Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

include/linux/bpf.h
  1f6e04a1c7b8 ("bpf: Fix offset calculation error in __copy_map_value and zero_map_value")
  aa3496accc41 ("bpf: Refactor kptr_off_tab into btf_record")
  f71b2f64177a ("bpf: Refactor map->off_arr handling")
https://lore.kernel.org/all/20221114095000.67a73239@canb.auug.org.au/

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

6 files changed, 262 insertions, 142 deletions

diff --git a/arch/x86/kvm/svm/pmu.c b/arch/x86/kvm/svm/pmu.c
index b68956299fa8..9d65cd095691 100644
--- a/arch/x86/kvm/svm/pmu.c
+++ b/arch/x86/kvm/svm/pmu.c
@@ -192,9 +192,10 @@ static void amd_pmu_init(struct kvm_vcpu *vcpu)
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 	int i;
 
-	BUILD_BUG_ON(AMD64_NUM_COUNTERS_CORE > INTEL_PMC_MAX_GENERIC);
+	BUILD_BUG_ON(KVM_AMD_PMC_MAX_GENERIC > AMD64_NUM_COUNTERS_CORE);
+	BUILD_BUG_ON(KVM_AMD_PMC_MAX_GENERIC > INTEL_PMC_MAX_GENERIC);
 
-	for (i = 0; i < AMD64_NUM_COUNTERS_CORE ; i++) {
+	for (i = 0; i < KVM_AMD_PMC_MAX_GENERIC ; i++) {
 		pmu->gp_counters[i].type = KVM_PMC_GP;
 		pmu->gp_counters[i].vcpu = vcpu;
 		pmu->gp_counters[i].idx = i;
@@ -207,7 +208,7 @@ static void amd_pmu_reset(struct kvm_vcpu *vcpu)
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 	int i;
 
-	for (i = 0; i < AMD64_NUM_COUNTERS_CORE; i++) {
+	for (i = 0; i < KVM_AMD_PMC_MAX_GENERIC; i++) {
 		struct kvm_pmc *pmc = &pmu->gp_counters[i];
 
 		pmc_stop_counter(pmc);
diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c
index 28064060413a..efaaef2b7ae1 100644
--- a/arch/x86/kvm/svm/sev.c
+++ b/arch/x86/kvm/svm/sev.c
@@ -196,7 +196,7 @@ static void sev_asid_free(struct kvm_sev_info *sev)
 	__set_bit(sev->asid, sev_reclaim_asid_bitmap);
 
 	for_each_possible_cpu(cpu) {
-		sd = per_cpu(svm_data, cpu);
+		sd = per_cpu_ptr(&svm_data, cpu);
 		sd->sev_vmcbs[sev->asid] = NULL;
 	}
 
@@ -605,7 +605,7 @@ static int sev_es_sync_vmsa(struct vcpu_svm *svm)
 	save->dr6  = svm->vcpu.arch.dr6;
 
 	pr_debug("Virtual Machine Save Area (VMSA):\n");
-	print_hex_dump(KERN_CONT, "", DUMP_PREFIX_NONE, 16, 1, save, sizeof(*save), false);
+	print_hex_dump_debug("", DUMP_PREFIX_NONE, 16, 1, save, sizeof(*save), false);
 
 	return 0;
 }
@@ -2600,7 +2600,7 @@ 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(svm_data, cpu);
+	struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, cpu);
 	int asid = sev_get_asid(svm->vcpu.kvm);
 
 	/* Assign the asid allocated with this SEV guest */
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index 58f0077d9357..4b6d2b050e57 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -245,7 +245,7 @@ struct kvm_ldttss_desc {
 	u32 zero1;
 } __attribute__((packed));
 
-DEFINE_PER_CPU(struct svm_cpu_data *, svm_data);
+DEFINE_PER_CPU(struct svm_cpu_data, svm_data);
 
