| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <CACzj_yU5x=EePe-HsMw7ABrL5F7Pf2R_GLevUJpMCCdr1HNUPw@mail.gmail.com>
Date: Sat, 24 Jan 2015 18:23:41 +0800
From: Wincy Van <fanwenyi0529@...il.com>
To: Paolo Bonzini <pbonzini@...hat.com>,
"gleb@...nel.org" <gleb@...nel.org>,
"Zhang, Yang Z" <yang.z.zhang@...el.com>
Cc: "kvm@...r.kernel.org" <kvm@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Wanpeng Li <wanpeng.li@...ux.intel.com>,
Jan Kiszka <jan.kiszka@....de>,
θζδΈ <fanwenyi0529@...il.com>
Subject: [PATCH v3 5/6] KVM: nVMX: Enable nested virtual interrupt delivery.
With virtual interrupt delivery, the hardware prevent KVM from
the low efficiency interrupt inject way. In nested vmx, it is
a important feature, we can reduce much more nested-vmexit,
especially in high throughput scenes.
Signed-off-by: Wincy Van <fanwenyi0529@...il.com>
---
arch/x86/kvm/vmx.c | 63 +++++++++++++++++++++++++++++++++++++++++++++++++++-
1 files changed, 62 insertions(+), 1 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index d57e370..5d8500c 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -216,6 +216,10 @@ struct __packed vmcs12 {
u64 virtual_apic_page_addr;
u64 apic_access_addr;
u64 ept_pointer;
+ u64 eoi_exit_bitmap0;
+ u64 eoi_exit_bitmap1;
+ u64 eoi_exit_bitmap2;
+ u64 eoi_exit_bitmap3;
u64 xss_exit_bitmap;
u64 guest_physical_address;
u64 vmcs_link_pointer;
@@ -338,6 +342,7 @@ struct __packed vmcs12 {
u16 guest_gs_selector;
u16 guest_ldtr_selector;
u16 guest_tr_selector;
+ u16 guest_intr_status;
u16 host_es_selector;
u16 host_cs_selector;
u16 host_ss_selector;
@@ -624,6 +629,7 @@ static const unsigned short vmcs_field_to_offset_table[] = {
FIELD(GUEST_GS_SELECTOR, guest_gs_selector),
FIELD(GUEST_LDTR_SELECTOR, guest_ldtr_selector),
FIELD(GUEST_TR_SELECTOR, guest_tr_selector),
+ FIELD(GUEST_INTR_STATUS, guest_intr_status),
FIELD(HOST_ES_SELECTOR, host_es_selector),
FIELD(HOST_CS_SELECTOR, host_cs_selector),
FIELD(HOST_SS_SELECTOR, host_ss_selector),
@@ -641,6 +647,10 @@ static const unsigned short
vmcs_field_to_offset_table[] = {
FIELD64(VIRTUAL_APIC_PAGE_ADDR, virtual_apic_page_addr),
FIELD64(APIC_ACCESS_ADDR, apic_access_addr),
FIELD64(EPT_POINTER, ept_pointer),
+ FIELD64(EOI_EXIT_BITMAP0, eoi_exit_bitmap0),
+ FIELD64(EOI_EXIT_BITMAP1, eoi_exit_bitmap1),
+ FIELD64(EOI_EXIT_BITMAP2, eoi_exit_bitmap2),
+ FIELD64(EOI_EXIT_BITMAP3, eoi_exit_bitmap3),
FIELD64(XSS_EXIT_BITMAP, xss_exit_bitmap),
FIELD64(GUEST_PHYSICAL_ADDRESS, guest_physical_address),
FIELD64(VMCS_LINK_POINTER, vmcs_link_pointer),
@@ -1135,6 +1145,11 @@ static inline bool
nested_cpu_has_apic_reg_virt(struct vmcs12 *vmcs12)
return nested_cpu_has2(vmcs12, SECONDARY_EXEC_APIC_REGISTER_VIRT);
}
+static inline bool nested_cpu_has_vid(struct vmcs12 *vmcs12)
+{
+ return nested_cpu_has2(vmcs12, SECONDARY_EXEC_VIRTUAL_INTR_DELIVERY);
+}
+
static inline bool is_exception(u32 intr_info)
{
return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VALID_MASK))
@@ -2429,6 +2444,7 @@ static void nested_vmx_setup_ctls_msrs(struct
