[PATCH 34/54] KVM: Replace #GP injection by the generalized exception queue

From: Avi Kivity
Date: Tue Jan 01 2008 - 10:47:05 EST


Signed-off-by: Avi Kivity <avi@xxxxxxxxxxxx>
---
drivers/kvm/svm.c | 17 ++---------------
drivers/kvm/vmx.c | 18 ++----------------
drivers/kvm/x86.c | 43 +++++++++++++++++++------------------------
drivers/kvm/x86.h | 7 +++++--
drivers/kvm/x86_emulate.c | 4 ++--
5 files changed, 30 insertions(+), 59 deletions(-)

diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c
index ce77f15..b896614 100644
--- a/drivers/kvm/svm.c
+++ b/drivers/kvm/svm.c
@@ -207,17 +207,6 @@ static bool svm_exception_injected(struct kvm_vcpu *vcpu)
return !(svm->vmcb->control.exit_int_info & SVM_EXITINTINFO_VALID);
}

-static void svm_inject_gp(struct kvm_vcpu *vcpu, unsigned error_code)
-{
- struct vcpu_svm *svm = to_svm(vcpu);
-
- svm->vmcb->control.event_inj = SVM_EVTINJ_VALID |
- SVM_EVTINJ_VALID_ERR |
- SVM_EVTINJ_TYPE_EXEPT |
- GP_VECTOR;
- svm->vmcb->control.event_inj_err = error_code;
-}
-
static void inject_ud(struct kvm_vcpu *vcpu)
{
to_svm(vcpu)->vmcb->control.event_inj = SVM_EVTINJ_VALID |
@@ -1115,7 +1104,7 @@ static int rdmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
u64 data;

if (svm_get_msr(&svm->vcpu, ecx, &data))
- svm_inject_gp(&svm->vcpu, 0);
+ kvm_inject_gp(&svm->vcpu, 0);
else {
svm->vmcb->save.rax = data & 0xffffffff;
svm->vcpu.regs[VCPU_REGS_RDX] = data >> 32;
@@ -1176,7 +1165,7 @@ static int wrmsr_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
| ((u64)(svm->vcpu.regs[VCPU_REGS_RDX] & -1u) << 32);
svm->next_rip = svm->vmcb->save.rip + 2;
if (svm_set_msr(&svm->vcpu, ecx, data))
- svm_inject_gp(&svm->vcpu, 0);
+ kvm_inject_gp(&svm->vcpu, 0);
else
skip_emulated_instruction(&svm->vcpu);
return 1;
@@ -1688,8 +1677,6 @@ static struct kvm_x86_ops svm_x86_ops = {

.tlb_flush = svm_flush_tlb,

- .inject_gp = svm_inject_gp,
-
.run = svm_vcpu_run,
.handle_exit = handle_exit,
.skip_emulated_instruction = skip_emulated_instruction,
diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c
index 20e9dfc..92660db 100644
--- a/drivers/kvm/vmx.c
+++ b/drivers/kvm/vmx.c
@@ -613,18 +613,6 @@ static bool vmx_exception_injected(struct kvm_vcpu *vcpu)
return !(vmx->idt_vectoring_info & VECTORING_INFO_VALID_MASK);
}

-static void vmx_inject_gp(struct kvm_vcpu *vcpu, unsigned error_code)
-{
- printk(KERN_DEBUG "inject_general_protection: rip 0x%lx\n",
- vmcs_readl(GUEST_RIP));
- vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, error_code);
- vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
- GP_VECTOR |
- INTR_TYPE_EXCEPTION |
- INTR_INFO_DELIEVER_CODE_MASK |
- INTR_INFO_VALID_MASK);
-}
-
static void vmx_inject_ud(struct kvm_vcpu *vcpu)
{
vmcs_write32(VM_ENTRY_INTR_INFO_FIELD,
@@ -2083,7 +2071,7 @@ static int handle_rdmsr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
u64 data;

if (vmx_get_msr(vcpu, ecx, &data)) {
- vmx_inject_gp(vcpu, 0);
+ kvm_inject_gp(vcpu, 0);
return 1;
}

