word_index]);
int irq = word_index * BITS_PER_LONG + bit_index;
- clear_bit(bit_index, &vcpu->irq_pending[word_index]);
- if (!vcpu->irq_pending[word_index])
- clear_bit(word_index, &vcpu->irq_summary);
+ clear_bit(bit_index, &vcpu->arch.irq_pending[word_index]);
+ if (!vcpu->arch.irq_pending[word_index])
+ clear_bit(word_index, &vcpu->arch.irq_summary);
svm_inject_irq(svm, irq);
}
@@ -1405,11 +1405,11 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu,
struct vcpu_svm *svm = to_svm(vcpu);
struct vmcb_control_area *control = &svm->vmcb->control;
- svm->vcpu.interrupt_window_open =
+ svm->vcpu.arch.interrupt_window_open =
(!(control->int_state & SVM_INTERRUPT_SHADOW_MASK) &&
(svm->vmcb->save.rflags & X86_EFLAGS_IF));
- if (svm->vcpu.interrupt_window_open && svm->vcpu.irq_summary)
+ if (svm->vcpu.arch.interrupt_window_open && svm->vcpu.arch.irq_summary)
/*
* If interrupts enabled, and not blocked by sti or mov ss. Good.
*/
@@ -1418,8 +1418,8 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu,
/*
* Interrupts blocked. Wait for unblock.
*/
- if (!svm->vcpu.interrupt_window_open &&
- (svm->vcpu.irq_summary || kvm_run->request_interrupt_window))
+ if (!svm->vcpu.arch.interrupt_window_open &&
+ (svm->vcpu.arch.irq_summary || kvm_run->request_interrupt_window))
control->intercept |= 1ULL << INTERCEPT_VINTR;
else
control->intercept &= ~(1ULL << INTERCEPT_VINTR);
@@ -1471,7 +1471,7 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
svm->host_cr2 = kvm_read_cr2();
svm->host_dr6 = read_dr6();
svm->host_dr7 = read_dr7();
- svm->vmcb->save.cr2 = vcpu->cr2;
+ svm->vmcb->save.cr2 = vcpu->arch.cr2;
if (svm->vmcb->save.dr7 & 0xff) {
write_dr7(0);
@@ -1563,21 +1563,21 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
:
: [svm]"a"(svm),
[vmcb]"i"(offsetof(struct vcpu_svm, vmcb_pa)),
- [rbx]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_RBX])),
- [rcx]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_RCX])),
- [rdx]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_RDX])),
- [rsi]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_RSI])),
- [rdi]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_RDI])),
- [rbp]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_RBP]))
+ [rbx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RBX])),
+ [rcx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RCX])),
+ [rdx]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RDX])),
+ [rsi]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RSI])),
+ [rdi]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RDI])),
+ [rbp]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_RBP]))
#ifdef CONFIG_X86_64
- , [r8]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_R8])),
- [r9]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_R9])),
- [r10]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_R10])),
- [r11]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_R11])),
- [r12]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_R12])),
- [r13]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_R13])),
- [r14]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_R14])),
- [r15]"i"(offsetof(struct vcpu_svm, vcpu.regs[VCPU_REGS_R15]))
+ , [r8]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R8])),
+ [r9]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R9])),
+ [r10]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R10])),
+ [r11]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R11])),
+ [r12]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R12])),
+ [r13]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R13])),
+ [r14]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R14])),
+ [r15]"i"(offsetof(struct vcpu_svm, vcpu.arch.regs[VCPU_REGS_R15]))
#endif
: "cc", "memory"
#ifdef CONFIG_X86_64
@@ -1591,7 +1591,7 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
if ((svm->vmcb->save.dr7 & 0xff))
load_db_regs(svm->host_db_regs);
- vcpu->cr2 = svm->vmcb->save.cr2;
+ vcpu->arch.cr2 = svm->vmcb->save.cr2;
write_dr6(svm->host_dr6);
write_dr7(svm->host_dr7);
diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c
index 548e3a5..e26e299 100644
--- a/drivers/kvm/vmx.c
+++ b/drivers/kvm/vmx.c
@@ -247,7 +247,7 @@ static void __vcpu_clear(void *arg)
vmcs_clear(vmx->vmcs);
if (per_cpu(current_vmcs, cpu) == vmx->vmcs)
per_cpu(current_vmcs, cpu) = NULL;
- rdtscll(vmx->vcpu.host_tsc);
+ rdtscll(vmx->vcpu.arch.host_tsc);
}
static void vcpu_clear(struct vcpu_vmx *vmx)
@@ -343,7 +343,7 @@ static void update_exception_bitmap(struct kvm_vcpu *vcpu)
eb |= 1u << NM_VECTOR;
if (vcpu->guest_debug.enabled)
eb |= 1u << 1;
- if (vcpu->rmode.active)
+ if (vcpu->arch.rmode.active)
eb = ~0;
vmcs_write32(EXCEPTION_BITMAP, eb);
}
@@ -528,7 +528,7 @@ static void vmx_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
* Make sure the time stamp counter is monotonous.
*/
rdtscll(tsc_this);
- delta = vcpu->host_tsc - tsc_this;
+ delta = vcpu->arch.host_tsc - tsc_this;
vmcs_write64(TSC_OFFSET, vmcs_read64(TSC_OFFSET) + delta);
}
}
@@ -544,7 +544,7 @@ static void vmx_fpu_activate(struct kvm_vcpu *vcpu)
return;
vcpu->fpu_active = 1;
vmcs_clear_bits(GUEST_CR0, X86_CR0_TS);
- if (vcpu->cr0 & X86_CR0_TS)
+ if (vcpu->arch.cr0 & X86_CR0_TS)
vmcs_set_bits(GUEST_CR0, X86_CR0_TS);
update_exception_bitmap(vcpu);
}
@@ -570,7 +570,7 @@ static unsigned long vmx_get_rflags(struct kvm_vcpu *vcpu)
static void vmx_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags)
{
- if (vcpu->rmode.active)
+ if (vcpu->arch.rmode.active)
rflags |= IOPL_MASK | X86_EFLAGS_VM;
vmcs_writel(GUEST_RFLAGS, rflags);
}
@@ -592,7 +592,7 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu)
if (interruptibility & 3)
vmcs_write32(GUEST_INTERRUPTIBILITY_INFO,
interruptibility & ~3);
- vcpu->interrupt_window_open = 1;
+ vcpu->arch.interrupt_window_open = 1;
}
static void vmx_queue_exception(struct kvm_vcpu *vcpu, unsigned nr,
@@ -661,7 +661,7 @@ static void setup_msrs(struct vcpu_vmx *vmx)
* if efer.sce is enabled.
*/
index = __find_msr_index(vmx, MSR_K6_STAR);
- if ((index >= 0) && (vmx->vcpu.shadow_efer & EFER_SCE))
+ if ((index >= 0) && (vmx->vcpu.arch.shadow_efer & EFER_SCE))
move_msr_up(vmx, index, save_nmsrs++);
}
#endif
@@ -805,12 +805,12 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data)
/*
* Sync the rsp and rip registers into the vcpu structure. This allows
- * registers to be accessed by indexing vcpu->regs.
+ * registers to be accessed by indexing vcpu->arch.regs.
*/
static void vcpu_load_rsp_rip(struct kvm_vcpu *vcpu)
{
- vcpu->regs[VCPU_REGS_RSP] = vmcs_readl(GUEST_RSP);
- vcpu->rip = vmcs_readl(GUEST_RIP);
+ vcpu->arch.regs[VCPU_REGS_RSP] = vmcs_readl(GUEST_RSP);
+ vcpu->arch.rip = vmcs_readl(GUEST_RIP);
}
/*
@@ -819,8 +819,8 @@ static void vcpu_load_rsp_rip(struct kvm_vcpu *vcpu)
*/
static void vcpu_put_rsp_rip(struct kvm_vcpu *vcpu)
{
- vmcs_writel(GUEST_RSP, vcpu->regs[VCPU_REGS_RSP]);
- vmcs_writel(GUEST_RIP, vcpu->rip);
+ vmcs_writel(GUEST_RSP, vcpu->arch.regs[VCPU_REGS_RSP]);
+ vmcs_writel(GUEST_RIP, vcpu->arch.rip);
}
static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg)
@@ -1111,15 +1111,15 @@ static void enter_pmode(struct kvm_vcpu *vcpu)
{
unsigned long flags;
- vcpu->rmode.active = 0;
+ vcpu->arch.rmode.active = 0;
- vmcs_writel(GUEST_TR_BASE, vcpu->rmode.tr.base);
- vmcs_write32(GUEST_TR_LIMIT, vcpu->rmode.tr.limit);
- vmcs_write32(GUEST_TR_AR_BYTES, vcpu->rmode.tr.ar);
+ vmcs_writel(GUEST_TR_BASE, vcpu->arch.rmode.tr.base);
+ vmcs_write32(GUEST_TR_LIMIT, vcpu->arch.rmode.tr.limit);
+ vmcs_write32(GUEST_TR_AR_BYTES, vcpu->arch.rmode.tr.ar);
flags = vmcs_readl(GUEST_RFLAGS);
flags &= ~(IOPL_MASK | X86_EFLAGS_VM);
- flags |= (vcpu->rmode.save_iopl << IOPL_SHIFT);
+ flags |= (vcpu->arch.rmode.save_iopl << IOPL_SHIFT);
vmcs_writel(GUEST_RFLAGS, flags);
vmcs_writel(GUEST_CR4, (vmcs_readl(GUEST_CR4) & ~X86_CR4_VME) |
@@ -1127,10 +1127,10 @@ static void enter_pmode(struct kvm_vcpu *vcpu)
update_exception_bitmap(vcpu);
- fix_pmode_dataseg(VCPU_SREG_ES, &vcpu->rmode.es);
- fix_pmode_dataseg(VCPU_SREG_DS, &vcpu->rmode.ds);
- fix_pmode_dataseg(VCPU_SREG_GS, &vcpu->rmode.gs);
- fix_pmode_dataseg(VCPU_SREG_FS, &vcpu->rmode.fs);
+ fix_pmode_dataseg(VCPU_SREG_ES, &vcpu->arch.rmode.es);
+ fix_pmode_dataseg(VCPU_SREG_DS, &vcpu->arch.rmode.ds);
+ fix_pmode_dataseg(VCPU_SREG_GS, &vcpu->arch.rmode.gs);
+ fix_pmode_dataseg(VCPU_SREG_FS, &vcpu->arch.rmode.fs);
vmcs_write16(GUEST_SS_SELECTOR, 0);
vmcs_write32(GUEST_SS_AR_BYTES, 0x93);
@@ -1168,19 +1168,19 @@ static void enter_rmode(struct kvm_vcpu *vcpu)
{
unsigned long flags;
- vcpu->rmode.active = 1;
+ vcpu->arch.rmode.active = 1;
- vcpu->rmode.tr.base = vmcs_readl(GUEST_TR_BASE);
+ vcpu->arch.rmode.tr.base = vmcs_readl(GUEST_TR_BASE);
vmcs_writel(GUEST_TR_BASE, rmode_tss_base(vcpu->kvm));
- vcpu->rmode.tr.limit = vmcs_read32(GUEST_TR_LIMIT);
+ vcpu->arch.rmode.tr.limit = vmcs_read32(GUEST_TR_LIMIT);
vmcs_write32(GUEST_TR_LIMIT, RMODE_TSS_SIZE - 1);
- vcpu->rmode.tr.ar = vmcs_read32(GUEST_TR_AR_BYTES);
+ vcpu->arch.rmode.tr.ar = vmcs_read32(GUEST_TR_AR_BYTES);
vmcs_write32(GUEST_TR_AR_BYTES, 0x008b);
flags = vmcs_readl(GUEST_RFLAGS);
- vcpu->rmode.save_iopl = (flags & IOPL_MASK) >> IOPL_SHIFT;
+ vcpu->arch.rmode.save_iopl = (flags & IOPL_MASK) >> IOPL_SHIFT;
flags |= IOPL_MASK | X86_EFLAGS_VM;
@@ -1198,10 +1198,10 @@ static void enter_rmode(struct kvm_vcpu *vcpu)
vmcs_writel(GUEST_CS_BASE, 0xf0000);
vmcs_write16(GUEST_CS_SELECTOR, vmcs_readl(GUEST_CS_BASE) >> 4);
- fix_rmode_seg(VCPU_SREG_ES, &vcpu->rmode.es);
- fix_rmode_seg(VCPU_SREG_DS, &vcpu->rmode.ds);
- fix_rmode_seg(VCPU_SREG_GS, &vcpu->rmode.gs);
- fix_rmode_seg(VCPU_SREG_FS, &vcpu->rmode.fs);
+ fix_rmode_seg(VCPU_SREG_ES, &vcpu->arch.rmode.es);
+ fix_rmode_seg(VCPU_SREG_DS, &vcpu->arch.rmode.ds);
+ fix_rmode_seg(VCPU_SREG_GS, &vcpu->arch.rmode.gs);
+ fix_rmode_seg(VCPU_SREG_FS, &vcpu->arch.rmode.fs);
kvm_mmu_reset_context(vcpu);
init_rmode_tss(vcpu->kvm);
@@ -1222,7 +1222,7 @@ static void enter_lmode(struct kvm_vcpu *vcpu)
| AR_TYPE_BUSY_64_TSS);
}
- vcpu->shadow_efer |= EFER_LMA;
+ vcpu->arch.shadow_efer |= EFER_LMA;
find_msr_entry(to_vmx(vcpu), MSR_EFER)->data |= EFER_LMA | EFER_LME;
vmcs_write32(VM_ENTRY_CONTROLS,
@@ -1232,7 +1232,7 @@ static void enter_lmode(struct kvm_vcpu *vcpu)
static void exit_lmode(struct kvm_vcpu *vcpu)
{
- vcpu->shadow_efer &= ~EFER_LMA;
+ vcpu->arch.shadow_efer &= ~EFER_LMA;
vmcs_write32(VM_ENTRY_CONTROLS,
vmcs_read32(VM_ENTRY_CONTROLS)
@@ -1243,22 +1243,22 @@ static void exit_lmode(struct kvm_vcpu *vcpu)
static void vmx_decache_cr4_guest_bits(struct kvm_vcpu *vcpu)
{
- vcpu->cr4 &= KVM_GUEST_CR4_MASK;
- vcpu->cr4 |= vmcs_readl(GUEST_CR4) & ~KVM_GUEST_CR4_MASK;
+ vcpu->arch.cr4 &= KVM_GUEST_CR4_MASK;
+ vcpu->arch.cr4 |= vmcs_readl(GUEST_CR4) & ~KVM_GUEST_CR4_MASK;
}
static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
{
vmx_fpu_deactivate(vcpu);
- if (vcpu->rmode.active && (cr0 & X86_CR0_PE))
+ if (vcpu->arch.rmode.active && (cr0 & X86_CR0_PE))
enter_pmode(vcpu);
- if (!vcpu->rmode.active && !(cr0 & X86_CR0_PE))
+ if (!vcpu->arch.rmode.active && !(cr0 & X86_CR0_PE))
enter_rmode(vcpu);
#ifdef CONFIG_X86_64
- if (vcpu->shadow_efer & EFER_LME) {
+ if (vcpu->arch.shadow_efer & EFER_LME) {
if (!is_paging(vcpu) && (cr0 & X86_CR0_PG))
enter_lmode(vcpu);
if (is_paging(vcpu) && !(cr0 & X86_CR0_PG))
@@ -1269,7 +1269,7 @@ static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
vmcs_writel(CR0_READ_SHADOW, cr0);
vmcs_writel(GUEST_CR0,
(cr0 & ~KVM_GUEST_CR0_MASK) | KVM_VM_CR0_ALWAYS_ON);
- vcpu->cr0 = cr0;
+ vcpu->arch.cr0 = cr0;
if (!(cr0 & X86_CR0_TS) || !(cr0 & X86_CR0_PE))
vmx_fpu_activate(vcpu);
@@ -1278,16 +1278,16 @@ static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
static void vmx_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
{
vmcs_writel(GUEST_CR3, cr3);
- if (vcpu->cr0 & X86_CR0_PE)
+ if (vcpu->arch.cr0 & X86_CR0_PE)
vmx_fpu_deactivate(vcpu);
}
static void vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
{
vmcs_writel(CR4_READ_SHADOW, cr4);
- vmcs_writel(GUEST_CR4, cr4 | (vcpu->rmode.active ?
+ vmcs_writel(GUEST_CR4, cr4 | (vcpu->arch.rmode.active ?
KVM_RMODE_VM_CR4_ALWAYS_ON : KVM_PMODE_VM_CR4_ALWAYS_ON));
- vcpu->cr4 = cr4;
+ vcpu->arch.cr4 = cr4;
}
#ifdef CONFIG_X86_64
@@ -1297,7 +1297,7 @@ static void vmx_set_efer(struct kvm_vcpu *vcpu, u64 efer)
struct vcpu_vmx *vmx = to_vmx(vcpu);
struct kvm_msr_entry *msr = find_msr_entry(vmx, MSR_EFER);
- vcpu->shadow_efer = efer;
+ vcpu->arch.shadow_efer = efer;
if (efer & EFER_LMA) {
vmcs_write32(VM_ENTRY_CONTROLS,
vmcs_read32(VM_ENTRY_CONTROLS) |
@@ -1374,17 +1374,17 @@ static void vmx_set_segment(struct kvm_vcpu *vcpu,
struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg];
u32 ar;
- if (vcpu->rmode.active && seg == VCPU_SREG_TR) {
- vcpu->rmode.tr.selector = var->selector;
- vcpu->rmode.tr.base = var->base;
- vcpu->rmode.tr.limit = var->limit;
- vcpu->rmode.tr.ar = vmx_segment_access_rights(var);
+ if (vcpu->arch.rmode.active && seg == VCPU_SREG_TR) {
+ vcpu->arch.rmode.tr.selector = var->selector;
+ vcpu->arch.rmode.tr.base = var->base;
+ vcpu->arch.rmode.tr.limit = var->limit;
+ vcpu->arch.rmode.tr.ar = vmx_segment_access_rights(var);
return;
}
vmcs_writel(sf->base, var->base);
vmcs_write32(sf->limit, var->limit);
vmcs_write16(sf->selector, var->selector);
- if (vcpu->rmode.active && var->s) {
+ if (vcpu->arch.rmode.active && var->s) {
/*
* Hack real-mode segments into vm86 compatibility.
*/
@@ -1613,9 +1613,9 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu)
goto out;
}
- vmx->vcpu.rmode.active = 0;
+ vmx->vcpu.arch.rmode.active = 0;
- vmx->vcpu.regs[VCPU_REGS_RDX] = get_rdx_init_val();
+ vmx->vcpu.arch.regs[VCPU_REGS_RDX] = get_rdx_init_val();
set_cr8(&vmx->vcpu, 0);
msr = 0xfee00000 | MSR_IA32_APICBASE_ENABLE;
if (vmx->vcpu.vcpu_id == 0)
@@ -1632,8 +1632,8 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu)
vmcs_write16(GUEST_CS_SELECTOR, 0xf000);
vmcs_writel(GUEST_CS_BASE, 0x000f0000);
} else {
- vmcs_write16(GUEST_CS_SELECTOR, vmx->vcpu.sipi_vector << 8);
- vmcs_writel(GUEST_CS_BASE, vmx->vcpu.sipi_vector << 12);
+ vmcs_write16(GUEST_CS_SELECTOR, vmx->vcpu.arch.sipi_vector << 8);
+ vmcs_writel(GUEST_CS_BASE, vmx->vcpu.arch.sipi_vector << 12);
}
vmcs_write32(GUEST_CS_LIMIT, 0xffff);
vmcs_write32(GUEST_CS_AR_BYTES, 0x9b);
@@ -1691,7 +1691,7 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu)
vmcs_write64(VIRTUAL_APIC_PAGE_ADDR, 0);
if (vm_need_tpr_shadow(vmx->vcpu.kvm))
vmcs_write64(VIRTUAL_APIC_PAGE_ADDR,
- page_to_phys(vmx->vcpu.apic->regs_page));
+ page_to_phys(vmx->vcpu.arch.apic->regs_page));
vmcs_write32(TPR_THRESHOLD, 0);
}
@@ -1699,8 +1699,8 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu)
vmcs_write64(APIC_ACCESS_ADDR,
page_to_phys(vmx->vcpu.kvm->apic_access_page));
- vmx->vcpu.cr0 = 0x60000010;
- vmx_set_cr0(&vmx->vcpu, vmx->vcpu.cr0); /* enter rmode */
+ vmx->vcpu.arch.cr0 = 0x60000010;
+ vmx_set_cr0(&vmx->vcpu, vmx->vcpu.arch.cr0); /* enter rmode */
vmx_set_cr4(&vmx->vcpu, 0);
#ifdef CONFIG_X86_64
vmx_set_efer(&vmx->vcpu, 0);
@@ -1718,7 +1718,7 @@ static void vmx_inject_irq(struct kvm_vcpu *vcpu, int irq)
{
struct vcpu_vmx *vmx = to_vmx(vcpu);
- if (vcpu->rmode.active) {
+ if (vcpu->arch.rmode.active) {
vmx->rmode.irq.pending = true;
vmx->rmode.irq.vector = irq;
vmx->rmode.irq.rip = vmcs_readl(GUEST_RIP);
@@ -1734,13 +1734,13 @@ static void vmx_inject_irq(struct kvm_vcpu *vcpu, int irq)
static void kvm_do_inject_irq(struct kvm_vcpu *vcpu)
{
- int word_index = __ffs(vcpu->irq_summary);
- int bit_index = __ffs(vcpu->irq_pending[word_index]);
+ int word_index = __ffs(vcpu->arch.irq_summary);
+ int bit_index = __ffs(vcpu->arch.irq_pending[word_index]);
int irq = word_index * BITS_PER_LONG + bit_index;
- clear_bit(bit_index, &vcpu->irq_pending[word_index]);
- if (!vcpu->irq_pending[word_index])
- clear_bit(word_index, &vcpu->irq_summary);
+ clear_bit(bit_index, &vcpu->arch.irq_pending[word_index]);
+ if (!vcpu->arch.irq_pending[word_index])
+ clear_bit(word_index, &vcpu->arch.irq_summary);
vmx_inject_irq(vcpu, irq);
}
@@ -1750,12 +1750,12 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu,
{
u32 cpu_based_vm_exec_control;
- vcpu->interrupt_window_open =
+ vcpu->arch.interrupt_window_open =
((vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF) &&
(vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0);
- if (vcpu->interrupt_window_open &&
- vcpu->irq_summary &&
+ if (vcpu->arch.interrupt_window_open &&
+ vcpu->arch.irq_summary &&
!(vmcs_read32(VM_ENTRY_INTR_INFO_FIELD) & INTR_INFO_VALID_MASK))
/*
* If interrupts enabled, and not blocked by sti or mov ss. Good.
@@ -1763,8 +1763,8 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu,
kvm_do_inject_irq(vcpu);
cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL);
- if (!vcpu->interrupt_window_open &&
- (vcpu->irq_summary || kvm_run->request_interrupt_window))
+ if (!vcpu->arch.interrupt_window_open &&
+ (vcpu->arch.irq_summary || kvm_run->request_interrupt_window))
/*
* Interrupts blocked. Wait for unblock.