vcpu_vmx *vmx)
SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES |
SECONDARY_EXEC_VIRTUALIZE_X2APIC_MODE |
SECONDARY_EXEC_APIC_REGISTER_VIRT |
+ SECONDARY_EXEC_VIRTUAL_INTR_DELIVERY |
SECONDARY_EXEC_WBINVD_EXITING |
SECONDARY_EXEC_XSAVES;
@@ -7389,6 +7405,10 @@ static bool nested_vmx_exit_handled(struct
kvm_vcpu *vcpu)
case EXIT_REASON_APIC_ACCESS:
return nested_cpu_has2(vmcs12,
SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES);
+ case EXIT_REASON_APIC_WRITE:
+ case EXIT_REASON_EOI_INDUCED:
+ /* apic_write and eoi_induced should exit unconditionally. */
+ return 1;
case EXIT_REASON_EPT_VIOLATION:
/*
* L0 always deals with the EPT violation. If nested EPT is
@@ -8474,6 +8494,19 @@ static inline bool
nested_vmx_merge_msr_bitmap(struct kvm_vcpu *vcpu,
vmx_msr_bitmap_nested,
APIC_BASE_MSR + (APIC_TASKPRI >> 4),
MSR_TYPE_R | MSR_TYPE_W);
+ if (nested_cpu_has_vid(vmcs12)) {
+ /* EOI and self-IPI are allowed */
+ nested_vmx_disable_intercept_for_msr(
+ msr_bitmap,
+ vmx_msr_bitmap_nested,
+ APIC_BASE_MSR + (APIC_EOI >> 4),
+ MSR_TYPE_W);
+ nested_vmx_disable_intercept_for_msr(
+ msr_bitmap,
+ vmx_msr_bitmap_nested,
+ APIC_BASE_MSR + (APIC_SELF_IPI >> 4),
+ MSR_TYPE_W);
+ }
}
kunmap(page);
nested_release_page_clean(page);
@@ -8489,17 +8522,29 @@ static inline int
nested_vmx_check_virt_x2apic(struct kvm_vcpu *vcpu,
return 0;
}
+static inline int nested_vmx_check_vid(struct kvm_vcpu *vcpu,
+ struct vmcs12 *vmcs12)
+{
+ if (!nested_exit_on_intr(vcpu))
+ return -EINVAL;
+ return 0;
+}
+
static int nested_vmx_check_apicv_controls(struct kvm_vcpu *vcpu,
struct vmcs12 *vmcs12)
{
int r = 0;
if (!nested_cpu_has_virt_x2apic_mode(vmcs12) &&
- !nested_cpu_has_apic_reg_virt(vmcs12))
+ !nested_cpu_has_apic_reg_virt(vmcs12) &&
+ !nested_cpu_has_vid(vmcs12))
return 0;
if (nested_cpu_has_virt_x2apic_mode(vmcs12))
r = nested_vmx_check_virt_x2apic(vcpu, vmcs12);
+ if (nested_cpu_has_vid(vmcs12))
+ r |= nested_vmx_check_vid(vcpu, vmcs12);
+
if (r)
goto fail;
@@ -8815,6 +8860,19 @@ static void prepare_vmcs02(struct kvm_vcpu
*vcpu, struct vmcs12 *vmcs12)
kvm_vcpu_reload_apic_access_page(vcpu);
}
+ if (exec_control & SECONDARY_EXEC_VIRTUAL_INTR_DELIVERY) {
+ vmcs_write64(EOI_EXIT_BITMAP0,
+ vmcs12->eoi_exit_bitmap0);
+ vmcs_write64(EOI_EXIT_BITMAP1,
+ vmcs12->eoi_exit_bitmap1);
+ vmcs_write64(EOI_EXIT_BITMAP2,
+ vmcs12->eoi_exit_bitmap2);
+ vmcs_write64(EOI_EXIT_BITMAP3,
+ vmcs12->eoi_exit_bitmap3);
+ vmcs_write16(GUEST_INTR_STATUS,
+ vmcs12->guest_intr_status);
+ }
+
vmcs_write32(SECONDARY_VM_EXEC_CONTROL, exec_control);
}
@@ -9386,6 +9444,9 @@ static void prepare_vmcs12(struct kvm_vcpu
*vcpu, struct vmcs12 *vmcs12,
vmcs12->guest_pdptr3 = vmcs_read64(GUEST_PDPTR3);
}
+ if (nested_cpu_has_vid(vmcs12))
+ vmcs12->guest_intr_status = vmcs_read16(GUEST_INTR_STATUS);
+
vmcs12->vm_entry_controls =
(vmcs12->vm_entry_controls & ~VM_ENTRY_IA32E_MODE) |
(vm_entry_controls_get(to_vmx(vcpu)) & VM_ENTRY_IA32E_MODE);
--
1.7.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists