| 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: <20160922231840.26746-1-rkrcmar@redhat.com>
Date: Fri, 23 Sep 2016 01:18:40 +0200
From: Radim Krčmář <rkrcmar@...hat.com>
To: linux-kernel@...r.kernel.org, kvm@...r.kernel.org
Cc: Paolo Bonzini <pbonzini@...hat.com>,
Wanpeng Li <kernellwp@...il.com>
Subject: [PATCH] KVM: VMX: refactor global page-sized bitmaps
We've had 10 page-sized bitmaps that were being allocated and freed one
by one when we could just use a cycle and MSR bitmaps had a lot of
useless code lying around.
This patch
* enumerates vmx bitmaps and uses an array to store them
* replaces vmx_enable_intercept_msr_read_x2apic() with a condition
* joins vmx_msr_disable_intercept_msr_{read,write}_x2apic()
* renames x2apic_apicv_inactive msr_bitmaps to x2apic and original
x2apic bitmaps to x2apic_apicv
Signed-off-by: Radim Krčmář <rkrcmar@...hat.com>
---
arch/x86/kvm/vmx.c | 297 +++++++++++++++++------------------------------------
1 file changed, 92 insertions(+), 205 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 1cca146f4341..dfbcd45fcb2b 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -921,16 +921,20 @@ static DEFINE_PER_CPU(struct desc_ptr, host_gdt);
static DEFINE_PER_CPU(struct list_head, blocked_vcpu_on_cpu);
static DEFINE_PER_CPU(spinlock_t, blocked_vcpu_on_cpu_lock);
-static unsigned long *vmx_io_bitmap_a;
-static unsigned long *vmx_io_bitmap_b;
-static unsigned long *vmx_msr_bitmap_legacy;
-static unsigned long *vmx_msr_bitmap_longmode;
-static unsigned long *vmx_msr_bitmap_legacy_x2apic;
-static unsigned long *vmx_msr_bitmap_longmode_x2apic;
-static unsigned long *vmx_msr_bitmap_legacy_x2apic_apicv_inactive;
-static unsigned long *vmx_msr_bitmap_longmode_x2apic_apicv_inactive;
-static unsigned long *vmx_vmread_bitmap;
-static unsigned long *vmx_vmwrite_bitmap;
+enum vmx_bitmap {
+ vmx_io_bitmap_a,
+ vmx_io_bitmap_b,
+ vmx_msr_bitmap_legacy,
+ vmx_msr_bitmap_legacy_x2apic,
+ vmx_msr_bitmap_legacy_x2apic_apicv,
+ vmx_msr_bitmap_longmode,
+ vmx_msr_bitmap_longmode_x2apic,
+ vmx_msr_bitmap_longmode_x2apic_apicv,
+ vmx_vmread_bitmap,
+ vmx_vmwrite_bitmap,
+ VMX_BITMAP_NR
+};
+static unsigned long *vmx_bitmap[VMX_BITMAP_NR];
static bool cpu_has_load_ia32_efer;
static bool cpu_has_load_perf_global_ctrl;
@@ -2519,23 +2523,26 @@ static void move_msr_up(struct vcpu_vmx *vmx, int from, int to)
static void vmx_set_msr_bitmap(struct kvm_vcpu *vcpu)
{
- unsigned long *msr_bitmap;
+ enum vmx_bitmap msr_bitmap;
- if (is_guest_mode(vcpu))
- msr_bitmap = to_vmx(vcpu)->nested.msr_bitmap;
- else if (cpu_has_secondary_exec_ctrls() &&
- (vmcs_read32(SECONDARY_VM_EXEC_CONTROL) &