 /*
  * Only MSR_TSC_AUX is switched via the user return hook.  EFER is switched via
@@ -581,12 +581,7 @@ static int svm_hardware_enable(void)
 		pr_err("%s: err EOPNOTSUPP on %d\n", __func__, me);
 		return -EINVAL;
 	}
-	sd = per_cpu(svm_data, me);
-	if (!sd) {
-		pr_err("%s: svm_data is NULL on %d\n", __func__, me);
-		return -EINVAL;
-	}
-
+	sd = per_cpu_ptr(&svm_data, me);
 	sd->asid_generation = 1;
 	sd->max_asid = cpuid_ebx(SVM_CPUID_FUNC) - 1;
 	sd->next_asid = sd->max_asid + 1;
@@ -597,7 +592,7 @@ static int svm_hardware_enable(void)
 
 	wrmsrl(MSR_EFER, efer | EFER_SVME);
 
-	wrmsrl(MSR_VM_HSAVE_PA, __sme_page_pa(sd->save_area));
+	wrmsrl(MSR_VM_HSAVE_PA, sd->save_area_pa);
 
 	if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) {
 		/*
@@ -646,42 +641,37 @@ static int svm_hardware_enable(void)
 
 static void svm_cpu_uninit(int cpu)
 {
-	struct svm_cpu_data *sd = per_cpu(svm_data, cpu);
+	struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, cpu);
 
-	if (!sd)
+	if (!sd->save_area)
 		return;
 
-	per_cpu(svm_data, cpu) = NULL;
 	kfree(sd->sev_vmcbs);
 	__free_page(sd->save_area);
-	kfree(sd);
+	sd->save_area_pa = 0;
+	sd->save_area = NULL;
 }
 
 static int svm_cpu_init(int cpu)
 {
-	struct svm_cpu_data *sd;
+	struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, cpu);
 	int ret = -ENOMEM;
 
-	sd = kzalloc(sizeof(struct svm_cpu_data), GFP_KERNEL);
-	if (!sd)
-		return ret;
-	sd->cpu = cpu;
+	memset(sd, 0, sizeof(struct svm_cpu_data));
 	sd->save_area = alloc_page(GFP_KERNEL | __GFP_ZERO);
 	if (!sd->save_area)
-		goto free_cpu_data;
+		return ret;
 
 	ret = sev_cpu_init(sd);
 	if (ret)
 		goto free_save_area;
 
-	per_cpu(svm_data, cpu) = sd;
-
+	sd->save_area_pa = __sme_page_pa(sd->save_area);
 	return 0;
 
 free_save_area:
 	__free_page(sd->save_area);
-free_cpu_data:
-	kfree(sd);
+	sd->save_area = NULL;
 	return ret;
 
 }
@@ -730,6 +720,15 @@ static bool msr_write_intercepted(struct kvm_vcpu *vcpu, u32 msr)
 	u32 offset;
 	u32 *msrpm;
 
+	/*
+	 * For non-nested case:
+	 * If the L01 MSR bitmap does not intercept the MSR, then we need to
+	 * save it.
+	 *
+	 * For nested case:
+	 * If the L02 MSR bitmap does not intercept the MSR, then we need to
+	 * save it.
+	 */
 	msrpm = is_guest_mode(vcpu) ? to_svm(vcpu)->nested.msrpm:
 				      to_svm(vcpu)->msrpm;
 
@@ -1425,7 +1424,7 @@ static void svm_clear_current_vmcb(struct vmcb *vmcb)
 	int i;
 
 	for_each_online_cpu(i)
-		cmpxchg(&per_cpu(svm_data, i)->current_vmcb, vmcb, NULL);
+		cmpxchg(per_cpu_ptr(&svm_data.current_vmcb, i), vmcb, NULL);
 }
 
 static void svm_vcpu_free(struct kvm_vcpu *vcpu)
@@ -1450,7 +1449,7 @@ static void svm_vcpu_free(struct kvm_vcpu *vcpu)
 static void svm_prepare_switch_to_guest(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
-	struct svm_cpu_data *sd = per_cpu(svm_data, vcpu->cpu);
+	struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, vcpu->cpu);
 
 	if (sev_es_guest(vcpu->kvm))
 		sev_es_unmap_ghcb(svm);
@@ -1462,7 +1461,7 @@ static void svm_prepare_switch_to_guest(struct kvm_vcpu *vcpu)
 	 * Save additional host state that will be restored on VMEXIT (sev-es)
 	 * or subsequent vmload of host save area.
 	 */
-	vmsave(__sme_page_pa(sd->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);
@@ -1487,7 +1486,7 @@ static void svm_prepare_host_switch(struct kvm_vcpu *vcpu)
 static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
-	struct svm_cpu_data *sd = per_cpu(svm_data, cpu);
+	struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, cpu);
 
 	if (sd->current_vmcb != svm->vmcb) {
 		sd->current_vmcb = svm->vmcb;
@@ -2710,9 +2709,9 @@ static int svm_get_msr_feature(struct kvm_msr_entry *msr)
 	msr->data = 0;
 
 	switch (msr->index) {
-	case MSR_F10H_DECFG:
-		if (boot_cpu_has(X86_FEATURE_LFENCE_RDTSC))
-			msr->data |= MSR_F10H_DECFG_LFENCE_SERIALIZE;
+	case MSR_AMD64_DE_CFG:
+		if (cpu_feature_enabled(X86_FEATURE_LFENCE_RDTSC))
+			msr->data |= MSR_AMD64_DE_CFG_LFENCE_SERIALIZE;
 		break;
 	case MSR_IA32_PERF_CAPABILITIES:
 		return 0;
@@ -2813,7 +2812,7 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 			msr_info->data = 0x1E;
 		}
 		break;
-	case MSR_F10H_DECFG:
+	case MSR_AMD64_DE_CFG:
 		msr_info->data = svm->msr_decfg;
 		break;
 	default:
@@ -3042,7 +3041,7 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr)
 	case MSR_VM_IGNNE:
 		vcpu_unimpl(vcpu, "unimplemented wrmsr: 0x%x data 0x%llx\n", ecx, data);
 		break;
-	case MSR_F10H_DECFG: {
+	case MSR_AMD64_DE_CFG: {
 		struct kvm_msr_entry msr_entry;
 
 		msr_entry.index = msr->index;
@@ -3443,7 +3442,7 @@ static int svm_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath)
 
 static void reload_tss(struct kvm_vcpu *vcpu)
 {
-	struct svm_cpu_data *sd = per_cpu(svm_data, vcpu->cpu);
+	struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, vcpu->cpu);
 
 	sd->tss_desc->type = 9; /* available 32/64-bit TSS */
 	load_TR_desc();
@@ -3451,7 +3450,7 @@ static void reload_tss(struct kvm_vcpu *vcpu)
 
 static void pre_svm_run(struct kvm_vcpu *vcpu)
 {
-	struct svm_cpu_data *sd = per_cpu(svm_data, vcpu->cpu);
+	struct svm_cpu_data *sd = per_cpu_ptr(&svm_data, vcpu->cpu);
 	struct vcpu_svm *svm = to_svm(vcpu);
 
 	/*
@@ -3911,30 +3910,16 @@ static fastpath_t svm_exit_handlers_fastpath(struct kvm_vcpu *vcpu)
 	return EXIT_FASTPATH_NONE;
 }
 
-static noinstr void svm_vcpu_enter_exit(struct kvm_vcpu *vcpu)
+static noinstr void svm_vcpu_enter_exit(struct kvm_vcpu *vcpu, bool spec_ctrl_intercepted)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
-	unsigned long vmcb_pa = svm->current_vmcb->pa;
 
 	guest_state_enter_irqoff();
 
-	if (sev_es_guest(vcpu->kvm)) {
-		__svm_sev_es_vcpu_run(vmcb_pa);
-	} else {
-		struct svm_cpu_data *sd = per_cpu(svm_data, vcpu->cpu);
-
-		/*
-		 * Use a single vmcb (vmcb01 because it's always valid) for
-		 * context switching guest state via VMLOAD/VMSAVE, that way
-		 * the state doesn't need to be copied between vmcb01 and
-		 * vmcb02 when switching vmcbs for nested virtualization.
-		 */
-		vmload(svm->vmcb01.pa);
-		__svm_vcpu_run(vmcb_pa, (unsigned long *)&vcpu->arch.regs);
-		vmsave(svm->vmcb01.pa);
-
-		vmload(__sme_page_pa(sd->save_area));
-	}
+	if (sev_es_guest(vcpu->kvm))
+		__svm_sev_es_vcpu_run(svm, spec_ctrl_intercepted);
+	else
+		__svm_vcpu_run(svm, spec_ctrl_intercepted);
 
 	guest_state_exit_irqoff();
 }
@@ -3942,6 +3927,7 @@ static noinstr void svm_vcpu_enter_exit(struct kvm_vcpu *vcpu)
 static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
+	bool spec_ctrl_intercepted = msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL);
 
 	trace_kvm_entry(vcpu);
 
@@ -3998,34 +3984,15 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu)
 	 * being speculatively taken.
 	 */
 	if (!static_cpu_has(X86_FEATURE_V_SPEC_CTRL))
-		x86_spec_ctrl_set_guest(svm->spec_ctrl, svm->virt_spec_ctrl);
+		x86_spec_ctrl_set_guest(svm->virt_spec_ctrl);
 
-	svm_vcpu_enter_exit(vcpu);
-
-	/*
-	 * We do not use IBRS in the kernel. If this vCPU has used the
-	 * SPEC_CTRL MSR it may have left it on; save the value and
-	 * turn it off. This is much more efficient than blindly adding
-	 * it to the atomic save/restore list. Especially as the former
-	 * (Saving guest MSRs on vmexit) doesn't even exist in KVM.
-	 *
-	 * For non-nested case:
-	 * If the L01 MSR bitmap does not intercept the MSR, then we need to
-	 * save it.
-	 *
-	 * For nested case:
-	 * If the L02 MSR bitmap does not intercept the MSR, then we need to
-	 * save it.
-	 */
-	if (!static_cpu_has(X86_FEATURE_V_SPEC_CTRL) &&
-	    unlikely(!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL)))
-		svm->spec_ctrl = native_read_msr(MSR_IA32_SPEC_CTRL);
+	svm_vcpu_enter_exit(vcpu, spec_ctrl_intercepted);
 
 	if (!sev_es_guest(vcpu->kvm))
 		reload_tss(vcpu);
 
 	if (!static_cpu_has(X86_FEATURE_V_SPEC_CTRL))
-		x86_spec_ctrl_restore_host(svm->spec_ctrl, svm->virt_spec_ctrl);
+		x86_spec_ctrl_restore_host(svm->virt_spec_ctrl);
 
 	if (!sev_es_guest(vcpu->kvm)) {
 		vcpu->arch.cr2 = svm->vmcb->save.cr2;
diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h
index 6a7686bf6900..199a2ecef1ce 100644
--- a/arch/x86/kvm/svm/svm.h
+++ b/arch/x86/kvm/svm/svm.h
@@ -209,7 +209,6 @@ struct vcpu_svm {
 	struct vmcb *vmcb;
 	struct kvm_vmcb_info vmcb01;
 	struct kvm_vmcb_info *current_vmcb;
-	struct svm_cpu_data *svm_data;
 	u32 asid;
 	u32 sysenter_esp_hi;
 	u32 sysenter_eip_hi;
@@ -281,8 +280,6 @@ struct vcpu_svm {
 };
 
 struct svm_cpu_data {
-	int cpu;
-
 	u64 asid_generation;
 	u32 max_asid;
 	u32 next_asid;
@@ -290,13 +287,15 @@ struct svm_cpu_data {
 	struct kvm_ldttss_desc *tss_desc;
 
 	struct page *save_area;
+	unsigned long save_area_pa;
+
 	struct vmcb *current_vmcb;
 
 	/* index = sev_asid, value = vmcb pointer */
 	struct vmcb **sev_vmcbs;
 };
 
-DECLARE_PER_CPU(struct svm_cpu_data *, svm_data);
+DECLARE_PER_CPU(struct svm_cpu_data, svm_data);
 
 void recalc_intercepts(struct vcpu_svm *svm);
 
@@ -683,7 +682,7 @@ void sev_es_unmap_ghcb(struct vcpu_svm *svm);
 
 /* vmenter.S */
 
-void __svm_sev_es_vcpu_run(unsigned long vmcb_pa);
-void __svm_vcpu_run(unsigned long vmcb_pa, unsigned long *regs);
+void __svm_sev_es_vcpu_run(struct vcpu_svm *svm, bool spec_ctrl_intercepted);
+void __svm_vcpu_run(struct vcpu_svm *svm, bool spec_ctrl_intercepted);
 