*/
@@ -1812,7 +1812,7 @@ static void kvm_guest_debug_pre(struct kvm_vcpu *vcpu)
static int handle_rmode_exception(struct kvm_vcpu *vcpu,
int vec, u32 err_code)
{
- if (!vcpu->rmode.active)
+ if (!vcpu->arch.rmode.active)
return 0;
/*
@@ -1843,8 +1843,8 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
if (!irqchip_in_kernel(vcpu->kvm) && is_external_interrupt(vect_info)) {
int irq = vect_info & VECTORING_INFO_VECTOR_MASK;
- set_bit(irq, vcpu->irq_pending);
- set_bit(irq / BITS_PER_LONG, &vcpu->irq_summary);
+ set_bit(irq, vcpu->arch.irq_pending);
+ set_bit(irq / BITS_PER_LONG, &vcpu->arch.irq_summary);
}
if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200) /* nmi */
@@ -1871,11 +1871,11 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
return kvm_mmu_page_fault(vcpu, cr2, error_code);
}
- if (vcpu->rmode.active &&
+ if (vcpu->arch.rmode.active &&
handle_rmode_exception(vcpu, intr_info & INTR_INFO_VECTOR_MASK,
error_code)) {
- if (vcpu->halt_request) {
- vcpu->halt_request = 0;
+ if (vcpu->arch.halt_request) {
+ vcpu->arch.halt_request = 0;
return kvm_emulate_halt(vcpu);
}
return 1;
@@ -1956,22 +1956,22 @@ static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
switch (cr) {
case 0:
vcpu_load_rsp_rip(vcpu);
- set_cr0(vcpu, vcpu->regs[reg]);
+ set_cr0(vcpu, vcpu->arch.regs[reg]);
skip_emulated_instruction(vcpu);
return 1;
case 3:
vcpu_load_rsp_rip(vcpu);
- set_cr3(vcpu, vcpu->regs[reg]);
+ set_cr3(vcpu, vcpu->arch.regs[reg]);
skip_emulated_instruction(vcpu);
return 1;
case 4:
vcpu_load_rsp_rip(vcpu);
- set_cr4(vcpu, vcpu->regs[reg]);
+ set_cr4(vcpu, vcpu->arch.regs[reg]);
skip_emulated_instruction(vcpu);
return 1;
case 8:
vcpu_load_rsp_rip(vcpu);
- set_cr8(vcpu, vcpu->regs[reg]);
+ set_cr8(vcpu, vcpu->arch.regs[reg]);
skip_emulated_instruction(vcpu);
if (irqchip_in_kernel(vcpu->kvm))
return 1;
@@ -1982,8 +1982,8 @@ static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
case 2: /* clts */
vcpu_load_rsp_rip(vcpu);
vmx_fpu_deactivate(vcpu);
- vcpu->cr0 &= ~X86_CR0_TS;
- vmcs_writel(CR0_READ_SHADOW, vcpu->cr0);
+ vcpu->arch.cr0 &= ~X86_CR0_TS;
+ vmcs_writel(CR0_READ_SHADOW, vcpu->arch.cr0);
vmx_fpu_activate(vcpu);
skip_emulated_instruction(vcpu);
return 1;
@@ -1991,13 +1991,13 @@ static int handle_cr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
switch (cr) {
case 3:
vcpu_load_rsp_rip(vcpu);
- vcpu->regs[reg] = vcpu->cr3;
+ vcpu->arch.regs[reg] = vcpu->arch.cr3;
vcpu_put_rsp_rip(vcpu);
skip_emulated_instruction(vcpu);
return 1;
case 8:
vcpu_load_rsp_rip(vcpu);
- vcpu->regs[reg] = get_cr8(vcpu);
+ vcpu->arch.regs[reg] = get_cr8(vcpu);
vcpu_put_rsp_rip(vcpu);
skip_emulated_instruction(vcpu);
return 1;
@@ -2043,7 +2043,7 @@ static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
default:
val = 0;
}
- vcpu->regs[reg] = val;
+ vcpu->arch.regs[reg] = val;
} else {
/* mov to dr */
}
@@ -2060,7 +2060,7 @@ static int handle_cpuid(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
static int handle_rdmsr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
- u32 ecx = vcpu->regs[VCPU_REGS_RCX];
+ u32 ecx = vcpu->arch.regs[VCPU_REGS_RCX];
u64 data;
if (vmx_get_msr(vcpu, ecx, &data)) {
@@ -2069,17 +2069,17 @@ static int handle_rdmsr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
}
/* FIXME: handling of bits 32:63 of rax, rdx */
- vcpu->regs[VCPU_REGS_RAX] = data & -1u;
- vcpu->regs[VCPU_REGS_RDX] = (data >> 32) & -1u;
+ vcpu->arch.regs[VCPU_REGS_RAX] = data & -1u;
+ vcpu->arch.regs[VCPU_REGS_RDX] = (data >> 32) & -1u;
skip_emulated_instruction(vcpu);
return 1;
}
static int handle_wrmsr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
- u32 ecx = vcpu->regs[VCPU_REGS_RCX];
- u64 data = (vcpu->regs[VCPU_REGS_RAX] & -1u)
- | ((u64)(vcpu->regs[VCPU_REGS_RDX] & -1u) << 32);
+ u32 ecx = vcpu->arch.regs[VCPU_REGS_RCX];
+ u64 data = (vcpu->arch.regs[VCPU_REGS_RAX] & -1u)
+ | ((u64)(vcpu->arch.regs[VCPU_REGS_RDX] & -1u) << 32);
if (vmx_set_msr(vcpu, ecx, data) != 0) {
kvm_inject_gp(vcpu, 0);
@@ -2110,7 +2110,7 @@ static int handle_interrupt_window(struct kvm_vcpu *vcpu,
* possible
*/
if (kvm_run->request_interrupt_window &&
- !vcpu->irq_summary) {
+ !vcpu->arch.irq_summary) {
kvm_run->exit_reason = KVM_EXIT_IRQ_WINDOW_OPEN;
++vcpu->stat.irq_window_exits;
return 0;
@@ -2270,7 +2270,7 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu)
if (unlikely(idtv_info_field & INTR_INFO_VALID_MASK)) {
if ((idtv_info_field & VECTORING_INFO_TYPE_MASK)
== INTR_TYPE_EXT_INTR
- && vcpu->rmode.active) {
+ && vcpu->arch.rmode.active) {
u8 vect = idtv_info_field & VECTORING_INFO_VECTOR_MASK;
vmx_inject_irq(vcpu, vect);
@@ -2424,24 +2424,24 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
: : "c"(vmx), "d"((unsigned long)HOST_RSP),
[launched]"i"(offsetof(struct vcpu_vmx, launched)),
[fail]"i"(offsetof(struct vcpu_vmx, fail)),
- [rax]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_RAX])),
- [rbx]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_RBX])),
- [rcx]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_RCX])),
- [rdx]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_RDX])),
- [rsi]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_RSI])),
- [rdi]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_RDI])),
- [rbp]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_RBP])),
+ [rax]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RAX])),
+ [rbx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RBX])),
+ [rcx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RCX])),
+ [rdx]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RDX])),
+ [rsi]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RSI])),
+ [rdi]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RDI])),
+ [rbp]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_RBP])),
#ifdef CONFIG_X86_64
- [r8]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_R8])),
- [r9]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_R9])),
- [r10]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_R10])),
- [r11]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_R11])),
- [r12]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_R12])),
- [r13]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_R13])),
- [r14]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_R14])),
- [r15]"i"(offsetof(struct vcpu_vmx, vcpu.regs[VCPU_REGS_R15])),
+ [r8]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R8])),
+ [r9]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R9])),
+ [r10]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R10])),
+ [r11]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R11])),
+ [r12]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R12])),
+ [r13]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R13])),
+ [r14]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R14])),
+ [r15]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R15])),
#endif
- [cr2]"i"(offsetof(struct vcpu_vmx, vcpu.cr2))
+ [cr2]"i"(offsetof(struct vcpu_vmx, vcpu.arch.cr2))
: "cc", "memory"
#ifdef CONFIG_X86_64
, "rbx", "rdi", "rsi"
@@ -2455,7 +2455,7 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
if (vmx->rmode.irq.pending)
fixup_rmode_irq(vmx);
- vcpu->interrupt_window_open =
+ vcpu->arch.interrupt_window_open =
(vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0;
asm("mov %0, %%ds; mov %0, %%es" : : "r"(__USER_DS));
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index 3b79684..5a2f33a 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -113,9 +113,9 @@ EXPORT_SYMBOL_GPL(segment_base);
u64 kvm_get_apic_base(struct kvm_vcpu *vcpu)
{
if (irqchip_in_kernel(vcpu->kvm))
- return vcpu->apic_base;
+ return vcpu->arch.apic_base;
else
- return vcpu->apic_base;
+ return vcpu->arch.apic_base;
}
EXPORT_SYMBOL_GPL(kvm_get_apic_base);
@@ -125,16 +125,16 @@ void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data)
if (irqchip_in_kernel(vcpu->kvm))
kvm_lapic_set_base(vcpu, data);
else
- vcpu->apic_base = data;
+ vcpu->arch.apic_base = data;
}
EXPORT_SYMBOL_GPL(kvm_set_apic_base);
void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr)
{
- WARN_ON(vcpu->exception.pending);
- vcpu->exception.pending = true;
- vcpu->exception.has_error_code = false;
- vcpu->exception.nr = nr;
+ WARN_ON(vcpu->arch.exception.pending);
+ vcpu->arch.exception.pending = true;
+ vcpu->arch.exception.has_error_code = false;
+ vcpu->arch.exception.nr = nr;
}
EXPORT_SYMBOL_GPL(kvm_queue_exception);
@@ -142,32 +142,32 @@ void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long addr,
u32 error_code)
{
++vcpu->stat.pf_guest;
- if (vcpu->exception.pending && vcpu->exception.nr == PF_VECTOR) {
+ if (vcpu->arch.exception.pending && vcpu->arch.exception.nr == PF_VECTOR) {
printk(KERN_DEBUG "kvm: inject_page_fault:"
" double fault 0x%lx\n", addr);
- vcpu->exception.nr = DF_VECTOR;
- vcpu->exception.error_code = 0;
+ vcpu->arch.exception.nr = DF_VECTOR;
+ vcpu->arch.exception.error_code = 0;
return;
}
- vcpu->cr2 = addr;
+ vcpu->arch.cr2 = addr;
kvm_queue_exception_e(vcpu, PF_VECTOR, error_code);
}
void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code)
{
- WARN_ON(vcpu->exception.pending);
- vcpu->exception.pending = true;
- vcpu->exception.has_error_code = true;
- vcpu->exception.nr = nr;
- vcpu->exception.error_code = error_code;
+ WARN_ON(vcpu->arch.exception.pending);
+ vcpu->arch.exception.pending = true;
+ vcpu->arch.exception.has_error_code = true;
+ vcpu->arch.exception.nr = nr;
+ vcpu->arch.exception.error_code = error_code;
}
EXPORT_SYMBOL_GPL(kvm_queue_exception_e);
static void __queue_exception(struct kvm_vcpu *vcpu)
{
- kvm_x86_ops->queue_exception(vcpu, vcpu->exception.nr,
- vcpu->exception.has_error_code,
- vcpu->exception.error_code);
+ kvm_x86_ops->queue_exception(vcpu, vcpu->arch.exception.nr,
+ vcpu->arch.exception.has_error_code,
+ vcpu->arch.exception.error_code);
}
/*
@@ -179,7 +179,7 @@ int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3)
unsigned offset = ((cr3 & (PAGE_SIZE-1)) >> 5) << 2;
int i;
int ret;
- u64 pdpte[ARRAY_SIZE(vcpu->pdptrs)];
+ u64 pdpte[ARRAY_SIZE(vcpu->arch.pdptrs)];
mutex_lock(&vcpu->kvm->lock);
ret = kvm_read_guest_page(vcpu->kvm, pdpt_gfn, pdpte,
@@ -196,7 +196,7 @@ int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3)
}
ret = 1;
- memcpy(vcpu->pdptrs, pdpte, sizeof(vcpu->pdptrs));
+ memcpy(vcpu->arch.pdptrs, pdpte, sizeof(vcpu->arch.pdptrs));
out:
mutex_unlock(&vcpu->kvm->lock);
@@ -205,7 +205,7 @@ out:
static bool pdptrs_changed(struct kvm_vcpu *vcpu)
{
- u64 pdpte[ARRAY_SIZE(vcpu->pdptrs)];
+ u64 pdpte[ARRAY_SIZE(vcpu->arch.pdptrs)];
bool changed = true;
int r;
@@ -213,10 +213,10 @@ static bool pdptrs_changed(struct kvm_vcpu *vcpu)
return false;
mutex_lock(&vcpu->kvm->lock);
- r = kvm_read_guest(vcpu->kvm, vcpu->cr3 & ~31u, pdpte, sizeof(pdpte));
+ r = kvm_read_guest(vcpu->kvm, vcpu->arch.cr3 & ~31u, pdpte, sizeof(pdpte));
if (r < 0)
goto out;
- changed = memcmp(pdpte, vcpu->pdptrs, sizeof(pdpte)) != 0;
+ changed = memcmp(pdpte, vcpu->arch.pdptrs, sizeof(pdpte)) != 0;
out:
mutex_unlock(&vcpu->kvm->lock);
@@ -227,7 +227,7 @@ void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
{
if (cr0 & CR0_RESERVED_BITS) {
printk(KERN_DEBUG "set_cr0: 0x%lx #GP, reserved bits 0x%lx\n",
- cr0, vcpu->cr0);
+ cr0, vcpu->arch.cr0);
kvm_inject_gp(vcpu, 0);
return;
}
@@ -247,7 +247,7 @@ void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
if (!is_paging(vcpu) && (cr0 & X86_CR0_PG)) {
#ifdef CONFIG_X86_64
- if ((vcpu->shadow_efer & EFER_LME)) {
+ if ((vcpu->arch.shadow_efer & EFER_LME)) {
int cs_db, cs_l;
if (!is_pae(vcpu)) {
@@ -266,7 +266,7 @@ void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
}
} else
#endif
- if (is_pae(vcpu) && !load_pdptrs(vcpu, vcpu->cr3)) {
+ if (is_pae(vcpu) && !load_pdptrs(vcpu, vcpu->arch.cr3)) {
printk(KERN_DEBUG "set_cr0: #GP, pdptrs "
"reserved bits\n");
kvm_inject_gp(vcpu, 0);
@@ -276,7 +276,7 @@ void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
}
kvm_x86_ops->set_cr0(vcpu, cr0);
- vcpu->cr0 = cr0;
+ vcpu->arch.cr0 = cr0;
mutex_lock(&vcpu->kvm->lock);
kvm_mmu_reset_context(vcpu);
@@ -287,7 +287,7 @@ EXPORT_SYMBOL_GPL(set_cr0);
void lmsw(struct kvm_vcpu *vcpu, unsigned long msw)
{
- set_cr0(vcpu, (vcpu->cr0 & ~0x0ful) | (msw & 0x0f));
+ set_cr0(vcpu, (vcpu->arch.cr0 & ~0x0ful) | (msw & 0x0f));
}
EXPORT_SYMBOL_GPL(lmsw);
@@ -307,7 +307,7 @@ void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
return;
}
} else if (is_paging(vcpu) && !is_pae(vcpu) && (cr4 & X86_CR4_PAE)
- && !load_pdptrs(vcpu, vcpu->cr3)) {
+ && !load_pdptrs(vcpu, vcpu->arch.cr3)) {
printk(KERN_DEBUG "set_cr4: #GP, pdptrs reserved bits\n");
kvm_inject_gp(vcpu, 0);
return;
@@ -319,7 +319,7 @@ void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
return;
}
kvm_x86_ops->set_cr4(vcpu, cr4);
- vcpu->cr4 = cr4;
+ vcpu->arch.cr4 = cr4;
mutex_lock(&vcpu->kvm->lock);
kvm_mmu_reset_context(vcpu);
mutex_unlock(&vcpu->kvm->lock);
@@ -328,7 +328,7 @@ EXPORT_SYMBOL_GPL(set_cr4);
void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
{
- if (cr3 == vcpu->cr3 && !pdptrs_changed(vcpu)) {
+ if (cr3 == vcpu->arch.cr3 && !pdptrs_changed(vcpu)) {
kvm_mmu_flush_tlb(vcpu);
return;
}
@@ -373,8 +373,8 @@ void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
if (unlikely(!gfn_to_memslot(vcpu->kvm, cr3 >> PAGE_SHIFT)))
kvm_inject_gp(vcpu, 0);
else {
- vcpu->cr3 = cr3;
- vcpu->mmu.new_cr3(vcpu);
+ vcpu->arch.cr3 = cr3;
+ vcpu->arch.mmu.new_cr3(vcpu);
}
mutex_unlock(&vcpu->kvm->lock);
}
@@ -390,7 +390,7 @@ void set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8)
if (irqchip_in_kernel(vcpu->kvm))
kvm_lapic_set_tpr(vcpu, cr8);
else
- vcpu->cr8 = cr8;
+ vcpu->arch.cr8 = cr8;
}
EXPORT_SYMBOL_GPL(set_cr8);
@@ -399,7 +399,7 @@ unsigned long get_cr8(struct kvm_vcpu *vcpu)
if (irqchip_in_kernel(vcpu->kvm))
return kvm_lapic_get_cr8(vcpu);
else
- return vcpu->cr8;
+ return vcpu->arch.cr8;
}
EXPORT_SYMBOL_GPL(get_cr8);
@@ -437,7 +437,7 @@ static void set_efer(struct kvm_vcpu *vcpu, u64 efer)
}
if (is_paging(vcpu)
- && (vcpu->shadow_efer & EFER_LME) != (efer & EFER_LME)) {
+ && (vcpu->arch.shadow_efer & EFER_LME) != (efer & EFER_LME)) {
printk(KERN_DEBUG "set_efer: #GP, change LME while paging\n");
kvm_inject_gp(vcpu, 0);
return;
@@ -446,9 +446,9 @@ static void set_efer(struct kvm_vcpu *vcpu, u64 efer)
kvm_x86_ops->set_efer(vcpu, efer);
efer &= ~EFER_LMA;
- efer |= vcpu->shadow_efer & EFER_LMA;
+ efer |= vcpu->arch.shadow_efer & EFER_LMA;
- vcpu->shadow_efer = efer;
+ vcpu->arch.shadow_efer = efer;
}
#endif
@@ -496,7 +496,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
kvm_set_apic_base(vcpu, data);
break;
case MSR_IA32_MISC_ENABLE:
- vcpu->ia32_misc_enable_msr = data;
+ vcpu->arch.ia32_misc_enable_msr = data;
break;
default:
pr_unimpl(vcpu, "unhandled wrmsr: 0x%x\n", msr);
@@ -550,11 +550,11 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
data = kvm_get_apic_base(vcpu);
break;
case MSR_IA32_MISC_ENABLE:
- data = vcpu->ia32_misc_enable_msr;
+ data = vcpu->arch.ia32_misc_enable_msr;
break;
#ifdef CONFIG_X86_64
case MSR_EFER:
- data = vcpu->shadow_efer;
+ data = vcpu->arch.shadow_efer;
break;
#endif
default:
@@ -760,8 +760,8 @@ static void cpuid_fix_nx_cap(struct kvm_vcpu *vcpu)
struct kvm_cpuid_entry2 *e, *entry;
entry = NULL;
- for (i = 0; i < vcpu->cpuid_nent; ++i) {
- e = &vcpu->cpuid_entries[i];
+ for (i = 0; i < vcpu->arch.cpuid_nent; ++i) {
+ e = &vcpu->arch.cpuid_entries[i];
if (e->function == 0x80000001) {
entry = e;
break;
@@ -793,18 +793,18 @@ static int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu,
cpuid->nent * sizeof(struct kvm_cpuid_entry)))
goto out_free;
for (i = 0; i < cpuid->nent; i++) {
- vcpu->cpuid_entries[i].function = cpuid_entries[i].function;
- vcpu->cpuid_entries[i].eax = cpuid_entries[i].eax;
- vcpu->cpuid_entries[i].ebx = cpuid_entries[i].ebx;
- vcpu->cpuid_entries[i].ecx = cpuid_entries[i].ecx;
- vcpu->cpuid_entries[i].edx = cpuid_entries[i].edx;
- vcpu->cpuid_entries[i].index = 0;
- vcpu->cpuid_entries[i].flags = 0;
- vcpu->cpuid_entries[i].padding[0] = 0;
- vcpu->cpuid_entries[i].padding[1] = 0;
- vcpu->cpuid_entries[i].padding[2] = 0;
- }
- vcpu->cpuid_nent = cpuid->nent;
+ vcpu->arch.cpuid_entries[i].function = cpuid_entries[i].function;
+ vcpu->arch.cpuid_entries[i].eax = cpuid_entries[i].eax;
+ vcpu->arch.cpuid_entries[i].ebx = cpuid_entries[i].ebx;
+ vcpu->arch.cpuid_entries[i].ecx = cpuid_entries[i].ecx;
+ vcpu->arch.cpuid_entries[i].edx = cpuid_entries[i].edx;
+ vcpu->arch.cpuid_entries[i].index = 0;
+ vcpu->arch.cpuid_entries[i].flags = 0;
+ vcpu->arch.cpuid_entries[i].padding[0] = 0;
+ vcpu->arch.cpuid_entries[i].padding[1] = 0;
+ vcpu->arch.cpuid_entries[i].padding[2] = 0;
+ }
+ vcpu->arch.cpuid_nent = cpuid->nent;
cpuid_fix_nx_cap(vcpu);
r = 0;
@@ -824,10 +824,10 @@ static int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu,
if (cpuid->nent > KVM_MAX_CPUID_ENTRIES)
goto out;
r = -EFAULT;
- if (copy_from_user(&vcpu->cpuid_entries, entries,
+ if (copy_from_user(&vcpu->arch.cpuid_entries, entries,
cpuid->nent * sizeof(struct kvm_cpuid_entry2)))
goto out;
- vcpu->cpuid_nent = cpuid->nent;
+ vcpu->arch.cpuid_nent = cpuid->nent;
return 0;
out:
@@ -841,16 +841,16 @@ static int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu,
int r;
r = -E2BIG;
- if (cpuid->nent < vcpu->cpuid_nent)
+ if (cpuid->nent < vcpu->arch.cpuid_nent)
goto out;
r = -EFAULT;
- if (copy_to_user(entries, &vcpu->cpuid_entries,
- vcpu->cpuid_nent * sizeof(struct kvm_cpuid_entry2)))
+ if (copy_to_user(entries, &vcpu->arch.cpuid_entries,
+ vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2)))
goto out;
return 0;
out:
- cpuid->nent = vcpu->cpuid_nent;
+ cpuid->nent = vcpu->arch.cpuid_nent;
return r;
}
@@ -1021,7 +1021,7 @@ static int kvm_vcpu_ioctl_get_lapic(struct kvm_vcpu *vcpu,
struct kvm_lapic_state *s)
{
vcpu_load(vcpu);
- memcpy(s->regs, vcpu->apic->regs, sizeof *s);
+ memcpy(s->regs, vcpu->arch.apic->regs, sizeof *s);
vcpu_put(vcpu);
return 0;
@@ -1031,7 +1031,7 @@ static int kvm_vcpu_ioctl_set_lapic(struct kvm_vcpu *vcpu,
struct kvm_lapic_state *s)
{
vcpu_load(vcpu);
- memcpy(vcpu->apic->regs, s->regs, sizeof *s);
+ memcpy(vcpu->arch.apic->regs, s->regs, sizeof *s);
kvm_apic_post_state_restore(vcpu);
vcpu_put(vcpu);
@@ -1047,8 +1047,8 @@ static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu,
return -ENXIO;
vcpu_load(vcpu);
- set_bit(irq->irq, vcpu->irq_pending);
- set_bit(irq->irq / BITS_PER_LONG, &vcpu->irq_summary);
+ set_bit(irq->irq, vcpu->arch.irq_pending);
+ set_bit(irq->irq / BITS_PER_LONG, &vcpu->arch.irq_summary);
vcpu_put(vcpu);
@@ -1499,8 +1499,8 @@ static struct kvm_io_device *vcpu_find_pervcpu_dev(struct kvm_vcpu *vcpu,
{
struct kvm_io_device *dev;
- if (vcpu->apic) {
- dev = &vcpu->apic->dev;
+ if (vcpu->arch.apic) {
+ dev = &vcpu->arch.apic->dev;
if (dev->in_range(dev, addr))
return dev;
}
@@ -1527,7 +1527,7 @@ int emulator_read_std(unsigned long addr,
void *data = val;
while (bytes) {
- gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr);
+ gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
unsigned offset = addr & (PAGE_SIZE-1);
unsigned tocopy = min(bytes, (unsigned)PAGE_SIZE - offset);
int ret;
@@ -1561,7 +1561,7 @@ static int emulator_read_emulated(unsigned long addr,
return X86EMUL_CONTINUE;
}
- gpa = vcpu->mmu.gva_to_gpa(vcpu, addr);
+ gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
/* For APIC access vmexit */
if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
@@ -1609,7 +1609,7 @@ static int emulator_write_emulated_onepage(unsigned long addr,
struct kvm_vcpu *vcpu)
{
struct kvm_io_device *mmio_dev;
- gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr);
+ gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
if (gpa == UNMAPPED_GVA) {
kvm_inject_page_fault(vcpu, addr, 2);
@@ -1678,7 +1678,7 @@ static int emulator_cmpxchg_emulated(unsigned long addr,
#ifndef CONFIG_X86_64
/* guests cmpxchg8b have to be emulated atomically */
if (bytes == 8) {
- gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr);
+ gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
struct page *page;
char *addr;
u64 val;
@@ -1715,7 +1715,7 @@ int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address)
int emulate_clts(struct kvm_vcpu *vcpu)
{
- kvm_x86_ops->set_cr0(vcpu, vcpu->cr0 & ~X86_CR0_TS);
+ kvm_x86_ops->set_cr0(vcpu, vcpu->arch.cr0 & ~X86_CR0_TS);
return X86EMUL_CONTINUE;
}
@@ -1750,7 +1750,7 @@ void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context)
{
static int reported;
u8 opcodes[4];
- unsigned long rip = vcpu->rip;
+ unsigned long rip = vcpu->arch.rip;
unsigned long rip_linear;
rip_linear = rip + get_segment_base(vcpu, VCPU_SREG_CS);
@@ -1781,46 +1781,46 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
{
int r;
- vcpu->mmio_fault_cr2 = cr2;
+ vcpu->arch.mmio_fault_cr2 = cr2;
kvm_x86_ops->cache_regs(vcpu);
vcpu->mmio_is_write = 0;
- vcpu->pio.string = 0;
+ vcpu->arch.pio.string = 0;
if (!no_decode) {
int cs_db, cs_l;
kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);
- vcpu->emulate_ctxt.vcpu = vcpu;
- vcpu->emulate_ctxt.eflags = kvm_x86_ops->get_rflags(vcpu);
- vcpu->emulate_ctxt.mode =
- (vcpu->emulate_ctxt.eflags & X86_EFLAGS_VM)
+ vcpu->arch.emulate_ctxt.vcpu = vcpu;
+ vcpu->arch.emulate_ctxt.eflags = kvm_x86_ops->get_rflags(vcpu);
+ vcpu->arch.emulate_ctxt.mode =
+ (vcpu->arch.emulate_ctxt.eflags & X86_EFLAGS_VM)
? X86EMUL_MODE_REAL : cs_l
? X86EMUL_MODE_PROT64 : cs_db
? X86EMUL_MODE_PROT32 : X86EMUL_MODE_PROT16;
- if (vcpu->emulate_ctxt.mode == X86EMUL_MODE_PROT64) {
- vcpu->emulate_ctxt.cs_base = 0;
- vcpu->emulate_ctxt.ds_base = 0;
- vcpu->emulate_ctxt.es_base = 0;
- vcpu->emulate_ctxt.ss_base = 0;
+ if (vcpu->arch.emulate_ctxt.mode == X86EMUL_MODE_PROT64) {
+ vcpu->arch.emulate_ctxt.cs_base = 0;
+ vcpu->arch.emulate_ctxt.ds_base = 0;
+ vcpu->arch.emulate_ctxt.es_base = 0;
+ vcpu->arch.emulate_ctxt.ss_base = 0;
} else {
- vcpu->emulate_ctxt.cs_base =
+ vcpu->arch.emulate_ctxt.cs_base =
get_segment_base(vcpu, VCPU_SREG_CS);
- vcpu->emulate_ctxt.ds_base =
+ vcpu->arch.emulate_ctxt.ds_base =
get_segment_base(vcpu, VCPU_SREG_DS);
- vcpu->emulate_ctxt.es_base =
+ vcpu->arch.emulate_ctxt.es_base =
get_segment_base(vcpu, VCPU_SREG_ES);
- vcpu->emulate_ctxt.ss_base =
+ vcpu->arch.emulate_ctxt.ss_base =
get_segment_base(vcpu, VCPU_SREG_SS);
}
- vcpu->emulate_ctxt.gs_base =
+ vcpu->arch.emulate_ctxt.gs_base =
get_segment_base(vcpu, VCPU_SREG_GS);
- vcpu->emulate_ctxt.fs_base =
+ vcpu->arch.emulate_ctxt.fs_base =
get_segment_base(vcpu, VCPU_SREG_FS);
- r = x86_decode_insn(&vcpu->emulate_ctxt, &emulate_ops);
+ r = x86_decode_insn(&vcpu->arch.emulate_ctxt, &emulate_ops);
++vcpu->stat.insn_emulation;
if (r) {
++vcpu->stat.insn_emulation_fail;
@@ -1830,9 +1830,9 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
}
}
- r = x86_emulate_insn(&vcpu->emulate_ctxt, &emulate_ops);
+ r = x86_emulate_insn(&vcpu->arch.emulate_ctxt, &emulate_ops);
- if (vcpu->pio.string)
+ if (vcpu->arch.pio.string)
return EMULATE_DO_MMIO;
if ((r || vcpu->mmio_is_write) && run) {
@@ -1854,7 +1854,7 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
}
kvm_x86_ops->decache_regs(vcpu);
- kvm_x86_ops->set_rflags(vcpu, vcpu->emulate_ctxt.eflags);
+ kvm_x86_ops->set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags);
if (vcpu->mmio_is_write) {
vcpu->mmio_needed = 0;
@@ -1869,33 +1869,33 @@ static void free_pio_guest_pages(struct kvm_vcpu *vcpu)
{
int i;
- for (i = 0; i < ARRAY_SIZE(vcpu->pio.guest_pages); ++i)
- if (vcpu->pio.guest_pages[i]) {
- kvm_release_page_dirty(vcpu->pio.guest_pages[i]);
- vcpu->pio.guest_pages[i] = NULL;
+ for (i = 0; i < ARRAY_SIZE(vcpu->arch.pio.guest_pages); ++i)
+ if (vcpu->arch.pio.guest_pages[i]) {
+ kvm_release_page_dirty(vcpu->arch.pio.guest_pages[i]);
+ vcpu->arch.pio.guest_pages[i] = NULL;
}
}
static int pio_copy_data(struct kvm_vcpu *vcpu)
{
- void *p = vcpu->pio_data;
+ void *p = vcpu->arch.pio_data;
void *q;
unsigned bytes;
- int nr_pages = vcpu->pio.guest_pages[1] ? 2 : 1;
+ int nr_pages = vcpu->arch.pio.guest_pages[1] ? 2 : 1;
- q = vmap(vcpu->pio.guest_pages, nr_pages, VM_READ|VM_WRITE,
+ q = vmap(vcpu->arch.pio.guest_pages, nr_pages, VM_READ|VM_WRITE,
PAGE_KERNEL);
if (!q) {
free_pio_guest_pages(vcpu);
return -ENOMEM;
}
- q += vcpu->pio.guest_page_offset;
- bytes = vcpu->pio.size * vcpu->pio.cur_count;
- if (vcpu->pio.in)
+ q += vcpu->arch.pio.guest_page_offset;
+ bytes = vcpu->arch.pio.size * vcpu->arch.pio.cur_count;
+ if (vcpu->arch.pio.in)
memcpy(q, p, bytes);
else
memcpy(p, q, bytes);
- q -= vcpu->pio.guest_page_offset;
+ q -= vcpu->arch.pio.guest_page_offset;
vunmap(q);
free_pio_guest_pages(vcpu);
return 0;
@@ -1903,7 +1903,7 @@ static int pio_copy_data(struct kvm_vcpu *vcpu)
int complete_pio(struct kvm_vcpu *vcpu)
{
- struct kvm_pio_request *io = &vcpu->pio;
+ struct kvm_pio_request *io = &vcpu->arch.pio;
long delta;
int r;
@@ -1911,7 +1911,7 @@ int complete_pio(struct kvm_vcpu *vcpu)
if (!io->string) {
if (io->in)
- memcpy(&vcpu->regs[VCPU_REGS_RAX], vcpu->pio_data,
+ memcpy(&vcpu->arch.regs[VCPU_REGS_RAX], vcpu->arch.pio_data,
io->size);
} else {
if (io->in) {
@@ -1929,15 +1929,15 @@ int complete_pio(struct kvm_vcpu *vcpu)
* The size of the register should really depend on
* current address size.
*/
- vcpu->regs[VCPU_REGS_RCX] -= delta;
+ vcpu->arch.regs[VCPU_REGS_RCX] -= delta;
}
if (io->down)
delta = -delta;
delta *= io->size;
if (io->in)
- vcpu->regs[VCPU_REGS_RDI] += delta;
+ vcpu->arch.regs[VCPU_REGS_RDI] += delta;
else
- vcpu->regs[VCPU_REGS_RSI] += delta;
+ vcpu->arch.regs[VCPU_REGS_RSI] += delta;
}
kvm_x86_ops->decache_regs(vcpu);
@@ -1955,13 +1955,13 @@ static void kernel_pio(struct kvm_io_device *pio_dev,
/* TODO: String I/O for in kernel device */
mutex_lock(&vcpu->kvm->lock);
- if (vcpu->pio.in)
- kvm_iodevice_read(pio_dev, vcpu->pio.port,
- vcpu->pio.size,
+ if (vcpu->arch.pio.in)
+ kvm_iodevice_read(pio_dev, vcpu->arch.pio.port,
+ vcpu->arch.pio.size,
pd);
else
- kvm_iodevice_write(pio_dev, vcpu->pio.port,
- vcpu->pio.size,
+ kvm_iodevice_write(pio_dev, vcpu->arch.pio.port,
+ vcpu->arch.pio.size,
pd);
mutex_unlock(&vcpu->kvm->lock);
}
@@ -1969,8 +1969,8 @@ static void kernel_pio(struct kvm_io_device *pio_dev,
static void pio_string_write(struct kvm_io_device *pio_dev,
struct kvm_vcpu *vcpu)
{
- struct kvm_pio_request *io = &vcpu->pio;
- void *pd = vcpu->pio_data;
+ struct kvm_pio_request *io = &vcpu->arch.pio;
+ void *pd = vcpu->arch.pio_data;
int i;
mutex_lock(&vcpu->kvm->lock);
@@ -1996,25 +1996,25 @@ int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
vcpu->run->exit_reason = KVM_EXIT_IO;
vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT;
- vcpu->run->io.size = vcpu->pio.size = size;
+ vcpu->run->io.size = vcpu->arch.pio.size = size;
vcpu->run->io.data_offset = KVM_PIO_PAGE_OFFSET * PAGE_SIZE;
- vcpu->run->io.count = vcpu->pio.count = vcpu->pio.cur_count = 1;
- vcpu->run->io.port = vcpu->pio.port = port;
- vcpu->pio.in = in;
- vcpu->pio.string = 0;
- vcpu->pio.down = 0;
- vcpu->pio.guest_page_offset = 0;
- vcpu->pio.rep = 0;
+ vcpu->run->io.count = vcpu->arch.pio.count = vcpu->arch.pio.cur_count = 1;
+ vcpu->run->io.port = vcpu->arch.pio.port = port;
+ vcpu->arch.pio.in = in;
+ vcpu->arch.pio.string = 0;
+ vcpu->arch.pio.down = 0;
+ vcpu->arch.pio.guest_page_offset = 0;
+ vcpu->arch.pio.rep = 0;
kvm_x86_ops->cache_regs(vcpu);
- memcpy(vcpu->pio_data, &vcpu->regs[VCPU_REGS_RAX], 4);
+ memcpy(vcpu->arch.pio_data, &vcpu->arch.regs[VCPU_REGS_RAX], 4);
kvm_x86_ops->decache_regs(vcpu);
kvm_x86_ops->skip_emulated_instruction(vcpu);
pio_dev = vcpu_find_pio_dev(vcpu, port);
if (pio_dev) {
- kernel_pio(pio_dev, vcpu, vcpu->pio_data);
+ kernel_pio(pio_dev, vcpu, vcpu->arch.pio_data);
complete_pio(vcpu);
return 1;
}
@@ -2034,15 +2034,15 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
vcpu->run->exit_reason = KVM_EXIT_IO;
vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT;
- vcpu->run->io.size = vcpu->pio.size = size;
+ vcpu->run->io.size = vcpu->arch.pio.size = size;
vcpu->run->io.data_offset = KVM_PIO_PAGE_OFFSET * PAGE_SIZE;
- vcpu->run->io.count = vcpu->pio.count = vcpu->pio.cur_count = count;
- vcpu->run->io.port = vcpu->pio.port = port;
- vcpu->pio.in = in;
- vcpu->pio.string = 1;
- vcpu->pio.down = down;
- vcpu->pio.guest_page_offset = offset_in_page(address);
- vcpu->pio.rep = rep;
+ vcpu->run->io.count = vcpu->arch.pio.count = vcpu->arch.pio.cur_count = count;
+ vcpu->run->io.port = vcpu->arch.pio.port = port;
+ vcpu->arch.pio.in = in;
+ vcpu->arch.pio.string = 1;
+ vcpu->arch.pio.down = down;
+ vcpu->arch.pio.guest_page_offset = offset_in_page(address);
+ vcpu->arch.pio.rep = rep;
if (!count) {
kvm_x86_ops->skip_emulated_instruction(vcpu);
@@ -2072,15 +2072,15 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
return 1;
}
vcpu->run->io.count = now;
- vcpu->pio.cur_count = now;
+ vcpu->arch.pio.cur_count = now;
- if (vcpu->pio.cur_count == vcpu->pio.count)
+ if (vcpu->arch.pio.cur_count == vcpu->arch.pio.count)
kvm_x86_ops->skip_emulated_instruction(vcpu);
for (i = 0; i < nr_pages; ++i) {
mutex_lock(&vcpu->kvm->lock);
page = gva_to_page(vcpu, address + i * PAGE_SIZE);
- vcpu->pio.guest_pages[i] = page;
+ vcpu->arch.pio.guest_pages[i] = page;
mutex_unlock(&vcpu->kvm->lock);
if (!page) {
kvm_inject_gp(vcpu, 0);
@@ -2090,13 +2090,13 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
}
pio_dev = vcpu_find_pio_dev(vcpu, port);
- if (!vcpu->pio.in) {
+ if (!vcpu->arch.pio.in) {
/* string PIO write */
ret = pio_copy_data(vcpu);
if (ret >= 0 && pio_dev) {
pio_string_write(pio_dev, vcpu);
complete_pio(vcpu);
- if (vcpu->pio.count == 0)
+ if (vcpu->arch.pio.count == 0)
ret = 1;
}
} else if (pio_dev)
@@ -2156,9 +2156,9 @@ int kvm_emulate_halt(struct kvm_vcpu *vcpu)
{
++vcpu->stat.halt_exits;
if (irqchip_in_kernel(vcpu->kvm)) {
- vcpu->mp_state = VCPU_MP_STATE_HALTED;
+ vcpu->arch.mp_state = VCPU_MP_STATE_HALTED;
kvm_vcpu_block(vcpu);
- if (vcpu->mp_state != VCPU_MP_STATE_RUNNABLE)
+ if (vcpu->arch.mp_state != VCPU_MP_STATE_RUNNABLE)
return -EINTR;
return 1;
} else {
@@ -2174,11 +2174,11 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu)
kvm_x86_ops->cache_regs(vcpu);
- nr = vcpu->regs[VCPU_REGS_RAX];
- a0 = vcpu->regs[VCPU_REGS_RBX];
- a1 = vcpu->regs[VCPU_REGS_RCX];
- a2 = vcpu->regs[VCPU_REGS_RDX];
- a3 = vcpu->regs[VCPU_REGS_RSI];
+ nr = vcpu->arch.regs[VCPU_REGS_RAX];
+ a0 = vcpu->arch.regs[VCPU_REGS_RBX];
+ a1 = vcpu->arch.regs[VCPU_REGS_RCX];
+ a2 = vcpu->arch.regs[VCPU_REGS_RDX];
+ a3 = vcpu->arch.regs[VCPU_REGS_RSI];
if (!is_long_mode(vcpu)) {
nr &= 0xFFFFFFFF;
@@ -2193,7 +2193,7 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu)
ret = -KVM_ENOSYS;
break;
}
- vcpu->regs[VCPU_REGS_RAX] = ret;
+ vcpu->arch.regs[VCPU_REGS_RAX] = ret;
kvm_x86_ops->decache_regs(vcpu);
return 0;
}
@@ -2215,7 +2215,7 @@ int kvm_fix_hypercall(struct kvm_vcpu *vcpu)
kvm_x86_ops->cache_regs(vcpu);
kvm_x86_ops->patch_hypercall(vcpu, instruction);
- if (emulator_write_emulated(vcpu->rip, instruction, 3, vcpu)
+ if (emulator_write_emulated(vcpu->arch.rip, instruction, 3, vcpu)
!= X86EMUL_CONTINUE)
ret = -EFAULT;
@@ -2255,13 +2255,13 @@ unsigned long realmode_get_cr(struct kvm_vcpu *vcpu, int cr)
kvm_x86_ops->decache_cr4_guest_bits(vcpu);
switch (cr) {
case 0:
- return vcpu->cr0;
+ return vcpu->arch.cr0;
case 2:
- return vcpu->cr2;
+ return vcpu->arch.cr2;
case 3:
- return vcpu->cr3;
+ return vcpu->arch.cr3;
case 4:
- return vcpu->cr4;
+ return vcpu->arch.cr4;
case 8:
return get_cr8(vcpu);
default:
@@ -2275,17 +2275,17 @@ void realmode_set_cr(struct kvm_vcpu *vcpu, int cr, unsigned long val,
{
switch (cr) {
case 0:
- set_cr0(vcpu, mk_cr_64(vcpu->cr0, val));
+ set_cr0(vcpu, mk_cr_64(vcpu->arch.cr0, val));
*rflags = kvm_x86_ops->get_rflags(vcpu);
break;
case 2:
- vcpu->cr2 = val;
+ vcpu->arch.cr2 = val;
break;
case 3:
set_cr3(vcpu, val);
break;
case 4:
- set_cr4(vcpu, mk_cr_64(vcpu->cr4, val));
+ set_cr4(vcpu, mk_cr_64(vcpu->arch.cr4, val));
break;
case 8:
set_cr8(vcpu, val & 0xfUL);
@@ -2297,13 +2297,13 @@ void realmode_set_cr(struct kvm_vcpu *vcpu, int cr, unsigned long val,
static int move_to_next_stateful_cpuid_entry(struct kvm_vcpu *vcpu, int i)
{
- struct kvm_cpuid_entry2 *e = &vcpu->cpuid_entries[i];
- int j, nent = vcpu->cpuid_nent;
+ struct kvm_cpuid_entry2 *e = &vcpu->arch.cpuid_entries[i];
+ int j, nent = vcpu->arch.cpuid_nent;
e->flags &= ~KVM_CPUID_FLAG_STATE_READ_NEXT;
/* when no next entry is found, the current entry[i] is reselected */
for (j = i + 1; j == i; j = (j + 1) % nent) {
- struct kvm_cpuid_entry2 *ej = &vcpu->cpuid_entries[j];
+ struct kvm_cpuid_entry2 *ej = &vcpu->arch.cpuid_entries[j];
if (ej->function == e->function) {
ej->flags |= KVM_CPUID_FLAG_STATE_READ_NEXT;
return j;
@@ -2334,15 +2334,15 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
struct kvm_cpuid_entry2 *e, *best;
kvm_x86_ops->cache_regs(vcpu);
- function = vcpu->regs[VCPU_REGS_RAX];
- index = vcpu->regs[VCPU_REGS_RCX];
- vcpu->regs[VCPU_REGS_RAX] = 0;
- vcpu->regs[VCPU_REGS_RBX] = 0;
- vcpu->regs[VCPU_REGS_RCX] = 0;
- vcpu->regs[VCPU_REGS_RDX] = 0;
+ function = vcpu->arch.regs[VCPU_REGS_RAX];
+ index = vcpu->arch.regs[VCPU_REGS_RCX];
+ vcpu->arch.regs[VCPU_REGS_RAX] = 0;
+ vcpu->arch.regs[VCPU_REGS_RBX] = 0;
+ vcpu->arch.regs[VCPU_REGS_RCX] = 0;
+ vcpu->arch.regs[VCPU_REGS_RDX] = 0;
best = NULL;
- for (i = 0; i < vcpu->cpuid_nent; ++i) {
- e = &vcpu->cpuid_entries[i];
+ for (i = 0; i < vcpu->arch.cpuid_nent; ++i) {
+ e = &vcpu->arch.cpuid_entries[i];
if (is_matching_cpuid_entry(e, function, index)) {
if (e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC)
move_to_next_stateful_cpuid_entry(vcpu, i);
@@ -2357,10 +2357,10 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu)
best = e;
}
if (best) {
- vcpu->regs[VCPU_REGS_RAX] = best->eax;
- vcpu->regs[VCPU_REGS_RBX] = best->ebx;
- vcpu->regs[VCPU_REGS_RCX] = best->ecx;
- vcpu->regs[VCPU_REGS_RDX] = best->edx;
+ vcpu->arch.regs[VCPU_REGS_RAX] = best->eax;
+ vcpu->arch.regs[VCPU_REGS_RBX] = best->ebx;
+ vcpu->arch.regs[VCPU_REGS_RCX] = best->ecx;
+ vcpu->arch.regs[VCPU_REGS_RDX] = best->edx;
}
kvm_x86_ops->decache_regs(vcpu);
kvm_x86_ops->skip_emulated_instruction(vcpu);
@@ -2376,9 +2376,9 @@ EXPORT_SYMBOL_GPL(kvm_emulate_cpuid);
static int dm_request_for_irq_injection(struct kvm_vcpu *vcpu,
struct kvm_run *kvm_run)
{
- return (!vcpu->irq_summary &&
+ return (!vcpu->arch.irq_summary &&
kvm_run->request_interrupt_window &&
- vcpu->interrupt_window_open &&
+ vcpu->arch.interrupt_window_open &&
(kvm_x86_ops->get_rflags(vcpu) & X86_EFLAGS_IF));
}
@@ -2392,22 +2392,22 @@ static void post_kvm_run_save(struct kvm_vcpu *vcpu,
kvm_run->ready_for_interrupt_injection = 1;
else
kvm_run->ready_for_interrupt_injection =
- (vcpu->interrupt_window_open &&
- vcpu->irq_summary == 0);
+ (vcpu->arch.interrupt_window_open &&
+ vcpu->arch.irq_summary == 0);
}
static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
int r;
- if (unlikely(vcpu->mp_state == VCPU_MP_STATE_SIPI_RECEIVED)) {
+ if (unlikely(vcpu->arch.mp_state == VCPU_MP_STATE_SIPI_RECEIVED)) {
pr_debug("vcpu %d received sipi with vector # %x\n",
- vcpu->vcpu_id, vcpu->sipi_vector);
+ vcpu->vcpu_id, vcpu->arch.sipi_vector);
kvm_lapic_reset(vcpu);
r = kvm_x86_ops->vcpu_reset(vcpu);
if (r)
return r;
- vcpu->mp_state = VCPU_MP_STATE_RUNNABLE;
+ vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
}
preempted:
@@ -2437,7 +2437,7 @@ again:
goto out;
}
- if (vcpu->exception.pending)
+ if (vcpu->arch.exception.pending)
__queue_exception(vcpu);
else if (irqchip_in_kernel(vcpu->kvm))
kvm_x86_ops->inject_pending_irq(vcpu);
@@ -2475,11 +2475,11 @@ again:
*/
if (unlikely(prof_on == KVM_PROFILING)) {
kvm_x86_ops->cache_regs(vcpu);
- profile_hit(KVM_PROFILING, (void *)vcpu->rip);
+ profile_hit(KVM_PROFILING, (void *)vcpu->arch.rip);
}
- if (vcpu->exception.pending && kvm_x86_ops->exception_injected(vcpu))
- vcpu->exception.pending = false;
+ if (vcpu->arch.exception.pending && kvm_x86_ops->exception_injected(vcpu))
+ vcpu->arch.exception.pending = false;
r = kvm_x86_ops->handle_exit(kvm_run, vcpu);
@@ -2512,7 +2512,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
vcpu_load(vcpu);
- if (unlikely(vcpu->mp_state == VCPU_MP_STATE_UNINITIALIZED)) {
+ if (unlikely(vcpu->arch.mp_state == VCPU_MP_STATE_UNINITIALIZED)) {
kvm_vcpu_block(vcpu);
vcpu_put(vcpu);
return -EAGAIN;
@@ -2525,7 +2525,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
if (!irqchip_in_kernel(vcpu->kvm))
set_cr8(vcpu, kvm_run->cr8);
- if (vcpu->pio.cur_count) {
+ if (vcpu->arch.pio.cur_count) {
r = complete_pio(vcpu);
if (r)
goto out;
@@ -2536,7 +2536,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
vcpu->mmio_read_completed = 1;
vcpu->mmio_needed = 0;
r = emulate_instruction(vcpu, kvm_run,
- vcpu->mmio_fault_cr2, 0, 1);
+ vcpu->arch.mmio_fault_cr2, 0, 1);
if (r == EMULATE_DO_MMIO) {
/*
* Read-modify-write. Back to userspace.