- SECONDARY_EXEC_VIRTUALIZE_X2APIC_MODE)) {
+ if (is_guest_mode(vcpu)) {
+ vmcs_write64(MSR_BITMAP, __pa(to_vmx(vcpu)->nested.msr_bitmap));
+ return;
+ }
+
+ if (cpu_has_secondary_exec_ctrls() &&
+ (vmcs_read32(SECONDARY_VM_EXEC_CONTROL) &
+ SECONDARY_EXEC_VIRTUALIZE_X2APIC_MODE)) {
if (enable_apicv && kvm_vcpu_apicv_active(vcpu)) {
if (is_long_mode(vcpu))
+ msr_bitmap = vmx_msr_bitmap_longmode_x2apic_apicv;
+ else
+ msr_bitmap = vmx_msr_bitmap_legacy_x2apic_apicv;
+ } else {
+ if (is_long_mode(vcpu))
msr_bitmap = vmx_msr_bitmap_longmode_x2apic;
else
msr_bitmap = vmx_msr_bitmap_legacy_x2apic;
- } else {
- if (is_long_mode(vcpu))
- msr_bitmap = vmx_msr_bitmap_longmode_x2apic_apicv_inactive;
- else
- msr_bitmap = vmx_msr_bitmap_legacy_x2apic_apicv_inactive;
}
} else {
if (is_long_mode(vcpu))
@@ -2544,7 +2551,7 @@ static void vmx_set_msr_bitmap(struct kvm_vcpu *vcpu)
msr_bitmap = vmx_msr_bitmap_legacy;
}
- vmcs_write64(MSR_BITMAP, __pa(msr_bitmap));
+ vmcs_write64(MSR_BITMAP, __pa(vmx_bitmap[msr_bitmap]));
}
/*
@@ -3600,13 +3607,13 @@ static void init_vmcs_shadow_fields(void)
/* shadowed fields guest access without vmexit */
for (i = 0; i < max_shadow_read_write_fields; i++) {
clear_bit(shadow_read_write_fields[i],
- vmx_vmwrite_bitmap);
+ vmx_bitmap[vmx_vmwrite_bitmap]);
clear_bit(shadow_read_write_fields[i],
- vmx_vmread_bitmap);
+ vmx_bitmap[vmx_vmread_bitmap]);
}
for (i = 0; i < max_shadow_read_only_fields; i++)
clear_bit(shadow_read_only_fields[i],
- vmx_vmread_bitmap);
+ vmx_bitmap[vmx_vmread_bitmap]);
}
static __init int alloc_kvm_area(void)
@@ -4601,41 +4608,6 @@ static void __vmx_disable_intercept_for_msr(unsigned long *msr_bitmap,
}
}
-static void __vmx_enable_intercept_for_msr(unsigned long *msr_bitmap,
- u32 msr, int type)
-{
- int f = sizeof(unsigned long);
-
- if (!cpu_has_vmx_msr_bitmap())
- return;
-
- /*
- * See Intel PRM Vol. 3, 20.6.9 (MSR-Bitmap Address). Early manuals
- * have the write-low and read-high bitmap offsets the wrong way round.
- * We can control MSRs 0x00000000-0x00001fff and 0xc0000000-0xc0001fff.
- */
- if (msr <= 0x1fff) {
- if (type & MSR_TYPE_R)
- /* read-low */
- __set_bit(msr, msr_bitmap + 0x000 / f);
-
- if (type & MSR_TYPE_W)
- /* write-low */
- __set_bit(msr, msr_bitmap + 0x800 / f);
-
- } else if ((msr >= 0xc0000000) && (msr <= 0xc0001fff)) {
- msr &= 0x1fff;
- if (type & MSR_TYPE_R)
- /* read-high */
- __set_bit(msr, msr_bitmap + 0x400 / f);
-
- if (type & MSR_TYPE_W)
- /* write-high */
- __set_bit(msr, msr_bitmap + 0xc00 / f);
-
- }
-}
-
/*
* If a msr is allowed by L0, we should check whether it is allowed by L1.
* The corresponding bit will be cleared unless both of L0 and L1 allow it.
