summaryrefslogtreecommitdiffstats
path: root/arch/x86/kvm
diff options
context:
space:
mode:
authorArnaldo Carvalho de Melo <acme@redhat.com>2024-04-22 13:35:18 -0300
committerArnaldo Carvalho de Melo <acme@redhat.com>2024-04-22 13:35:18 -0300
commit173b0b5b0e865348684c02bd9cb1d22b5d46e458 (patch)
treea42a2cc8b724b94a9478aedfcbfabcec0197b246 /arch/x86/kvm
parent61ba075d9911d2a6db181fbb5602762a24d5d0a9 (diff)
parented30a4a51bb196781c8058073ea720133a65596f (diff)
downloadlinux-stable-173b0b5b0e865348684c02bd9cb1d22b5d46e458.tar.gz
linux-stable-173b0b5b0e865348684c02bd9cb1d22b5d46e458.tar.bz2
linux-stable-173b0b5b0e865348684c02bd9cb1d22b5d46e458.zip
Merge remote-tracking branch 'torvalds/master' into perf-tools-next
To pick up fixes sent via perf-tools, by Namhyung Kim. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Diffstat (limited to 'arch/x86/kvm')
-rw-r--r--arch/x86/kvm/Kconfig3
-rw-r--r--arch/x86/kvm/Makefile4
-rw-r--r--arch/x86/kvm/cpuid.c43
-rw-r--r--arch/x86/kvm/cpuid.h10
-rw-r--r--arch/x86/kvm/lapic.c3
-rw-r--r--arch/x86/kvm/mmu/mmu.c11
-rw-r--r--arch/x86/kvm/mmu/tdp_mmu.c51
-rw-r--r--arch/x86/kvm/pmu.c16
-rw-r--r--arch/x86/kvm/reverse_cpuid.h5
-rw-r--r--arch/x86/kvm/svm/sev.c62
-rw-r--r--arch/x86/kvm/svm/svm.c17
-rw-r--r--arch/x86/kvm/svm/svm.h3
-rw-r--r--arch/x86/kvm/svm/vmenter.S97
-rw-r--r--arch/x86/kvm/trace.h10
-rw-r--r--arch/x86/kvm/vmx/pmu_intel.c2
-rw-r--r--arch/x86/kvm/vmx/vmenter.S2
-rw-r--r--arch/x86/kvm/vmx/vmx.c41
-rw-r--r--arch/x86/kvm/vmx/vmx.h6
-rw-r--r--arch/x86/kvm/x86.c4
19 files changed, 229 insertions, 161 deletions
diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig
index 8c3032a96caf..0ebdd088f28b 100644
--- a/arch/x86/kvm/Kconfig
+++ b/arch/x86/kvm/Kconfig
@@ -118,10 +118,11 @@ config KVM_AMD
will be called kvm-amd.
config KVM_AMD_SEV
- def_bool y
bool "AMD Secure Encrypted Virtualization (SEV) support"
+ default y
depends on KVM_AMD && X86_64
depends on CRYPTO_DEV_SP_PSP && !(KVM_AMD=y && CRYPTO_DEV_CCP_DD=m)
+ select ARCH_HAS_CC_PLATFORM
help
Provides support for launching Encrypted VMs (SEV) and Encrypted VMs
with Encrypted State (SEV-ES) on AMD processors.
diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile
index 475b5fa917a6..addc44fc7187 100644
--- a/arch/x86/kvm/Makefile
+++ b/arch/x86/kvm/Makefile
@@ -3,10 +3,6 @@
ccflags-y += -I $(srctree)/arch/x86/kvm
ccflags-$(CONFIG_KVM_WERROR) += -Werror
-ifeq ($(CONFIG_FRAME_POINTER),y)
-OBJECT_FILES_NON_STANDARD_vmenter.o := y
-endif
-
include $(srctree)/virt/kvm/Makefile.kvm
kvm-y += x86.o emulate.o i8259.o irq.o lapic.o \
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index adba49afb5fe..77352a4abd87 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -189,15 +189,15 @@ static int kvm_cpuid_check_equal(struct kvm_vcpu *vcpu, struct kvm_cpuid_entry2
return 0;
}
-static struct kvm_hypervisor_cpuid kvm_get_hypervisor_cpuid(struct kvm_vcpu *vcpu,
- const char *sig)
+static struct kvm_hypervisor_cpuid __kvm_get_hypervisor_cpuid(struct kvm_cpuid_entry2 *entries,
+ int nent, const char *sig)
{
struct kvm_hypervisor_cpuid cpuid = {};
struct kvm_cpuid_entry2 *entry;
u32 base;
for_each_possible_hypervisor_cpuid_base(base) {
- entry = kvm_find_cpuid_entry(vcpu, base);
+ entry = cpuid_entry2_find(entries, nent, base, KVM_CPUID_INDEX_NOT_SIGNIFICANT);
if (entry) {
u32 signature[3];
@@ -217,22 +217,29 @@ static struct kvm_hypervisor_cpuid kvm_get_hypervisor_cpuid(struct kvm_vcpu *vcp
return cpuid;
}
-static struct kvm_cpuid_entry2 *__kvm_find_kvm_cpuid_features(struct kvm_vcpu *vcpu,
- struct kvm_cpuid_entry2 *entries, int nent)
+static struct kvm_hypervisor_cpuid kvm_get_hypervisor_cpuid(struct kvm_vcpu *vcpu,
+ const char *sig)
{
- u32 base = vcpu->arch.kvm_cpuid.base;
-
- if (!base)
- return NULL;
+ return __kvm_get_hypervisor_cpuid(vcpu->arch.cpuid_entries,
+ vcpu->arch.cpuid_nent, sig);
+}
- return cpuid_entry2_find(entries, nent, base | KVM_CPUID_FEATURES,
+static struct kvm_cpuid_entry2 *__kvm_find_kvm_cpuid_features(struct kvm_cpuid_entry2 *entries,
+ int nent, u32 kvm_cpuid_base)
+{
+ return cpuid_entry2_find(entries, nent, kvm_cpuid_base | KVM_CPUID_FEATURES,
KVM_CPUID_INDEX_NOT_SIGNIFICANT);
}
static struct kvm_cpuid_entry2 *kvm_find_kvm_cpuid_features(struct kvm_vcpu *vcpu)
{
- return __kvm_find_kvm_cpuid_features(vcpu, vcpu->arch.cpuid_entries,
- vcpu->arch.cpuid_nent);
+ u32 base = vcpu->arch.kvm_cpuid.base;
+
+ if (!base)
+ return NULL;
+
+ return __kvm_find_kvm_cpuid_features(vcpu->arch.cpuid_entries,
+ vcpu->arch.cpuid_nent, base);
}
void kvm_update_pv_runtime(struct kvm_vcpu *vcpu)
@@ -266,6 +273,7 @@ static void __kvm_update_cpuid_runtime(struct kvm_vcpu *vcpu, struct kvm_cpuid_e
int nent)
{
struct kvm_cpuid_entry2 *best;
+ struct kvm_hypervisor_cpuid kvm_cpuid;
best = cpuid_entry2_find(entries, nent, 1, KVM_CPUID_INDEX_NOT_SIGNIFICANT);
if (best) {
@@ -292,10 +300,12 @@ static void __kvm_update_cpuid_runtime(struct kvm_vcpu *vcpu, struct kvm_cpuid_e
cpuid_entry_has(best, X86_FEATURE_XSAVEC)))
best->ebx = xstate_required_size(vcpu->arch.xcr0, true);
- best = __kvm_find_kvm_cpuid_features(vcpu, entries, nent);
- if (kvm_hlt_in_guest(vcpu->kvm) && best &&
- (best->eax & (1 << KVM_FEATURE_PV_UNHALT)))
- best->eax &= ~(1 << KVM_FEATURE_PV_UNHALT);
+ kvm_cpuid = __kvm_get_hypervisor_cpuid(entries, nent, KVM_SIGNATURE);
+ if (kvm_cpuid.base) {
+ best = __kvm_find_kvm_cpuid_features(entries, nent, kvm_cpuid.base);
+ if (kvm_hlt_in_guest(vcpu->kvm) && best)
+ best->eax &= ~(1 << KVM_FEATURE_PV_UNHALT);
+ }
if (!kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT)) {
best = cpuid_entry2_find(entries, nent, 0x1, KVM_CPUID_INDEX_NOT_SIGNIFICANT);
@@ -366,6 +376,7 @@ static void kvm_vcpu_after_set_cpuid(struct kvm_vcpu *vcpu)
kvm_update_pv_runtime(vcpu);
+ vcpu->arch.is_amd_compatible = guest_cpuid_is_amd_or_hygon(vcpu);
vcpu->arch.maxphyaddr = cpuid_query_maxphyaddr(vcpu);
vcpu->arch.reserved_gpa_bits = kvm_vcpu_reserved_gpa_bits_raw(vcpu);
diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h
index 856e3037e74f..23dbb9eb277c 100644
--- a/arch/x86/kvm/cpuid.h
+++ b/arch/x86/kvm/cpuid.h
@@ -120,6 +120,16 @@ static inline bool guest_cpuid_is_intel(struct kvm_vcpu *vcpu)
return best && is_guest_vendor_intel(best->ebx, best->ecx, best->edx);
}
+static inline bool guest_cpuid_is_amd_compatible(struct kvm_vcpu *vcpu)
+{
+ return vcpu->arch.is_amd_compatible;
+}
+
+static inline bool guest_cpuid_is_intel_compatible(struct kvm_vcpu *vcpu)
+{
+ return !guest_cpuid_is_amd_compatible(vcpu);
+}
+
static inline int guest_cpuid_family(struct kvm_vcpu *vcpu)
{
struct kvm_cpuid_entry2 *best;
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index cf37586f0466..ebf41023be38 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -2776,7 +2776,8 @@ int kvm_apic_local_deliver(struct kvm_lapic *apic, int lvt_type)
trig_mode = reg & APIC_LVT_LEVEL_TRIGGER;
r = __apic_accept_irq(apic, mode, vector, 1, trig_mode, NULL);
- if (r && lvt_type == APIC_LVTPC)
+ if (r && lvt_type == APIC_LVTPC &&
+ guest_cpuid_is_intel_compatible(apic->vcpu))
kvm_lapic_set_reg(apic, APIC_LVTPC, reg | APIC_LVT_MASKED);
return r;
}
diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index 992e651540e8..db007a4dffa2 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -4935,7 +4935,7 @@ static void reset_guest_rsvds_bits_mask(struct kvm_vcpu *vcpu,
context->cpu_role.base.level, is_efer_nx(context),
guest_can_use(vcpu, X86_FEATURE_GBPAGES),
is_cr4_pse(context),
- guest_cpuid_is_amd_or_hygon(vcpu));
+ guest_cpuid_is_amd_compatible(vcpu));
}
static void __reset_rsvds_bits_mask_ept(struct rsvd_bits_validate *rsvd_check,
@@ -5576,9 +5576,9 @@ void kvm_mmu_after_set_cpuid(struct kvm_vcpu *vcpu)
* that problem is swept under the rug; KVM's CPUID API is horrific and
* it's all but impossible to solve it without introducing a new API.
*/
- vcpu->arch.root_mmu.root_role.word = 0;
- vcpu->arch.guest_mmu.root_role.word = 0;
- vcpu->arch.nested_mmu.root_role.word = 0;
+ vcpu->arch.root_mmu.root_role.invalid = 1;
+ vcpu->arch.guest_mmu.root_role.invalid = 1;
+ vcpu->arch.nested_mmu.root_role.invalid = 1;
vcpu->arch.root_mmu.cpu_role.ext.valid = 0;
vcpu->arch.guest_mmu.cpu_role.ext.valid = 0;
vcpu->arch.nested_mmu.cpu_role.ext.valid = 0;
@@ -7399,7 +7399,8 @@ bool kvm_arch_post_set_memory_attributes(struct kvm *kvm,
* by the memslot, KVM can't use a hugepage due to the
* misaligned address regardless of memory attributes.
*/
- if (gfn >= slot->base_gfn) {
+ if (gfn >= slot->base_gfn &&
+ gfn + nr_pages <= slot->base_gfn + slot->npages) {
if (hugepage_has_attrs(kvm, slot, gfn, level, attrs))
hugepage_clear_mixed(slot, gfn, level);
else
diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
index d078157e62aa..04c1f0957fea 100644
--- a/arch/x86/kvm/mmu/tdp_mmu.c
+++ b/arch/x86/kvm/mmu/tdp_mmu.c
@@ -1548,17 +1548,21 @@ void kvm_tdp_mmu_try_split_huge_pages(struct kvm *kvm,
}
}
-/*
- * Clear the dirty status of all the SPTEs mapping GFNs in the memslot. If
- * AD bits are enabled, this will involve clearing the dirty bit on each SPTE.
- * If AD bits are not enabled, this will require clearing the writable bit on
- * each SPTE. Returns true if an SPTE has been changed and the TLBs need to
- * be flushed.
- */
+static bool tdp_mmu_need_write_protect(struct kvm_mmu_page *sp)
+{
+ /*
+ * All TDP MMU shadow pages share the same role as their root, aside
+ * from level, so it is valid to key off any shadow page to determine if
+ * write protection is needed for an entire tree.
+ */
+ return kvm_mmu_page_ad_need_write_protect(sp) || !kvm_ad_enabled();
+}
+
static bool clear_dirty_gfn_range(struct kvm *kvm, struct kvm_mmu_page *root,
gfn_t start, gfn_t end)
{
- u64 dbit = kvm_ad_enabled() ? shadow_dirty_mask : PT_WRITABLE_MASK;
+ const u64 dbit = tdp_mmu_need_write_protect(root) ? PT_WRITABLE_MASK :
+ shadow_dirty_mask;
struct tdp_iter iter;
bool spte_set = false;
@@ -1573,7 +1577,7 @@ retry:
if (tdp_mmu_iter_cond_resched(kvm, &iter, false, true))
continue;
- KVM_MMU_WARN_ON(kvm_ad_enabled() &&
+ KVM_MMU_WARN_ON(dbit == shadow_dirty_mask &&
spte_ad_need_write_protect(iter.old_spte));
if (!(iter.old_spte & dbit))
@@ -1590,11 +1594,9 @@ retry:
}
/*
- * Clear the dirty status of all the SPTEs mapping GFNs in the memslot. If
- * AD bits are enabled, this will involve clearing the dirty bit on each SPTE.
- * If AD bits are not enabled, this will require clearing the writable bit on
- * each SPTE. Returns true if an SPTE has been changed and the TLBs need to
- * be flushed.
+ * Clear the dirty status (D-bit or W-bit) of all the SPTEs mapping GFNs in the
+ * memslot. Returns true if an SPTE has been changed and the TLBs need to be
+ * flushed.
*/
bool kvm_tdp_mmu_clear_dirty_slot(struct kvm *kvm,
const struct kvm_memory_slot *slot)
@@ -1610,18 +1612,11 @@ bool kvm_tdp_mmu_clear_dirty_slot(struct kvm *kvm,
return spte_set;
}
-/*
- * Clears the dirty status of all the 4k SPTEs mapping GFNs for which a bit is
- * set in mask, starting at gfn. The given memslot is expected to contain all
- * the GFNs represented by set bits in the mask. If AD bits are enabled,
- * clearing the dirty status will involve clearing the dirty bit on each SPTE
- * or, if AD bits are not enabled, clearing the writable bit on each SPTE.
- */
static void clear_dirty_pt_masked(struct kvm *kvm, struct kvm_mmu_page *root,
gfn_t gfn, unsigned long mask, bool wrprot)
{
- u64 dbit = (wrprot || !kvm_ad_enabled()) ? PT_WRITABLE_MASK :
- shadow_dirty_mask;
+ const u64 dbit = (wrprot || tdp_mmu_need_write_protect(root)) ? PT_WRITABLE_MASK :
+ shadow_dirty_mask;
struct tdp_iter iter;
lockdep_assert_held_write(&kvm->mmu_lock);
@@ -1633,7 +1628,7 @@ static void clear_dirty_pt_masked(struct kvm *kvm, struct kvm_mmu_page *root,
if (!mask)
break;
- KVM_MMU_WARN_ON(kvm_ad_enabled() &&
+ KVM_MMU_WARN_ON(dbit == shadow_dirty_mask &&
spte_ad_need_write_protect(iter.old_spte));
if (iter.level > PG_LEVEL_4K ||
@@ -1659,11 +1654,9 @@ static void clear_dirty_pt_masked(struct kvm *kvm, struct kvm_mmu_page *root,
}
/*
- * Clears the dirty status of all the 4k SPTEs mapping GFNs for which a bit is
- * set in mask, starting at gfn. The given memslot is expected to contain all
- * the GFNs represented by set bits in the mask. If AD bits are enabled,
- * clearing the dirty status will involve clearing the dirty bit on each SPTE
- * or, if AD bits are not enabled, clearing the writable bit on each SPTE.
+ * Clear the dirty status (D-bit or W-bit) of all the 4k SPTEs mapping GFNs for
+ * which a bit is set in mask, starting at gfn. The given memslot is expected to
+ * contain all the GFNs represented by set bits in the mask.
*/
void kvm_tdp_mmu_clear_dirty_pt_masked(struct kvm *kvm,
struct kvm_memory_slot *slot,
diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c
index c397b28e3d1b..a593b03c9aed 100644
--- a/arch/x86/kvm/pmu.c
+++ b/arch/x86/kvm/pmu.c
@@ -775,8 +775,20 @@ void kvm_pmu_refresh(struct kvm_vcpu *vcpu)
pmu->pebs_data_cfg_mask = ~0ull;
bitmap_zero(pmu->all_valid_pmc_idx, X86_PMC_IDX_MAX);
- if (vcpu->kvm->arch.enable_pmu)
- static_call(kvm_x86_pmu_refresh)(vcpu);
+ if (!vcpu->kvm->arch.enable_pmu)
+ return;
+
+ static_call(kvm_x86_pmu_refresh)(vcpu);
+
+ /*
+ * At RESET, both Intel and AMD CPUs set all enable bits for general
+ * purpose counters in IA32_PERF_GLOBAL_CTRL (so that software that
+ * was written for v1 PMUs don't unknowingly leave GP counters disabled
+ * in the global controls). Emulate that behavior when refreshing the
+ * PMU so that userspace doesn't need to manually set PERF_GLOBAL_CTRL.
+ */
+ if (kvm_pmu_has_perf_global_ctrl(pmu) && pmu->nr_arch_gp_counters)
+ pmu->global_ctrl = GENMASK_ULL(pmu->nr_arch_gp_counters - 1, 0);
}
void kvm_pmu_init(struct kvm_vcpu *vcpu)
diff --git a/arch/x86/kvm/reverse_cpuid.h b/arch/x86/kvm/reverse_cpuid.h
index aadefcaa9561..2f4e155080ba 100644
--- a/arch/x86/kvm/reverse_cpuid.h
+++ b/arch/x86/kvm/reverse_cpuid.h
@@ -52,7 +52,7 @@ enum kvm_only_cpuid_leafs {
#define X86_FEATURE_IPRED_CTRL KVM_X86_FEATURE(CPUID_7_2_EDX, 1)
#define KVM_X86_FEATURE_RRSBA_CTRL KVM_X86_FEATURE(CPUID_7_2_EDX, 2)
#define X86_FEATURE_DDPD_U KVM_X86_FEATURE(CPUID_7_2_EDX, 3)
-#define X86_FEATURE_BHI_CTRL KVM_X86_FEATURE(CPUID_7_2_EDX, 4)
+#define KVM_X86_FEATURE_BHI_CTRL KVM_X86_FEATURE(CPUID_7_2_EDX, 4)
#define X86_FEATURE_MCDT_NO KVM_X86_FEATURE(CPUID_7_2_EDX, 5)
/* CPUID level 0x80000007 (EDX). */
@@ -102,10 +102,12 @@ static const struct cpuid_reg reverse_cpuid[] = {
*/
static __always_inline void reverse_cpuid_check(unsigned int x86_leaf)
{
+ BUILD_BUG_ON(NR_CPUID_WORDS != NCAPINTS);
BUILD_BUG_ON(x86_leaf == CPUID_LNX_1);
BUILD_BUG_ON(x86_leaf == CPUID_LNX_2);
BUILD_BUG_ON(x86_leaf == CPUID_LNX_3);
BUILD_BUG_ON(x86_leaf == CPUID_LNX_4);
+ BUILD_BUG_ON(x86_leaf == CPUID_LNX_5);
BUILD_BUG_ON(x86_leaf >= ARRAY_SIZE(reverse_cpuid));
BUILD_BUG_ON(reverse_cpuid[x86_leaf].function == 0);
}
@@ -126,6 +128,7 @@ static __always_inline u32 __feature_translate(int x86_feature)
KVM_X86_TRANSLATE_FEATURE(CONSTANT_TSC);
KVM_X86_TRANSLATE_FEATURE(PERFMON_V2);
KVM_X86_TRANSLATE_FEATURE(RRSBA_CTRL);
+ KVM_X86_TRANSLATE_FEATURE(BHI_CTRL);
default:
return x86_feature;
}
diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c
index ae0ac12382b9..759581bb2128 100644
--- a/arch/x86/kvm/svm/sev.c
+++ b/arch/x86/kvm/svm/sev.c
@@ -84,9 +84,10 @@ struct enc_region {
};
/* Called with the sev_bitmap_lock held, or on shutdown */
-static int sev_flush_asids(int min_asid, int max_asid)
+static int sev_flush_asids(unsigned int min_asid, unsigned int max_asid)
{
- int ret, asid, error = 0;
+ int ret, error = 0;
+ unsigned int asid;
/* Check if there are any ASIDs to reclaim before performing a flush */
asid = find_next_bit(sev_reclaim_asid_bitmap, nr_asids, min_asid);
@@ -116,7 +117,7 @@ static inline bool is_mirroring_enc_context(struct kvm *kvm)
}
/* Must be called with the sev_bitmap_lock held */
-static bool __sev_recycle_asids(int min_asid, int max_asid)
+static bool __sev_recycle_asids(unsigned int min_asid, unsigned int max_asid)
{
if (sev_flush_asids(min_asid, max_asid))
return false;
@@ -143,8 +144,20 @@ static void sev_misc_cg_uncharge(struct kvm_sev_info *sev)
static int sev_asid_new(struct kvm_sev_info *sev)
{
- int asid, min_asid, max_asid, ret;
+ /*
+ * SEV-enabled guests must use asid from min_sev_asid to max_sev_asid.
+ * SEV-ES-enabled guest can use from 1 to min_sev_asid - 1.
+ * Note: min ASID can end up larger than the max if basic SEV support is
+ * effectively disabled by disallowing use of ASIDs for SEV guests.
+ */
+ unsigned int min_asid = sev->es_active ? 1 : min_sev_asid;
+ unsigned int max_asid = sev->es_active ? min_sev_asid - 1 : max_sev_asid;
+ unsigned int asid;
bool retry = true;
+ int ret;
+
+ if (min_asid > max_asid)
+ return -ENOTTY;
WARN_ON(sev->misc_cg);
sev->misc_cg = get_current_misc_cg();
@@ -157,12 +170,6 @@ static int sev_asid_new(struct kvm_sev_info *sev)
mutex_lock(&sev_bitmap_lock);
- /*
- * SEV-enabled guests must use asid from min_sev_asid to max_sev_asid.
- * SEV-ES-enabled guest can use from 1 to min_sev_asid - 1.
- */
- min_asid = sev->es_active ? 1 : min_sev_asid;
- max_asid = sev->es_active ? min_sev_asid - 1 : max_sev_asid;
again:
asid = find_next_zero_bit(sev_asid_bitmap, max_asid + 1, min_asid);
if (asid > max_asid) {
@@ -179,7 +186,8 @@ again:
mutex_unlock(&sev_bitmap_lock);
- return asid;
+ sev->asid = asid;
+ return 0;
e_uncharge:
sev_misc_cg_uncharge(sev);
put_misc_cg(sev->misc_cg);
@@ -187,7 +195,7 @@ e_uncharge:
return ret;
}
-static int sev_get_asid(struct kvm *kvm)
+static unsigned int sev_get_asid(struct kvm *kvm)
{
struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info;
@@ -247,21 +255,19 @@ static int sev_guest_init(struct kvm *kvm, struct kvm_sev_cmd *argp)
{
struct kvm_sev_info *sev = &to_kvm_svm(kvm)->sev_info;
struct sev_platform_init_args init_args = {0};
- int asid, ret;
+ int ret;
if (kvm->created_vcpus)
return -EINVAL;
- ret = -EBUSY;
if (unlikely(sev->active))
- return ret;
+ return -EINVAL;
sev->active = true;
sev->es_active = argp->id == KVM_SEV_ES_INIT;
- asid = sev_asid_new(sev);
- if (asid < 0)
+ ret = sev_asid_new(sev);
+ if (ret)
goto e_no_asid;
- sev->asid = asid;
init_args.probe = false;
ret = sev_platform_init(&init_args);
@@ -287,8 +293,8 @@ e_no_asid:
static int sev_bind_asid(struct kvm *kvm, unsigned int handle, int *error)
{
+ unsigned int asid = sev_get_asid(kvm);
struct sev_data_activate activate;
- int asid = sev_get_asid(kvm);
int ret;
/* activate ASID on the given handle */
@@ -428,7 +434,7 @@ static struct page **sev_pin_memory(struct kvm *kvm, unsigned long uaddr,
/* Avoid using vmalloc for smaller buffers. */
size = npages * sizeof(struct page *);
if (size > PAGE_SIZE)
- pages = __vmalloc(size, GFP_KERNEL_ACCOUNT | __GFP_ZERO);
+ pages = __vmalloc(size, GFP_KERNEL_ACCOUNT);
else
pages = kmalloc(size, GFP_KERNEL_ACCOUNT);
@@ -2240,8 +2246,10 @@ void __init sev_hardware_setup(void)
goto out;
}
- sev_asid_count = max_sev_asid - min_sev_asid + 1;
- WARN_ON_ONCE(misc_cg_set_capacity(MISC_CG_RES_SEV, sev_asid_count));
+ if (min_sev_asid <= max_sev_asid) {
+ sev_asid_count = max_sev_asid - min_sev_asid + 1;
+ WARN_ON_ONCE(misc_cg_set_capacity(MISC_CG_RES_SEV, sev_asid_count));
+ }
sev_supported = true;
/* SEV-ES support requested? */
@@ -2272,7 +2280,9 @@ void __init sev_hardware_setup(void)
out:
if (boot_cpu_has(X86_FEATURE_SEV))
pr_info("SEV %s (ASIDs %u - %u)\n",
- sev_supported ? "enabled" : "disabled",
+ sev_supported ? min_sev_asid <= max_sev_asid ? "enabled" :
+ "unusable" :
+ "disabled",
min_sev_asid, max_sev_asid);
if (boot_cpu_has(X86_FEATURE_SEV_ES))
pr_info("SEV-ES %s (ASIDs %u - %u)\n",
@@ -2320,7 +2330,7 @@ int sev_cpu_init(struct svm_cpu_data *sd)
*/
static void sev_flush_encrypted_page(struct kvm_vcpu *vcpu, void *va)
{
- int asid = to_kvm_svm(vcpu->kvm)->sev_info.asid;
+ unsigned int asid = sev_get_asid(vcpu->kvm);
/*
* Note! The address must be a kernel address, as regular page walk
@@ -2638,7 +2648,7 @@ void sev_es_unmap_ghcb(struct vcpu_svm *svm)
void pre_sev_run(struct vcpu_svm *svm, int cpu)
{
struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, cpu);
- int asid = sev_get_asid(svm->vcpu.kvm);
+ unsigned int asid = sev_get_asid(svm->vcpu.kvm);
/* Assign the asid allocated with this SEV guest */
svm->asid = asid;
@@ -3174,7 +3184,7 @@ struct page *snp_safe_alloc_page(struct kvm_vcpu *vcpu)
unsigned long pfn;
struct page *p;
- if (!cpu_feature_enabled(X86_FEATURE_SEV_SNP))
+ if (!cc_platform_has(CC_ATTR_HOST_SEV_SNP))
return alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO);
/*
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index d1a9f9951635..9aaf83c8d57d 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -1503,6 +1503,11 @@ static void svm_vcpu_free(struct kvm_vcpu *vcpu)
__free_pages(virt_to_page(svm->msrpm), get_order(MSRPM_SIZE));
}
+static struct sev_es_save_area *sev_es_host_save_area(struct svm_cpu_data *sd)
+{
+ return page_address(sd->save_area) + 0x400;
+}
+
static void svm_prepare_switch_to_guest(struct kvm_vcpu *vcpu)
{
struct vcpu_svm *svm = to_svm(vcpu);
@@ -1519,12 +1524,8 @@ static void svm_prepare_switch_to_guest(struct kvm_vcpu *vcpu)
* or subsequent vmload of host save area.
*/
vmsave(sd->save_area_pa);
- if (sev_es_guest(vcpu->kvm)) {
- struct sev_es_save_area *hostsa;
- hostsa = (struct sev_es_save_area *)(page_address(sd->save_area) + 0x400);
-
- sev_es_prepare_switch_to_guest(hostsa);
- }
+ if (sev_es_guest(vcpu->kvm))
+ sev_es_prepare_switch_to_guest(sev_es_host_save_area(sd));
if (tsc_scaling)
__svm_write_tsc_multiplier(vcpu->arch.tsc_scaling_ratio);
@@ -4101,6 +4102,7 @@ static fastpath_t svm_exit_handlers_fastpath(struct kvm_vcpu *vcpu)
static noinstr void svm_vcpu_enter_exit(struct kvm_vcpu *vcpu, bool spec_ctrl_intercepted)
{
+ struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, vcpu->cpu);
struct vcpu_svm *svm = to_svm(vcpu);
guest_state_enter_irqoff();
@@ -4108,7 +4110,8 @@ static noinstr void svm_vcpu_enter_exit(struct kvm_vcpu *vcpu, bool spec_ctrl_in
amd_clear_divider();
if (sev_es_guest(vcpu->kvm))
- __svm_sev_es_vcpu_run(svm, spec_ctrl_intercepted);
+ __svm_sev_es_vcpu_run(svm, spec_ctrl_intercepted,
+ sev_es_host_save_area(sd));
else
__svm_vcpu_run(svm, spec_ctrl_intercepted);
diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h
index 7f1fbd874c45..33878efdebc8 100644
--- a/arch/x86/kvm/svm/svm.h
+++ b/arch/x86/kvm/svm/svm.h
@@ -698,7 +698,8 @@ struct page *snp_safe_alloc_page(struct kvm_vcpu *vcpu);
/* vmenter.S */
-void __svm_sev_es_vcpu_run(struct vcpu_svm *svm, bool spec_ctrl_intercepted);
+void __svm_sev_es_vcpu_run(struct vcpu_svm *svm, bool spec_ctrl_intercepted,
+ struct sev_es_save_area *hostsa);
void __svm_vcpu_run(struct vcpu_svm *svm, bool spec_ctrl_intercepted);
#define DEFINE_KVM_GHCB_ACCESSORS(field) \
diff --git a/arch/x86/kvm/svm/vmenter.S b/arch/x86/kvm/svm/vmenter.S
index 187018c424bf..a0c8eb37d3e1 100644
--- a/arch/x86/kvm/svm/vmenter.S
+++ b/arch/x86/kvm/svm/vmenter.S
@@ -3,6 +3,7 @@
#include <asm/asm.h>
#include <asm/asm-offsets.h>
#include <asm/bitsperlong.h>
+#include <asm/frame.h>
#include <asm/kvm_vcpu_regs.h>
#include <asm/nospec-branch.h>
#include "kvm-asm-offsets.h"
@@ -67,7 +68,7 @@
"", X86_FEATURE_V_SPEC_CTRL
901:
.endm
-.macro RESTORE_HOST_SPEC_CTRL_BODY
+.macro RESTORE_HOST_SPEC_CTRL_BODY spec_ctrl_intercepted:req
900:
/* Same for after vmexit. */
mov $MSR_IA32_SPEC_CTRL, %ecx
@@ -76,7 +77,7 @@
* Load the value that the guest had written into MSR_IA32_SPEC_CTRL,
* if it was not intercepted during guest execution.
*/
- cmpb $0, (%_ASM_SP)
+ cmpb $0, \spec_ctrl_intercepted
jnz 998f
rdmsr
movl %eax, SVM_spec_ctrl(%_ASM_DI)
@@ -99,6 +100,7 @@
*/
SYM_FUNC_START(__svm_vcpu_run)
push %_ASM_BP
+ mov %_ASM_SP, %_ASM_BP
#ifdef CONFIG_X86_64
push %r15
push %r14
@@ -268,7 +270,7 @@ SYM_FUNC_START(__svm_vcpu_run)
RET
RESTORE_GUEST_SPEC_CTRL_BODY
- RESTORE_HOST_SPEC_CTRL_BODY
+ RESTORE_HOST_SPEC_CTRL_BODY (%_ASM_SP)
10: cmpb $0, _ASM_RIP(kvm_rebooting)
jne 2b
@@ -290,66 +292,68 @@ SYM_FUNC_START(__svm_vcpu_run)
SYM_FUNC_END(__svm_vcpu_run)
+#ifdef CONFIG_KVM_AMD_SEV
+
+
+#ifdef CONFIG_X86_64
+#define SEV_ES_GPRS_BASE 0x300
+#define SEV_ES_RBX (SEV_ES_GPRS_BASE + __VCPU_REGS_RBX * WORD_SIZE)
+#define SEV_ES_RBP (SEV_ES_GPRS_BASE + __VCPU_REGS_RBP * WORD_SIZE)
+#define SEV_ES_RSI (SEV_ES_GPRS_BASE + __VCPU_REGS_RSI * WORD_SIZE)
+#define SEV_ES_RDI (SEV_ES_GPRS_BASE + __VCPU_REGS_RDI * WORD_SIZE)
+#define SEV_ES_R12 (SEV_ES_GPRS_BASE + __VCPU_REGS_R12 * WORD_SIZE)
+#define SEV_ES_R13 (SEV_ES_GPRS_BASE + __VCPU_REGS_R13 * WORD_SIZE)
+#define SEV_ES_R14 (SEV_ES_GPRS_BASE + __VCPU_REGS_R14 * WORD_SIZE)
+#define SEV_ES_R15 (SEV_ES_GPRS_BASE + __VCPU_REGS_R15 * WORD_SIZE)
+#endif
+
/**
* __svm_sev_es_vcpu_run - Run a SEV-ES vCPU via a transition to SVM guest mode
* @svm: struct vcpu_svm *
* @spec_ctrl_intercepted: bool
*/
SYM_FUNC_START(__svm_sev_es_vcpu_run)
- push %_ASM_BP
-#ifdef CONFIG_X86_64
- push %r15
- push %r14
- push %r13
- push %r12
-#else
- push %edi
- push %esi
-#endif
- push %_ASM_BX
+ FRAME_BEGIN
/*
- * Save variables needed after vmexit on the stack, in inverse
- * order compared to when they are needed.
+ * Save non-volatile (callee-saved) registers to the host save area.
+ * Except for RAX and RSP, all GPRs are restored on #VMEXIT, but not
+ * saved on VMRUN.