@@ -2548,7 +2548,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
#endif
if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL) {
kvm_x86_ops->cache_regs(vcpu);
- vcpu->regs[VCPU_REGS_RAX] = kvm_run->hypercall.ret;
+ vcpu->arch.regs[VCPU_REGS_RAX] = kvm_run->hypercall.ret;
kvm_x86_ops->decache_regs(vcpu);
}
@@ -2568,26 +2568,26 @@ int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
kvm_x86_ops->cache_regs(vcpu);
- regs->rax = vcpu->regs[VCPU_REGS_RAX];
- regs->rbx = vcpu->regs[VCPU_REGS_RBX];
- regs->rcx = vcpu->regs[VCPU_REGS_RCX];
- regs->rdx = vcpu->regs[VCPU_REGS_RDX];
- regs->rsi = vcpu->regs[VCPU_REGS_RSI];
- regs->rdi = vcpu->regs[VCPU_REGS_RDI];
- regs->rsp = vcpu->regs[VCPU_REGS_RSP];
- regs->rbp = vcpu->regs[VCPU_REGS_RBP];
+ regs->rax = vcpu->arch.regs[VCPU_REGS_RAX];
+ regs->rbx = vcpu->arch.regs[VCPU_REGS_RBX];
+ regs->rcx = vcpu->arch.regs[VCPU_REGS_RCX];
+ regs->rdx = vcpu->arch.regs[VCPU_REGS_RDX];
+ regs->rsi = vcpu->arch.regs[VCPU_REGS_RSI];
+ regs->rdi = vcpu->arch.regs[VCPU_REGS_RDI];
+ regs->rsp = vcpu->arch.regs[VCPU_REGS_RSP];
+ regs->rbp = vcpu->arch.regs[VCPU_REGS_RBP];
#ifdef CONFIG_X86_64
- regs->r8 = vcpu->regs[VCPU_REGS_R8];
- regs->r9 = vcpu->regs[VCPU_REGS_R9];
- regs->r10 = vcpu->regs[VCPU_REGS_R10];
- regs->r11 = vcpu->regs[VCPU_REGS_R11];
- regs->r12 = vcpu->regs[VCPU_REGS_R12];
- regs->r13 = vcpu->regs[VCPU_REGS_R13];
- regs->r14 = vcpu->regs[VCPU_REGS_R14];
- regs->r15 = vcpu->regs[VCPU_REGS_R15];
+ regs->r8 = vcpu->arch.regs[VCPU_REGS_R8];
+ regs->r9 = vcpu->arch.regs[VCPU_REGS_R9];
+ regs->r10 = vcpu->arch.regs[VCPU_REGS_R10];
+ regs->r11 = vcpu->arch.regs[VCPU_REGS_R11];
+ regs->r12 = vcpu->arch.regs[VCPU_REGS_R12];
+ regs->r13 = vcpu->arch.regs[VCPU_REGS_R13];
+ regs->r14 = vcpu->arch.regs[VCPU_REGS_R14];
+ regs->r15 = vcpu->arch.regs[VCPU_REGS_R15];
#endif
- regs->rip = vcpu->rip;
+ regs->rip = vcpu->arch.rip;
regs->rflags = kvm_x86_ops->get_rflags(vcpu);
/*
@@ -2605,26 +2605,26 @@ int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
vcpu_load(vcpu);
- vcpu->regs[VCPU_REGS_RAX] = regs->rax;
- vcpu->regs[VCPU_REGS_RBX] = regs->rbx;
- vcpu->regs[VCPU_REGS_RCX] = regs->rcx;
- vcpu->regs[VCPU_REGS_RDX] = regs->rdx;
- vcpu->regs[VCPU_REGS_RSI] = regs->rsi;
- vcpu->regs[VCPU_REGS_RDI] = regs->rdi;
- vcpu->regs[VCPU_REGS_RSP] = regs->rsp;
- vcpu->regs[VCPU_REGS_RBP] = regs->rbp;
+ vcpu->arch.regs[VCPU_REGS_RAX] = regs->rax;
+ vcpu->arch.regs[VCPU_REGS_RBX] = regs->rbx;
+ vcpu->arch.regs[VCPU_REGS_RCX] = regs->rcx;
+ vcpu->arch.regs[VCPU_REGS_RDX] = regs->rdx;
+ vcpu->arch.regs[VCPU_REGS_RSI] = regs->rsi;
+ vcpu->arch.regs[VCPU_REGS_RDI] = regs->rdi;
+ vcpu->arch.regs[VCPU_REGS_RSP] = regs->rsp;
+ vcpu->arch.regs[VCPU_REGS_RBP] = regs->rbp;
#ifdef CONFIG_X86_64
- vcpu->regs[VCPU_REGS_R8] = regs->r8;
- vcpu->regs[VCPU_REGS_R9] = regs->r9;
- vcpu->regs[VCPU_REGS_R10] = regs->r10;
- vcpu->regs[VCPU_REGS_R11] = regs->r11;
- vcpu->regs[VCPU_REGS_R12] = regs->r12;
- vcpu->regs[VCPU_REGS_R13] = regs->r13;
- vcpu->regs[VCPU_REGS_R14] = regs->r14;
- vcpu->regs[VCPU_REGS_R15] = regs->r15;
+ vcpu->arch.regs[VCPU_REGS_R8] = regs->r8;
+ vcpu->arch.regs[VCPU_REGS_R9] = regs->r9;
+ vcpu->arch.regs[VCPU_REGS_R10] = regs->r10;
+ vcpu->arch.regs[VCPU_REGS_R11] = regs->r11;
+ vcpu->arch.regs[VCPU_REGS_R12] = regs->r12;
+ vcpu->arch.regs[VCPU_REGS_R13] = regs->r13;
+ vcpu->arch.regs[VCPU_REGS_R14] = regs->r14;
+ vcpu->arch.regs[VCPU_REGS_R15] = regs->r15;
#endif
- vcpu->rip = regs->rip;
+ vcpu->arch.rip = regs->rip;
kvm_x86_ops->set_rflags(vcpu, regs->rflags);
kvm_x86_ops->decache_regs(vcpu);
@@ -2676,12 +2676,12 @@ int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
sregs->gdt.base = dt.base;
kvm_x86_ops->decache_cr4_guest_bits(vcpu);
- sregs->cr0 = vcpu->cr0;
- sregs->cr2 = vcpu->cr2;
- sregs->cr3 = vcpu->cr3;
- sregs->cr4 = vcpu->cr4;
+ sregs->cr0 = vcpu->arch.cr0;
+ sregs->cr2 = vcpu->arch.cr2;
+ sregs->cr3 = vcpu->arch.cr3;
+ sregs->cr4 = vcpu->arch.cr4;
sregs->cr8 = get_cr8(vcpu);
- sregs->efer = vcpu->shadow_efer;
+ sregs->efer = vcpu->arch.shadow_efer;
sregs->apic_base = kvm_get_apic_base(vcpu);
if (irqchip_in_kernel(vcpu->kvm)) {
@@ -2692,7 +2692,7 @@ int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
set_bit(pending_vec,
(unsigned long *)sregs->interrupt_bitmap);
} else
- memcpy(sregs->interrupt_bitmap, vcpu->irq_pending,
+ memcpy(sregs->interrupt_bitmap, vcpu->arch.irq_pending,
sizeof sregs->interrupt_bitmap);
vcpu_put(vcpu);
@@ -2722,13 +2722,13 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
dt.base = sregs->gdt.base;
kvm_x86_ops->set_gdt(vcpu, &dt);
- vcpu->cr2 = sregs->cr2;
- mmu_reset_needed |= vcpu->cr3 != sregs->cr3;
- vcpu->cr3 = sregs->cr3;
+ vcpu->arch.cr2 = sregs->cr2;
+ mmu_reset_needed |= vcpu->arch.cr3 != sregs->cr3;
+ vcpu->arch.cr3 = sregs->cr3;
set_cr8(vcpu, sregs->cr8);
- mmu_reset_needed |= vcpu->shadow_efer != sregs->efer;
+ mmu_reset_needed |= vcpu->arch.shadow_efer != sregs->efer;
#ifdef CONFIG_X86_64
kvm_x86_ops->set_efer(vcpu, sregs->efer);
#endif
@@ -2736,25 +2736,25 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
kvm_x86_ops->decache_cr4_guest_bits(vcpu);
- mmu_reset_needed |= vcpu->cr0 != sregs->cr0;
- vcpu->cr0 = sregs->cr0;
+ mmu_reset_needed |= vcpu->arch.cr0 != sregs->cr0;
+ vcpu->arch.cr0 = sregs->cr0;
kvm_x86_ops->set_cr0(vcpu, sregs->cr0);
- mmu_reset_needed |= vcpu->cr4 != sregs->cr4;
+ mmu_reset_needed |= vcpu->arch.cr4 != sregs->cr4;
kvm_x86_ops->set_cr4(vcpu, sregs->cr4);
if (!is_long_mode(vcpu) && is_pae(vcpu))
- load_pdptrs(vcpu, vcpu->cr3);
+ load_pdptrs(vcpu, vcpu->arch.cr3);
if (mmu_reset_needed)
kvm_mmu_reset_context(vcpu);
if (!irqchip_in_kernel(vcpu->kvm)) {
- memcpy(vcpu->irq_pending, sregs->interrupt_bitmap,
- sizeof vcpu->irq_pending);
- vcpu->irq_summary = 0;
- for (i = 0; i < ARRAY_SIZE(vcpu->irq_pending); ++i)
- if (vcpu->irq_pending[i])
- __set_bit(i, &vcpu->irq_summary);
+ memcpy(vcpu->arch.irq_pending, sregs->interrupt_bitmap,
+ sizeof vcpu->arch.irq_pending);
+ vcpu->arch.irq_summary = 0;
+ for (i = 0; i < ARRAY_SIZE(vcpu->arch.irq_pending); ++i)
+ if (vcpu->arch.irq_pending[i])
+ __set_bit(i, &vcpu->arch.irq_summary);
} else {
max_bits = (sizeof sregs->interrupt_bitmap) << 3;
pending_vec = find_first_bit(
@@ -2829,7 +2829,7 @@ int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
vcpu_load(vcpu);
mutex_lock(&vcpu->kvm->lock);
- gpa = vcpu->mmu.gva_to_gpa(vcpu, vaddr);
+ gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, vaddr);
tr->physical_address = gpa;
tr->valid = gpa != UNMAPPED_GVA;
tr->writeable = 1;
@@ -2842,7 +2842,7 @@ int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
- struct fxsave *fxsave = (struct fxsave *)&vcpu->guest_fx_image;
+ struct fxsave *fxsave = (struct fxsave *)&vcpu->arch.guest_fx_image;
vcpu_load(vcpu);
@@ -2862,7 +2862,7 @@ int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
- struct fxsave *fxsave = (struct fxsave *)&vcpu->guest_fx_image;
+ struct fxsave *fxsave = (struct fxsave *)&vcpu->arch.guest_fx_image;
vcpu_load(vcpu);
@@ -2886,16 +2886,16 @@ void fx_init(struct kvm_vcpu *vcpu)
/* Initialize guest FPU by resetting ours and saving into guest's */
preempt_disable();
- fx_save(&vcpu->host_fx_image);
+ fx_save(&vcpu->arch.host_fx_image);
fpu_init();
- fx_save(&vcpu->guest_fx_image);
- fx_restore(&vcpu->host_fx_image);
+ fx_save(&vcpu->arch.guest_fx_image);
+ fx_restore(&vcpu->arch.host_fx_image);
preempt_enable();
- vcpu->cr0 |= X86_CR0_ET;
+ vcpu->arch.cr0 |= X86_CR0_ET;
after_mxcsr_mask = offsetof(struct i387_fxsave_struct, st_space);
- vcpu->guest_fx_image.mxcsr = 0x1f80;
- memset((void *)&vcpu->guest_fx_image + after_mxcsr_mask,
+ vcpu->arch.guest_fx_image.mxcsr = 0x1f80;
+ memset((void *)&vcpu->arch.guest_fx_image + after_mxcsr_mask,
0, sizeof(struct i387_fxsave_struct) - after_mxcsr_mask);
}
EXPORT_SYMBOL_GPL(fx_init);
@@ -2906,8 +2906,8 @@ void kvm_load_guest_fpu(struct kvm_vcpu *vcpu)
return;
vcpu->guest_fpu_loaded = 1;
- fx_save(&vcpu->host_fx_image);
- fx_restore(&vcpu->guest_fx_image);
+ fx_save(&vcpu->arch.host_fx_image);
+ fx_restore(&vcpu->arch.guest_fx_image);
}
EXPORT_SYMBOL_GPL(kvm_load_guest_fpu);
@@ -2917,8 +2917,8 @@ void kvm_put_guest_fpu(struct kvm_vcpu *vcpu)
return;
vcpu->guest_fpu_loaded = 0;
- fx_save(&vcpu->guest_fx_image);
- fx_restore(&vcpu->host_fx_image);
+ fx_save(&vcpu->arch.guest_fx_image);
+ fx_restore(&vcpu->arch.host_fx_image);
++vcpu->stat.fpu_reload;
}
EXPORT_SYMBOL_GPL(kvm_put_guest_fpu);
@@ -2939,7 +2939,7 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
int r;
/* We do fxsave: this must be aligned. */
- BUG_ON((unsigned long)&vcpu->host_fx_image & 0xF);
+ BUG_ON((unsigned long)&vcpu->arch.host_fx_image & 0xF);
vcpu_load(vcpu);
r = kvm_arch_vcpu_reset(vcpu);
@@ -3003,18 +3003,18 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
BUG_ON(vcpu->kvm == NULL);
kvm = vcpu->kvm;
- vcpu->mmu.root_hpa = INVALID_PAGE;
+ vcpu->arch.mmu.root_hpa = INVALID_PAGE;
if (!irqchip_in_kernel(kvm) || vcpu->vcpu_id == 0)
- vcpu->mp_state = VCPU_MP_STATE_RUNNABLE;
+ vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
else
- vcpu->mp_state = VCPU_MP_STATE_UNINITIALIZED;
+ vcpu->arch.mp_state = VCPU_MP_STATE_UNINITIALIZED;
page = alloc_page(GFP_KERNEL | __GFP_ZERO);
if (!page) {
r = -ENOMEM;
goto fail;
}
- vcpu->pio_data = page_address(page);
+ vcpu->arch.pio_data = page_address(page);
r = kvm_mmu_create(vcpu);
if (r < 0)
@@ -3031,7 +3031,7 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
fail_mmu_destroy:
kvm_mmu_destroy(vcpu);
fail_free_pio_data:
- free_page((unsigned long)vcpu->pio_data);
+ free_page((unsigned long)vcpu->arch.pio_data);
fail:
return r;
}
@@ -3040,7 +3040,7 @@ void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
{
kvm_free_lapic(vcpu);
kvm_mmu_destroy(vcpu);
- free_page((unsigned long)vcpu->pio_data);
+ free_page((unsigned long)vcpu->arch.pio_data);
}
struct kvm *kvm_arch_create_vm(void)
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index 0fc7020..0e01ac7 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -92,8 +92,7 @@ enum {
#include "x86_emulate.h"
-struct kvm_vcpu {
- KVM_VCPU_COMM;
+struct kvm_vcpu_arch {
u64 host_tsc;
int interrupt_window_open;
unsigned long irq_summary; /* bit vector: 1 per word in irq_pending */
@@ -130,7 +129,6 @@ struct kvm_vcpu {
int last_pt_write_count;
u64 *last_pte_updated;
-
struct i387_fxsave_struct host_fx_image;
struct i387_fxsave_struct guest_fx_image;
@@ -159,12 +157,17 @@ struct kvm_vcpu {
int cpuid_nent;
struct kvm_cpuid_entry2 cpuid_entries[KVM_MAX_CPUID_ENTRIES];
-
/* emulate context */
struct x86_emulate_ctxt emulate_ctxt;
};
+struct kvm_vcpu {
+ KVM_VCPU_COMM;
+
+ struct kvm_vcpu_arch arch;
+};
+
struct descriptor_table {
u16 limit;
unsigned long base;
@@ -339,7 +342,7 @@ static inline void kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
static inline int kvm_mmu_reload(struct kvm_vcpu *vcpu)
{
- if (likely(vcpu->mmu.root_hpa != INVALID_PAGE))
+ if (likely(vcpu->arch.mmu.root_hpa != INVALID_PAGE))
return 0;
return kvm_mmu_load(vcpu);
@@ -348,7 +351,7 @@ static inline int kvm_mmu_reload(struct kvm_vcpu *vcpu)
static inline int is_long_mode(struct kvm_vcpu *vcpu)
{
#ifdef CONFIG_X86_64
- return vcpu->shadow_efer & EFER_LME;
+ return vcpu->arch.shadow_efer & EFER_LME;
#else
return 0;
#endif
@@ -356,17 +359,17 @@ static inline int is_long_mode(struct kvm_vcpu *vcpu)
static inline int is_pae(struct kvm_vcpu *vcpu)
{
- return vcpu->cr4 & X86_CR4_PAE;
+ return vcpu->arch.cr4 & X86_CR4_PAE;
}
static inline int is_pse(struct kvm_vcpu *vcpu)
{
- return vcpu->cr4 & X86_CR4_PSE;
+ return vcpu->arch.cr4 & X86_CR4_PSE;
}
static inline int is_paging(struct kvm_vcpu *vcpu)
{
- return vcpu->cr0 & X86_CR0_PG;
+ return vcpu->arch.cr0 & X86_CR0_PG;
}
int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3);
@@ -489,8 +492,8 @@ static inline void kvm_inject_gp(struct kvm_vcpu *vcpu, u32 error_code)
static inline int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
{
- return vcpu->mp_state == VCPU_MP_STATE_RUNNABLE
- || vcpu->mp_state == VCPU_MP_STATE_SIPI_RECEIVED;
+ return vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE
+ || vcpu->arch.mp_state == VCPU_MP_STATE_SIPI_RECEIVED;
}
#endif
diff --git a/drivers/kvm/x86_emulate.c b/drivers/kvm/x86_emulate.c
index 0a6ab06..50b133f 100644
--- a/drivers/kvm/x86_emulate.c
+++ b/drivers/kvm/x86_emulate.c
@@ -769,8 +769,8 @@ x86_decode_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops)
/* Shadow copy of register state. Committed on successful emulation. */
memset(c, 0, sizeof(struct decode_cache));
- c->eip = ctxt->vcpu->rip;
- memcpy(c->regs, ctxt->vcpu->regs, sizeof c->regs);
+ c->eip = ctxt->vcpu->arch.rip;
+ memcpy(c->regs, ctxt->vcpu->arch.regs, sizeof c->regs);
switch (mode) {
case X86EMUL_MODE_REAL:
@@ -1226,7 +1226,7 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops)
* modify them.
*/
- memcpy(c->regs, ctxt->vcpu->regs, sizeof c->regs);
+ memcpy(c->regs, ctxt->vcpu->arch.regs, sizeof c->regs);
saved_eip = c->eip;
if (((c->d & ModRM) && (c->modrm_mod != 3)) || (c->d & MemAbs))
@@ -1235,7 +1235,7 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops)
if (c->rep_prefix && (c->d & String)) {
/* All REP prefixes have the same first termination condition */
if (c->regs[VCPU_REGS_RCX] == 0) {
- ctxt->vcpu->rip = c->eip;
+ ctxt->vcpu->arch.rip = c->eip;
goto done;
}
/* The second termination condition only applies for REPE
@@ -1249,17 +1249,17 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops)
(c->b == 0xae) || (c->b == 0xaf)) {
if ((c->rep_prefix == REPE_PREFIX) &&
((ctxt->eflags & EFLG_ZF) == 0)) {
- ctxt->vcpu->rip = c->eip;
+ ctxt->vcpu->arch.rip = c->eip;
goto done;
}
if ((c->rep_prefix == REPNE_PREFIX) &&
((ctxt->eflags & EFLG_ZF) == EFLG_ZF)) {
- ctxt->vcpu->rip = c->eip;
+ ctxt->vcpu->arch.rip = c->eip;
goto done;
}
}
c->regs[VCPU_REGS_RCX]--;
- c->eip = ctxt->vcpu->rip;
+ c->eip = ctxt->vcpu->arch.rip;
}
if (c->src.type == OP_MEM) {
@@ -1628,7 +1628,7 @@ special_insn:
c->dst.type = OP_NONE; /* Disable writeback. */
break;
case 0xf4: /* hlt */
- ctxt->vcpu->halt_request = 1;
+ ctxt->vcpu->arch.halt_request = 1;
goto done;
case 0xf5: /* cmc */
/* complement carry flag from eflags reg */
@@ -1665,8 +1665,8 @@ writeback:
goto done;
/* Commit shadow register state. */
- memcpy(ctxt->vcpu->regs, c->regs, sizeof c->regs);
- ctxt->vcpu->rip = c->eip;
+ memcpy(ctxt->vcpu->arch.regs, c->regs, sizeof c->regs);
+ ctxt->vcpu->arch.rip = c->eip;
done:
if (rc == X86EMUL_UNHANDLEABLE) {
@@ -1783,7 +1783,7 @@ twobyte_insn:
rc = kvm_set_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], msr_data);
if (rc) {
kvm_inject_gp(ctxt->vcpu, 0);
- c->eip = ctxt->vcpu->rip;
+ c->eip = ctxt->vcpu->arch.rip;
}
rc = X86EMUL_CONTINUE;
c->dst.type = OP_NONE;
@@ -1793,7 +1793,7 @@ twobyte_insn:
rc = kvm_get_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], &msr_data);
if (rc) {
kvm_inject_gp(ctxt->vcpu, 0);
- c->eip = ctxt->vcpu->rip;
+ c->eip = ctxt->vcpu->arch.rip;
} else {
c->regs[VCPU_REGS_RAX] = (u32)msr_data;
c->regs[VCPU_REGS_RDX] = msr_data >> 32;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 05/33] KVM: Portability: Split mmu-related static inline functions to mmu.h
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (3 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 04/33] KVM: Portability: Introduce kvm_vcpu_arch Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 06/33] KVM: Portability: Move kvm_vcpu definition back to kvm.h Avi Kivity
` (27 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
Since these functions need to know the details of kvm or kvm_vcpu structure,
it can't be put in x86.h. Create mmu.h to hold them.
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>
---
drivers/kvm/kvm.h | 1 +
drivers/kvm/mmu.c | 1 +
drivers/kvm/mmu.h | 44 ++++++++++++++++++++++++++++++++++++++++++++
drivers/kvm/svm.c | 1 +
drivers/kvm/vmx.c | 1 +
drivers/kvm/x86.c | 7 +++++++
drivers/kvm/x86.h | 44 --------------------------------------------
7 files changed, 55 insertions(+), 44 deletions(-)
create mode 100644 drivers/kvm/mmu.h
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index 668a830..c040bb3 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -396,6 +396,7 @@ 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);
+int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu);
void kvm_free_physmem(struct kvm *kvm);
diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c
index da1dedb..1dc0e8c 100644
--- a/drivers/kvm/mmu.c
+++ b/drivers/kvm/mmu.c
@@ -20,6 +20,7 @@
#include "vmx.h"
#include "kvm.h"
#include "x86.h"
+#include "mmu.h"
#include <linux/types.h>
#include <linux/string.h>
diff --git a/drivers/kvm/mmu.h b/drivers/kvm/mmu.h
new file mode 100644
index 0000000..9ebfd1c
--- /dev/null
+++ b/drivers/kvm/mmu.h
@@ -0,0 +1,44 @@
+#ifndef __KVM_X86_MMU_H
+#define __KVM_X86_MMU_H
+
+#include "kvm.h"
+
+static inline void kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
+{
+ if (unlikely(vcpu->kvm->n_free_mmu_pages < KVM_MIN_FREE_MMU_PAGES))
+ __kvm_mmu_free_some_pages(vcpu);
+}
+
+static inline int kvm_mmu_reload(struct kvm_vcpu *vcpu)
+{
+ if (likely(vcpu->arch.mmu.root_hpa != INVALID_PAGE))
+ return 0;
+
+ return kvm_mmu_load(vcpu);
+}
+
+static inline int is_long_mode(struct kvm_vcpu *vcpu)
+{
+#ifdef CONFIG_X86_64
+ return vcpu->arch.shadow_efer & EFER_LME;
+#else
+ return 0;
+#endif
+}
+
+static inline int is_pae(struct kvm_vcpu *vcpu)
+{
+ return vcpu->arch.cr4 & X86_CR4_PAE;
+}
+
+static inline int is_pse(struct kvm_vcpu *vcpu)
+{
+ return vcpu->arch.cr4 & X86_CR4_PSE;
+}
+
+static inline int is_paging(struct kvm_vcpu *vcpu)
+{
+ return vcpu->arch.cr0 & X86_CR0_PG;
+}
+
+#endif
diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c
index 6c100b1..35ad96d 100644
--- a/drivers/kvm/svm.c
+++ b/drivers/kvm/svm.c
@@ -17,6 +17,7 @@
#include "kvm_svm.h"
#include "x86_emulate.h"
#include "irq.h"
+#include "mmu.h"
#include <linux/module.h>
#include <linux/kernel.h>
diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c
index e26e299..456eb9c 100644
--- a/drivers/kvm/vmx.c
+++ b/drivers/kvm/vmx.c
@@ -21,6 +21,7 @@
#include "irq.h"
#include "vmx.h"
#include "segment_descriptor.h"
+#include "mmu.h"
#include <linux/module.h>
#include <linux/kernel.h>
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index 5a2f33a..60f9722 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -19,6 +19,7 @@
#include "x86_emulate.h"
#include "segment_descriptor.h"
#include "irq.h"
+#include "mmu.h"
#include <linux/kvm.h>
#include <linux/fs.h>
@@ -3139,3 +3140,9 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
return 0;
}
+
+int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
+{
+ return vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE
+ || vcpu->arch.mp_state == VCPU_MP_STATE_SIPI_RECEIVED;
+}
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index 0e01ac7..0da4be9 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -334,44 +334,6 @@ int kvm_fix_hypercall(struct kvm_vcpu *vcpu);
int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t gva, u32 error_code);
-static inline void kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
-{
- if (unlikely(vcpu->kvm->n_free_mmu_pages < KVM_MIN_FREE_MMU_PAGES))
- __kvm_mmu_free_some_pages(vcpu);
-}
-
-static inline int kvm_mmu_reload(struct kvm_vcpu *vcpu)
-{
- if (likely(vcpu->arch.mmu.root_hpa != INVALID_PAGE))
- return 0;
-
- return kvm_mmu_load(vcpu);
-}
-
-static inline int is_long_mode(struct kvm_vcpu *vcpu)
-{
-#ifdef CONFIG_X86_64
- return vcpu->arch.shadow_efer & EFER_LME;
-#else
- return 0;
-#endif
-}
-
-static inline int is_pae(struct kvm_vcpu *vcpu)
-{
- return vcpu->arch.cr4 & X86_CR4_PAE;
-}
-
-static inline int is_pse(struct kvm_vcpu *vcpu)
-{
- return vcpu->arch.cr4 & X86_CR4_PSE;
-}
-
-static inline int is_paging(struct kvm_vcpu *vcpu)
-{
- return vcpu->arch.cr0 & X86_CR0_PG;
-}
-
int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3);
int complete_pio(struct kvm_vcpu *vcpu);
@@ -490,10 +452,4 @@ static inline void kvm_inject_gp(struct kvm_vcpu *vcpu, u32 error_code)
#define TSS_REDIRECTION_SIZE (256 / 8)
#define RMODE_TSS_SIZE (TSS_BASE_SIZE + TSS_REDIRECTION_SIZE + TSS_IOPB_SIZE + 1)
-static inline int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
-{
- return vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE
- || vcpu->arch.mp_state == VCPU_MP_STATE_SIPI_RECEIVED;
-}
-
#endif
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 06/33] KVM: Portability: Move kvm_vcpu definition back to kvm.h
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (4 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 05/33] KVM: Portability: Split mmu-related static inline functions to mmu.h Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 07/33] KVM: Portability: Expand the KVM_VCPU_COMM in kvm_vcpu structure Avi Kivity
` (26 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
This patches moves kvm_vcpu definition to kvm.h, and finally
kvm.h includes x86.h.
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>
---
drivers/kvm/irq.h | 2 +-
drivers/kvm/kvm.h | 113 ++++-----------------------------------------------
drivers/kvm/types.h | 13 ++++++
drivers/kvm/x86.h | 102 ++++++++++++++++++++++++++++++++++++++++++---
4 files changed, 117 insertions(+), 113 deletions(-)
diff --git a/drivers/kvm/irq.h b/drivers/kvm/irq.h
index e88d939..a688c51 100644
--- a/drivers/kvm/irq.h
+++ b/drivers/kvm/irq.h
@@ -26,7 +26,7 @@
#include <linux/hrtimer.h>
#include <asm/kvm.h>
#include "iodev.h"
-#include "x86.h"
+#include "kvm.h"
struct kvm;
struct kvm_vcpu;
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index c040bb3..53717be 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -22,17 +22,13 @@
#include "types.h"
+#include "x86.h"
+
#define KVM_MAX_VCPUS 4
#define KVM_ALIAS_SLOTS 4
#define KVM_MEMORY_SLOTS 8
/* memory slots that does not exposed to userspace */
#define KVM_PRIVATE_MEM_SLOTS 4
-#define KVM_PERMILLE_MMU_PAGES 20
-#define KVM_MIN_ALLOC_MMU_PAGES 64
-#define KVM_NUM_MMU_PAGES 1024
-#define KVM_MIN_FREE_MMU_PAGES 5
-#define KVM_REFILL_PAGES 25
-#define KVM_MAX_CPUID_ENTRIES 40
#define KVM_PIO_PAGE_OFFSET 1
@@ -41,111 +37,16 @@
*/
#define KVM_REQ_TLB_FLUSH 0
-#define NR_PTE_CHAIN_ENTRIES 5
-
-struct kvm_pte_chain {
- u64 *parent_ptes[NR_PTE_CHAIN_ENTRIES];
- struct hlist_node link;
-};
-
-/*
- * kvm_mmu_page_role, below, is defined as:
- *
- * bits 0:3 - total guest paging levels (2-4, or zero for real mode)
- * bits 4:7 - page table level for this shadow (1-4)
- * bits 8:9 - page table quadrant for 2-level guests
- * bit 16 - "metaphysical" - gfn is not a real page (huge page/real mode)
- * bits 17:19 - common access permissions for all ptes in this shadow page
- */
-union kvm_mmu_page_role {
- unsigned word;
- struct {
- unsigned glevels : 4;
- unsigned level : 4;
- unsigned quadrant : 2;
- unsigned pad_for_nice_hex_output : 6;
- unsigned metaphysical : 1;
- unsigned access : 3;
- };
-};
-
-struct kvm_mmu_page {
- struct list_head link;
- struct hlist_node hash_link;
-
- /*
- * The following two entries are used to key the shadow page in the
- * hash table.
- */
- gfn_t gfn;
- union kvm_mmu_page_role role;
-
- u64 *spt;
- /* hold the gfn of each spte inside spt */
- gfn_t *gfns;
- unsigned long slot_bitmap; /* One bit set per slot which has memory
- * in this shadow page.
- */
- int multimapped; /* More than one parent_pte? */
- int root_count; /* Currently serving as active root */
- union {
- u64 *parent_pte; /* !multimapped */
- struct hlist_head parent_ptes; /* multimapped, kvm_pte_chain */
- };
-};
struct kvm_vcpu;
extern struct kmem_cache *kvm_vcpu_cache;
-/*
- * x86 supports 3 paging modes (4-level 64-bit, 3-level 64-bit, and 2-level
- * 32-bit). The kvm_mmu structure abstracts the details of the current mmu
- * mode.
- */
-struct kvm_mmu {
- void (*new_cr3)(struct kvm_vcpu *vcpu);
- int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err);
- void (*free)(struct kvm_vcpu *vcpu);
- gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva);
- void (*prefetch_page)(struct kvm_vcpu *vcpu,
- struct kvm_mmu_page *page);
- hpa_t root_hpa;
- int root_level;
- int shadow_root_level;
-
- u64 *pae_root;
-};
-
-#define KVM_NR_MEM_OBJS 40
-
-/*
- * We don't want allocation failures within the mmu code, so we preallocate
- * enough memory for a single page fault in a cache.
- */
-struct kvm_mmu_memory_cache {
- int nobjs;
- void *objects[KVM_NR_MEM_OBJS];
-};
-
struct kvm_guest_debug {
int enabled;
unsigned long bp[4];
int singlestep;
};
-struct kvm_pio_request {
- unsigned long count;
- int cur_count;
- struct page *guest_pages[2];
- unsigned guest_page_offset;
- int in;
- int port;
- int size;
- int string;
- int down;
- int rep;
-};
-
struct kvm_vcpu_stat {
u32 pf_fixed;
u32 pf_guest;
@@ -218,6 +119,12 @@ void kvm_io_bus_register_dev(struct kvm_io_bus *bus,
struct kvm_vcpu_stat stat; \
KVM_VCPU_MMIO
+struct kvm_vcpu {
+ KVM_VCPU_COMM;
+
+ struct kvm_vcpu_arch arch;
+};
+
struct kvm_mem_alias {
gfn_t base_gfn;
unsigned long npages;
@@ -441,8 +348,4 @@ struct kvm_stats_debugfs_item {
};
extern struct kvm_stats_debugfs_item debugfs_entries[];
-#if defined(CONFIG_X86)
-#include "x86.h"
-#endif
-
#endif
diff --git a/drivers/kvm/types.h b/drivers/kvm/types.h
index 6ad7623..1c4e46d 100644
--- a/drivers/kvm/types.h
+++ b/drivers/kvm/types.h
@@ -38,4 +38,17 @@ typedef unsigned long hva_t;
typedef u64 hpa_t;
typedef unsigned long hfn_t;
+struct kvm_pio_request {
+ unsigned long count;
+ int cur_count;
+ struct page *guest_pages[2];
+ unsigned guest_page_offset;
+ int in;
+ int port;
+ int size;
+ int string;
+ int down;
+ int rep;
+};
+
#endif /* __KVM_TYPES_H__ */
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index 0da4be9..a0dd447 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -11,8 +11,6 @@
#ifndef KVM_X86_H
#define KVM_X86_H
-#include "kvm.h"
-
#include <linux/types.h>
#include <linux/mm.h>
@@ -21,6 +19,8 @@
#include <asm/desc.h>
+#include "types.h"
+
#define CR3_PAE_RESERVED_BITS ((X86_CR3_PWT | X86_CR3_PCD) - 1)
#define CR3_NONPAE_RESERVED_BITS ((PAGE_SIZE-1) & ~(X86_CR3_PWT | X86_CR3_PCD))
#define CR3_L_MODE_RESERVED_BITS (CR3_NONPAE_RESERVED_BITS|0xFFFFFF0000000000ULL)
@@ -54,9 +54,19 @@
#define IOPL_SHIFT 12
+#define KVM_PERMILLE_MMU_PAGES 20
+#define KVM_MIN_ALLOC_MMU_PAGES 64
+#define KVM_NUM_MMU_PAGES 1024
+#define KVM_MIN_FREE_MMU_PAGES 5
+#define KVM_REFILL_PAGES 25
+#define KVM_MAX_CPUID_ENTRIES 40
+
extern spinlock_t kvm_lock;
extern struct list_head vm_list;
+struct kvm_vcpu;
+struct kvm;
+
enum {
VCPU_REGS_RAX = 0,
VCPU_REGS_RCX = 1,
@@ -92,6 +102,89 @@ enum {
#include "x86_emulate.h"
+#define KVM_NR_MEM_OBJS 40
+
+/*
+ * We don't want allocation failures within the mmu code, so we preallocate
+ * enough memory for a single page fault in a cache.
+ */
+struct kvm_mmu_memory_cache {
+ int nobjs;
+ void *objects[KVM_NR_MEM_OBJS];
+};
+
+#define NR_PTE_CHAIN_ENTRIES 5
+
+struct kvm_pte_chain {
+ u64 *parent_ptes[NR_PTE_CHAIN_ENTRIES];
+ struct hlist_node link;
+};
+
+/*
+ * kvm_mmu_page_role, below, is defined as:
+ *
+ * bits 0:3 - total guest paging levels (2-4, or zero for real mode)
+ * bits 4:7 - page table level for this shadow (1-4)
+ * bits 8:9 - page table quadrant for 2-level guests
+ * bit 16 - "metaphysical" - gfn is not a real page (huge page/real mode)
+ * bits 17:19 - common access permissions for all ptes in this shadow page
+ */
+union kvm_mmu_page_role {
+ unsigned word;
+ struct {
+ unsigned glevels : 4;
+ unsigned level : 4;
+ unsigned quadrant : 2;
+ unsigned pad_for_nice_hex_output : 6;
+ unsigned metaphysical : 1;
+ unsigned access : 3;
+ };
+};
+
+struct kvm_mmu_page {
+ struct list_head link;
+ struct hlist_node hash_link;
+
+ /*
+ * The following two entries are used to key the shadow page in the
+ * hash table.
+ */
+ gfn_t gfn;
+ union kvm_mmu_page_role role;
+
+ u64 *spt;
+ /* hold the gfn of each spte inside spt */
+ gfn_t *gfns;
+ unsigned long slot_bitmap; /* One bit set per slot which has memory
+ * in this shadow page.
+ */
+ int multimapped; /* More than one parent_pte? */
+ int root_count; /* Currently serving as active root */
+ union {
+ u64 *parent_pte; /* !multimapped */
+ struct hlist_head parent_ptes; /* multimapped, kvm_pte_chain */
+ };
+};
+
+/*
+ * x86 supports 3 paging modes (4-level 64-bit, 3-level 64-bit, and 2-level
+ * 32-bit). The kvm_mmu structure abstracts the details of the current mmu
+ * mode.
+ */
+struct kvm_mmu {
+ void (*new_cr3)(struct kvm_vcpu *vcpu);
+ int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err);
+ void (*free)(struct kvm_vcpu *vcpu);
+ gpa_t (*gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t gva);
+ void (*prefetch_page)(struct kvm_vcpu *vcpu,
+ struct kvm_mmu_page *page);
+ hpa_t root_hpa;
+ int root_level;
+ int shadow_root_level;
+
+ u64 *pae_root;
+};
+
struct kvm_vcpu_arch {
u64 host_tsc;
int interrupt_window_open;
@@ -162,11 +255,6 @@ struct kvm_vcpu_arch {
struct x86_emulate_ctxt emulate_ctxt;
};
-struct kvm_vcpu {
- KVM_VCPU_COMM;
-
- struct kvm_vcpu_arch arch;
-};
struct descriptor_table {
u16 limit;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 07/33] KVM: Portability: Expand the KVM_VCPU_COMM in kvm_vcpu structure.
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (5 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 06/33] KVM: Portability: Move kvm_vcpu definition back to kvm.h Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 08/33] KVM: Portability: Move kvm_vcpu_stat to x86.h Avi Kivity
` (25 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
This patches removes KVM_COMM macro, original it is hold
kvm_vcpu common fields.
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>
---
drivers/kvm/kvm.h | 53 ++++++++++++++++++++++-------------------------------
1 files changed, 22 insertions(+), 31 deletions(-)
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index 53717be..039faa7 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -87,41 +87,32 @@ struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus, gpa_t addr);
void kvm_io_bus_register_dev(struct kvm_io_bus *bus,
struct kvm_io_device *dev);
+struct kvm_vcpu {
+ struct kvm *kvm;
+ struct preempt_notifier preempt_notifier;
+ int vcpu_id;
+ struct mutex mutex;
+ int cpu;
+ struct kvm_run *run;
+ int guest_mode;
+ unsigned long requests;
+ struct kvm_guest_debug guest_debug;
+ int fpu_active;
+ int guest_fpu_loaded;
+ wait_queue_head_t wq;
+ int sigset_active;
+ sigset_t sigset;
+ struct kvm_vcpu_stat stat;
+
#ifdef CONFIG_HAS_IOMEM
-#define KVM_VCPU_MMIO \
- int mmio_needed; \
- int mmio_read_completed; \
- int mmio_is_write; \
- int mmio_size; \
- unsigned char mmio_data[8]; \
+ int mmio_needed;
+ int mmio_read_completed;
+ int mmio_is_write;
+ int mmio_size;
+ unsigned char mmio_data[8];
gpa_t mmio_phys_addr;
-
-#else
-#define KVM_VCPU_MMIO
-
#endif
-#define KVM_VCPU_COMM \
- struct kvm *kvm; \
- struct preempt_notifier preempt_notifier; \
- int vcpu_id; \
- struct mutex mutex; \
- int cpu; \
- struct kvm_run *run; \
- int guest_mode; \
- unsigned long requests; \
- struct kvm_guest_debug guest_debug; \
- int fpu_active; \
- int guest_fpu_loaded; \
- wait_queue_head_t wq; \
- int sigset_active; \
- sigset_t sigset; \
- struct kvm_vcpu_stat stat; \
- KVM_VCPU_MMIO
-
-struct kvm_vcpu {
- KVM_VCPU_COMM;
-
struct kvm_vcpu_arch arch;
};
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 08/33] KVM: Portability: Move kvm_vcpu_stat to x86.h
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (6 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 07/33] KVM: Portability: Expand the KVM_VCPU_COMM in kvm_vcpu structure Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 09/33] KVM: Portability: Move memslot aliases to new struct kvm_arch Avi Kivity
` (24 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
This patches moves kvm_vcpu_stat to x86.h, so every
arch can define its own kvm_vcpu_stat structure.
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>
---
drivers/kvm/kvm.h | 22 ----------------------
drivers/kvm/x86.h | 21 +++++++++++++++++++++
2 files changed, 21 insertions(+), 22 deletions(-)
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index 039faa7..9ea7149 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -47,28 +47,6 @@ struct kvm_guest_debug {
int singlestep;
};
-struct kvm_vcpu_stat {
- u32 pf_fixed;
- u32 pf_guest;
- u32 tlb_flush;
- u32 invlpg;
-
- u32 exits;
- u32 io_exits;
- u32 mmio_exits;
- u32 signal_exits;
- u32 irq_window_exits;
- u32 halt_exits;
- u32 halt_wakeup;
- u32 request_irq_exits;
- u32 irq_exits;
- u32 host_state_reload;
- u32 efer_reload;
- u32 fpu_reload;
- u32 insn_emulation;
- u32 insn_emulation_fail;
-};
-
/*
* It would be nice to use something smarter than a linear search, TBD...
* Thankfully we dont expect many devices to register (famous last words :),
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index a0dd447..4ca8485 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -255,6 +255,27 @@ struct kvm_vcpu_arch {
struct x86_emulate_ctxt emulate_ctxt;
};
+struct kvm_vcpu_stat {
+ u32 pf_fixed;
+ u32 pf_guest;
+ u32 tlb_flush;
+ u32 invlpg;
+
+ u32 exits;
+ u32 io_exits;
+ u32 mmio_exits;
+ u32 signal_exits;
+ u32 irq_window_exits;
+ u32 halt_exits;
+ u32 halt_wakeup;
+ u32 request_irq_exits;
+ u32 irq_exits;
+ u32 host_state_reload;
+ u32 efer_reload;
+ u32 fpu_reload;
+ u32 insn_emulation;
+ u32 insn_emulation_fail;
+};
struct descriptor_table {
u16 limit;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 09/33] KVM: Portability: Move memslot aliases to new struct kvm_arch
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (7 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 08/33] KVM: Portability: Move kvm_vcpu_stat to x86.h Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 10/33] KVM: Portability: Move mmu-related fields to kvm_arch Avi Kivity
` (23 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
This patches create kvm_arch to hold arch-specific kvm fileds
and moves fields naliases and aliases 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>
---
drivers/kvm/kvm.h | 10 +---------
drivers/kvm/x86.c | 10 +++++-----
drivers/kvm/x86.h | 13 +++++++++++++
3 files changed, 19 insertions(+), 14 deletions(-)
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index 9ea7149..bf5b85c 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -25,7 +25,6 @@
#include "x86.h"
#define KVM_MAX_VCPUS 4
-#define KVM_ALIAS_SLOTS 4
#define KVM_MEMORY_SLOTS 8
/* memory slots that does not exposed to userspace */
#define KVM_PRIVATE_MEM_SLOTS 4
@@ -94,12 +93,6 @@ struct kvm_vcpu {
struct kvm_vcpu_arch arch;
};
-struct kvm_mem_alias {
- gfn_t base_gfn;
- unsigned long npages;
- gfn_t target_gfn;
-};
-
struct kvm_memory_slot {
gfn_t base_gfn;
unsigned long npages;
@@ -123,8 +116,6 @@ struct kvm_vm_stat {
struct kvm {
struct mutex lock; /* protects everything except vcpus */
struct mm_struct *mm; /* userspace tied to this vm */
- int naliases;
- struct kvm_mem_alias aliases[KVM_ALIAS_SLOTS];
int nmemslots;
struct kvm_memory_slot memslots[KVM_MEMORY_SLOTS +
KVM_PRIVATE_MEM_SLOTS];
@@ -147,6 +138,7 @@ struct kvm {
unsigned int tss_addr;
struct page *apic_access_page;
struct kvm_vm_stat stat;
+ struct kvm_arch arch;
};
/* The guest did something we don't support. */
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index 60f9722..7e1bd52 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -1191,8 +1191,8 @@ gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn)
int i;
struct kvm_mem_alias *alias;
- for (i = 0; i < kvm->naliases; ++i) {
- alias = &kvm->aliases[i];
+ for (i = 0; i < kvm->arch.naliases; ++i) {
+ alias = &kvm->arch.aliases[i];
if (gfn >= alias->base_gfn
&& gfn < alias->base_gfn + alias->npages)
return alias->target_gfn + gfn - alias->base_gfn;
@@ -1228,15 +1228,15 @@ static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm,
mutex_lock(&kvm->lock);
- p = &kvm->aliases[alias->slot];
+ p = &kvm->arch.aliases[alias->slot];
p->base_gfn = alias->guest_phys_addr >> PAGE_SHIFT;
p->npages = alias->memory_size >> PAGE_SHIFT;
p->target_gfn = alias->target_phys_addr >> PAGE_SHIFT;
for (n = KVM_ALIAS_SLOTS; n > 0; --n)
- if (kvm->aliases[n - 1].npages)
+ if (kvm->arch.aliases[n - 1].npages)
break;
- kvm->naliases = n;
+ kvm->arch.naliases = n;
kvm_mmu_zap_all(kvm);
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index 4ca8485..be84f2b 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -54,6 +54,8 @@
#define IOPL_SHIFT 12
+#define KVM_ALIAS_SLOTS 4
+
#define KVM_PERMILLE_MMU_PAGES 20
#define KVM_MIN_ALLOC_MMU_PAGES 64
#define KVM_NUM_MMU_PAGES 1024
@@ -255,6 +257,17 @@ struct kvm_vcpu_arch {
struct x86_emulate_ctxt emulate_ctxt;
};
+struct kvm_mem_alias {
+ gfn_t base_gfn;
+ unsigned long npages;
+ gfn_t target_gfn;
+};
+
+struct kvm_arch{
+ int naliases;
+ struct kvm_mem_alias aliases[KVM_ALIAS_SLOTS];
+};
+
struct kvm_vcpu_stat {
u32 pf_fixed;
u32 pf_guest;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 10/33] KVM: Portability: Move mmu-related fields to kvm_arch
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (8 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 09/33] KVM: Portability: Move memslot aliases to new struct kvm_arch Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 11/33] KVM: Portability: move vpic and vioapic " Avi Kivity
` (22 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
This patches moves mmu-related fields 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>
---
drivers/kvm/kvm.h | 8 -------
drivers/kvm/mmu.c | 58 +++++++++++++++++++++++++++-------------------------
drivers/kvm/mmu.h | 2 +-
drivers/kvm/x86.c | 8 +++---
drivers/kvm/x86.h | 9 ++++++++
5 files changed, 44 insertions(+), 41 deletions(-)
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index bf5b85c..65de5e4 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -119,14 +119,6 @@ struct kvm {
int nmemslots;
struct kvm_memory_slot memslots[KVM_MEMORY_SLOTS +
KVM_PRIVATE_MEM_SLOTS];
- /*
- * Hash table of struct kvm_mmu_page.
- */
- struct list_head active_mmu_pages;
- unsigned int n_free_mmu_pages;
- unsigned int n_requested_mmu_pages;
- unsigned int n_alloc_mmu_pages;
- struct hlist_head mmu_page_hash[KVM_NUM_MMU_PAGES];
struct kvm_vcpu *vcpus[KVM_MAX_VCPUS];
struct list_head vm_list;
struct file *filp;
diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c
index 1dc0e8c..c26d83f 100644
--- a/drivers/kvm/mmu.c
+++ b/drivers/kvm/mmu.c
@@ -553,7 +553,7 @@ static void kvm_mmu_free_page(struct kvm *kvm, struct kvm_mmu_page *sp)
__free_page(virt_to_page(sp->spt));
__free_page(virt_to_page(sp->gfns));
kfree(sp);
- ++kvm->n_free_mmu_pages;
+ ++kvm->arch.n_free_mmu_pages;
}
static unsigned kvm_page_table_hashfn(gfn_t gfn)
@@ -566,19 +566,19 @@ static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu,
{
struct kvm_mmu_page *sp;
- if (!vcpu->kvm->n_free_mmu_pages)
+ if (!vcpu->kvm->arch.n_free_mmu_pages)
return NULL;
sp = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_header_cache, sizeof *sp);
sp->spt = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
sp->gfns = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
set_page_private(virt_to_page(sp->spt), (unsigned long)sp);
- list_add(&sp->link, &vcpu->kvm->active_mmu_pages);
+ list_add(&sp->link, &vcpu->kvm->arch.active_mmu_pages);
ASSERT(is_empty_shadow_page(sp->spt));
sp->slot_bitmap = 0;
sp->multimapped = 0;
sp->parent_pte = parent_pte;
- --vcpu->kvm->n_free_mmu_pages;
+ --vcpu->kvm->arch.n_free_mmu_pages;
return sp;
}
@@ -666,7 +666,7 @@ static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn)
pgprintk("%s: looking for gfn %lx\n", __FUNCTION__, gfn);
index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
- bucket = &kvm->mmu_page_hash[index];
+ bucket = &kvm->arch.mmu_page_hash[index];
hlist_for_each_entry(sp, node, bucket, hash_link)
if (sp->gfn == gfn && !sp->role.metaphysical) {
pgprintk("%s: found role %x\n",
@@ -705,7 +705,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
pgprintk("%s: looking gfn %lx role %x\n", __FUNCTION__,
gfn, role.word);
index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
- bucket = &vcpu->kvm->mmu_page_hash[index];
+ bucket = &vcpu->kvm->arch.mmu_page_hash[index];
hlist_for_each_entry(sp, node, bucket, hash_link)
if (sp->gfn == gfn && sp->role.word == role.word) {
mmu_page_add_parent_pte(vcpu, sp, parent_pte);
@@ -796,7 +796,7 @@ static void kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp)
hlist_del(&sp->hash_link);
kvm_mmu_free_page(kvm, sp);
} else
- list_move(&sp->link, &kvm->active_mmu_pages);
+ list_move(&sp->link, &kvm->arch.active_mmu_pages);
kvm_mmu_reset_last_pte_updated(kvm);
}
@@ -812,26 +812,26 @@ void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages)
* change the value
*/
- if ((kvm->n_alloc_mmu_pages - kvm->n_free_mmu_pages) >
+ if ((kvm->arch.n_alloc_mmu_pages - kvm->arch.n_free_mmu_pages) >
kvm_nr_mmu_pages) {
- int n_used_mmu_pages = kvm->n_alloc_mmu_pages
- - kvm->n_free_mmu_pages;
+ int n_used_mmu_pages = kvm->arch.n_alloc_mmu_pages
+ - kvm->arch.n_free_mmu_pages;
while (n_used_mmu_pages > kvm_nr_mmu_pages) {
struct kvm_mmu_page *page;
- page = container_of(kvm->active_mmu_pages.prev,
+ page = container_of(kvm->arch.active_mmu_pages.prev,
struct kvm_mmu_page, link);
kvm_mmu_zap_page(kvm, page);
n_used_mmu_pages--;
}
- kvm->n_free_mmu_pages = 0;
+ kvm->arch.n_free_mmu_pages = 0;
}
else
- kvm->n_free_mmu_pages += kvm_nr_mmu_pages
- - kvm->n_alloc_mmu_pages;
+ kvm->arch.n_free_mmu_pages += kvm_nr_mmu_pages
+ - kvm->arch.n_alloc_mmu_pages;
- kvm->n_alloc_mmu_pages = kvm_nr_mmu_pages;
+ kvm->arch.n_alloc_mmu_pages = kvm_nr_mmu_pages;
}
static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn)
@@ -845,7 +845,7 @@ static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn)
pgprintk("%s: looking for gfn %lx\n", __FUNCTION__, gfn);
r = 0;
index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
- bucket = &kvm->mmu_page_hash[index];
+ bucket = &kvm->arch.mmu_page_hash[index];
hlist_for_each_entry_safe(sp, node, n, bucket, hash_link)
if (sp->gfn == gfn && !sp->role.metaphysical) {
pgprintk("%s: gfn %lx role %x\n", __FUNCTION__, gfn,
@@ -1362,7 +1362,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
vcpu->arch.last_pte_updated = NULL;
}
index = kvm_page_table_hashfn(gfn) % KVM_NUM_MMU_PAGES;
- bucket = &vcpu->kvm->mmu_page_hash[index];
+ bucket = &vcpu->kvm->arch.mmu_page_hash[index];
hlist_for_each_entry_safe(sp, node, n, bucket, hash_link) {
if (sp->gfn != gfn || sp->role.metaphysical)
continue;
@@ -1428,10 +1428,10 @@ int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
{
- while (vcpu->kvm->n_free_mmu_pages < KVM_REFILL_PAGES) {
+ while (vcpu->kvm->arch.n_free_mmu_pages < KVM_REFILL_PAGES) {
struct kvm_mmu_page *sp;
- sp = container_of(vcpu->kvm->active_mmu_pages.prev,
+ sp = container_of(vcpu->kvm->arch.active_mmu_pages.prev,
struct kvm_mmu_page, link);
kvm_mmu_zap_page(vcpu->kvm, sp);
++vcpu->kvm->stat.mmu_recycled;
@@ -1482,8 +1482,8 @@ static void free_mmu_pages(struct kvm_vcpu *vcpu)
{
struct kvm_mmu_page *sp;
- while (!list_empty(&vcpu->kvm->active_mmu_pages)) {
- sp = container_of(vcpu->kvm->active_mmu_pages.next,
+ while (!list_empty(&vcpu->kvm->arch.active_mmu_pages)) {
+ sp = container_of(vcpu->kvm->arch.active_mmu_pages.next,
struct kvm_mmu_page, link);
kvm_mmu_zap_page(vcpu->kvm, sp);
}
@@ -1497,10 +1497,12 @@ static int alloc_mmu_pages(struct kvm_vcpu *vcpu)
ASSERT(vcpu);
- if (vcpu->kvm->n_requested_mmu_pages)
- vcpu->kvm->n_free_mmu_pages = vcpu->kvm->n_requested_mmu_pages;
+ if (vcpu->kvm->arch.n_requested_mmu_pages)
+ vcpu->kvm->arch.n_free_mmu_pages =
+ vcpu->kvm->arch.n_requested_mmu_pages;
else
- vcpu->kvm->n_free_mmu_pages = vcpu->kvm->n_alloc_mmu_pages;
+ vcpu->kvm->arch.n_free_mmu_pages =
+ vcpu->kvm->arch.n_alloc_mmu_pages;
/*
* When emulating 32-bit mode, cr3 is only 32 bits even on x86_64.
* Therefore we need to allocate shadow page tables in the first
@@ -1549,7 +1551,7 @@ void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot)
{
struct kvm_mmu_page *sp;
- list_for_each_entry(sp, &kvm->active_mmu_pages, link) {
+ list_for_each_entry(sp, &kvm->arch.active_mmu_pages, link) {
int i;
u64 *pt;
@@ -1568,7 +1570,7 @@ void kvm_mmu_zap_all(struct kvm *kvm)
{
struct kvm_mmu_page *sp, *node;
- list_for_each_entry_safe(sp, node, &kvm->active_mmu_pages, link)
+ list_for_each_entry_safe(sp, node, &kvm->arch.active_mmu_pages, link)
kvm_mmu_zap_page(kvm, sp);
kvm_flush_remote_tlbs(kvm);
@@ -1738,7 +1740,7 @@ static int count_writable_mappings(struct kvm_vcpu *vcpu)
struct kvm_mmu_page *sp;
int i;
- list_for_each_entry(sp, &vcpu->kvm->active_mmu_pages, link) {
+ list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) {
u64 *pt = sp->spt;
if (sp->role.level != PT_PAGE_TABLE_LEVEL)
@@ -1774,7 +1776,7 @@ static void audit_write_protection(struct kvm_vcpu *vcpu)
unsigned long *rmapp;
gfn_t gfn;
- list_for_each_entry(sp, &vcpu->kvm->active_mmu_pages, link) {
+ list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) {
if (sp->role.metaphysical)
continue;
diff --git a/drivers/kvm/mmu.h b/drivers/kvm/mmu.h
index 9ebfd1c..cbfc272 100644
--- a/drivers/kvm/mmu.h
+++ b/drivers/kvm/mmu.h
@@ -5,7 +5,7 @@
static inline void kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
{
- if (unlikely(vcpu->kvm->n_free_mmu_pages < KVM_MIN_FREE_MMU_PAGES))
+ if (unlikely(vcpu->kvm->arch.n_free_mmu_pages < KVM_MIN_FREE_MMU_PAGES))
__kvm_mmu_free_some_pages(vcpu);
}
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index 7e1bd52..c0e95fb 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -1175,7 +1175,7 @@ static int kvm_vm_ioctl_set_nr_mmu_pages(struct kvm *kvm,
mutex_lock(&kvm->lock);
kvm_mmu_change_mmu_pages(kvm, kvm_nr_mmu_pages);
- kvm->n_requested_mmu_pages = kvm_nr_mmu_pages;
+ kvm->arch.n_requested_mmu_pages = kvm_nr_mmu_pages;
mutex_unlock(&kvm->lock);
return 0;
@@ -1183,7 +1183,7 @@ static int kvm_vm_ioctl_set_nr_mmu_pages(struct kvm *kvm,
static int kvm_vm_ioctl_get_nr_mmu_pages(struct kvm *kvm)
{
- return kvm->n_alloc_mmu_pages;
+ return kvm->arch.n_alloc_mmu_pages;
}
gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn)
@@ -3051,7 +3051,7 @@ struct kvm *kvm_arch_create_vm(void)
if (!kvm)
return ERR_PTR(-ENOMEM);
- INIT_LIST_HEAD(&kvm->active_mmu_pages);
+ INIT_LIST_HEAD(&kvm->arch.active_mmu_pages);
return kvm;
}
@@ -3130,7 +3130,7 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
}
}
- if (!kvm->n_requested_mmu_pages) {
+ if (!kvm->arch.n_requested_mmu_pages) {
unsigned int nr_mmu_pages = kvm_mmu_calculate_mmu_pages(kvm);
kvm_mmu_change_mmu_pages(kvm, nr_mmu_pages);
}
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index be84f2b..5cdc366 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -266,6 +266,15 @@ struct kvm_mem_alias {
struct kvm_arch{
int naliases;
struct kvm_mem_alias aliases[KVM_ALIAS_SLOTS];
+
+ unsigned int n_free_mmu_pages;
+ unsigned int n_requested_mmu_pages;
+ unsigned int n_alloc_mmu_pages;
+ struct hlist_head mmu_page_hash[KVM_NUM_MMU_PAGES];
+ /*
+ * Hash table of struct kvm_mmu_page.
+ */
+ struct list_head active_mmu_pages;
};
struct kvm_vcpu_stat {
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 11/33] KVM: Portability: move vpic and vioapic to kvm_arch
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (9 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 10/33] KVM: Portability: Move mmu-related fields to kvm_arch Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 12/33] KVM: Portability: Move round_robin_prev_vcpu and tss_addr " Avi Kivity
` (21 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
This patches moves two fields vpid and vioapic 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>
---
drivers/kvm/ioapic.c | 4 ++--
drivers/kvm/irq.h | 4 ++--
drivers/kvm/kvm.h | 2 --
drivers/kvm/x86.c | 14 +++++++-------
drivers/kvm/x86.h | 2 ++
5 files changed, 13 insertions(+), 13 deletions(-)
diff --git a/drivers/kvm/ioapic.c b/drivers/kvm/ioapic.c
index 0491036..f823677 100644
--- a/drivers/kvm/ioapic.c
+++ b/drivers/kvm/ioapic.c
@@ -271,7 +271,7 @@ static int get_eoi_gsi(struct kvm_ioapic *ioapic, int vector)
void kvm_ioapic_update_eoi(struct kvm *kvm, int vector)
{
- struct kvm_ioapic *ioapic = kvm->vioapic;
+ struct kvm_ioapic *ioapic = kvm->arch.vioapic;
union ioapic_redir_entry *ent;
int gsi;
@@ -390,7 +390,7 @@ int kvm_ioapic_init(struct kvm *kvm)
ioapic = kzalloc(sizeof(struct kvm_ioapic), GFP_KERNEL);
if (!ioapic)
return -ENOMEM;
- kvm->vioapic = ioapic;
+ kvm->arch.vioapic = ioapic;
kvm_ioapic_reset(ioapic);
ioapic->dev.read = ioapic_mmio_read;
ioapic->dev.write = ioapic_mmio_write;
diff --git a/drivers/kvm/irq.h b/drivers/kvm/irq.h
index a688c51..6e023dc 100644
--- a/drivers/kvm/irq.h
+++ b/drivers/kvm/irq.h
@@ -148,12 +148,12 @@ do { \
static inline struct kvm_pic *pic_irqchip(struct kvm *kvm)
{
- return kvm->vpic;
+ return kvm->arch.vpic;
}
static inline struct kvm_ioapic *ioapic_irqchip(struct kvm *kvm)
{
- return kvm->vioapic;
+ return kvm->arch.vioapic;
}
static inline int irqchip_in_kernel(struct kvm *kvm)
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index 65de5e4..d9ce916 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -124,8 +124,6 @@ struct kvm {
struct file *filp;
struct kvm_io_bus mmio_bus;
struct kvm_io_bus pio_bus;
- struct kvm_pic *vpic;
- struct kvm_ioapic *vioapic;
int round_robin_prev_vcpu;
unsigned int tss_addr;
struct page *apic_access_page;
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index c0e95fb..b37c009 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -1386,12 +1386,12 @@ long kvm_arch_vm_ioctl(struct file *filp,
}
case KVM_CREATE_IRQCHIP:
r = -ENOMEM;
- kvm->vpic = kvm_create_pic(kvm);
- if (kvm->vpic) {
+ kvm->arch.vpic = kvm_create_pic(kvm);
+ if (kvm->arch.vpic) {
r = kvm_ioapic_init(kvm);
if (r) {
- kfree(kvm->vpic);
- kvm->vpic = NULL;
+ kfree(kvm->arch.vpic);
+ kvm->arch.vpic = NULL;
goto out;
}
} else
@@ -1409,7 +1409,7 @@ long kvm_arch_vm_ioctl(struct file *filp,
kvm_pic_set_irq(pic_irqchip(kvm),
irq_event.irq,
irq_event.level);
- kvm_ioapic_set_irq(kvm->vioapic,
+ kvm_ioapic_set_irq(kvm->arch.vioapic,
irq_event.irq,
irq_event.level);
mutex_unlock(&kvm->lock);
@@ -3084,8 +3084,8 @@ static void kvm_free_vcpus(struct kvm *kvm)
void kvm_arch_destroy_vm(struct kvm *kvm)
{
- kfree(kvm->vpic);
- kfree(kvm->vioapic);
+ kfree(kvm->arch.vpic);
+ kfree(kvm->arch.vioapic);
kvm_free_vcpus(kvm);
kvm_free_physmem(kvm);
kfree(kvm);
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index 5cdc366..ef23a30 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -275,6 +275,8 @@ struct kvm_arch{
* Hash table of struct kvm_mmu_page.
*/
struct list_head active_mmu_pages;
+ struct kvm_pic *vpic;
+ struct kvm_ioapic *vioapic;
};
struct kvm_vcpu_stat {
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 12/33] KVM: Portability: Move round_robin_prev_vcpu and tss_addr to kvm_arch
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (10 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 11/33] KVM: Portability: move vpic and vioapic " Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 13/33] KVM: Portability: Move kvm_vm_stat to x86.h Avi Kivity
` (20 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
This patches moves two fields round_robin_prev_vcpu and tss 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>
---
drivers/kvm/kvm.h | 3 ---
drivers/kvm/lapic.c | 4 ++--
drivers/kvm/vmx.c | 12 ++++++------
drivers/kvm/x86.h | 4 ++++
4 files changed, 12 insertions(+), 11 deletions(-)
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index d9ce916..ab9657a 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -124,9 +124,6 @@ struct kvm {
struct file *filp;
struct kvm_io_bus mmio_bus;
struct kvm_io_bus pio_bus;
- int round_robin_prev_vcpu;
- unsigned int tss_addr;
- struct page *apic_access_page;
struct kvm_vm_stat stat;
struct kvm_arch arch;
};
diff --git a/drivers/kvm/lapic.c b/drivers/kvm/lapic.c
index 5c9f467..8c74bf1 100644
--- a/drivers/kvm/lapic.c
+++ b/drivers/kvm/lapic.c
@@ -404,7 +404,7 @@ static struct kvm_lapic *kvm_apic_round_robin(struct kvm *kvm, u8 vector,
int next;
struct kvm_lapic *apic = NULL;
- last = kvm->round_robin_prev_vcpu;
+ last = kvm->arch.round_robin_prev_vcpu;
next = last;
do {
@@ -417,7 +417,7 @@ static struct kvm_lapic *kvm_apic_round_robin(struct kvm *kvm, u8 vector,
break;
apic = NULL;
} while (next != last);
- kvm->round_robin_prev_vcpu = next;
+ kvm->arch.round_robin_prev_vcpu = next;
if (!apic)
printk(KERN_DEBUG "vcpu not ready for apic_round_robin\n");
diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c
index 456eb9c..13c6a51 100644
--- a/drivers/kvm/vmx.c
+++ b/drivers/kvm/vmx.c
@@ -1143,12 +1143,12 @@ static void enter_pmode(struct kvm_vcpu *vcpu)
static gva_t rmode_tss_base(struct kvm *kvm)
{
- if (!kvm->tss_addr) {
+ if (!kvm->arch.tss_addr) {
gfn_t base_gfn = kvm->memslots[0].base_gfn +
kvm->memslots[0].npages - 3;
return base_gfn << PAGE_SHIFT;
}
- return kvm->tss_addr;
+ return kvm->arch.tss_addr;
}
static void fix_rmode_seg(int seg, struct kvm_save_segment *save)
@@ -1472,7 +1472,7 @@ static int alloc_apic_access_page(struct kvm *kvm)
int r = 0;
mutex_lock(&kvm->lock);
- if (kvm->apic_access_page)
+ if (kvm->arch.apic_access_page)
goto out;
kvm_userspace_mem.slot = APIC_ACCESS_PAGE_PRIVATE_MEMSLOT;
kvm_userspace_mem.flags = 0;
@@ -1481,7 +1481,7 @@ static int alloc_apic_access_page(struct kvm *kvm)
r = __kvm_set_memory_region(kvm, &kvm_userspace_mem, 0);
if (r)
goto out;
- kvm->apic_access_page = gfn_to_page(kvm, 0xfee00);
+ kvm->arch.apic_access_page = gfn_to_page(kvm, 0xfee00);
out:
mutex_unlock(&kvm->lock);
return r;
@@ -1698,7 +1698,7 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu)
if (vm_need_virtualize_apic_accesses(vmx->vcpu.kvm))
vmcs_write64(APIC_ACCESS_ADDR,
- page_to_phys(vmx->vcpu.kvm->apic_access_page));
+ page_to_phys(vmx->vcpu.kvm->arch.apic_access_page));
vmx->vcpu.arch.cr0 = 0x60000010;
vmx_set_cr0(&vmx->vcpu, vmx->vcpu.arch.cr0); /* enter rmode */
@@ -1788,7 +1788,7 @@ static int vmx_set_tss_addr(struct kvm *kvm, unsigned int addr)
ret = kvm_set_memory_region(kvm, &tss_mem, 0);
if (ret)
return ret;
- kvm->tss_addr = addr;
+ kvm->arch.tss_addr = addr;
return 0;
}
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index ef23a30..9455839 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -277,6 +277,10 @@ struct kvm_arch{
struct list_head active_mmu_pages;
struct kvm_pic *vpic;
struct kvm_ioapic *vioapic;
+
+ int round_robin_prev_vcpu;
+ unsigned int tss_addr;
+ struct page *apic_access_page;
};
struct kvm_vcpu_stat {
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 13/33] KVM: Portability: Move kvm_vm_stat to x86.h
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (11 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 12/33] KVM: Portability: Move round_robin_prev_vcpu and tss_addr " Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 14/33] KVM: VMX: Add printk_ratelimit in vmx_intr_assist Avi Kivity
` (19 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
This patch moves kvm_vm_stat to x86.h, and every arch
can define its own kvm_vm_stat in $arch.h
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>
---
drivers/kvm/kvm.h | 10 ----------
drivers/kvm/x86.h | 10 ++++++++++
2 files changed, 10 insertions(+), 10 deletions(-)
diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h
index ab9657a..bf6a3b3 100644
--- a/drivers/kvm/kvm.h
+++ b/drivers/kvm/kvm.h
@@ -103,16 +103,6 @@ struct kvm_memory_slot {
int user_alloc;
};
-struct kvm_vm_stat {
- u32 mmu_shadow_zapped;
- u32 mmu_pte_write;
- u32 mmu_pte_updated;
- u32 mmu_pde_zapped;
- u32 mmu_flooded;
- u32 mmu_recycled;
- u32 remote_tlb_flush;
-};
-
struct kvm {
struct mutex lock; /* protects everything except vcpus */
struct mm_struct *mm; /* userspace tied to this vm */
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index 9455839..dfb8091 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -283,6 +283,16 @@ struct kvm_arch{
struct page *apic_access_page;
};
+struct kvm_vm_stat {
+ u32 mmu_shadow_zapped;
+ u32 mmu_pte_write;
+ u32 mmu_pte_updated;
+ u32 mmu_pde_zapped;
+ u32 mmu_flooded;
+ u32 mmu_recycled;
+ u32 remote_tlb_flush;
+};
+
struct kvm_vcpu_stat {
u32 pf_fixed;
u32 pf_guest;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 14/33] KVM: VMX: Add printk_ratelimit in vmx_intr_assist
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (12 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 13/33] KVM: Portability: Move kvm_vm_stat to x86.h Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 15/33] KVM: Move arch dependent files to new directory arch/x86/kvm/ Avi Kivity
` (18 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Ryan Harper
From: Ryan Harper <ryanh@us.ibm.com>
Add printk_ratelimit check in front of printk. This prevents spamming
of the message during 32-bit ubuntu 6.06server install. Previously, it
would hang during the partition formatting stage.
Signed-off-by: Ryan Harper <ryanh@us.ibm.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
drivers/kvm/vmx.c | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c
index 13c6a51..4407998 100644
--- a/drivers/kvm/vmx.c
+++ b/drivers/kvm/vmx.c
@@ -2262,7 +2262,8 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu)
if (intr_info_field & INTR_INFO_VALID_MASK) {
if (idtv_info_field & INTR_INFO_VALID_MASK) {
/* TODO: fault when IDT_Vectoring */
- printk(KERN_ERR "Fault when IDT_Vectoring\n");
+ if (printk_ratelimit())
+ printk(KERN_ERR "Fault when IDT_Vectoring\n");
}
if (has_ext_irq)
enable_irq_window(vcpu);
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 15/33] KVM: Move arch dependent files to new directory arch/x86/kvm/
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (13 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 14/33] KVM: VMX: Add printk_ratelimit in vmx_intr_assist Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 16/33] KVM: Move drivers/kvm/* to virt/kvm/ Avi Kivity
` (17 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel
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>
---
arch/x86/Kconfig | 2 ++
arch/x86/Makefile | 2 ++
{drivers => arch/x86}/kvm/Kconfig | 0
{drivers => arch/x86}/kvm/Makefile | 7 ++++++-
{drivers => arch/x86}/kvm/i8259.c | 3 ++-
{drivers => arch/x86}/kvm/ioapic.c | 4 +---
{drivers => arch/x86}/kvm/irq.c | 3 +--
{drivers => arch/x86}/kvm/irq.h | 3 +--
{drivers => arch/x86}/kvm/kvm_svm.h | 2 +-
{drivers => arch/x86}/kvm/lapic.c | 4 +---
{drivers => arch/x86}/kvm/mmu.c | 3 +--
{drivers => arch/x86}/kvm/mmu.h | 2 +-
{drivers => arch/x86}/kvm/paging_tmpl.h | 0
{drivers => arch/x86}/kvm/segment_descriptor.h | 0
{drivers => arch/x86}/kvm/svm.c | 4 ++--
{drivers => arch/x86}/kvm/svm.h | 0
{drivers => arch/x86}/kvm/vmx.c | 4 +---
{drivers => arch/x86}/kvm/vmx.h | 0
{drivers => arch/x86}/kvm/x86.c | 4 +---
{drivers => arch/x86}/kvm/x86_emulate.c | 5 ++---
drivers/Kconfig | 2 --
drivers/Makefile | 1 -
drivers/kvm/iodev.h | 2 +-
drivers/kvm/kvm_main.c | 2 +-
drivers/kvm/x86.h => include/asm-x86/kvm_host.h | 9 ++++-----
.../asm-x86/kvm_x86_emulate.h | 0
drivers/kvm/kvm.h => include/linux/kvm_host.h | 8 ++++----
drivers/kvm/types.h => include/linux/kvm_types.h | 0
28 files changed, 35 insertions(+), 41 deletions(-)
rename {drivers => arch/x86}/kvm/Kconfig (100%)
rename {drivers => arch/x86}/kvm/Makefile (52%)
rename {drivers => arch/x86}/kvm/i8259.c (99%)
rename {drivers => arch/x86}/kvm/ioapic.c (99%)
rename {drivers => arch/x86}/kvm/irq.c (98%)
rename {drivers => arch/x86}/kvm/irq.h (99%)
rename {drivers => arch/x86}/kvm/kvm_svm.h (96%)
rename {drivers => arch/x86}/kvm/lapic.c (99%)
rename {drivers => arch/x86}/kvm/mmu.c (99%)
rename {drivers => arch/x86}/kvm/mmu.h (96%)
rename {drivers => arch/x86}/kvm/paging_tmpl.h (100%)
rename {drivers => arch/x86}/kvm/segment_descriptor.h (100%)
rename {drivers => arch/x86}/kvm/svm.c (99%)
rename {drivers => arch/x86}/kvm/svm.h (100%)
rename {drivers => arch/x86}/kvm/vmx.c (99%)
rename {drivers => arch/x86}/kvm/vmx.h (100%)
rename {drivers => arch/x86}/kvm/x86.c (99%)
rename {drivers => arch/x86}/kvm/x86_emulate.c (99%)
rename drivers/kvm/x86.h => include/asm-x86/kvm_host.h (99%)
rename drivers/kvm/x86_emulate.h => include/asm-x86/kvm_x86_emulate.h (100%)
rename drivers/kvm/kvm.h => include/linux/kvm_host.h (98%)
rename drivers/kvm/types.h => include/linux/kvm_types.h (100%)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index e029a93..ab2df55 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1623,4 +1623,6 @@ source "security/Kconfig"
source "crypto/Kconfig"
+source "arch/x86/kvm/Kconfig"
+
source "lib/Kconfig"
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 7aa1dc6..96f79eb 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -17,3 +17,5 @@ else
UTS_MACHINE := x86_64
include $(srctree)/arch/x86/Makefile_64
endif
+
+core-$(CONFIG_KVM) += arch/x86/kvm/
diff --git a/drivers/kvm/Kconfig b/arch/x86/kvm/Kconfig
similarity index 100%
rename from drivers/kvm/Kconfig
rename to arch/x86/kvm/Kconfig
diff --git a/drivers/kvm/Makefile b/arch/x86/kvm/Makefile
similarity index 52%
rename from drivers/kvm/Makefile
rename to arch/x86/kvm/Makefile
index cf18ad4..880ffe4 100644
--- a/drivers/kvm/Makefile
+++ b/arch/x86/kvm/Makefile
@@ -2,7 +2,12 @@
# Makefile for Kernel-based Virtual Machine module
#
-kvm-objs := kvm_main.o x86.o mmu.o x86_emulate.o i8259.o irq.o lapic.o ioapic.o
+common-objs = $(addprefix ../../../drivers/kvm/, kvm_main.o)
+
+EXTRA_CFLAGS += -I drivers/kvm
+
+kvm-objs := $(common-objs) x86.o mmu.o x86_emulate.o i8259.o irq.o lapic.o \
+ ioapic.o
obj-$(CONFIG_KVM) += kvm.o
kvm-intel-objs = vmx.o
obj-$(CONFIG_KVM_INTEL) += kvm-intel.o
diff --git a/drivers/kvm/i8259.c b/arch/x86/kvm/i8259.c
similarity index 99%
rename from drivers/kvm/i8259.c
rename to arch/x86/kvm/i8259.c
index b3cad63..ab29cf2 100644
--- a/drivers/kvm/i8259.c
+++ b/arch/x86/kvm/i8259.c
@@ -27,7 +27,8 @@
*/
#include <linux/mm.h>
#include "irq.h"
-#include "kvm.h"
+
+#include <linux/kvm_host.h>
/*
* set irq level. If an edge is detected, then the IRR is set to 1
diff --git a/drivers/kvm/ioapic.c b/arch/x86/kvm/ioapic.c
similarity index 99%
rename from drivers/kvm/ioapic.c
rename to arch/x86/kvm/ioapic.c
index f823677..72f12f7 100644
--- a/drivers/kvm/ioapic.c
+++ b/arch/x86/kvm/ioapic.c
@@ -26,9 +26,7 @@
* Based on Xen 3.1 code.
*/
-#include "kvm.h"
-#include "x86.h"
-
+#include <linux/kvm_host.h>
#include <linux/kvm.h>
#include <linux/mm.h>
#include <linux/highmem.h>
diff --git a/drivers/kvm/irq.c b/arch/x86/kvm/irq.c
similarity index 98%
rename from drivers/kvm/irq.c
rename to arch/x86/kvm/irq.c
index 59b47c5..07a09aa 100644
--- a/drivers/kvm/irq.c
+++ b/arch/x86/kvm/irq.c
@@ -20,9 +20,8 @@
*/
#include <linux/module.h>
+#include <linux/kvm_host.h>
-#include "kvm.h"
-#include "x86.h"
#include "irq.h"
/*
diff --git a/drivers/kvm/irq.h b/arch/x86/kvm/irq.h
similarity index 99%
rename from drivers/kvm/irq.h
rename to arch/x86/kvm/irq.h
index 6e023dc..6316638 100644
--- a/drivers/kvm/irq.h
+++ b/arch/x86/kvm/irq.h
@@ -24,9 +24,8 @@
#include <linux/mm_types.h>
#include <linux/hrtimer.h>
-#include <asm/kvm.h>
+#include <linux/kvm_host.h>
#include "iodev.h"
-#include "kvm.h"
struct kvm;
struct kvm_vcpu;
diff --git a/drivers/kvm/kvm_svm.h b/arch/x86/kvm/kvm_svm.h
similarity index 96%
rename from drivers/kvm/kvm_svm.h
rename to arch/x86/kvm/kvm_svm.h
index a0e415d..ecdfe97 100644
--- a/drivers/kvm/kvm_svm.h
+++ b/arch/x86/kvm/kvm_svm.h
@@ -4,10 +4,10 @@
#include <linux/kernel.h>
#include <linux/types.h>
#include <linux/list.h>
+#include <linux/kvm_host.h>
#include <asm/msr.h>
#include "svm.h"
-#include "kvm.h"
static const u32 host_save_user_msrs[] = {
#ifdef CONFIG_X86_64
diff --git a/drivers/kvm/lapic.c b/arch/x86/kvm/lapic.c
similarity index 99%
rename from drivers/kvm/lapic.c
rename to arch/x86/kvm/lapic.c
index 8c74bf1..4076331 100644
--- a/drivers/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -17,9 +17,7 @@
* the COPYING file in the top-level directory.
*/
-#include "kvm.h"
-#include "x86.h"
-
+#include <linux/kvm_host.h>
#include <linux/kvm.h>
#include <linux/mm.h>
#include <linux/highmem.h>
diff --git a/drivers/kvm/mmu.c b/arch/x86/kvm/mmu.c
similarity index 99%
rename from drivers/kvm/mmu.c
rename to arch/x86/kvm/mmu.c
index c26d83f..401eb7c 100644
--- a/drivers/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -18,10 +18,9 @@
*/
#include "vmx.h"
-#include "kvm.h"
-#include "x86.h"
#include "mmu.h"
+#include <linux/kvm_host.h>
#include <linux/types.h>
#include <linux/string.h>
#include <linux/mm.h>
diff --git a/drivers/kvm/mmu.h b/arch/x86/kvm/mmu.h
similarity index 96%
rename from drivers/kvm/mmu.h
rename to arch/x86/kvm/mmu.h
index cbfc272..1fce19e 100644
--- a/drivers/kvm/mmu.h
+++ b/arch/x86/kvm/mmu.h
@@ -1,7 +1,7 @@
#ifndef __KVM_X86_MMU_H
#define __KVM_X86_MMU_H
-#include "kvm.h"
+#include <linux/kvm_host.h>
static inline void kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
{
diff --git a/drivers/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
similarity index 100%
rename from drivers/kvm/paging_tmpl.h
rename to arch/x86/kvm/paging_tmpl.h
diff --git a/drivers/kvm/segment_descriptor.h b/arch/x86/kvm/segment_descriptor.h
similarity index 100%
rename from drivers/kvm/segment_descriptor.h
rename to arch/x86/kvm/segment_descriptor.h
diff --git a/drivers/kvm/svm.c b/arch/x86/kvm/svm.c
similarity index 99%
rename from drivers/kvm/svm.c
rename to arch/x86/kvm/svm.c
index 35ad96d..745b1ec 100644
--- a/drivers/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -13,9 +13,9 @@
* the COPYING file in the top-level directory.
*
*/
-#include "x86.h"
+#include <linux/kvm_host.h>
+
#include "kvm_svm.h"
-#include "x86_emulate.h"
#include "irq.h"
#include "mmu.h"
diff --git a/drivers/kvm/svm.h b/arch/x86/kvm/svm.h
similarity index 100%
rename from drivers/kvm/svm.h
rename to arch/x86/kvm/svm.h
diff --git a/drivers/kvm/vmx.c b/arch/x86/kvm/vmx.c
similarity index 99%
rename from drivers/kvm/vmx.c
rename to arch/x86/kvm/vmx.c
index 4407998..20c0f5e 100644
--- a/drivers/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -15,14 +15,12 @@
*
*/
-#include "kvm.h"
-#include "x86.h"
-#include "x86_emulate.h"
#include "irq.h"
#include "vmx.h"
#include "segment_descriptor.h"
#include "mmu.h"
+#include <linux/kvm_host.h>
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/mm.h>
diff --git a/drivers/kvm/vmx.h b/arch/x86/kvm/vmx.h
similarity index 100%
rename from drivers/kvm/vmx.h
rename to arch/x86/kvm/vmx.h
diff --git a/drivers/kvm/x86.c b/arch/x86/kvm/x86.c
similarity index 99%
rename from drivers/kvm/x86.c
rename to arch/x86/kvm/x86.c
index b37c009..5902c5c 100644
--- a/drivers/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -14,9 +14,7 @@
*
*/
-#include "kvm.h"
-#include "x86.h"
-#include "x86_emulate.h"
+#include <linux/kvm_host.h>
#include "segment_descriptor.h"
#include "irq.h"
#include "mmu.h"
diff --git a/drivers/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c
similarity index 99%
rename from drivers/kvm/x86_emulate.c
rename to arch/x86/kvm/x86_emulate.c
index 50b133f..7958600 100644
--- a/drivers/kvm/x86_emulate.c
+++ b/arch/x86/kvm/x86_emulate.c
@@ -25,12 +25,11 @@
#include <public/xen.h>
#define DPRINTF(_f, _a ...) printf(_f , ## _a)
#else
-#include "kvm.h"
-#include "x86.h"
+#include <linux/kvm_host.h>
#define DPRINTF(x...) do {} while (0)
#endif
-#include "x86_emulate.h"
#include <linux/module.h>
+#include <asm/kvm_x86_emulate.h>
/*
* Opcode effective-address decode tables.
diff --git a/drivers/Kconfig b/drivers/Kconfig
index f4076d9..08d4ae2 100644
--- a/drivers/Kconfig
+++ b/drivers/Kconfig
@@ -90,8 +90,6 @@ source "drivers/dca/Kconfig"
source "drivers/auxdisplay/Kconfig"
-source "drivers/kvm/Kconfig"
-
source "drivers/uio/Kconfig"
source "drivers/virtio/Kconfig"
diff --git a/drivers/Makefile b/drivers/Makefile
index 8cb37e3..513ae86 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -47,7 +47,6 @@ obj-$(CONFIG_SPI) += spi/
obj-$(CONFIG_PCCARD) += pcmcia/
obj-$(CONFIG_DIO) += dio/
obj-$(CONFIG_SBUS) += sbus/
-obj-$(CONFIG_KVM) += kvm/
obj-$(CONFIG_ZORRO) += zorro/
obj-$(CONFIG_MAC) += macintosh/
obj-$(CONFIG_ATA_OVER_ETH) += block/aoe/
diff --git a/drivers/kvm/iodev.h b/drivers/kvm/iodev.h
index eb9e8a7..c14e642 100644
--- a/drivers/kvm/iodev.h
+++ b/drivers/kvm/iodev.h
@@ -16,7 +16,7 @@
#ifndef __KVM_IODEV_H__
#define __KVM_IODEV_H__
-#include "types.h"
+#include <linux/kvm_types.h>
struct kvm_io_device {
void (*read)(struct kvm_io_device *this,
diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
index 5bfa200..a5ee518 100644
--- a/drivers/kvm/kvm_main.c
+++ b/drivers/kvm/kvm_main.c
@@ -15,9 +15,9 @@
*
*/
-#include "kvm.h"
#include "iodev.h"
+#include <linux/kvm_host.h>
#include <linux/kvm.h>
#include <linux/module.h>
#include <linux/errno.h>
diff --git a/drivers/kvm/x86.h b/include/asm-x86/kvm_host.h
similarity index 99%
rename from drivers/kvm/x86.h
rename to include/asm-x86/kvm_host.h
index dfb8091..28940e1 100644
--- a/drivers/kvm/x86.h
+++ b/include/asm-x86/kvm_host.h
@@ -8,19 +8,18 @@
*
*/
-#ifndef KVM_X86_H
-#define KVM_X86_H
+#ifndef ASM_KVM_HOST_H
+#define ASM_KVM_HOST_H
#include <linux/types.h>
#include <linux/mm.h>
#include <linux/kvm.h>
#include <linux/kvm_para.h>
+#include <linux/kvm_types.h>
#include <asm/desc.h>
-#include "types.h"
-
#define CR3_PAE_RESERVED_BITS ((X86_CR3_PWT | X86_CR3_PCD) - 1)
#define CR3_NONPAE_RESERVED_BITS ((PAGE_SIZE-1) & ~(X86_CR3_PWT | X86_CR3_PCD))
#define CR3_L_MODE_RESERVED_BITS (CR3_NONPAE_RESERVED_BITS|0xFFFFFF0000000000ULL)
@@ -102,7 +101,7 @@ enum {
VCPU_SREG_LDTR,
};
-#include "x86_emulate.h"
+#include <asm/kvm_x86_emulate.h>
#define KVM_NR_MEM_OBJS 40
diff --git a/drivers/kvm/x86_emulate.h b/include/asm-x86/kvm_x86_emulate.h
similarity index 100%
rename from drivers/kvm/x86_emulate.h
rename to include/asm-x86/kvm_x86_emulate.h
diff --git a/drivers/kvm/kvm.h b/include/linux/kvm_host.h
similarity index 98%
rename from drivers/kvm/kvm.h
rename to include/linux/kvm_host.h
index bf6a3b3..a85d5b6 100644
--- a/drivers/kvm/kvm.h
+++ b/include/linux/kvm_host.h
@@ -1,5 +1,5 @@
-#ifndef __KVM_H
-#define __KVM_H
+#ifndef __KVM_HOST_H
+#define __KVM_HOST_H
/*
* This work is licensed under the terms of the GNU GPL, version 2. See
@@ -20,9 +20,9 @@
#include <linux/kvm.h>
#include <linux/kvm_para.h>
-#include "types.h"
+#include <linux/kvm_types.h>
-#include "x86.h"
+#include <asm/kvm_host.h>
#define KVM_MAX_VCPUS 4
#define KVM_MEMORY_SLOTS 8
diff --git a/drivers/kvm/types.h b/include/linux/kvm_types.h
similarity index 100%
rename from drivers/kvm/types.h
rename to include/linux/kvm_types.h
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 16/33] KVM: Move drivers/kvm/* to virt/kvm/
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (14 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 15/33] KVM: Move arch dependent files to new directory arch/x86/kvm/ Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 17/33] KVM: Move irqchip declarations into new ioapic.h and lapic.h Avi Kivity
` (16 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/Makefile | 4 ++--
{drivers => virt}/kvm/iodev.h | 0
{drivers => virt}/kvm/kvm_main.c | 0
3 files changed, 2 insertions(+), 2 deletions(-)
rename {drivers => virt}/kvm/iodev.h (100%)
rename {drivers => virt}/kvm/kvm_main.c (100%)
diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile
index 880ffe4..af41534 100644
--- a/arch/x86/kvm/Makefile
+++ b/arch/x86/kvm/Makefile
@@ -2,9 +2,9 @@
# Makefile for Kernel-based Virtual Machine module
#
-common-objs = $(addprefix ../../../drivers/kvm/, kvm_main.o)
+common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o)
-EXTRA_CFLAGS += -I drivers/kvm
+EXTRA_CFLAGS += -I virt/kvm
kvm-objs := $(common-objs) x86.o mmu.o x86_emulate.o i8259.o irq.o lapic.o \
ioapic.o
diff --git a/drivers/kvm/iodev.h b/virt/kvm/iodev.h
similarity index 100%
rename from drivers/kvm/iodev.h
rename to virt/kvm/iodev.h
diff --git a/drivers/kvm/kvm_main.c b/virt/kvm/kvm_main.c
similarity index 100%
rename from drivers/kvm/kvm_main.c
rename to virt/kvm/kvm_main.c
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 17/33] KVM: Move irqchip declarations into new ioapic.h and lapic.h
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (15 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 16/33] KVM: Move drivers/kvm/* to virt/kvm/ Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 18/33] KVM: Move ioapic code to common directory Avi Kivity
` (15 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
This allows reuse of ioapic in ia64.
Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/ioapic.c | 5 ++-
arch/x86/kvm/ioapic.h | 95 ++++++++++++++++++++++++++++++++++++++++
arch/x86/kvm/irq.h | 115 ++----------------------------------------------
arch/x86/kvm/lapic.h | 44 +++++++++++++++++++
4 files changed, 148 insertions(+), 111 deletions(-)
create mode 100644 arch/x86/kvm/ioapic.h
create mode 100644 arch/x86/kvm/lapic.h
diff --git a/arch/x86/kvm/ioapic.c b/arch/x86/kvm/ioapic.c
index 72f12f7..40cd53e 100644
--- a/arch/x86/kvm/ioapic.c
+++ b/arch/x86/kvm/ioapic.c
@@ -36,7 +36,10 @@
#include <asm/processor.h>
#include <asm/page.h>
#include <asm/current.h>
-#include "irq.h"
+
+#include "ioapic.h"
+#include "lapic.h"
+
#if 0
#define ioapic_debug(fmt,arg...) printk(KERN_WARNING fmt,##arg)
#else
diff --git a/arch/x86/kvm/ioapic.h b/arch/x86/kvm/ioapic.h
new file mode 100644
index 0000000..7f16675
--- /dev/null
+++ b/arch/x86/kvm/ioapic.h
@@ -0,0 +1,95 @@
+#ifndef __KVM_IO_APIC_H
+#define __KVM_IO_APIC_H
+
+#include <linux/kvm_host.h>
+
+#include "iodev.h"
+
+struct kvm;
+struct kvm_vcpu;
+
+#define IOAPIC_NUM_PINS KVM_IOAPIC_NUM_PINS
+#define IOAPIC_VERSION_ID 0x11 /* IOAPIC version */
+#define IOAPIC_EDGE_TRIG 0
+#define IOAPIC_LEVEL_TRIG 1
+
+#define IOAPIC_DEFAULT_BASE_ADDRESS 0xfec00000
+#define IOAPIC_MEM_LENGTH 0x100
+
+/* Direct registers. */
+#define IOAPIC_REG_SELECT 0x00
+#define IOAPIC_REG_WINDOW 0x10
+#define IOAPIC_REG_EOI 0x40 /* IA64 IOSAPIC only */
+
+/* Indirect registers. */
+#define IOAPIC_REG_APIC_ID 0x00 /* x86 IOAPIC only */
+#define IOAPIC_REG_VERSION 0x01
+#define IOAPIC_REG_ARB_ID 0x02 /* x86 IOAPIC only */
+
+/*ioapic delivery mode*/
+#define IOAPIC_FIXED 0x0
+#define IOAPIC_LOWEST_PRIORITY 0x1
+#define IOAPIC_PMI 0x2
+#define IOAPIC_NMI 0x4
+#define IOAPIC_INIT 0x5
+#define IOAPIC_EXTINT 0x7
+
+struct kvm_ioapic {
+ u64 base_address;
+ u32 ioregsel;
+ u32 id;
+ u32 irr;
+ u32 pad;
+ union ioapic_redir_entry {
+ u64 bits;
+ struct {
+ u8 vector;
+ u8 delivery_mode:3;
+ u8 dest_mode:1;
+ u8 delivery_status:1;
+ u8 polarity:1;
+ u8 remote_irr:1;
+ u8 trig_mode:1;
+ u8 mask:1;
+ u8 reserve:7;
+ u8 reserved[4];
+ u8 dest_id;
+ } fields;
+ } redirtbl[IOAPIC_NUM_PINS];
+ struct kvm_io_device dev;
+ struct kvm *kvm;
+};
+
+#ifdef DEBUG
+#define ASSERT(x) \
+do { \
+ if (!(x)) { \
+ printk(KERN_EMERG "assertion failed %s: %d: %s\n", \
+ __FILE__, __LINE__, #x); \
+ BUG(); \
+ } \
+} while (0)
+#else
+#define ASSERT(x) do { } while (0)
+#endif
+
+static inline struct kvm_ioapic *ioapic_irqchip(struct kvm *kvm)
+{
+ return kvm->arch.vioapic;
+}
+
+#ifdef CONFIG_IA64
+static inline int irqchip_in_kernel(struct kvm *kvm)
+{
+ return 1;
+}
+#endif
+
+struct kvm_vcpu *kvm_get_lowest_prio_vcpu(struct kvm *kvm, u8 vector,
+ unsigned long bitmap);
+void kvm_ioapic_update_eoi(struct kvm *kvm, int vector);
+int kvm_ioapic_init(struct kvm *kvm);
+void kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level);
+void kvm_ioapic_reset(struct kvm_ioapic *ioapic);
+
+#endif
diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h
index 6316638..c376b5b 100644
--- a/arch/x86/kvm/irq.h
+++ b/arch/x86/kvm/irq.h
@@ -25,7 +25,10 @@
#include <linux/mm_types.h>
#include <linux/hrtimer.h>
#include <linux/kvm_host.h>
+
#include "iodev.h"
+#include "ioapic.h"
+#include "lapic.h"
struct kvm;
struct kvm_vcpu;
@@ -65,131 +68,23 @@ void kvm_pic_set_irq(void *opaque, int irq, int level);
int kvm_pic_read_irq(struct kvm_pic *s);
void kvm_pic_update_irq(struct kvm_pic *s);
-#define IOAPIC_NUM_PINS KVM_IOAPIC_NUM_PINS
-#define IOAPIC_VERSION_ID 0x11 /* IOAPIC version */
-#define IOAPIC_EDGE_TRIG 0
-#define IOAPIC_LEVEL_TRIG 1
-
-#define IOAPIC_DEFAULT_BASE_ADDRESS 0xfec00000
-#define IOAPIC_MEM_LENGTH 0x100
-
-/* Direct registers. */
-#define IOAPIC_REG_SELECT 0x00
-#define IOAPIC_REG_WINDOW 0x10
-#define IOAPIC_REG_EOI 0x40 /* IA64 IOSAPIC only */
-
-/* Indirect registers. */
-#define IOAPIC_REG_APIC_ID 0x00 /* x86 IOAPIC only */
-#define IOAPIC_REG_VERSION 0x01
-#define IOAPIC_REG_ARB_ID 0x02 /* x86 IOAPIC only */
-
-/*ioapic delivery mode*/
-#define IOAPIC_FIXED 0x0
-#define IOAPIC_LOWEST_PRIORITY 0x1
-#define IOAPIC_PMI 0x2
-#define IOAPIC_NMI 0x4
-#define IOAPIC_INIT 0x5
-#define IOAPIC_EXTINT 0x7
-
-struct kvm_ioapic {
- u64 base_address;
- u32 ioregsel;
- u32 id;
- u32 irr;
- u32 pad;
- union ioapic_redir_entry {
- u64 bits;
- struct {
- u8 vector;
- u8 delivery_mode:3;
- u8 dest_mode:1;
- u8 delivery_status:1;
- u8 polarity:1;
- u8 remote_irr:1;
- u8 trig_mode:1;
- u8 mask:1;
- u8 reserve:7;
- u8 reserved[4];
- u8 dest_id;
- } fields;
- } redirtbl[IOAPIC_NUM_PINS];
- struct kvm_io_device dev;
- struct kvm *kvm;
-};
-
-struct kvm_lapic {
- unsigned long base_address;
- struct kvm_io_device dev;
- struct {
- atomic_t pending;
- s64 period; /* unit: ns */
- u32 divide_count;
- ktime_t last_update;
- struct hrtimer dev;
- } timer;
- struct kvm_vcpu *vcpu;
- struct page *regs_page;
- void *regs;
-};
-
-#ifdef DEBUG
-#define ASSERT(x) \
-do { \
- if (!(x)) { \
- printk(KERN_EMERG "assertion failed %s: %d: %s\n", \
- __FILE__, __LINE__, #x); \
- BUG(); \
- } \
-} while (0)
-#else
-#define ASSERT(x) do { } while (0)
-#endif
-
static inline struct kvm_pic *pic_irqchip(struct kvm *kvm)
{
return kvm->arch.vpic;
}
-static inline struct kvm_ioapic *ioapic_irqchip(struct kvm *kvm)
-{
- return kvm->arch.vioapic;
-}
-
static inline int irqchip_in_kernel(struct kvm *kvm)
{
return pic_irqchip(kvm) != NULL;
}
-void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
-int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu);
-int kvm_apic_accept_pic_intr(struct kvm_vcpu *vcpu);
-int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu);
-int kvm_create_lapic(struct kvm_vcpu *vcpu);
-void kvm_lapic_reset(struct kvm_vcpu *vcpu);
void kvm_pic_reset(struct kvm_kpic_state *s);
-void kvm_ioapic_reset(struct kvm_ioapic *ioapic);
-void kvm_free_lapic(struct kvm_vcpu *vcpu);
-u64 kvm_lapic_get_cr8(struct kvm_vcpu *vcpu);
-void kvm_lapic_set_tpr(struct kvm_vcpu *vcpu, unsigned long cr8);
-void kvm_lapic_set_base(struct kvm_vcpu *vcpu, u64 value);
-struct kvm_vcpu *kvm_get_lowest_prio_vcpu(struct kvm *kvm, u8 vector,
- unsigned long bitmap);
-u64 kvm_get_apic_base(struct kvm_vcpu *vcpu);
-void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data);
-int kvm_apic_match_physical_addr(struct kvm_lapic *apic, u16 dest);
-void kvm_ioapic_update_eoi(struct kvm *kvm, int vector);
-int kvm_apic_match_logical_addr(struct kvm_lapic *apic, u8 mda);
-int kvm_apic_set_irq(struct kvm_vcpu *vcpu, u8 vec, u8 trig);
-void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu);
-int kvm_ioapic_init(struct kvm *kvm);
-void kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level);
-int kvm_lapic_enabled(struct kvm_vcpu *vcpu);
-int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu);
-void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec);
void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec);
void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu);
void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu);
void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu);
+void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
+
#endif
diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h
new file mode 100644
index 0000000..447b654
--- /dev/null
+++ b/arch/x86/kvm/lapic.h
@@ -0,0 +1,44 @@
+#ifndef __KVM_X86_LAPIC_H
+#define __KVM_X86_LAPIC_H
+
+#include "iodev.h"
+
+#include <linux/kvm_host.h>
+
+struct kvm_lapic {
+ unsigned long base_address;
+ struct kvm_io_device dev;
+ struct {
+ atomic_t pending;
+ s64 period; /* unit: ns */
+ u32 divide_count;
+ ktime_t last_update;
+ struct hrtimer dev;
+ } timer;
+ struct kvm_vcpu *vcpu;
+ struct page *regs_page;
+ void *regs;
+};
+int kvm_create_lapic(struct kvm_vcpu *vcpu);
+void kvm_free_lapic(struct kvm_vcpu *vcpu);
+
+int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu);
+int kvm_apic_accept_pic_intr(struct kvm_vcpu *vcpu);
+int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu);
+void kvm_lapic_reset(struct kvm_vcpu *vcpu);
+u64 kvm_lapic_get_cr8(struct kvm_vcpu *vcpu);
+void kvm_lapic_set_tpr(struct kvm_vcpu *vcpu, unsigned long cr8);
+void kvm_lapic_set_base(struct kvm_vcpu *vcpu, u64 value);
+
+int kvm_apic_match_physical_addr(struct kvm_lapic *apic, u16 dest);
+int kvm_apic_match_logical_addr(struct kvm_lapic *apic, u8 mda);
+int kvm_apic_set_irq(struct kvm_vcpu *vcpu, u8 vec, u8 trig);
+
+u64 kvm_get_apic_base(struct kvm_vcpu *vcpu);
+void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data);
+void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu);
+int kvm_lapic_enabled(struct kvm_vcpu *vcpu);
+int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu);
+void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec);
+
+#endif
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 18/33] KVM: Move ioapic code to common directory.
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (16 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 17/33] KVM: Move irqchip declarations into new ioapic.h and lapic.h Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 19/33] KVM: Move kvm_vcpu_kick() to x86.c Avi Kivity
` (14 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
Move ioapic code to common, since IA64 also needs it.
Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/Makefile | 7 +++----
{arch/x86 => virt}/kvm/ioapic.c | 2 +-
{arch/x86 => virt}/kvm/ioapic.h | 0
3 files changed, 4 insertions(+), 5 deletions(-)
rename {arch/x86 => virt}/kvm/ioapic.c (99%)
rename {arch/x86 => virt}/kvm/ioapic.h (100%)
diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile
index af41534..ffdd0b3 100644
--- a/arch/x86/kvm/Makefile
+++ b/arch/x86/kvm/Makefile
@@ -2,12 +2,11 @@
# Makefile for Kernel-based Virtual Machine module
#
-common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o)
+common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o)
-EXTRA_CFLAGS += -I virt/kvm
+EXTRA_CFLAGS += -Ivirt/kvm -Iarch/x86/kvm
-kvm-objs := $(common-objs) x86.o mmu.o x86_emulate.o i8259.o irq.o lapic.o \
- ioapic.o
+kvm-objs := $(common-objs) x86.o mmu.o x86_emulate.o i8259.o irq.o lapic.o
obj-$(CONFIG_KVM) += kvm.o
kvm-intel-objs = vmx.o
obj-$(CONFIG_KVM_INTEL) += kvm-intel.o
diff --git a/arch/x86/kvm/ioapic.c b/virt/kvm/ioapic.c
similarity index 99%
rename from arch/x86/kvm/ioapic.c
rename to virt/kvm/ioapic.c
index 40cd53e..317f8e2 100644
--- a/arch/x86/kvm/ioapic.c
+++ b/virt/kvm/ioapic.c
@@ -363,7 +363,7 @@ static void ioapic_mmio_write(struct kvm_io_device *this, gpa_t addr, int len,
break;
#ifdef CONFIG_IA64
case IOAPIC_REG_EOI:
- kvm_ioapic_update_eoi(ioapic, data);
+ kvm_ioapic_update_eoi(ioapic->kvm, data);
break;
#endif
diff --git a/arch/x86/kvm/ioapic.h b/virt/kvm/ioapic.h
similarity index 100%
rename from arch/x86/kvm/ioapic.h
rename to virt/kvm/ioapic.h
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 19/33] KVM: Move kvm_vcpu_kick() to x86.c
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (17 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 18/33] KVM: Move ioapic code to common directory Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 20/33] KVM: Expose ioapic to ia64 save/restore APIs Avi Kivity
` (13 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
Moving kvm_vcpu_kick() to x86.c. Since it should be
common for all archs, put its declarations in <linux/kvm_host.h>
Signed-off-by: Zhang Xiantao <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/irq.c | 20 --------------------
arch/x86/kvm/irq.h | 2 --
arch/x86/kvm/x86.c | 20 ++++++++++++++++++++
include/linux/kvm_host.h | 1 +
4 files changed, 21 insertions(+), 22 deletions(-)
diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c
index 07a09aa..e571475 100644
--- a/arch/x86/kvm/irq.c
+++ b/arch/x86/kvm/irq.c
@@ -63,26 +63,6 @@ int kvm_cpu_get_interrupt(struct kvm_vcpu *v)
}
EXPORT_SYMBOL_GPL(kvm_cpu_get_interrupt);
-static void vcpu_kick_intr(void *info)
-{
-#ifdef DEBUG
- struct kvm_vcpu *vcpu = (struct kvm_vcpu *)info;
- printk(KERN_DEBUG "vcpu_kick_intr %p \n", vcpu);
-#endif
-}
-
-void kvm_vcpu_kick(struct kvm_vcpu *vcpu)
-{
- int ipi_pcpu = vcpu->cpu;
-
- if (waitqueue_active(&vcpu->wq)) {
- wake_up_interruptible(&vcpu->wq);
- ++vcpu->stat.halt_wakeup;
- }
- if (vcpu->guest_mode)
- smp_call_function_single(ipi_pcpu, vcpu_kick_intr, vcpu, 0, 0);
-}
-
void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu)
{
kvm_inject_apic_timer_irqs(vcpu);
diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h
index c376b5b..53c7f48 100644
--- a/arch/x86/kvm/irq.h
+++ b/arch/x86/kvm/irq.h
@@ -85,6 +85,4 @@ void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu);
void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu);
void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu);
-void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
-
#endif
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 5902c5c..733bff6 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3144,3 +3144,23 @@ int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
return vcpu->arch.mp_state == VCPU_MP_STATE_RUNNABLE
|| vcpu->arch.mp_state == VCPU_MP_STATE_SIPI_RECEIVED;
}
+
+static void vcpu_kick_intr(void *info)
+{
+#ifdef DEBUG
+ struct kvm_vcpu *vcpu = (struct kvm_vcpu *)info;
+ printk(KERN_DEBUG "vcpu_kick_intr %p \n", vcpu);
+#endif
+}
+
+void kvm_vcpu_kick(struct kvm_vcpu *vcpu)
+{
+ int ipi_pcpu = vcpu->cpu;
+
+ if (waitqueue_active(&vcpu->wq)) {
+ wake_up_interruptible(&vcpu->wq);
+ ++vcpu->stat.halt_wakeup;
+ }
+ if (vcpu->guest_mode)
+ smp_call_function_single(ipi_pcpu, vcpu_kick_intr, vcpu, 0, 0);
+}
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index a85d5b6..953b50a 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -250,6 +250,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm);
int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
int kvm_cpu_has_interrupt(struct kvm_vcpu *v);
+void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
static inline void kvm_guest_enter(void)
{
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 20/33] KVM: Expose ioapic to ia64 save/restore APIs
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (18 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 19/33] KVM: Move kvm_vcpu_kick() to x86.c Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 21/33] KVM: MMU: Coalesce remote tlb flushes Avi Kivity
` (12 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Zhang Xiantao
From: Zhang Xiantao <xiantao.zhang@intel.com>
IA64 also needs to see ioapic structure in irqchip.
Signed-off-by: xiantao.zhang@intel.com <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
include/linux/kvm.h | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index f0bebd6..de9f28d 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -53,6 +53,8 @@ struct kvm_irqchip {
char dummy[512]; /* reserving space */
#ifdef CONFIG_X86
struct kvm_pic_state pic;
+#endif
+#if defined(CONFIG_X86) || defined(CONFIG_IA64)
struct kvm_ioapic_state ioapic;
#endif
} chip;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 21/33] KVM: MMU: Coalesce remote tlb flushes
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (19 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 20/33] KVM: Expose ioapic to ia64 save/restore APIs Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 22/33] KVM: MMU: Add cache miss statistic Avi Kivity
` (11 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Eddie Dong
From: Eddie Dong <eddie.dong@intel.com>
Host side TLB flush can be merged together if multiple
spte need to be write-protected.
Signed-off-by: Yaozu (Eddie) Dong <eddie.dong@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/mmu.c | 8 ++++++--
1 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 401eb7c..9a57e1a 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -513,6 +513,7 @@ static void rmap_write_protect(struct kvm *kvm, u64 gfn)
{
unsigned long *rmapp;
u64 *spte;
+ int write_protected = 0;
gfn = unalias_gfn(kvm, gfn);
rmapp = gfn_to_rmap(kvm, gfn);
@@ -522,11 +523,14 @@ static void rmap_write_protect(struct kvm *kvm, u64 gfn)
BUG_ON(!spte);
BUG_ON(!(*spte & PT_PRESENT_MASK));
rmap_printk("rmap_write_protect: spte %p %llx\n", spte, *spte);
- if (is_writeble_pte(*spte))
+ if (is_writeble_pte(*spte)) {
set_shadow_pte(spte, *spte & ~PT_WRITABLE_MASK);
- kvm_flush_remote_tlbs(kvm);
+ write_protected = 1;
+ }
spte = rmap_next(kvm, rmapp, spte);
}
+ if (write_protected)
+ kvm_flush_remote_tlbs(kvm);
}
#ifdef MMU_DEBUG
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 22/33] KVM: MMU: Add cache miss statistic
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (20 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 21/33] KVM: MMU: Coalesce remote tlb flushes Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 23/33] KVM: Print data for unimplemented wrmsr Avi Kivity
` (10 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/mmu.c | 1 +
arch/x86/kvm/x86.c | 1 +
include/asm-x86/kvm_host.h | 1 +
3 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 9a57e1a..8f12ec5 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -715,6 +715,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
pgprintk("%s: found\n", __FUNCTION__);
return sp;
}
+ ++vcpu->kvm->stat.mmu_cache_miss;
sp = kvm_mmu_alloc_page(vcpu, parent_pte);
if (!sp)
return sp;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 733bff6..fa9e42c 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -73,6 +73,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
{ "mmu_pde_zapped", VM_STAT(mmu_pde_zapped) },
{ "mmu_flooded", VM_STAT(mmu_flooded) },
{ "mmu_recycled", VM_STAT(mmu_recycled) },
+ { "mmu_cache_miss", VM_STAT(mmu_cache_miss) },
{ "remote_tlb_flush", VM_STAT(remote_tlb_flush) },
{ NULL }
};
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 28940e1..ced1beb 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -289,6 +289,7 @@ struct kvm_vm_stat {
u32 mmu_pde_zapped;
u32 mmu_flooded;
u32 mmu_recycled;
+ u32 mmu_cache_miss;
u32 remote_tlb_flush;
};
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 23/33] KVM: Print data for unimplemented wrmsr
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (21 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 22/33] KVM: MMU: Add cache miss statistic Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 24/33] KVM: local APIC TPR access reporting facility Avi Kivity
` (9 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel
This can help diagnosing what the guest is trying to do. In many cases
we can get away with partial emulation of msrs.
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/x86.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index fa9e42c..513258c 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -499,7 +499,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
vcpu->arch.ia32_misc_enable_msr = data;
break;
default:
- pr_unimpl(vcpu, "unhandled wrmsr: 0x%x\n", msr);
+ pr_unimpl(vcpu, "unhandled wrmsr: 0x%x data %llx\n", msr, data);
return 1;
}
return 0;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 24/33] KVM: local APIC TPR access reporting facility
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (22 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 23/33] KVM: Print data for unimplemented wrmsr Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 25/33] KVM: Accelerated apic support Avi Kivity
` (8 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel
Add a facility to report on accesses to the local apic tpr even if the
local apic is emulated in the kernel. This is basically a hack that
allows userspace to patch Windows which tends to bang on the tpr a lot.
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/lapic.c | 21 +++++++++++++++++++++
arch/x86/kvm/x86.c | 25 +++++++++++++++++++++++++
include/asm-x86/kvm_host.h | 1 +
include/linux/kvm.h | 17 +++++++++++++++++
include/linux/kvm_host.h | 2 +-
5 files changed, 65 insertions(+), 1 deletions(-)
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 4076331..50c3f3a 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -551,6 +551,23 @@ static u32 apic_get_tmcct(struct kvm_lapic *apic)
return tmcct;
}
+static void __report_tpr_access(struct kvm_lapic *apic, bool write)
+{
+ struct kvm_vcpu *vcpu = apic->vcpu;
+ struct kvm_run *run = vcpu->run;
+
+ set_bit(KVM_REQ_REPORT_TPR_ACCESS, &vcpu->requests);
+ kvm_x86_ops->cache_regs(vcpu);
+ run->tpr_access.rip = vcpu->arch.rip;
+ run->tpr_access.is_write = write;
+}
+
+static inline void report_tpr_access(struct kvm_lapic *apic, bool write)
+{
+ if (apic->vcpu->arch.tpr_access_reporting)
+ __report_tpr_access(apic, write);
+}
+
static u32 __apic_read(struct kvm_lapic *apic, unsigned int offset)
{
u32 val = 0;
@@ -568,6 +585,9 @@ static u32 __apic_read(struct kvm_lapic *apic, unsigned int offset)
val = apic_get_tmcct(apic);
break;
+ case APIC_TASKPRI:
+ report_tpr_access(apic, false);
+ /* fall thru */
default:
apic_update_ppr(apic);
val = apic_get_reg(apic, offset);
@@ -677,6 +697,7 @@ static void apic_mmio_write(struct kvm_io_device *this,
break;
case APIC_TASKPRI:
+ report_tpr_access(apic, true);
apic_set_tpr(apic, val & 0xff);
break;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 513258c..c2b8088 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -684,6 +684,7 @@ int kvm_dev_ioctl_check_extension(long ext)
case KVM_CAP_USER_MEMORY:
case KVM_CAP_SET_TSS_ADDR:
case KVM_CAP_EXT_CPUID:
+ case KVM_CAP_VAPIC:
r = 1;
break;
default:
@@ -1055,6 +1056,15 @@ static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu,
return 0;
}
+static int vcpu_ioctl_tpr_access_reporting(struct kvm_vcpu *vcpu,
+ struct kvm_tpr_access_ctl *tac)
+{
+ if (tac->flags)
+ return -EINVAL;
+ vcpu->arch.tpr_access_reporting = !!tac->enabled;
+ return 0;
+}
+
long kvm_arch_vcpu_ioctl(struct file *filp,
unsigned int ioctl, unsigned long arg)
{
@@ -1148,6 +1158,21 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
case KVM_SET_MSRS:
r = msr_io(vcpu, argp, do_set_msr, 0);
break;
+ case KVM_TPR_ACCESS_REPORTING: {
+ struct kvm_tpr_access_ctl tac;
+
+ r = -EFAULT;
+ if (copy_from_user(&tac, argp, sizeof tac))
+ goto out;
+ r = vcpu_ioctl_tpr_access_reporting(vcpu, &tac);
+ if (r)
+ goto out;
+ r = -EFAULT;
+ if (copy_to_user(argp, &tac, sizeof tac))
+ goto out;
+ r = 0;
+ break;
+ };
default:
r = -EINVAL;
}
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index ced1beb..6e649af 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -211,6 +211,7 @@ struct kvm_vcpu_arch {
int mp_state;
int sipi_vector;
u64 ia32_misc_enable_msr;
+ bool tpr_access_reporting;
struct kvm_mmu mmu;
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index de9f28d..850f5ef 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -72,6 +72,7 @@ struct kvm_irqchip {
#define KVM_EXIT_FAIL_ENTRY 9
#define KVM_EXIT_INTR 10
#define KVM_EXIT_SET_TPR 11
+#define KVM_EXIT_TPR_ACCESS 12
/* for KVM_RUN, returned by mmap(vcpu_fd, offset=0) */
struct kvm_run {
@@ -130,6 +131,12 @@ struct kvm_run {
__u32 longmode;
__u32 pad;
} hypercall;
+ /* KVM_EXIT_TPR_ACCESS */
+ struct {
+ __u64 rip;
+ __u32 is_write;
+ __u32 pad;
+ } tpr_access;
/* Fix the size of the union. */
char padding[256];
};
@@ -202,6 +209,13 @@ struct kvm_signal_mask {
__u8 sigset[0];
};
+/* for KVM_TPR_ACCESS_REPORTING */
+struct kvm_tpr_access_ctl {
+ __u32 enabled;
+ __u32 flags;
+ __u32 reserved[8];
+};
+
#define KVMIO 0xAE
/*
@@ -229,6 +243,7 @@ struct kvm_signal_mask {
#define KVM_CAP_USER_MEMORY 3
#define KVM_CAP_SET_TSS_ADDR 4
#define KVM_CAP_EXT_CPUID 5
+#define KVM_CAP_VAPIC 6
/*
* ioctls for VM fds
@@ -274,5 +289,7 @@ struct kvm_signal_mask {
#define KVM_SET_LAPIC _IOW(KVMIO, 0x8f, struct kvm_lapic_state)
#define KVM_SET_CPUID2 _IOW(KVMIO, 0x90, struct kvm_cpuid2)
#define KVM_GET_CPUID2 _IOWR(KVMIO, 0x91, struct kvm_cpuid2)
+/* Available with KVM_CAP_VAPIC */
+#define KVM_TPR_ACCESS_REPORTING _IOWR(KVMIO, 0x92, struct kvm_tpr_access_ctl)
#endif
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 953b50a..9ff5904 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -35,7 +35,7 @@
* vcpu->requests bit members
*/
#define KVM_REQ_TLB_FLUSH 0
-
+#define KVM_REQ_REPORT_TPR_ACCESS 2
struct kvm_vcpu;
extern struct kmem_cache *kvm_vcpu_cache;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 25/33] KVM: Accelerated apic support
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (23 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 24/33] KVM: local APIC TPR access reporting facility Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 26/33] KVM: Disable vapic support on Intel machines with FlexPriority Avi Kivity
` (7 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel
This adds a mechanism for exposing the virtual apic tpr to the guest, and a
protocol for letting the guest update the tpr without causing a vmexit if
conditions allow (e.g. there is no interrupt pending with a higher priority
than the new tpr).