@@ -4684,55 +4656,31 @@ static void nested_vmx_disable_intercept_for_msr(unsigned long *msr_bitmap_l1,
static void vmx_disable_intercept_for_msr(u32 msr, bool longmode_only)
{
- if (!longmode_only)
- __vmx_disable_intercept_for_msr(vmx_msr_bitmap_legacy,
- msr, MSR_TYPE_R | MSR_TYPE_W);
- __vmx_disable_intercept_for_msr(vmx_msr_bitmap_longmode,
- msr, MSR_TYPE_R | MSR_TYPE_W);
+ __vmx_disable_intercept_for_msr(vmx_bitmap[vmx_msr_bitmap_longmode],
+ msr, MSR_TYPE_R | MSR_TYPE_W);
+
+ if (longmode_only)
+ return;
+ __vmx_disable_intercept_for_msr(vmx_bitmap[vmx_msr_bitmap_legacy],
+ msr, MSR_TYPE_R | MSR_TYPE_W);
}
-static void vmx_enable_intercept_msr_read_x2apic(u32 msr, bool apicv_active)
+static void vmx_disable_intercept_msr_x2apic(u32 msr, int type, bool apicv_active)
{
if (apicv_active) {
- __vmx_enable_intercept_for_msr(vmx_msr_bitmap_legacy_x2apic,
- msr, MSR_TYPE_R);
- __vmx_enable_intercept_for_msr(vmx_msr_bitmap_longmode_x2apic,
- msr, MSR_TYPE_R);
+ __vmx_disable_intercept_for_msr(
+ vmx_bitmap[vmx_msr_bitmap_legacy_x2apic_apicv],
+ msr, type);
+ __vmx_disable_intercept_for_msr(
+ vmx_bitmap[vmx_msr_bitmap_longmode_x2apic_apicv],
+ msr, type);
} else {
- __vmx_enable_intercept_for_msr(vmx_msr_bitmap_legacy_x2apic_apicv_inactive,
- msr, MSR_TYPE_R);
- __vmx_enable_intercept_for_msr(vmx_msr_bitmap_longmode_x2apic_apicv_inactive,
- msr, MSR_TYPE_R);
- }
-}
-
-static void vmx_disable_intercept_msr_read_x2apic(u32 msr, bool apicv_active)
-{
- if (apicv_active) {
- __vmx_disable_intercept_for_msr(vmx_msr_bitmap_legacy_x2apic,
- msr, MSR_TYPE_R);
- __vmx_disable_intercept_for_msr(vmx_msr_bitmap_longmode_x2apic,
- msr, MSR_TYPE_R);
- } else {
- __vmx_disable_intercept_for_msr(vmx_msr_bitmap_legacy_x2apic_apicv_inactive,
- msr, MSR_TYPE_R);
- __vmx_disable_intercept_for_msr(vmx_msr_bitmap_longmode_x2apic_apicv_inactive,
- msr, MSR_TYPE_R);
- }
-}
-
-static void vmx_disable_intercept_msr_write_x2apic(u32 msr, bool apicv_active)
-{
- if (apicv_active) {
- __vmx_disable_intercept_for_msr(vmx_msr_bitmap_legacy_x2apic,
- msr, MSR_TYPE_W);
- __vmx_disable_intercept_for_msr(vmx_msr_bitmap_longmode_x2apic,
- msr, MSR_TYPE_W);
- } else {
- __vmx_disable_intercept_for_msr(vmx_msr_bitmap_legacy_x2apic_apicv_inactive,
- msr, MSR_TYPE_W);
- __vmx_disable_intercept_for_msr(vmx_msr_bitmap_longmode_x2apic_apicv_inactive,
- msr, MSR_TYPE_W);
+ __vmx_disable_intercept_for_msr(
+ vmx_bitmap[vmx_msr_bitmap_legacy_x2apic],
+ msr, type);
+ __vmx_disable_intercept_for_msr(
+ vmx_bitmap[vmx_msr_bitmap_longmode_x2apic],
+ msr, type);
}
}
@@ -5034,12 +4982,12 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx)
int i;
/* I/O */
- vmcs_write64(IO_BITMAP_A, __pa(vmx_io_bitmap_a));
- vmcs_write64(IO_BITMAP_B, __pa(vmx_io_bitmap_b));
+ vmcs_write64(IO_BITMAP_A, __pa(vmx_bitmap[vmx_io_bitmap_a]));
+ vmcs_write64(IO_BITMAP_B, __pa(vmx_bitmap[vmx_io_bitmap_b]));
if (enable_shadow_vmcs) {
- vmcs_write64(VMREAD_BITMAP, __pa(vmx_vmread_bitmap));
- vmcs_write64(VMWRITE_BITMAP, __pa(vmx_vmwrite_bitmap));
+ vmcs_write64(VMREAD_BITMAP, __pa(vmx_bitmap[vmx_vmread_bitmap]));
+ vmcs_write64(VMWRITE_BITMAP, __pa(vmx_bitmap[vmx_vmwrite_bitmap]));