@@ -2101,7 +2089,7 @@ static int handle_wrmsr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
| ((u64)(vcpu->regs[VCPU_REGS_RDX] & -1u) << 32);

if (vmx_set_msr(vcpu, ecx, data) != 0) {
- vmx_inject_gp(vcpu, 0);
+ kvm_inject_gp(vcpu, 0);
return 1;
}

@@ -2619,8 +2607,6 @@ static struct kvm_x86_ops vmx_x86_ops = {

.tlb_flush = vmx_flush_tlb,

- .inject_gp = vmx_inject_gp,
-
.run = vmx_vcpu_run,
.handle_exit = kvm_handle_exit,
.skip_emulated_instruction = skip_emulated_instruction,
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index dc007a3..6deb052 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -128,11 +128,6 @@ void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data)
}
EXPORT_SYMBOL_GPL(kvm_set_apic_base);

-static void inject_gp(struct kvm_vcpu *vcpu)
-{
- kvm_x86_ops->inject_gp(vcpu, 0);
-}
-
void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr)
{
WARN_ON(vcpu->exception.pending);
@@ -232,20 +227,20 @@ 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);
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}

if ((cr0 & X86_CR0_NW) && !(cr0 & X86_CR0_CD)) {
printk(KERN_DEBUG "set_cr0: #GP, CD == 0 && NW == 1\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}

if ((cr0 & X86_CR0_PG) && !(cr0 & X86_CR0_PE)) {
printk(KERN_DEBUG "set_cr0: #GP, set PG flag "
"and a clear PE flag\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}

@@ -257,14 +252,14 @@ void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
if (!is_pae(vcpu)) {
printk(KERN_DEBUG "set_cr0: #GP, start paging "
"in long mode while PAE is disabled\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}
kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);
if (cs_l) {
printk(KERN_DEBUG "set_cr0: #GP, start paging "
"in long mode while CS.L == 1\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;

}
@@ -273,7 +268,7 @@ void set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
if (is_pae(vcpu) && !load_pdptrs(vcpu, vcpu->cr3)) {
printk(KERN_DEBUG "set_cr0: #GP, pdptrs "
"reserved bits\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}

@@ -299,7 +294,7 @@ void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
{
if (cr4 & CR4_RESERVED_BITS) {
printk(KERN_DEBUG "set_cr4: #GP, reserved bits\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}

@@ -307,19 +302,19 @@ void set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
if (!(cr4 & X86_CR4_PAE)) {
printk(KERN_DEBUG "set_cr4: #GP, clearing PAE while "
"in long mode\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}
} else if (is_paging(vcpu) && !is_pae(vcpu) && (cr4 & X86_CR4_PAE)
&& !load_pdptrs(vcpu, vcpu->cr3)) {
printk(KERN_DEBUG "set_cr4: #GP, pdptrs reserved bits\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}

if (cr4 & X86_CR4_VMXE) {
printk(KERN_DEBUG "set_cr4: #GP, setting VMXE\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}
kvm_x86_ops->set_cr4(vcpu, cr4);
@@ -340,7 +335,7 @@ void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
if (is_long_mode(vcpu)) {
if (cr3 & CR3_L_MODE_RESERVED_BITS) {
printk(KERN_DEBUG "set_cr3: #GP, reserved bits\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}
} else {
@@ -348,13 +343,13 @@ void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
if (cr3 & CR3_PAE_RESERVED_BITS) {
printk(KERN_DEBUG
"set_cr3: #GP, reserved bits\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}
if (is_paging(vcpu) && !load_pdptrs(vcpu, cr3)) {
printk(KERN_DEBUG "set_cr3: #GP, pdptrs "
"reserved bits\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}
}
@@ -375,7 +370,7 @@ void set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
* to debug) behavior on the guest side.
*/
if (unlikely(!gfn_to_memslot(vcpu->kvm, cr3 >> PAGE_SHIFT)))
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
else {
vcpu->cr3 = cr3;
vcpu->mmu.new_cr3(vcpu);
@@ -388,7 +383,7 @@ void set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8)
{
if (cr8 & CR8_RESERVED_BITS) {
printk(KERN_DEBUG "set_cr8: #GP, reserved bits 0x%lx\n", cr8);
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}
if (irqchip_in_kernel(vcpu->kvm))
@@ -436,14 +431,14 @@ static void set_efer(struct kvm_vcpu *vcpu, u64 efer)
if (efer & EFER_RESERVED_BITS) {
printk(KERN_DEBUG "set_efer: 0x%llx #GP, reserved bits\n",
efer);
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}

if (is_paging(vcpu)
&& (vcpu->shadow_efer & EFER_LME) != (efer & EFER_LME)) {
printk(KERN_DEBUG "set_efer: #GP, change LME while paging\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return;
}

@@ -2047,7 +2042,7 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
* String I/O in reverse. Yuck. Kill the guest, fix later.
*/
pr_unimpl(vcpu, "guest string pio down\n");
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
return 1;
}
vcpu->run->io.count = now;
@@ -2062,7 +2057,7 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in,
vcpu->pio.guest_pages[i] = page;
mutex_unlock(&vcpu->kvm->lock);
if (!page) {
- inject_gp(vcpu);
+ kvm_inject_gp(vcpu, 0);
free_pio_guest_pages(vcpu);
return 1;
}
diff --git a/drivers/kvm/x86.h b/drivers/kvm/x86.h
index d3ac4e2..78396d6 100644
--- a/drivers/kvm/x86.h
+++ b/drivers/kvm/x86.h
@@ -220,8 +220,6 @@ struct kvm_x86_ops {

void (*tlb_flush)(struct kvm_vcpu *vcpu);

- void (*inject_gp)(struct kvm_vcpu *vcpu, unsigned err_code);
-
void (*run)(struct kvm_vcpu *vcpu, struct kvm_run *run);
int (*handle_exit)(struct kvm_run *run, struct kvm_vcpu *vcpu);
void (*skip_emulated_instruction)(struct kvm_vcpu *vcpu);
@@ -467,6 +465,11 @@ static inline u32 get_rdx_init_val(void)
return 0x600; /* P6 family */
}

+static inline void kvm_inject_gp(struct kvm_vcpu *vcpu, u32 error_code)
+{
+ kvm_queue_exception_e(vcpu, GP_VECTOR, error_code);
+}
+
#define ASM_VMX_VMCLEAR_RAX ".byte 0x66, 0x0f, 0xc7, 0x30"
#define ASM_VMX_VMLAUNCH ".byte 0x0f, 0x01, 0xc2"
#define ASM_VMX_VMRESUME ".byte 0x0f, 0x01, 0xc3"
diff --git a/drivers/kvm/x86_emulate.c b/drivers/kvm/x86_emulate.c
index 3e3eba7..2e259a8 100644
--- a/drivers/kvm/x86_emulate.c
+++ b/drivers/kvm/x86_emulate.c
@@ -1779,7 +1779,7 @@ twobyte_insn:
| ((u64)c->regs[VCPU_REGS_RDX] << 32);
rc = kvm_set_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], msr_data);
if (rc) {
- kvm_x86_ops->inject_gp(ctxt->vcpu, 0);
+ kvm_inject_gp(ctxt->vcpu, 0);
c->eip = ctxt->vcpu->rip;
}
rc = X86EMUL_CONTINUE;
@@ -1789,7 +1789,7 @@ twobyte_insn:
/* rdmsr */
rc = kvm_get_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], &msr_data);
if (rc) {
- kvm_x86_ops->inject_gp(ctxt->vcpu, 0);
+ kvm_inject_gp(ctxt->vcpu, 0);
c->eip = ctxt->vcpu->rip;
} else {
c->regs[VCPU_REGS_RAX] = (u32)msr_data;
--
1.5.3.7

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/