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/lapic.c | 51 +++++++++++++++++++++++++++++++++++++++++-
arch/x86/kvm/lapic.h | 6 +++++
arch/x86/kvm/x86.c | 55 ++++++++++++++++++++++++++++++++++++++++++++++
include/linux/kvm.h | 7 ++++++
include/linux/kvm_para.h | 2 +
5 files changed, 120 insertions(+), 1 deletions(-)
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 50c3f3a..e7513bb 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -815,7 +815,8 @@ void kvm_lapic_set_tpr(struct kvm_vcpu *vcpu, unsigned long cr8)
if (!apic)
return;
- apic_set_tpr(apic, ((cr8 & 0x0f) << 4));
+ apic_set_tpr(apic, ((cr8 & 0x0f) << 4)
+ | (apic_get_reg(apic, APIC_TASKPRI) & 4));
}
u64 kvm_lapic_get_cr8(struct kvm_vcpu *vcpu)
@@ -1104,3 +1105,51 @@ void kvm_migrate_apic_timer(struct kvm_vcpu *vcpu)
hrtimer_start(timer, timer->expires, HRTIMER_MODE_ABS);
}
EXPORT_SYMBOL_GPL(kvm_migrate_apic_timer);
+
+void kvm_lapic_sync_from_vapic(struct kvm_vcpu *vcpu)
+{
+ u32 data;
+ void *vapic;
+
+ if (!irqchip_in_kernel(vcpu->kvm) || !vcpu->arch.apic->vapic_addr)
+ return;
+
+ vapic = kmap_atomic(vcpu->arch.apic->vapic_page, KM_USER0);
+ data = *(u32 *)(vapic + offset_in_page(vcpu->arch.apic->vapic_addr));
+ kunmap_atomic(vapic, KM_USER0);
+
+ apic_set_tpr(vcpu->arch.apic, data & 0xff);
+}
+
+void kvm_lapic_sync_to_vapic(struct kvm_vcpu *vcpu)
+{
+ u32 data, tpr;
+ int max_irr, max_isr;
+ struct kvm_lapic *apic;
+ void *vapic;
+
+ if (!irqchip_in_kernel(vcpu->kvm) || !vcpu->arch.apic->vapic_addr)
+ return;
+
+ apic = vcpu->arch.apic;
+ tpr = apic_get_reg(apic, APIC_TASKPRI) & 0xff;
+ max_irr = apic_find_highest_irr(apic);
+ if (max_irr < 0)
+ max_irr = 0;
+ max_isr = apic_find_highest_isr(apic);
+ if (max_isr < 0)
+ max_isr = 0;
+ data = (tpr & 0xff) | ((max_isr & 0xf0) << 8) | (max_irr << 24);
+
+ vapic = kmap_atomic(vcpu->arch.apic->vapic_page, KM_USER0);
+ *(u32 *)(vapic + offset_in_page(vcpu->arch.apic->vapic_addr)) = data;
+ kunmap_atomic(vapic, KM_USER0);
+}
+
+void kvm_lapic_set_vapic_addr(struct kvm_vcpu *vcpu, gpa_t vapic_addr)
+{
+ if (!irqchip_in_kernel(vcpu->kvm))
+ return;
+
+ vcpu->arch.apic->vapic_addr = vapic_addr;
+}
diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h
index 447b654..676c396 100644
--- a/arch/x86/kvm/lapic.h
+++ b/arch/x86/kvm/lapic.h
@@ -18,6 +18,8 @@ struct kvm_lapic {
struct kvm_vcpu *vcpu;
struct page *regs_page;
void *regs;
+ gpa_t vapic_addr;
+ struct page *vapic_page;
};
int kvm_create_lapic(struct kvm_vcpu *vcpu);
void kvm_free_lapic(struct kvm_vcpu *vcpu);
@@ -41,4 +43,8 @@ int kvm_lapic_enabled(struct kvm_vcpu *vcpu);
int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu);
void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec);
+void kvm_lapic_set_vapic_addr(struct kvm_vcpu *vcpu, gpa_t vapic_addr);
+void kvm_lapic_sync_from_vapic(struct kvm_vcpu *vcpu);
+void kvm_lapic_sync_to_vapic(struct kvm_vcpu *vcpu);
+
#endif
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index c2b8088..e7eac27 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1173,6 +1173,19 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
r = 0;
break;
};
+ case KVM_SET_VAPIC_ADDR: {
+ struct kvm_vapic_addr va;
+
+ r = -EINVAL;
+ if (!irqchip_in_kernel(vcpu->kvm))
+ goto out;
+ r = -EFAULT;
+ if (copy_from_user(&va, argp, sizeof va))
+ goto out;
+ r = 0;
+ kvm_lapic_set_vapic_addr(vcpu, va.vapic_addr);
+ break;
+ }
default:
r = -EINVAL;
}
@@ -2214,6 +2227,9 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu)
}
switch (nr) {
+ case KVM_HC_VAPIC_POLL_IRQ:
+ ret = 0;
+ break;
default:
ret = -KVM_ENOSYS;
break;
@@ -2421,6 +2437,29 @@ static void post_kvm_run_save(struct kvm_vcpu *vcpu,
vcpu->arch.irq_summary == 0);
}
+static void vapic_enter(struct kvm_vcpu *vcpu)
+{
+ struct kvm_lapic *apic = vcpu->arch.apic;
+ struct page *page;
+
+ if (!apic || !apic->vapic_addr)
+ return;
+
+ page = gfn_to_page(vcpu->kvm, apic->vapic_addr >> PAGE_SHIFT);
+ vcpu->arch.apic->vapic_page = page;
+}
+
+static void vapic_exit(struct kvm_vcpu *vcpu)
+{
+ struct kvm_lapic *apic = vcpu->arch.apic;
+
+ if (!apic || !apic->vapic_addr)
+ return;
+
+ kvm_release_page_dirty(apic->vapic_page);
+ mark_page_dirty(vcpu->kvm, apic->vapic_addr >> PAGE_SHIFT);
+}
+
static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
int r;
@@ -2435,6 +2474,8 @@ static int __vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
vcpu->arch.mp_state = VCPU_MP_STATE_RUNNABLE;
}
+ vapic_enter(vcpu);
+
preempted:
if (vcpu->guest_debug.enabled)
kvm_x86_ops->guest_debug_pre(vcpu);
@@ -2444,6 +2485,14 @@ again:
if (unlikely(r))
goto out;
+ if (vcpu->requests)
+ if (test_and_clear_bit(KVM_REQ_REPORT_TPR_ACCESS,
+ &vcpu->requests)) {
+ kvm_run->exit_reason = KVM_EXIT_TPR_ACCESS;
+ r = 0;
+ goto out;
+ }
+
kvm_inject_pending_timer_irqs(vcpu);
preempt_disable();
@@ -2469,6 +2518,8 @@ again:
else
kvm_x86_ops->inject_pending_vectors(vcpu, kvm_run);
+ kvm_lapic_sync_to_vapic(vcpu);
+
vcpu->guest_mode = 1;
kvm_guest_enter();
@@ -2506,6 +2557,8 @@ again:
if (vcpu->arch.exception.pending && kvm_x86_ops->exception_injected(vcpu))
vcpu->arch.exception.pending = false;
+ kvm_lapic_sync_from_vapic(vcpu);
+
r = kvm_x86_ops->handle_exit(kvm_run, vcpu);
if (r > 0) {
@@ -2527,6 +2580,8 @@ out:
post_kvm_run_save(vcpu, kvm_run);
+ vapic_exit(vcpu);
+
return r;
}
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 850f5ef..b27a381 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -216,6 +216,11 @@ struct kvm_tpr_access_ctl {
__u32 reserved[8];
};
+/* for KVM_SET_VAPIC_ADDR */
+struct kvm_vapic_addr {
+ __u64 vapic_addr;
+};
+
#define KVMIO 0xAE
/*
@@ -291,5 +296,7 @@ struct kvm_tpr_access_ctl {
#define KVM_GET_CPUID2 _IOWR(KVMIO, 0x91, struct kvm_cpuid2)
/* Available with KVM_CAP_VAPIC */
#define KVM_TPR_ACCESS_REPORTING _IOWR(KVMIO, 0x92, struct kvm_tpr_access_ctl)
+/* Available with KVM_CAP_VAPIC */
+#define KVM_SET_VAPIC_ADDR _IOW(KVMIO, 0x93, struct kvm_vapic_addr)
#endif
diff --git a/include/linux/kvm_para.h b/include/linux/kvm_para.h
index e4db25f..6af91a5 100644
--- a/include/linux/kvm_para.h
+++ b/include/linux/kvm_para.h
@@ -12,6 +12,8 @@
/* Return values for hypercalls */
#define KVM_ENOSYS 1000
+#define KVM_HC_VAPIC_POLL_IRQ 1
+
#ifdef __KERNEL__
/*
* hypercalls use architecture specific
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 26/33] KVM: Disable vapic support on Intel machines with FlexPriority
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (24 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 25/33] KVM: Accelerated apic support Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 27/33] KVM: MMU: Concurrent guest walkers Avi Kivity
` (6 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel
FlexPriority accelerates the tpr without any patching.
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/svm.c | 6 ++++++
arch/x86/kvm/vmx.c | 3 ++-
arch/x86/kvm/x86.c | 4 +++-
include/asm-x86/kvm_host.h | 1 +
4 files changed, 12 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 745b1ec..11e5baf 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -1653,6 +1653,11 @@ static void svm_check_processor_compat(void *rtn)
*(int *)rtn = 0;
}
+static bool svm_cpu_has_accelerated_tpr(void)
+{
+ return false;
+}
+
static struct kvm_x86_ops svm_x86_ops = {
.cpu_has_kvm_support = has_svm,
.disabled_by_bios = is_disabled,
@@ -1661,6 +1666,7 @@ static struct kvm_x86_ops svm_x86_ops = {
.check_processor_compatibility = svm_check_processor_compat,
.hardware_enable = svm_hardware_enable,
.hardware_disable = svm_hardware_disable,
+ .cpu_has_accelerated_tpr = svm_cpu_has_accelerated_tpr,
.vcpu_create = svm_create_vcpu,
.vcpu_free = svm_free_vcpu,
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 20c0f5e..fa72f72 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -192,7 +192,7 @@ static inline int cpu_has_secondary_exec_ctrls(void)
CPU_BASED_ACTIVATE_SECONDARY_CONTROLS);
}
-static inline int cpu_has_vmx_virtualize_apic_accesses(void)
+static inline bool cpu_has_vmx_virtualize_apic_accesses(void)
{
return (vmcs_config.cpu_based_2nd_exec_ctrl &
SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES);
@@ -2564,6 +2564,7 @@ static struct kvm_x86_ops vmx_x86_ops = {
.check_processor_compatibility = vmx_check_processor_compat,
.hardware_enable = hardware_enable,
.hardware_disable = hardware_disable,
+ .cpu_has_accelerated_tpr = cpu_has_vmx_virtualize_apic_accesses,
.vcpu_create = vmx_create_vcpu,
.vcpu_free = vmx_free_vcpu,
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index e7eac27..1f48ec8 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -684,9 +684,11 @@ int kvm_dev_ioctl_check_extension(long ext)
case KVM_CAP_USER_MEMORY:
case KVM_CAP_SET_TSS_ADDR:
case KVM_CAP_EXT_CPUID:
- case KVM_CAP_VAPIC:
r = 1;
break;
+ case KVM_CAP_VAPIC:
+ r = !kvm_x86_ops->cpu_has_accelerated_tpr();
+ break;
default:
r = 0;
break;
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 6e649af..44b8925 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -329,6 +329,7 @@ struct kvm_x86_ops {
void (*check_processor_compatibility)(void *rtn);
int (*hardware_setup)(void); /* __init */
void (*hardware_unsetup)(void); /* __exit */
+ bool (*cpu_has_accelerated_tpr)(void);
/* Create, but do not attach this VCPU */
struct kvm_vcpu *(*vcpu_create)(struct kvm *kvm, unsigned id);
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 27/33] KVM: MMU: Concurrent guest walkers
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (25 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 26/33] KVM: Disable vapic support on Intel machines with FlexPriority Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 28/33] KVM: Add kvm_read_guest_atomic() Avi Kivity
` (5 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Marcelo Tosatti
From: Marcelo Tosatti <mtosatti@redhat.com>
Do not hold kvm->lock mutex across the entire pagefault code,
only acquire it in places where it is necessary, such as mmu
hash list, active list, rmap and parent pte handling.
Allow concurrent guest walkers by switching walk_addr() to use
mmap_sem in read-mode.
And get rid of the lockless __gfn_to_page.
[avi: move kvm_mmu_pte_write() locking inside the function]
[avi: add locking for real mode]
[avi: fix cmpxchg locking]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/mmu.c | 41 ++++++++++++++++----
arch/x86/kvm/paging_tmpl.h | 8 +++-
arch/x86/kvm/vmx.c | 25 ++++++++----
arch/x86/kvm/x86.c | 92 ++++++++++++++++++++++++++------------------
virt/kvm/kvm_main.c | 22 ++--------
5 files changed, 117 insertions(+), 71 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 8f12ec5..3b91227 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -974,7 +974,7 @@ static void nonpaging_new_cr3(struct kvm_vcpu *vcpu)
{
}
-static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
+static int __nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
{
int level = PT32E_ROOT_LEVEL;
hpa_t table_addr = vcpu->arch.mmu.root_hpa;
@@ -1015,6 +1015,17 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
}
}
+static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
+{
+ int r;
+
+ mutex_lock(&vcpu->kvm->lock);
+ r = __nonpaging_map(vcpu, v, write, gfn);
+ mutex_unlock(&vcpu->kvm->lock);
+ return r;
+}
+
+
static void nonpaging_prefetch_page(struct kvm_vcpu *vcpu,
struct kvm_mmu_page *sp)
{
@@ -1031,6 +1042,7 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu)
if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
return;
+ mutex_lock(&vcpu->kvm->lock);
#ifdef CONFIG_X86_64
if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
hpa_t root = vcpu->arch.mmu.root_hpa;
@@ -1038,6 +1050,7 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu)
sp = page_header(root);
--sp->root_count;
vcpu->arch.mmu.root_hpa = INVALID_PAGE;
+ mutex_unlock(&vcpu->kvm->lock);
return;
}
#endif
@@ -1051,6 +1064,7 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu)
}
vcpu->arch.mmu.pae_root[i] = INVALID_PAGE;
}
+ mutex_unlock(&vcpu->kvm->lock);
vcpu->arch.mmu.root_hpa = INVALID_PAGE;
}
@@ -1250,15 +1264,15 @@ int kvm_mmu_load(struct kvm_vcpu *vcpu)
{
int r;
- mutex_lock(&vcpu->kvm->lock);
r = mmu_topup_memory_caches(vcpu);
if (r)
goto out;
+ mutex_lock(&vcpu->kvm->lock);
mmu_alloc_roots(vcpu);
+ mutex_unlock(&vcpu->kvm->lock);
kvm_x86_ops->set_cr3(vcpu, vcpu->arch.mmu.root_hpa);
kvm_mmu_flush_tlb(vcpu);
out:
- mutex_unlock(&vcpu->kvm->lock);
return r;
}
EXPORT_SYMBOL_GPL(kvm_mmu_load);
@@ -1353,6 +1367,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
int npte;
pgprintk("%s: gpa %llx bytes %d\n", __FUNCTION__, gpa, bytes);
+ mutex_lock(&vcpu->kvm->lock);
++vcpu->kvm->stat.mmu_pte_write;
kvm_mmu_audit(vcpu, "pre pte write");
if (gfn == vcpu->arch.last_pt_write_gfn
@@ -1421,17 +1436,27 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
}
}
kvm_mmu_audit(vcpu, "post pte write");
+ mutex_unlock(&vcpu->kvm->lock);
}
int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
{
- gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
+ gpa_t gpa;
+ int r;
- return kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT);
+ down_read(¤t->mm->mmap_sem);
+ gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
+ up_read(¤t->mm->mmap_sem);
+
+ mutex_lock(&vcpu->kvm->lock);
+ r = kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT);
+ mutex_unlock(&vcpu->kvm->lock);
+ return r;
}
void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
{
+ mutex_lock(&vcpu->kvm->lock);
while (vcpu->kvm->arch.n_free_mmu_pages < KVM_REFILL_PAGES) {
struct kvm_mmu_page *sp;
@@ -1440,6 +1465,7 @@ void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
kvm_mmu_zap_page(vcpu->kvm, sp);
++vcpu->kvm->stat.mmu_recycled;
}
+ mutex_unlock(&vcpu->kvm->lock);
}
int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code)
@@ -1447,7 +1473,6 @@ int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code)
int r;
enum emulation_result er;
- mutex_lock(&vcpu->kvm->lock);
r = vcpu->arch.mmu.page_fault(vcpu, cr2, error_code);
if (r < 0)
goto out;
@@ -1462,7 +1487,6 @@ int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code)
goto out;
er = emulate_instruction(vcpu, vcpu->run, cr2, error_code, 0);
- mutex_unlock(&vcpu->kvm->lock);
switch (er) {
case EMULATE_DONE:
@@ -1477,7 +1501,6 @@ int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code)
BUG();
}
out:
- mutex_unlock(&vcpu->kvm->lock);
return r;
}
EXPORT_SYMBOL_GPL(kvm_mmu_page_fault);
@@ -1574,8 +1597,10 @@ void kvm_mmu_zap_all(struct kvm *kvm)
{
struct kvm_mmu_page *sp, *node;
+ mutex_lock(&kvm->lock);
list_for_each_entry_safe(sp, node, &kvm->arch.active_mmu_pages, link)
kvm_mmu_zap_page(kvm, sp);
+ mutex_unlock(&kvm->lock);
kvm_flush_remote_tlbs(kvm);
}
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 56b88f7..7f83f55 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -368,11 +368,13 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
if (r)
return r;
+ down_read(¤t->mm->mmap_sem);
/*
* Look up the shadow pte for the faulting address.
*/
r = FNAME(walk_addr)(&walker, vcpu, addr, write_fault, user_fault,
fetch_fault);
+ up_read(¤t->mm->mmap_sem);
/*
* The page is not mapped by the guest. Let the guest handle it.
@@ -384,6 +386,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
return 0;
}
+ mutex_lock(&vcpu->kvm->lock);
shadow_pte = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault,
&write_pt);
pgprintk("%s: shadow pte %p %llx ptwrite %d\n", __FUNCTION__,
@@ -395,11 +398,14 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
/*
* mmio: emulate if accessible, otherwise its a guest fault.
*/
- if (shadow_pte && is_io_pte(*shadow_pte))
+ if (shadow_pte && is_io_pte(*shadow_pte)) {
+ mutex_unlock(&vcpu->kvm->lock);
return 1;
+ }
++vcpu->stat.pf_fixed;
kvm_mmu_audit(vcpu, "post page fault (fixed)");
+ mutex_unlock(&vcpu->kvm->lock);
return write_pt;
}
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index fa72f72..99fec63 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -1431,27 +1431,34 @@ static int init_rmode_tss(struct kvm *kvm)
{
gfn_t fn = rmode_tss_base(kvm) >> PAGE_SHIFT;
u16 data = 0;
+ int ret = 0;
int r;
+ down_read(¤t->mm->mmap_sem);
r = kvm_clear_guest_page(kvm, fn, 0, PAGE_SIZE);
if (r < 0)
- return 0;
+ goto out;
data = TSS_BASE_SIZE + TSS_REDIRECTION_SIZE;
r = kvm_write_guest_page(kvm, fn++, &data, 0x66, sizeof(u16));
if (r < 0)
- return 0;
+ goto out;
r = kvm_clear_guest_page(kvm, fn++, 0, PAGE_SIZE);
if (r < 0)
- return 0;
+ goto out;
r = kvm_clear_guest_page(kvm, fn, 0, PAGE_SIZE);
if (r < 0)
- return 0;
+ goto out;
data = ~0;
- r = kvm_write_guest_page(kvm, fn, &data, RMODE_TSS_SIZE - 2 * PAGE_SIZE - 1,
- sizeof(u8));
+ r = kvm_write_guest_page(kvm, fn, &data,
+ RMODE_TSS_SIZE - 2 * PAGE_SIZE - 1,
+ sizeof(u8));
if (r < 0)
- return 0;
- return 1;
+ goto out;
+
+ ret = 1;
+out:
+ up_read(¤t->mm->mmap_sem);
+ return ret;
}
static void seg_setup(int seg)
@@ -1470,6 +1477,7 @@ static int alloc_apic_access_page(struct kvm *kvm)
int r = 0;
mutex_lock(&kvm->lock);
+ down_write(¤t->mm->mmap_sem);
if (kvm->arch.apic_access_page)
goto out;
kvm_userspace_mem.slot = APIC_ACCESS_PAGE_PRIVATE_MEMSLOT;
@@ -1481,6 +1489,7 @@ static int alloc_apic_access_page(struct kvm *kvm)
goto out;
kvm->arch.apic_access_page = gfn_to_page(kvm, 0xfee00);
out:
+ up_write(¤t->mm->mmap_sem);
mutex_unlock(&kvm->lock);
return r;
}
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 1f48ec8..e3b3141 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -181,7 +181,7 @@ int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3)
int ret;
u64 pdpte[ARRAY_SIZE(vcpu->arch.pdptrs)];
- mutex_lock(&vcpu->kvm->lock);
+ down_read(¤t->mm->mmap_sem);
ret = kvm_read_guest_page(vcpu->kvm, pdpt_gfn, pdpte,
offset * sizeof(u64), sizeof(pdpte));
if (ret < 0) {
@@ -198,7 +198,7 @@ int load_pdptrs(struct kvm_vcpu *vcpu, unsigned long cr3)
memcpy(vcpu->arch.pdptrs, pdpte, sizeof(vcpu->arch.pdptrs));
out:
- mutex_unlock(&vcpu->kvm->lock);
+ up_read(¤t->mm->mmap_sem);
return ret;
}
@@ -212,13 +212,13 @@ static bool pdptrs_changed(struct kvm_vcpu *vcpu)
if (is_long_mode(vcpu) || !is_pae(vcpu))
return false;
- mutex_lock(&vcpu->kvm->lock);
+ down_read(¤t->mm->mmap_sem);
r = kvm_read_guest(vcpu->kvm, vcpu->arch.cr3 & ~31u, pdpte, sizeof(pdpte));
if (r < 0)
goto out;
changed = memcmp(pdpte, vcpu->arch.pdptrs, sizeof(pdpte)) != 0;
out:
- mutex_unlock(&vcpu->kvm->lock);
+ up_read(¤t->mm->mmap_sem);
return changed;
}
@@ -278,9 +278,7 @@ void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
kvm_x86_ops->set_cr0(vcpu, cr0);
vcpu->arch.cr0 = cr0;
- mutex_lock(&vcpu->kvm->lock);
kvm_mmu_reset_context(vcpu);
- mutex_unlock(&vcpu->kvm->lock);
return;
}
EXPORT_SYMBOL_GPL(set_cr0);
@@ -320,9 +318,7 @@ void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
}
kvm_x86_ops->set_cr4(vcpu, cr4);
vcpu->arch.cr4 = cr4;
- mutex_lock(&vcpu->kvm->lock);
kvm_mmu_reset_context(vcpu);
- mutex_unlock(&vcpu->kvm->lock);
}
EXPORT_SYMBOL_GPL(set_cr4);
@@ -360,7 +356,7 @@ void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
*/
}
- mutex_lock(&vcpu->kvm->lock);
+ down_read(¤t->mm->mmap_sem);
/*
* Does the new cr3 value map to physical memory? (Note, we
* catch an invalid cr3 even in real-mode, because it would
@@ -376,7 +372,7 @@ void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
vcpu->arch.cr3 = cr3;
vcpu->arch.mmu.new_cr3(vcpu);
}
- mutex_unlock(&vcpu->kvm->lock);
+ up_read(¤t->mm->mmap_sem);
}
EXPORT_SYMBOL_GPL(set_cr3);
@@ -1211,12 +1207,12 @@ static int kvm_vm_ioctl_set_nr_mmu_pages(struct kvm *kvm,
if (kvm_nr_mmu_pages < KVM_MIN_ALLOC_MMU_PAGES)
return -EINVAL;
- mutex_lock(&kvm->lock);
+ down_write(¤t->mm->mmap_sem);
kvm_mmu_change_mmu_pages(kvm, kvm_nr_mmu_pages);
kvm->arch.n_requested_mmu_pages = kvm_nr_mmu_pages;
- mutex_unlock(&kvm->lock);
+ up_write(¤t->mm->mmap_sem);
return 0;
}
@@ -1265,7 +1261,7 @@ static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm,
< alias->target_phys_addr)
goto out;
- mutex_lock(&kvm->lock);
+ down_write(¤t->mm->mmap_sem);
p = &kvm->arch.aliases[alias->slot];
p->base_gfn = alias->guest_phys_addr >> PAGE_SHIFT;
@@ -1279,7 +1275,7 @@ static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm,
kvm_mmu_zap_all(kvm);
- mutex_unlock(&kvm->lock);
+ up_write(¤t->mm->mmap_sem);
return 0;
@@ -1355,7 +1351,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
struct kvm_memory_slot *memslot;
int is_dirty = 0;
- mutex_lock(&kvm->lock);
+ down_write(¤t->mm->mmap_sem);
r = kvm_get_dirty_log(kvm, log, &is_dirty);
if (r)
@@ -1371,7 +1367,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
}
r = 0;
out:
- mutex_unlock(&kvm->lock);
+ up_write(¤t->mm->mmap_sem);
return r;
}
@@ -1565,25 +1561,32 @@ int emulator_read_std(unsigned long addr,
struct kvm_vcpu *vcpu)
{
void *data = val;
+ int r = X86EMUL_CONTINUE;
+ down_read(¤t->mm->mmap_sem);
while (bytes) {
gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
unsigned offset = addr & (PAGE_SIZE-1);
unsigned tocopy = min(bytes, (unsigned)PAGE_SIZE - offset);
int ret;
- if (gpa == UNMAPPED_GVA)
- return X86EMUL_PROPAGATE_FAULT;
+ if (gpa == UNMAPPED_GVA) {
+ r = X86EMUL_PROPAGATE_FAULT;
+ goto out;
+ }
ret = kvm_read_guest(vcpu->kvm, gpa, data, tocopy);
- if (ret < 0)
- return X86EMUL_UNHANDLEABLE;
+ if (ret < 0) {
+ r = X86EMUL_UNHANDLEABLE;
+ goto out;
+ }
bytes -= tocopy;
data += tocopy;
addr += tocopy;
}
-
- return X86EMUL_CONTINUE;
+out:
+ up_read(¤t->mm->mmap_sem);
+ return r;
}
EXPORT_SYMBOL_GPL(emulator_read_std);
@@ -1601,7 +1604,9 @@ static int emulator_read_emulated(unsigned long addr,
return X86EMUL_CONTINUE;
}
+ down_read(¤t->mm->mmap_sem);
gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
+ up_read(¤t->mm->mmap_sem);
/* For APIC access vmexit */
if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
@@ -1617,11 +1622,14 @@ mmio:
/*
* Is this MMIO handled locally?
*/
+ mutex_lock(&vcpu->kvm->lock);
mmio_dev = vcpu_find_mmio_dev(vcpu, gpa);
if (mmio_dev) {
kvm_iodevice_read(mmio_dev, gpa, bytes, val);
+ mutex_unlock(&vcpu->kvm->lock);
return X86EMUL_CONTINUE;
}
+ mutex_unlock(&vcpu->kvm->lock);
vcpu->mmio_needed = 1;
vcpu->mmio_phys_addr = gpa;
@@ -1636,10 +1644,14 @@ static int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa,
{
int ret;
+ down_read(¤t->mm->mmap_sem);
ret = kvm_write_guest(vcpu->kvm, gpa, val, bytes);
- if (ret < 0)
+ if (ret < 0) {
+ up_read(¤t->mm->mmap_sem);
return 0;
+ }
kvm_mmu_pte_write(vcpu, gpa, val, bytes);
+ up_read(¤t->mm->mmap_sem);
return 1;
}
@@ -1649,7 +1661,11 @@ static int emulator_write_emulated_onepage(unsigned long addr,
struct kvm_vcpu *vcpu)
{
struct kvm_io_device *mmio_dev;
- gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
+ gpa_t gpa;
+
+ down_read(¤t->mm->mmap_sem);
+ gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
+ up_read(¤t->mm->mmap_sem);
if (gpa == UNMAPPED_GVA) {
kvm_inject_page_fault(vcpu, addr, 2);
@@ -1667,11 +1683,14 @@ mmio:
/*
* Is this MMIO handled locally?
*/
+ mutex_lock(&vcpu->kvm->lock);
mmio_dev = vcpu_find_mmio_dev(vcpu, gpa);
if (mmio_dev) {
kvm_iodevice_write(mmio_dev, gpa, bytes, val);
+ mutex_unlock(&vcpu->kvm->lock);
return X86EMUL_CONTINUE;
}
+ mutex_unlock(&vcpu->kvm->lock);
vcpu->mmio_needed = 1;
vcpu->mmio_phys_addr = gpa;
@@ -1718,11 +1737,14 @@ static int emulator_cmpxchg_emulated(unsigned long addr,
#ifndef CONFIG_X86_64
/* guests cmpxchg8b have to be emulated atomically */
if (bytes == 8) {
- gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
+ gpa_t gpa;
struct page *page;
char *addr;
u64 val;
+ down_read(¤t->mm->mmap_sem);
+ gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr);
+
if (gpa == UNMAPPED_GVA ||
(gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
goto emul_write;
@@ -1736,8 +1758,9 @@ static int emulator_cmpxchg_emulated(unsigned long addr,
set_64bit((u64 *)(addr + offset_in_page(gpa)), val);
kunmap_atomic(addr, KM_USER0);
kvm_release_page_dirty(page);
+ emul_write:
+ up_read(¤t->mm->mmap_sem);
}
-emul_write:
#endif
return emulator_write_emulated(addr, new, bytes, vcpu);
@@ -2118,10 +2141,10 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
kvm_x86_ops->skip_emulated_instruction(vcpu);
for (i = 0; i < nr_pages; ++i) {
- mutex_lock(&vcpu->kvm->lock);
+ down_read(¤t->mm->mmap_sem);
page = gva_to_page(vcpu, address + i * PAGE_SIZE);
vcpu->arch.pio.guest_pages[i] = page;
- mutex_unlock(&vcpu->kvm->lock);
+ up_read(¤t->mm->mmap_sem);
if (!page) {
kvm_inject_gp(vcpu, 0);
free_pio_guest_pages(vcpu);
@@ -2247,7 +2270,6 @@ int kvm_fix_hypercall(struct kvm_vcpu *vcpu)
char instruction[3];
int ret = 0;
- mutex_lock(&vcpu->kvm->lock);
/*
* Blow out the MMU to ensure that no other VCPU has an active mapping
@@ -2262,8 +2284,6 @@ int kvm_fix_hypercall(struct kvm_vcpu *vcpu)
!= X86EMUL_CONTINUE)
ret = -EFAULT;
- mutex_unlock(&vcpu->kvm->lock);
-
return ret;
}
@@ -2447,8 +2467,10 @@ static void vapic_enter(struct kvm_vcpu *vcpu)
if (!apic || !apic->vapic_addr)
return;
+ down_read(¤t->mm->mmap_sem);
page = gfn_to_page(vcpu->kvm, apic->vapic_addr >> PAGE_SHIFT);
vcpu->arch.apic->vapic_page = page;
+ up_read(¤t->mm->mmap_sem);
}
static void vapic_exit(struct kvm_vcpu *vcpu)
@@ -2910,13 +2932,13 @@ int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
gpa_t gpa;
vcpu_load(vcpu);
- mutex_lock(&vcpu->kvm->lock);
+ down_read(¤t->mm->mmap_sem);
gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, vaddr);
+ up_read(¤t->mm->mmap_sem);
tr->physical_address = gpa;
tr->valid = gpa != UNMAPPED_GVA;
tr->writeable = 1;
tr->usermode = 0;
- mutex_unlock(&vcpu->kvm->lock);
vcpu_put(vcpu);
return 0;
@@ -3185,13 +3207,11 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
*/
if (!user_alloc) {
if (npages && !old.rmap) {
- down_write(¤t->mm->mmap_sem);
memslot->userspace_addr = do_mmap(NULL, 0,
npages * PAGE_SIZE,
PROT_READ | PROT_WRITE,
MAP_SHARED | MAP_ANONYMOUS,
0);
- up_write(¤t->mm->mmap_sem);
if (IS_ERR((void *)memslot->userspace_addr))
return PTR_ERR((void *)memslot->userspace_addr);
@@ -3199,10 +3219,8 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
if (!old.user_alloc && old.rmap) {
int ret;
- down_write(¤t->mm->mmap_sem);
ret = do_munmap(current->mm, old.userspace_addr,
old.npages * PAGE_SIZE);
- up_write(¤t->mm->mmap_sem);
if (ret < 0)
printk(KERN_WARNING
"kvm_vm_ioctl_set_memory_region: "
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index a5ee518..396c619 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -227,7 +227,7 @@ static int kvm_vm_release(struct inode *inode, struct file *filp)
*
* Discontiguous memory is allowed, mostly for framebuffers.
*
- * Must be called holding kvm->lock.
+ * Must be called holding mmap_sem for write.
*/
int __kvm_set_memory_region(struct kvm *kvm,
struct kvm_userspace_memory_region *mem,
@@ -338,9 +338,9 @@ int kvm_set_memory_region(struct kvm *kvm,
{
int r;
- mutex_lock(&kvm->lock);
+ down_write(¤t->mm->mmap_sem);
r = __kvm_set_memory_region(kvm, mem, user_alloc);
- mutex_unlock(&kvm->lock);
+ up_write(¤t->mm->mmap_sem);
return r;
}
EXPORT_SYMBOL_GPL(kvm_set_memory_region);
@@ -456,7 +456,7 @@ static unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn)
/*
* Requires current->mm->mmap_sem to be held
*/
-static struct page *__gfn_to_page(struct kvm *kvm, gfn_t gfn)
+struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn)
{
struct page *page[1];
unsigned long addr;
@@ -481,17 +481,6 @@ static struct page *__gfn_to_page(struct kvm *kvm, gfn_t gfn)
return page[0];
}
-struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn)
-{
- struct page *page;
-
- down_read(¤t->mm->mmap_sem);
- page = __gfn_to_page(kvm, gfn);
- up_read(¤t->mm->mmap_sem);
-
- return page;
-}
-
EXPORT_SYMBOL_GPL(gfn_to_page);
void kvm_release_page_clean(struct page *page)
@@ -977,8 +966,7 @@ static int kvm_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
if (!kvm_is_visible_gfn(kvm, vmf->pgoff))
return VM_FAULT_SIGBUS;
- /* current->mm->mmap_sem is already held so call lockless version */
- page = __gfn_to_page(kvm, vmf->pgoff);
+ page = gfn_to_page(kvm, vmf->pgoff);
if (is_error_page(page)) {
kvm_release_page_clean(page);
return VM_FAULT_SIGBUS;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 28/33] KVM: Add kvm_read_guest_atomic()
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (26 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 27/33] KVM: MMU: Concurrent guest walkers Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 29/33] KVM: MMU: Avoid calling gfn_to_page() in mmu_set_spte() Avi Kivity
` (4 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Marcelo Tosatti
From: Marcelo Tosatti <mtosatti@redhat.com>
In preparation for a mmu spinlock, add kvm_read_guest_atomic()
and use it in fetch() and prefetch_page().
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/paging_tmpl.h | 28 ++++++++++++++++------------
include/linux/kvm_host.h | 2 ++
virt/kvm/kvm_main.c | 20 ++++++++++++++++++++
3 files changed, 38 insertions(+), 12 deletions(-)
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 7f83f55..136a65d 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -316,10 +316,12 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
metaphysical, access,
shadow_ent, &new_page);
if (new_page && !metaphysical) {
+ int r;
pt_element_t curr_pte;
- kvm_read_guest(vcpu->kvm, walker->pte_gpa[level - 2],
- &curr_pte, sizeof(curr_pte));
- if (curr_pte != walker->ptes[level - 2])
+ r = kvm_read_guest_atomic(vcpu->kvm,
+ walker->pte_gpa[level - 2],
+ &curr_pte, sizeof(curr_pte));
+ if (r || curr_pte != walker->ptes[level - 2])
return NULL;
}
shadow_addr = __pa(shadow_page->spt);
@@ -429,9 +431,8 @@ static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t vaddr)
static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu,
struct kvm_mmu_page *sp)
{
- int i, offset = 0;
- pt_element_t *gpt;
- struct page *page;
+ int i, offset = 0, r = 0;
+ pt_element_t pt;
if (sp->role.metaphysical
|| (PTTYPE == 32 && sp->role.level > PT_PAGE_TABLE_LEVEL)) {
@@ -441,15 +442,18 @@ static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu,
if (PTTYPE == 32)
offset = sp->role.quadrant << PT64_LEVEL_BITS;
- page = gfn_to_page(vcpu->kvm, sp->gfn);
- gpt = kmap_atomic(page, KM_USER0);
- for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
- if (is_present_pte(gpt[offset + i]))
+
+ for (i = 0; i < PT64_ENT_PER_PAGE; ++i) {
+ gpa_t pte_gpa = gfn_to_gpa(sp->gfn);
+ pte_gpa += (i+offset) * sizeof(pt_element_t);
+
+ r = kvm_read_guest_atomic(vcpu->kvm, pte_gpa, &pt,
+ sizeof(pt_element_t));
+ if (r || is_present_pte(pt))
sp->spt[i] = shadow_trap_nonpresent_pte;
else
sp->spt[i] = shadow_notrap_nonpresent_pte;
- kunmap_atomic(gpt, KM_USER0);
- kvm_release_page_clean(page);
+ }
}
#undef pt_element_t
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 9ff5904..a020fb2 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -167,6 +167,8 @@ void kvm_release_page_clean(struct page *page);
void kvm_release_page_dirty(struct page *page);
int kvm_read_guest_page(struct kvm *kvm, gfn_t gfn, void *data, int offset,
int len);
+int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data,
+ unsigned long len);
int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len);
int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn, const void *data,
int offset, int len);
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 396c619..c462d7e 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -541,6 +541,26 @@ int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len)
}
EXPORT_SYMBOL_GPL(kvm_read_guest);
+int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data,
+ unsigned long len)
+{
+ int r;
+ unsigned long addr;
+ gfn_t gfn = gpa >> PAGE_SHIFT;
+ int offset = offset_in_page(gpa);
+
+ addr = gfn_to_hva(kvm, gfn);
+ if (kvm_is_error_hva(addr))
+ return -EFAULT;
+ pagefault_disable();
+ r = __copy_from_user_inatomic(data, (void __user *)addr + offset, len);
+ pagefault_enable();
+ if (r)
+ return -EFAULT;
+ return 0;
+}
+EXPORT_SYMBOL(kvm_read_guest_atomic);
+
int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn, const void *data,
int offset, int len)
{
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 29/33] KVM: MMU: Avoid calling gfn_to_page() in mmu_set_spte()
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (27 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 28/33] KVM: Add kvm_read_guest_atomic() Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 30/33] KVM: MMU: Switch to mmu spinlock Avi Kivity
` (3 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel
Since gfn_to_page() is a sleeping function, and we want to make the core mmu
spinlocked, we need to pass the page from the walker context (which can sleep)
to the shadow context (which cannot).
[marcelo: avoid recursive locking of mmap_sem]
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/mmu.c | 55 ++++++++++++++++++++++++++++++++++++++++----
arch/x86/kvm/paging_tmpl.h | 23 ++++++++++++++----
include/asm-x86/kvm_host.h | 5 ++++
3 files changed, 73 insertions(+), 10 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 3b91227..c0b757b 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -890,11 +890,10 @@ struct page *gva_to_page(struct kvm_vcpu *vcpu, gva_t gva)
static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
unsigned pt_access, unsigned pte_access,
int user_fault, int write_fault, int dirty,
- int *ptwrite, gfn_t gfn)
+ int *ptwrite, gfn_t gfn, struct page *page)
{
u64 spte;
int was_rmapped = is_rmap_pte(*shadow_pte);
- struct page *page;
pgprintk("%s: spte %llx access %x write_fault %d"
" user_fault %d gfn %lx\n",
@@ -912,8 +911,6 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte,
if (!(pte_access & ACC_EXEC_MASK))
spte |= PT64_NX_MASK;
- page = gfn_to_page(vcpu->kvm, gfn);
-
spte |= PT_PRESENT_MASK;
if (pte_access & ACC_USER_MASK)
spte |= PT_USER_MASK;
@@ -979,6 +976,11 @@ static int __nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
int level = PT32E_ROOT_LEVEL;
hpa_t table_addr = vcpu->arch.mmu.root_hpa;
int pt_write = 0;
+ struct page *page;
+
+ down_read(¤t->mm->mmap_sem);
+ page = gfn_to_page(vcpu->kvm, gfn);
+ up_read(¤t->mm->mmap_sem);
for (; ; level--) {
u32 index = PT64_INDEX(v, level);
@@ -989,7 +991,7 @@ static int __nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
if (level == 1) {
mmu_set_spte(vcpu, &table[index], ACC_ALL, ACC_ALL,
- 0, write, 1, &pt_write, gfn);
+ 0, write, 1, &pt_write, gfn, page);
return pt_write || is_io_pte(table[index]);
}
@@ -1005,6 +1007,7 @@ static int __nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
NULL);
if (!new_table) {
pgprintk("nonpaging_map: ENOMEM\n");
+ kvm_release_page_clean(page);
return -ENOMEM;
}
@@ -1347,6 +1350,43 @@ static bool last_updated_pte_accessed(struct kvm_vcpu *vcpu)
return !!(spte && (*spte & PT_ACCESSED_MASK));
}
+static void mmu_guess_page_from_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
+ const u8 *new, int bytes)
+{
+ gfn_t gfn;
+ int r;
+ u64 gpte = 0;
+
+ if (bytes != 4 && bytes != 8)
+ return;
+
+ /*
+ * Assume that the pte write on a page table of the same type
+ * as the current vcpu paging mode. This is nearly always true
+ * (might be false while changing modes). Note it is verified later
+ * by update_pte().
+ */
+ if (is_pae(vcpu)) {
+ /* Handle a 32-bit guest writing two halves of a 64-bit gpte */
+ if ((bytes == 4) && (gpa % 4 == 0)) {
+ r = kvm_read_guest(vcpu->kvm, gpa & ~(u64)7, &gpte, 8);
+ if (r)
+ return;
+ memcpy((void *)&gpte + (gpa % 8), new, 4);
+ } else if ((bytes == 8) && (gpa % 8 == 0)) {
+ memcpy((void *)&gpte, new, 8);
+ }
+ } else {
+ if ((bytes == 4) && (gpa % 4 == 0))
+ memcpy((void *)&gpte, new, 4);
+ }
+ if (!is_present_pte(gpte))
+ return;
+ gfn = (gpte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT;
+ vcpu->arch.update_pte.gfn = gfn;
+ vcpu->arch.update_pte.page = gfn_to_page(vcpu->kvm, gfn);
+}
+
void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
const u8 *new, int bytes)
{
@@ -1367,6 +1407,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
int npte;
pgprintk("%s: gpa %llx bytes %d\n", __FUNCTION__, gpa, bytes);
+ mmu_guess_page_from_pte_write(vcpu, gpa, new, bytes);
mutex_lock(&vcpu->kvm->lock);
++vcpu->kvm->stat.mmu_pte_write;
kvm_mmu_audit(vcpu, "pre pte write");
@@ -1437,6 +1478,10 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
}
kvm_mmu_audit(vcpu, "post pte write");
mutex_unlock(&vcpu->kvm->lock);
+ if (vcpu->arch.update_pte.page) {
+ kvm_release_page_clean(vcpu->arch.update_pte.page);
+ vcpu->arch.update_pte.page = NULL;
+ }
}
int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 136a65d..3d7846b 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -245,6 +245,7 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
{
pt_element_t gpte;
unsigned pte_access;
+ struct page *npage;
gpte = *(const pt_element_t *)pte;
if (~gpte & (PT_PRESENT_MASK | PT_ACCESSED_MASK)) {
@@ -256,8 +257,14 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
return;
pgprintk("%s: gpte %llx spte %p\n", __FUNCTION__, (u64)gpte, spte);
pte_access = page->role.access & FNAME(gpte_access)(vcpu, gpte);
+ if (gpte_to_gfn(gpte) != vcpu->arch.update_pte.gfn)
+ return;
+ npage = vcpu->arch.update_pte.page;
+ if (!npage)
+ return;
+ get_page(npage);
mmu_set_spte(vcpu, spte, page->role.access, pte_access, 0, 0,
- gpte & PT_DIRTY_MASK, NULL, gpte_to_gfn(gpte));
+ gpte & PT_DIRTY_MASK, NULL, gpte_to_gfn(gpte), npage);
}
/*
@@ -265,7 +272,8 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page,
*/
static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
struct guest_walker *walker,
- int user_fault, int write_fault, int *ptwrite)
+ int user_fault, int write_fault, int *ptwrite,
+ struct page *page)
{
hpa_t shadow_addr;
int level;
@@ -321,8 +329,10 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
r = kvm_read_guest_atomic(vcpu->kvm,
walker->pte_gpa[level - 2],
&curr_pte, sizeof(curr_pte));
- if (r || curr_pte != walker->ptes[level - 2])
+ if (r || curr_pte != walker->ptes[level - 2]) {
+ kvm_release_page_clean(page);
return NULL;
+ }
}
shadow_addr = __pa(shadow_page->spt);
shadow_pte = shadow_addr | PT_PRESENT_MASK | PT_ACCESSED_MASK
@@ -333,7 +343,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
mmu_set_spte(vcpu, shadow_ent, access, walker->pte_access & access,
user_fault, write_fault,
walker->ptes[walker->level-1] & PT_DIRTY_MASK,
- ptwrite, walker->gfn);
+ ptwrite, walker->gfn, page);
return shadow_ent;
}
@@ -362,6 +372,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
u64 *shadow_pte;
int write_pt = 0;
int r;
+ struct page *page;
pgprintk("%s: addr %lx err %x\n", __FUNCTION__, addr, error_code);
kvm_mmu_audit(vcpu, "pre page fault");
@@ -388,9 +399,11 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
return 0;
}
+ page = gfn_to_page(vcpu->kvm, walker.gfn);
+
mutex_lock(&vcpu->kvm->lock);
shadow_pte = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault,
- &write_pt);
+ &write_pt, page);
pgprintk("%s: shadow pte %p %llx ptwrite %d\n", __FUNCTION__,
shadow_pte, *shadow_pte, write_pt);
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 44b8925..20597bc 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -224,6 +224,11 @@ struct kvm_vcpu_arch {
int last_pt_write_count;
u64 *last_pte_updated;
+ struct {
+ gfn_t gfn; /* presumed gfn during guest pte update */
+ struct page *page; /* page corresponding to that gfn */
+ } update_pte;
+
struct i387_fxsave_struct host_fx_image;
struct i387_fxsave_struct guest_fx_image;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 30/33] KVM: MMU: Switch to mmu spinlock
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (28 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 29/33] KVM: MMU: Avoid calling gfn_to_page() in mmu_set_spte() Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 31/33] KVM: MMU: Move kvm_free_some_pages() into critical section Avi Kivity
` (2 subsequent siblings)
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Marcelo Tosatti
From: Marcelo Tosatti <mtosatti@redhat.com>
Convert the synchronization of the shadow handling to a separate mmu_lock
spinlock.
Also guard fetch() by mmap_sem in read-mode to protect against alias
and memslot changes.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/mmu.c | 48 +++++++++++++++++++++++--------------------
arch/x86/kvm/paging_tmpl.h | 10 +++++---
arch/x86/kvm/vmx.c | 2 -
include/linux/kvm_host.h | 3 +-
virt/kvm/kvm_main.c | 3 +-
5 files changed, 35 insertions(+), 31 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index c0b757b..834698d 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -971,16 +971,12 @@ static void nonpaging_new_cr3(struct kvm_vcpu *vcpu)
{
}
-static int __nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
+static int __nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write,
+ gfn_t gfn, struct page *page)
{
int level = PT32E_ROOT_LEVEL;
hpa_t table_addr = vcpu->arch.mmu.root_hpa;
int pt_write = 0;
- struct page *page;
-
- down_read(¤t->mm->mmap_sem);
- page = gfn_to_page(vcpu->kvm, gfn);
- up_read(¤t->mm->mmap_sem);
for (; ; level--) {
u32 index = PT64_INDEX(v, level);
@@ -1022,9 +1018,17 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
{
int r;
- mutex_lock(&vcpu->kvm->lock);
- r = __nonpaging_map(vcpu, v, write, gfn);
- mutex_unlock(&vcpu->kvm->lock);
+ struct page *page;
+
+ down_read(¤t->mm->mmap_sem);
+ page = gfn_to_page(vcpu->kvm, gfn);
+
+ spin_lock(&vcpu->kvm->mmu_lock);
+ r = __nonpaging_map(vcpu, v, write, gfn, page);
+ spin_unlock(&vcpu->kvm->mmu_lock);
+
+ up_read(¤t->mm->mmap_sem);
+
return r;
}
@@ -1045,7 +1049,7 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu)
if (!VALID_PAGE(vcpu->arch.mmu.root_hpa))
return;
- mutex_lock(&vcpu->kvm->lock);
+ spin_lock(&vcpu->kvm->mmu_lock);
#ifdef CONFIG_X86_64
if (vcpu->arch.mmu.shadow_root_level == PT64_ROOT_LEVEL) {
hpa_t root = vcpu->arch.mmu.root_hpa;
@@ -1053,7 +1057,7 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu)
sp = page_header(root);
--sp->root_count;
vcpu->arch.mmu.root_hpa = INVALID_PAGE;
- mutex_unlock(&vcpu->kvm->lock);
+ spin_unlock(&vcpu->kvm->mmu_lock);
return;
}
#endif
@@ -1067,7 +1071,7 @@ static void mmu_free_roots(struct kvm_vcpu *vcpu)
}
vcpu->arch.mmu.pae_root[i] = INVALID_PAGE;
}
- mutex_unlock(&vcpu->kvm->lock);
+ spin_unlock(&vcpu->kvm->mmu_lock);
vcpu->arch.mmu.root_hpa = INVALID_PAGE;
}
@@ -1270,9 +1274,9 @@ int kvm_mmu_load(struct kvm_vcpu *vcpu)
r = mmu_topup_memory_caches(vcpu);
if (r)
goto out;
- mutex_lock(&vcpu->kvm->lock);
+ spin_lock(&vcpu->kvm->mmu_lock);
mmu_alloc_roots(vcpu);
- mutex_unlock(&vcpu->kvm->lock);
+ spin_unlock(&vcpu->kvm->mmu_lock);
kvm_x86_ops->set_cr3(vcpu, vcpu->arch.mmu.root_hpa);
kvm_mmu_flush_tlb(vcpu);
out:
@@ -1408,7 +1412,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
pgprintk("%s: gpa %llx bytes %d\n", __FUNCTION__, gpa, bytes);
mmu_guess_page_from_pte_write(vcpu, gpa, new, bytes);
- mutex_lock(&vcpu->kvm->lock);
+ spin_lock(&vcpu->kvm->mmu_lock);
++vcpu->kvm->stat.mmu_pte_write;
kvm_mmu_audit(vcpu, "pre pte write");
if (gfn == vcpu->arch.last_pt_write_gfn
@@ -1477,7 +1481,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
}
}
kvm_mmu_audit(vcpu, "post pte write");
- mutex_unlock(&vcpu->kvm->lock);
+ spin_unlock(&vcpu->kvm->mmu_lock);
if (vcpu->arch.update_pte.page) {
kvm_release_page_clean(vcpu->arch.update_pte.page);
vcpu->arch.update_pte.page = NULL;
@@ -1493,15 +1497,15 @@ int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, gva);
up_read(¤t->mm->mmap_sem);
- mutex_lock(&vcpu->kvm->lock);
+ spin_lock(&vcpu->kvm->mmu_lock);
r = kvm_mmu_unprotect_page(vcpu->kvm, gpa >> PAGE_SHIFT);
- mutex_unlock(&vcpu->kvm->lock);
+ spin_unlock(&vcpu->kvm->mmu_lock);
return r;
}
void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
{
- mutex_lock(&vcpu->kvm->lock);
+ spin_lock(&vcpu->kvm->mmu_lock);
while (vcpu->kvm->arch.n_free_mmu_pages < KVM_REFILL_PAGES) {
struct kvm_mmu_page *sp;
@@ -1510,7 +1514,7 @@ void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
kvm_mmu_zap_page(vcpu->kvm, sp);
++vcpu->kvm->stat.mmu_recycled;
}
- mutex_unlock(&vcpu->kvm->lock);
+ spin_unlock(&vcpu->kvm->mmu_lock);
}
int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code)
@@ -1642,10 +1646,10 @@ void kvm_mmu_zap_all(struct kvm *kvm)
{
struct kvm_mmu_page *sp, *node;
- mutex_lock(&kvm->lock);
+ spin_lock(&kvm->mmu_lock);
list_for_each_entry_safe(sp, node, &kvm->arch.active_mmu_pages, link)
kvm_mmu_zap_page(kvm, sp);
- mutex_unlock(&kvm->lock);
+ spin_unlock(&kvm->mmu_lock);
kvm_flush_remote_tlbs(kvm);
}
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 3d7846b..a35b83a 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -387,7 +387,6 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
*/
r = FNAME(walk_addr)(&walker, vcpu, addr, write_fault, user_fault,
fetch_fault);
- up_read(¤t->mm->mmap_sem);
/*
* The page is not mapped by the guest. Let the guest handle it.
@@ -396,12 +395,13 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
pgprintk("%s: guest page fault\n", __FUNCTION__);
inject_page_fault(vcpu, addr, walker.error_code);
vcpu->arch.last_pt_write_count = 0; /* reset fork detector */
+ up_read(¤t->mm->mmap_sem);
return 0;
}
page = gfn_to_page(vcpu->kvm, walker.gfn);
- mutex_lock(&vcpu->kvm->lock);
+ spin_lock(&vcpu->kvm->mmu_lock);
shadow_pte = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault,
&write_pt, page);
pgprintk("%s: shadow pte %p %llx ptwrite %d\n", __FUNCTION__,
@@ -414,13 +414,15 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
* mmio: emulate if accessible, otherwise its a guest fault.
*/
if (shadow_pte && is_io_pte(*shadow_pte)) {
- mutex_unlock(&vcpu->kvm->lock);
+ spin_unlock(&vcpu->kvm->mmu_lock);
+ up_read(¤t->mm->mmap_sem);
return 1;
}
++vcpu->stat.pf_fixed;
kvm_mmu_audit(vcpu, "post page fault (fixed)");
- mutex_unlock(&vcpu->kvm->lock);
+ spin_unlock(&vcpu->kvm->mmu_lock);
+ up_read(¤t->mm->mmap_sem);
return write_pt;
}
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 99fec63..4741806 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -1476,7 +1476,6 @@ static int alloc_apic_access_page(struct kvm *kvm)
struct kvm_userspace_memory_region kvm_userspace_mem;
int r = 0;
- mutex_lock(&kvm->lock);
down_write(¤t->mm->mmap_sem);
if (kvm->arch.apic_access_page)
goto out;
@@ -1490,7 +1489,6 @@ static int alloc_apic_access_page(struct kvm *kvm)
kvm->arch.apic_access_page = gfn_to_page(kvm, 0xfee00);
out:
up_write(¤t->mm->mmap_sem);
- mutex_unlock(&kvm->lock);
return r;
}
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index a020fb2..2714068 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -104,7 +104,8 @@ struct kvm_memory_slot {
};
struct kvm {
- struct mutex lock; /* protects everything except vcpus */
+ struct mutex lock; /* protects the vcpus array and APIC accesses */
+ spinlock_t mmu_lock;
struct mm_struct *mm; /* userspace tied to this vm */
int nmemslots;
struct kvm_memory_slot memslots[KVM_MEMORY_SLOTS +
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index c462d7e..4295623 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -165,6 +165,7 @@ static struct kvm *kvm_create_vm(void)
kvm->mm = current->mm;
atomic_inc(&kvm->mm->mm_count);
+ spin_lock_init(&kvm->mmu_lock);
kvm_io_bus_init(&kvm->pio_bus);
mutex_init(&kvm->lock);
kvm_io_bus_init(&kvm->mmio_bus);
@@ -552,9 +553,7 @@ int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data,
addr = gfn_to_hva(kvm, gfn);
if (kvm_is_error_hva(addr))
return -EFAULT;
- pagefault_disable();
r = __copy_from_user_inatomic(data, (void __user *)addr + offset, len);
- pagefault_enable();
if (r)
return -EFAULT;
return 0;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 31/33] KVM: MMU: Move kvm_free_some_pages() into critical section
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (29 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 30/33] KVM: MMU: Switch to mmu spinlock Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 32/33] KVM: MMU: Merge shadow level check in FNAME(fetch) Avi Kivity
2008-01-06 14:39 ` [PATCH 33/33] KVM: x86 emulator: Only allow VMCALL/VMMCALL trapped by #UD Avi Kivity
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel
If some other cpu steals mmu pages between our check and an attempt to
allocate, we can run out of mmu pages. Fix by moving the check into the
same critical section as the allocation.
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/mmu.c | 9 +++------
arch/x86/kvm/paging_tmpl.h | 1 +
2 files changed, 4 insertions(+), 6 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 834698d..c478ee2 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -291,7 +291,6 @@ static int mmu_topup_memory_caches(struct kvm_vcpu *vcpu)
{
int r;
- kvm_mmu_free_some_pages(vcpu);
r = mmu_topup_memory_cache(&vcpu->arch.mmu_pte_chain_cache,
pte_chain_cache, 4);
if (r)
@@ -569,9 +568,6 @@ static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu,
{
struct kvm_mmu_page *sp;
- if (!vcpu->kvm->arch.n_free_mmu_pages)
- return NULL;
-
sp = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_header_cache, sizeof *sp);
sp->spt = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
sp->gfns = mmu_memory_cache_alloc(&vcpu->arch.mmu_page_cache, PAGE_SIZE);
@@ -1024,6 +1020,7 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn)
page = gfn_to_page(vcpu->kvm, gfn);
spin_lock(&vcpu->kvm->mmu_lock);
+ kvm_mmu_free_some_pages(vcpu);
r = __nonpaging_map(vcpu, v, write, gfn, page);
spin_unlock(&vcpu->kvm->mmu_lock);
@@ -1275,6 +1272,7 @@ int kvm_mmu_load(struct kvm_vcpu *vcpu)
if (r)
goto out;
spin_lock(&vcpu->kvm->mmu_lock);
+ kvm_mmu_free_some_pages(vcpu);
mmu_alloc_roots(vcpu);
spin_unlock(&vcpu->kvm->mmu_lock);
kvm_x86_ops->set_cr3(vcpu, vcpu->arch.mmu.root_hpa);
@@ -1413,6 +1411,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa,
pgprintk("%s: gpa %llx bytes %d\n", __FUNCTION__, gpa, bytes);
mmu_guess_page_from_pte_write(vcpu, gpa, new, bytes);
spin_lock(&vcpu->kvm->mmu_lock);
+ kvm_mmu_free_some_pages(vcpu);
++vcpu->kvm->stat.mmu_pte_write;
kvm_mmu_audit(vcpu, "pre pte write");
if (gfn == vcpu->arch.last_pt_write_gfn
@@ -1505,7 +1504,6 @@ int kvm_mmu_unprotect_page_virt(struct kvm_vcpu *vcpu, gva_t gva)
void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
{
- spin_lock(&vcpu->kvm->mmu_lock);
while (vcpu->kvm->arch.n_free_mmu_pages < KVM_REFILL_PAGES) {
struct kvm_mmu_page *sp;
@@ -1514,7 +1512,6 @@ void __kvm_mmu_free_some_pages(struct kvm_vcpu *vcpu)
kvm_mmu_zap_page(vcpu->kvm, sp);
++vcpu->kvm->stat.mmu_recycled;
}
- spin_unlock(&vcpu->kvm->mmu_lock);
}
int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code)
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index a35b83a..3499205 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -402,6 +402,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr,
page = gfn_to_page(vcpu->kvm, walker.gfn);
spin_lock(&vcpu->kvm->mmu_lock);
+ kvm_mmu_free_some_pages(vcpu);
shadow_pte = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault,
&write_pt, page);
pgprintk("%s: shadow pte %p %llx ptwrite %d\n", __FUNCTION__,
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 32/33] KVM: MMU: Merge shadow level check in FNAME(fetch)
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (30 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 31/33] KVM: MMU: Move kvm_free_some_pages() into critical section Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 33/33] KVM: x86 emulator: Only allow VMCALL/VMMCALL trapped by #UD Avi Kivity
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Dong, Eddie
From: Dong, Eddie <eddie.dong@intel.com>
Remove the redundant level check when fetching
shadow pte for present & non-present spte.
Signed-off-by: Yaozu (Eddie) Dong <eddie.dong@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
arch/x86/kvm/paging_tmpl.h | 7 ++-----
1 files changed, 2 insertions(+), 5 deletions(-)
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 3499205..03ba860 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -300,16 +300,13 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
bool new_page = 0;
shadow_ent = ((u64 *)__va(shadow_addr)) + index;
+ if (level == PT_PAGE_TABLE_LEVEL)
+ break;
if (is_shadow_present_pte(*shadow_ent)) {
- if (level == PT_PAGE_TABLE_LEVEL)
- break;
shadow_addr = *shadow_ent & PT64_BASE_ADDR_MASK;
continue;
}
- if (level == PT_PAGE_TABLE_LEVEL)
- break;
-
if (level - 1 == PT_PAGE_TABLE_LEVEL
&& walker->level == PT_DIRECTORY_LEVEL) {
metaphysical = 1;
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread* [PATCH 33/33] KVM: x86 emulator: Only allow VMCALL/VMMCALL trapped by #UD
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
` (31 preceding siblings ...)
2008-01-06 14:39 ` [PATCH 32/33] KVM: MMU: Merge shadow level check in FNAME(fetch) Avi Kivity
@ 2008-01-06 14:39 ` Avi Kivity
32 siblings, 0 replies; 36+ messages in thread
From: Avi Kivity @ 2008-01-06 14:39 UTC (permalink / raw)
To: linux-kernel, kvm-devel; +Cc: Sheng Yang
From: Sheng Yang <sheng.yang@intel.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>
---
arch/x86/kvm/svm.c | 2 +-
arch/x86/kvm/vmx.c | 2 +-
arch/x86/kvm/x86.c | 18 +++++++++++++++---
include/asm-x86/kvm_host.h | 4 +++-
4 files changed, 20 insertions(+), 6 deletions(-)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 11e5baf..2da9e11 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -942,7 +942,7 @@ static int ud_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
{
int er;
- er = emulate_instruction(&svm->vcpu, kvm_run, 0, 0, 0);
+ er = emulate_instruction(&svm->vcpu, kvm_run, 0, 0, EMULTYPE_TRAP_UD);
if (er != EMULATE_DONE)
kvm_queue_exception(&svm->vcpu, UD_VECTOR);
return 1;
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 4741806..5346e42 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -1862,7 +1862,7 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
}
if (is_invalid_opcode(intr_info)) {
- er = emulate_instruction(vcpu, kvm_run, 0, 0, 0);
+ er = emulate_instruction(vcpu, kvm_run, 0, 0, EMULTYPE_TRAP_UD);
if (er != EMULATE_DONE)
kvm_queue_exception(vcpu, UD_VECTOR);
return 1;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index e3b3141..8a90403 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1840,9 +1840,10 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
struct kvm_run *run,
unsigned long cr2,
u16 error_code,
- int no_decode)
+ int emulation_type)
{
int r;
+ struct decode_cache *c;
vcpu->arch.mmio_fault_cr2 = cr2;
kvm_x86_ops->cache_regs(vcpu);
@@ -1850,7 +1851,7 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
vcpu->mmio_is_write = 0;
vcpu->arch.pio.string = 0;
- if (!no_decode) {
+ if (!(emulation_type & EMULTYPE_NO_DECODE)) {
int cs_db, cs_l;
kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);
@@ -1884,6 +1885,16 @@ int emulate_instruction(struct kvm_vcpu *vcpu,
get_segment_base(vcpu, VCPU_SREG_FS);
r = x86_decode_insn(&vcpu->arch.emulate_ctxt, &emulate_ops);
+
+ /* Reject the instructions other than VMCALL/VMMCALL when
+ * try to emulate invalid opcode */
+ c = &vcpu->arch.emulate_ctxt.decode;
+ if ((emulation_type & EMULTYPE_TRAP_UD) &&
+ (!(c->twobyte && c->b == 0x01 &&
+ (c->modrm_reg == 0 || c->modrm_reg == 3) &&
+ c->modrm_mod == 3 && c->modrm_rm == 1)))
+ return EMULATE_FAIL;
+
++vcpu->stat.insn_emulation;
if (r) {
++vcpu->stat.insn_emulation_fail;
@@ -2640,7 +2651,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
vcpu->mmio_read_completed = 1;
vcpu->mmio_needed = 0;
r = emulate_instruction(vcpu, kvm_run,
- vcpu->arch.mmio_fault_cr2, 0, 1);
+ vcpu->arch.mmio_fault_cr2, 0,
+ EMULTYPE_NO_DECODE);
if (r == EMULATE_DO_MMIO) {
/*
* Read-modify-write. Back to userspace.
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 20597bc..4702b04 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -415,8 +415,10 @@ enum emulation_result {
EMULATE_FAIL, /* can't emulate this instruction */
};
+#define EMULTYPE_NO_DECODE (1 << 0)
+#define EMULTYPE_TRAP_UD (1 << 1)
int emulate_instruction(struct kvm_vcpu *vcpu, struct kvm_run *run,
- unsigned long cr2, u16 error_code, int no_decode);
+ unsigned long cr2, u16 error_code, int emulation_type);
void kvm_report_emulation_failure(struct kvm_vcpu *cvpu, const char *context);
void realmode_lgdt(struct kvm_vcpu *vcpu, u16 size, unsigned long address);
void realmode_lidt(struct kvm_vcpu *vcpu, u16 size, unsigned long address);
--
1.5.3.7
^ permalink raw reply related [flat|nested] 36+ messages in thread