*/
+ mov %rbp, SEV_ES_RBP (%rdx)
+ mov %r15, SEV_ES_R15 (%rdx)
+ mov %r14, SEV_ES_R14 (%rdx)
+ mov %r13, SEV_ES_R13 (%rdx)
+ mov %r12, SEV_ES_R12 (%rdx)
+ mov %rbx, SEV_ES_RBX (%rdx)
- /* Accessed directly from the stack in RESTORE_HOST_SPEC_CTRL. */
- push %_ASM_ARG2
-
- /* Save @svm. */
- push %_ASM_ARG1
-
-.ifnc _ASM_ARG1, _ASM_DI
/*
- * Stash @svm in RDI early. On 32-bit, arguments are in RAX, RCX
- * and RDX which are clobbered by RESTORE_GUEST_SPEC_CTRL.
+ * Save volatile registers that hold arguments that are needed after
+ * #VMEXIT (RDI=@svm and RSI=@spec_ctrl_intercepted).
*/
- mov %_ASM_ARG1, %_ASM_DI
-.endif
+ mov %rdi, SEV_ES_RDI (%rdx)
+ mov %rsi, SEV_ES_RSI (%rdx)
- /* Clobbers RAX, RCX, RDX. */
+ /* Clobbers RAX, RCX, RDX (@hostsa). */
RESTORE_GUEST_SPEC_CTRL
/* Get svm->current_vmcb->pa into RAX. */
- mov SVM_current_vmcb(%_ASM_DI), %_ASM_AX
- mov KVM_VMCB_pa(%_ASM_AX), %_ASM_AX
+ mov SVM_current_vmcb(%rdi), %rax
+ mov KVM_VMCB_pa(%rax), %rax
/* Enter guest mode */
sti
-1: vmrun %_ASM_AX
+1: vmrun %rax
2: cli
- /* Pop @svm to RDI, guest registers have been saved already. */
- pop %_ASM_DI
-
#ifdef CONFIG_MITIGATION_RETPOLINE
/* IMPORTANT: Stuff the RSB immediately after VM-Exit, before RET! */
- FILL_RETURN_BUFFER %_ASM_AX, RSB_CLEAR_LOOPS, X86_FEATURE_RETPOLINE
+ FILL_RETURN_BUFFER %rax, RSB_CLEAR_LOOPS, X86_FEATURE_RETPOLINE
#endif
- /* Clobbers RAX, RCX, RDX. */
+ /* Clobbers RAX, RCX, RDX, consumes RDI (@svm) and RSI (@spec_ctrl_intercepted). */
RESTORE_HOST_SPEC_CTRL
/*
@@ -361,30 +365,17 @@ SYM_FUNC_START(__svm_sev_es_vcpu_run)
*/
UNTRAIN_RET_VM
- /* "Pop" @spec_ctrl_intercepted. */
- pop %_ASM_BX
-
- pop %_ASM_BX
-
-#ifdef CONFIG_X86_64
- pop %r12
- pop %r13
- pop %r14
- pop %r15
-#else
- pop %esi
- pop %edi
-#endif
- pop %_ASM_BP
+ FRAME_END
RET
RESTORE_GUEST_SPEC_CTRL_BODY
- RESTORE_HOST_SPEC_CTRL_BODY
+ RESTORE_HOST_SPEC_CTRL_BODY %sil
-3: cmpb $0, _ASM_RIP(kvm_rebooting)
+3: cmpb $0, kvm_rebooting(%rip)
jne 2b
ud2
_ASM_EXTABLE(1b, 3b)
SYM_FUNC_END(__svm_sev_es_vcpu_run)
+#endif /* CONFIG_KVM_AMD_SEV */
diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
index 88659de4d2a7..c6b4b1728006 100644
--- a/arch/x86/kvm/trace.h
+++ b/arch/x86/kvm/trace.h
@@ -735,13 +735,13 @@ TRACE_EVENT(kvm_nested_intr_vmexit,
* Tracepoint for nested #vmexit because of interrupt pending
*/
TRACE_EVENT(kvm_invlpga,
- TP_PROTO(__u64 rip, int asid, u64 address),
+ TP_PROTO(__u64 rip, unsigned int asid, u64 address),
TP_ARGS(rip, asid, address),
TP_STRUCT__entry(
- __field( __u64, rip )
- __field( int, asid )
- __field( __u64, address )
+ __field( __u64, rip )
+ __field( unsigned int, asid )
+ __field( __u64, address )
),
TP_fast_assign(
@@ -750,7 +750,7 @@ TRACE_EVENT(kvm_invlpga,
__entry->address = address;
),
- TP_printk("rip: 0x%016llx asid: %d address: 0x%016llx",
+ TP_printk("rip: 0x%016llx asid: %u address: 0x%016llx",
__entry->rip, __entry->asid, __entry->address)
);
diff --git a/arch/x86/kvm/vmx/pmu_intel.c b/arch/x86/kvm/vmx/pmu_intel.c
index 12ade343a17e..be40474de6e4 100644
--- a/arch/x86/kvm/vmx/pmu_intel.c
+++ b/arch/x86/kvm/vmx/pmu_intel.c
@@ -535,7 +535,7 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
perf_capabilities = vcpu_get_perf_capabilities(vcpu);
if (cpuid_model_is_consistent(vcpu) &&
(perf_capabilities & PMU_CAP_LBR_FMT))
- x86_perf_get_lbr(&lbr_desc->records);
+ memcpy(&lbr_desc->records, &vmx_lbr_caps, sizeof(vmx_lbr_caps));
else
lbr_desc->records.nr = 0;
diff --git a/arch/x86/kvm/vmx/vmenter.S b/arch/x86/kvm/vmx/vmenter.S
index 2bfbf758d061..f6986dee6f8c 100644
--- a/arch/x86/kvm/vmx/vmenter.S
+++ b/arch/x86/kvm/vmx/vmenter.S
@@ -275,6 +275,8 @@ SYM_INNER_LABEL_ALIGN(vmx_vmexit, SYM_L_GLOBAL)
call vmx_spec_ctrl_restore_host
+ CLEAR_BRANCH_HISTORY_VMEXIT
+
/* Put return value in AX */
mov %_ASM_BX, %_ASM_AX
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index c37a89eda90f..22411f4aff53 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -218,6 +218,8 @@ module_param(ple_window_max, uint, 0444);
int __read_mostly pt_mode = PT_MODE_SYSTEM;
module_param(pt_mode, int, S_IRUGO);
+struct x86_pmu_lbr __ro_after_init vmx_lbr_caps;
+
static DEFINE_STATIC_KEY_FALSE(vmx_l1d_should_flush);
static DEFINE_STATIC_KEY_FALSE(vmx_l1d_flush_cond);
static DEFINE_MUTEX(vmx_l1d_flush_mutex);
@@ -7862,10 +7864,9 @@ static void vmx_vcpu_after_set_cpuid(struct kvm_vcpu *vcpu)
vmx_update_exception_bitmap(vcpu);
}
-static u64 vmx_get_perf_capabilities(void)
+static __init u64 vmx_get_perf_capabilities(void)
{
u64 perf_cap = PMU_CAP_FW_WRITES;
- struct x86_pmu_lbr lbr;
u64 host_perf_cap = 0;
if (!enable_pmu)
@@ -7875,15 +7876,43 @@ static u64 vmx_get_perf_capabilities(void)
rdmsrl(MSR_IA32_PERF_CAPABILITIES, host_perf_cap);
if (!cpu_feature_enabled(X86_FEATURE_ARCH_LBR)) {
- x86_perf_get_lbr(&lbr);
- if (lbr.nr)
+ x86_perf_get_lbr(&vmx_lbr_caps);
+
+ /*
+ * KVM requires LBR callstack support, as the overhead due to
+ * context switching LBRs without said support is too high.