}
if (cpu_has_vmx_msr_bitmap())
vmcs_write64(MSR_BITMAP, __pa(vmx_msr_bitmap_legacy));
@@ -6377,68 +6325,30 @@ static __init int hardware_setup(void)
for (i = 0; i < ARRAY_SIZE(vmx_msr_index); ++i)
kvm_define_shared_msr(i, vmx_msr_index[i]);
- vmx_io_bitmap_a = (unsigned long *)__get_free_page(GFP_KERNEL);
- if (!vmx_io_bitmap_a)
- return r;
+ for (i = 0; i < ARRAY_SIZE(vmx_bitmap); i++) {
+ vmx_bitmap[i] = (unsigned long *)__get_free_page(GFP_KERNEL);
+ if (!vmx_bitmap[i])
+ goto out;
+ }
- vmx_io_bitmap_b = (unsigned long *)__get_free_page(GFP_KERNEL);
- if (!vmx_io_bitmap_b)
- goto out;
-
- vmx_msr_bitmap_legacy = (unsigned long *)__get_free_page(GFP_KERNEL);
- if (!vmx_msr_bitmap_legacy)
- goto out1;
-
- vmx_msr_bitmap_legacy_x2apic =
- (unsigned long *)__get_free_page(GFP_KERNEL);
- if (!vmx_msr_bitmap_legacy_x2apic)
- goto out2;
-
- vmx_msr_bitmap_legacy_x2apic_apicv_inactive =
- (unsigned long *)__get_free_page(GFP_KERNEL);
- if (!vmx_msr_bitmap_legacy_x2apic_apicv_inactive)
- goto out3;
-
- vmx_msr_bitmap_longmode = (unsigned long *)__get_free_page(GFP_KERNEL);
- if (!vmx_msr_bitmap_longmode)
- goto out4;
-
- vmx_msr_bitmap_longmode_x2apic =
- (unsigned long *)__get_free_page(GFP_KERNEL);
- if (!vmx_msr_bitmap_longmode_x2apic)
- goto out5;
-
- vmx_msr_bitmap_longmode_x2apic_apicv_inactive =
- (unsigned long *)__get_free_page(GFP_KERNEL);
- if (!vmx_msr_bitmap_longmode_x2apic_apicv_inactive)
- goto out6;
-
- vmx_vmread_bitmap = (unsigned long *)__get_free_page(GFP_KERNEL);
- if (!vmx_vmread_bitmap)
- goto out7;
-
- vmx_vmwrite_bitmap = (unsigned long *)__get_free_page(GFP_KERNEL);
- if (!vmx_vmwrite_bitmap)
- goto out8;
-
- memset(vmx_vmread_bitmap, 0xff, PAGE_SIZE);
- memset(vmx_vmwrite_bitmap, 0xff, PAGE_SIZE);
+ memset(vmx_bitmap[vmx_vmread_bitmap], 0xff, PAGE_SIZE);
+ memset(vmx_bitmap[vmx_vmwrite_bitmap], 0xff, PAGE_SIZE);
/*
* Allow direct access to the PC debug port (it is often used for I/O
* delays, but the vmexits simply slow things down).
*/
- memset(vmx_io_bitmap_a, 0xff, PAGE_SIZE);
- clear_bit(0x80, vmx_io_bitmap_a);
+ memset(vmx_bitmap[vmx_io_bitmap_a], 0xff, PAGE_SIZE);
+ clear_bit(0x80, vmx_bitmap[vmx_io_bitmap_a]);
- memset(vmx_io_bitmap_b, 0xff, PAGE_SIZE);
+ memset(vmx_bitmap[vmx_io_bitmap_b], 0xff, PAGE_SIZE);
- memset(vmx_msr_bitmap_legacy, 0xff, PAGE_SIZE);
- memset(vmx_msr_bitmap_longmode, 0xff, PAGE_SIZE);
+ memset(vmx_bitmap[vmx_msr_bitmap_legacy], 0xff, PAGE_SIZE);
+ memset(vmx_bitmap[vmx_msr_bitmap_longmode], 0xff, PAGE_SIZE);
if (setup_vmcs_config(&vmcs_config) < 0) {
r = -EIO;
- goto out9;
+ goto out;
}
if (boot_cpu_has(X86_FEATURE_NX))
@@ -6501,39 +6411,39 @@ static __init int hardware_setup(void)
vmx_disable_intercept_for_msr(MSR_IA32_SYSENTER_EIP, false);
vmx_disable_intercept_for_msr(MSR_IA32_BNDCFGS, true);
- memcpy(vmx_msr_bitmap_legacy_x2apic,
- vmx_msr_bitmap_legacy, PAGE_SIZE);
- memcpy(vmx_msr_bitmap_longmode_x2apic,
- vmx_msr_bitmap_longmode, PAGE_SIZE);
- memcpy(vmx_msr_bitmap_legacy_x2apic_apicv_inactive,
- vmx_msr_bitmap_legacy, PAGE_SIZE);
- memcpy(vmx_msr_bitmap_longmode_x2apic_apicv_inactive,
- vmx_msr_bitmap_longmode, PAGE_SIZE);
+ memcpy(vmx_bitmap[vmx_msr_bitmap_legacy_x2apic],
+ vmx_bitmap[vmx_msr_bitmap_legacy], PAGE_SIZE);
+ memcpy(vmx_bitmap[vmx_msr_bitmap_legacy_x2apic_apicv],
+ vmx_bitmap[vmx_msr_bitmap_legacy], PAGE_SIZE);
+ memcpy(vmx_bitmap[vmx_msr_bitmap_longmode_x2apic],
+ vmx_bitmap[vmx_msr_bitmap_longmode], PAGE_SIZE);
+ memcpy(vmx_bitmap[vmx_msr_bitmap_longmode_x2apic_apicv],
+ vmx_bitmap[vmx_msr_bitmap_longmode], PAGE_SIZE);
set_bit(0, vmx_vpid_bitmap); /* 0 is reserved for host */
/*
* enable_apicv && kvm_vcpu_apicv_active()
*/
- for (msr = 0x800; msr <= 0x8ff; msr++)
- vmx_disable_intercept_msr_read_x2apic(msr, true);
+ for (msr = 0x800; msr <= 0x8ff; msr++) {
+ if (msr == 0x839 /* TMCCT */)
+ continue;
+ vmx_disable_intercept_msr_x2apic(msr, MSR_TYPE_R, true);
+ }
- /* TMCCT */
- vmx_enable_intercept_msr_read_x2apic(0x839, true);
/* TPR */
- vmx_disable_intercept_msr_write_x2apic(0x808, true);
+ vmx_disable_intercept_msr_x2apic(0x808, MSR_TYPE_W, true);
/* EOI */
- vmx_disable_intercept_msr_write_x2apic(0x80b, true);
+ vmx_disable_intercept_msr_x2apic(0x80b, MSR_TYPE_W, true);
/* SELF-IPI */
- vmx_disable_intercept_msr_write_x2apic(0x83f, true);
+ vmx_disable_intercept_msr_x2apic(0x83f, MSR_TYPE_W, true);
/*
* (enable_apicv && !kvm_vcpu_apicv_active()) ||
* !enable_apicv
*/
/* TPR */
- vmx_disable_intercept_msr_read_x2apic(0x808, false);
- vmx_disable_intercept_msr_write_x2apic(0x808, false);
+ vmx_disable_intercept_msr_x2apic(0x808, MSR_TYPE_R | MSR_TYPE_W, false);
if (enable_ept) {
kvm_mmu_set_mask_ptes(VMX_EPT_READABLE_MASK,
@@ -6580,42 +6490,19 @@ static __init int hardware_setup(void)
return alloc_kvm_area();
-out9:
- free_page((unsigned long)vmx_vmwrite_bitmap);
-out8:
- free_page((unsigned long)vmx_vmread_bitmap);
-out7:
- free_page((unsigned long)vmx_msr_bitmap_longmode_x2apic_apicv_inactive);
-out6:
- free_page((unsigned long)vmx_msr_bitmap_longmode_x2apic);
-out5:
- free_page((unsigned long)vmx_msr_bitmap_longmode);
-out4:
- free_page((unsigned long)vmx_msr_bitmap_legacy_x2apic_apicv_inactive);
-out3:
- free_page((unsigned long)vmx_msr_bitmap_legacy_x2apic);
-out2:
- free_page((unsigned long)vmx_msr_bitmap_legacy);
-out1:
- free_page((unsigned long)vmx_io_bitmap_b);
out:
- free_page((unsigned long)vmx_io_bitmap_a);
+ for (i = 0; i < ARRAY_SIZE(vmx_bitmap); i++)
+ free_page((unsigned long)vmx_bitmap[i]);
return r;
}
static __exit void hardware_unsetup(void)
{
- free_page((unsigned long)vmx_msr_bitmap_legacy_x2apic);
- free_page((unsigned long)vmx_msr_bitmap_legacy_x2apic_apicv_inactive);
- free_page((unsigned long)vmx_msr_bitmap_longmode_x2apic);
- free_page((unsigned long)vmx_msr_bitmap_longmode_x2apic_apicv_inactive);
- free_page((unsigned long)vmx_msr_bitmap_legacy);
- free_page((unsigned long)vmx_msr_bitmap_longmode);
- free_page((unsigned long)vmx_io_bitmap_b);
- free_page((unsigned long)vmx_io_bitmap_a);
- free_page((unsigned long)vmx_vmwrite_bitmap);
- free_page((unsigned long)vmx_vmread_bitmap);
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(vmx_bitmap); i++)
+ free_page((unsigned long)vmx_bitmap[i]);
free_kvm_area();
}
--
2.10.0
Powered by blists - more mailing lists