+ * See intel_pmu_create_guest_lbr_event() for more info.
+ */
+ if (!vmx_lbr_caps.has_callstack)
+ memset(&vmx_lbr_caps, 0, sizeof(vmx_lbr_caps));
+ else if (vmx_lbr_caps.nr)
perf_cap |= host_perf_cap & PMU_CAP_LBR_FMT;
}
if (vmx_pebs_supported()) {
perf_cap |= host_perf_cap & PERF_CAP_PEBS_MASK;
- if ((perf_cap & PERF_CAP_PEBS_FORMAT) < 4)
- perf_cap &= ~PERF_CAP_PEBS_BASELINE;
+
+ /*
+ * Disallow adaptive PEBS as it is functionally broken, can be
+ * used by the guest to read *host* LBRs, and can be used to
+ * bypass userspace event filters. To correctly and safely
+ * support adaptive PEBS, KVM needs to:
+ *
+ * 1. Account for the ADAPTIVE flag when (re)programming fixed
+ * counters.
+ *
+ * 2. Gain support from perf (or take direct control of counter
+ * programming) to support events without adaptive PEBS
+ * enabled for the hardware counter.
+ *
+ * 3. Ensure LBR MSRs cannot hold host data on VM-Entry with
+ * adaptive PEBS enabled and MSR_PEBS_DATA_CFG.LBRS=1.
+ *
+ * 4. Document which PMU events are effectively exposed to the
+ * guest via adaptive PEBS, and make adaptive PEBS mutually
+ * exclusive with KVM_SET_PMU_EVENT_FILTER if necessary.
+ */
+ perf_cap &= ~PERF_CAP_PEBS_BASELINE;
}
return perf_cap;
diff --git a/arch/x86/kvm/vmx/vmx.h b/arch/x86/kvm/vmx/vmx.h
index 65786dbe7d60..90f9e4434646 100644
--- a/arch/x86/kvm/vmx/vmx.h
+++ b/arch/x86/kvm/vmx/vmx.h
@@ -15,6 +15,7 @@
#include "vmx_ops.h"
#include "../cpuid.h"
#include "run_flags.h"
+#include "../mmu.h"
#define MSR_TYPE_R 1
#define MSR_TYPE_W 2
@@ -109,6 +110,8 @@ struct lbr_desc {
bool msr_passthrough;
};
+extern struct x86_pmu_lbr vmx_lbr_caps;
+
/*
* The nested_vmx structure is part of vcpu_vmx, and holds information we need
* for correct emulation of VMX (i.e., nested VMX) on this vcpu.
@@ -719,7 +722,8 @@ static inline bool vmx_need_pf_intercept(struct kvm_vcpu *vcpu)
if (!enable_ept)
return true;
- return allow_smaller_maxphyaddr && cpuid_maxphyaddr(vcpu) < boot_cpu_data.x86_phys_bits;
+ return allow_smaller_maxphyaddr &&
+ cpuid_maxphyaddr(vcpu) < kvm_get_shadow_phys_bits();
}
static inline bool is_unrestricted_guest(struct kvm_vcpu *vcpu)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 47d9f03b7778..91478b769af0 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1621,7 +1621,7 @@ static bool kvm_is_immutable_feature_msr(u32 msr)
ARCH_CAP_PSCHANGE_MC_NO | ARCH_CAP_TSX_CTRL_MSR | ARCH_CAP_TAA_NO | \
ARCH_CAP_SBDR_SSDP_NO | ARCH_CAP_FBSDP_NO | ARCH_CAP_PSDP_NO | \
ARCH_CAP_FB_CLEAR | ARCH_CAP_RRSBA | ARCH_CAP_PBRSB_NO | ARCH_CAP_GDS_NO | \
- ARCH_CAP_RFDS_NO | ARCH_CAP_RFDS_CLEAR)
+ ARCH_CAP_RFDS_NO | ARCH_CAP_RFDS_CLEAR | ARCH_CAP_BHI_NO)
static u64 kvm_get_arch_capabilities(void)
{
@@ -3470,7 +3470,7 @@ static bool is_mci_status_msr(u32 msr)
static bool can_set_mci_status(struct kvm_vcpu *vcpu)
{
/* McStatusWrEn enabled? */
- if (guest_cpuid_is_amd_or_hygon(vcpu))
+ if (guest_cpuid_is_amd_compatible(vcpu))
return !!(vcpu->arch.msr_hwcr & BIT_ULL(18));
return false;