 #endif
diff --git a/arch/x86/kvm/svm/svm_ops.h b/arch/x86/kvm/svm/svm_ops.h
index 9430d6437c9f..36c8af87a707 100644
--- a/arch/x86/kvm/svm/svm_ops.h
+++ b/arch/x86/kvm/svm/svm_ops.h
@@ -61,9 +61,4 @@ static __always_inline void vmsave(unsigned long pa)
 	svm_asm1(vmsave, "a" (pa), "memory");
 }
 
-static __always_inline void vmload(unsigned long pa)
-{
-	svm_asm1(vmload, "a" (pa), "memory");
-}
-
 #endif /* __KVM_X86_SVM_OPS_H */
diff --git a/arch/x86/kvm/svm/vmenter.S b/arch/x86/kvm/svm/vmenter.S
index 723f8534986c..34367dc203f2 100644
--- a/arch/x86/kvm/svm/vmenter.S
+++ b/arch/x86/kvm/svm/vmenter.S
@@ -4,35 +4,97 @@
 #include <asm/bitsperlong.h>
 #include <asm/kvm_vcpu_regs.h>
 #include <asm/nospec-branch.h>
+#include "kvm-asm-offsets.h"
 
 #define WORD_SIZE (BITS_PER_LONG / 8)
 
 /* Intentionally omit RAX as it's context switched by hardware */
-#define VCPU_RCX	__VCPU_REGS_RCX * WORD_SIZE
-#define VCPU_RDX	__VCPU_REGS_RDX * WORD_SIZE
-#define VCPU_RBX	__VCPU_REGS_RBX * WORD_SIZE
+#define VCPU_RCX	(SVM_vcpu_arch_regs + __VCPU_REGS_RCX * WORD_SIZE)
+#define VCPU_RDX	(SVM_vcpu_arch_regs + __VCPU_REGS_RDX * WORD_SIZE)
+#define VCPU_RBX	(SVM_vcpu_arch_regs + __VCPU_REGS_RBX * WORD_SIZE)
 /* Intentionally omit RSP as it's context switched by hardware */
-#define VCPU_RBP	__VCPU_REGS_RBP * WORD_SIZE
-#define VCPU_RSI	__VCPU_REGS_RSI * WORD_SIZE
-#define VCPU_RDI	__VCPU_REGS_RDI * WORD_SIZE
+#define VCPU_RBP	(SVM_vcpu_arch_regs + __VCPU_REGS_RBP * WORD_SIZE)
+#define VCPU_RSI	(SVM_vcpu_arch_regs + __VCPU_REGS_RSI * WORD_SIZE)
+#define VCPU_RDI	(SVM_vcpu_arch_regs + __VCPU_REGS_RDI * WORD_SIZE)
 
 #ifdef CONFIG_X86_64
-#define VCPU_R8		__VCPU_REGS_R8  * WORD_SIZE
-#define VCPU_R9		__VCPU_REGS_R9  * WORD_SIZE
-#define VCPU_R10	__VCPU_REGS_R10 * WORD_SIZE
-#define VCPU_R11	__VCPU_REGS_R11 * WORD_SIZE
-#define VCPU_R12	__VCPU_REGS_R12 * WORD_SIZE
-#define VCPU_R13	__VCPU_REGS_R13 * WORD_SIZE
-#define VCPU_R14	__VCPU_REGS_R14 * WORD_SIZE
-#define VCPU_R15	__VCPU_REGS_R15 * WORD_SIZE
+#define VCPU_R8		(SVM_vcpu_arch_regs + __VCPU_REGS_R8  * WORD_SIZE)
+#define VCPU_R9		(SVM_vcpu_arch_regs + __VCPU_REGS_R9  * WORD_SIZE)
+#define VCPU_R10	(SVM_vcpu_arch_regs + __VCPU_REGS_R10 * WORD_SIZE)
+#define VCPU_R11	(SVM_vcpu_arch_regs + __VCPU_REGS_R11 * WORD_SIZE)
+#define VCPU_R12	(SVM_vcpu_arch_regs + __VCPU_REGS_R12 * WORD_SIZE)
+#define VCPU_R13	(SVM_vcpu_arch_regs + __VCPU_REGS_R13 * WORD_SIZE)
+#define VCPU_R14	(SVM_vcpu_arch_regs + __VCPU_REGS_R14 * WORD_SIZE)
+#define VCPU_R15	(SVM_vcpu_arch_regs + __VCPU_REGS_R15 * WORD_SIZE)
 #endif
 
+#define SVM_vmcb01_pa	(SVM_vmcb01 + KVM_VMCB_pa)
+
 .section .noinstr.text, "ax"
 
+.macro RESTORE_GUEST_SPEC_CTRL
+	/* No need to do anything if SPEC_CTRL is unset or V_SPEC_CTRL is set */
+	ALTERNATIVE_2 "", \
+		"jmp 800f", X86_FEATURE_MSR_SPEC_CTRL, \
+		"", X86_FEATURE_V_SPEC_CTRL
+801:
+.endm
+.macro RESTORE_GUEST_SPEC_CTRL_BODY
+800:
+	/*
+	 * SPEC_CTRL handling: if the guest's SPEC_CTRL value differs from the
+	 * host's, write the MSR.  This is kept out-of-line so that the common
+	 * case does not have to jump.
+	 *
+	 * IMPORTANT: To avoid RSB underflow attacks and any other nastiness,
+	 * there must not be any returns or indirect branches between this code
+	 * and vmentry.
+	 */
+	movl SVM_spec_ctrl(%_ASM_DI), %eax
+	cmp PER_CPU_VAR(x86_spec_ctrl_current), %eax
+	je 801b
+	mov $MSR_IA32_SPEC_CTRL, %ecx
+	xor %edx, %edx
+	wrmsr
+	jmp 801b
+.endm
+
+.macro RESTORE_HOST_SPEC_CTRL
+	/* No need to do anything if SPEC_CTRL is unset or V_SPEC_CTRL is set */
+	ALTERNATIVE_2 "", \
+		"jmp 900f", X86_FEATURE_MSR_SPEC_CTRL, \
+		"", X86_FEATURE_V_SPEC_CTRL
+901:
+.endm
+.macro RESTORE_HOST_SPEC_CTRL_BODY
+900:
+	/* Same for after vmexit.  */
+	mov $MSR_IA32_SPEC_CTRL, %ecx
+
+	/*
+	 * 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)
+	jnz 998f
+	rdmsr
+	movl %eax, SVM_spec_ctrl(%_ASM_DI)
+998:
+
+	/* Now restore the host value of the MSR if different from the guest's.  */
+	movl PER_CPU_VAR(x86_spec_ctrl_current), %eax
+	cmp SVM_spec_ctrl(%_ASM_DI), %eax
+	je 901b
+	xor %edx, %edx
+	wrmsr
+	jmp 901b
+.endm
+
+
 /**
  * __svm_vcpu_run - Run a vCPU via a transition to SVM guest mode
- * @vmcb_pa:	unsigned long
- * @regs:	unsigned long * (to guest registers)
+ * @svm:	struct vcpu_svm *
+ * @spec_ctrl_intercepted: bool
  */
 SYM_FUNC_START(__svm_vcpu_run)
 	push %_ASM_BP
@@ -47,49 +109,71 @@ SYM_FUNC_START(__svm_vcpu_run)
 #endif
 	push %_ASM_BX
 
-	/* Save @regs. */
+	/*
+	 * Save variables needed after vmexit on the stack, in inverse
+	 * order compared to when they are needed.
+	 */
+
+	/* Accessed directly from the stack in RESTORE_HOST_SPEC_CTRL.  */
 	push %_ASM_ARG2
 
-	/* Save @vmcb. */
+	/* Needed to restore access to percpu variables.  */
+	__ASM_SIZE(push) PER_CPU_VAR(svm_data + SD_save_area_pa)
+
+	/* Finally save @svm. */
 	push %_ASM_ARG1
 
-	/* Move @regs to RAX. */
-	mov %_ASM_ARG2, %_ASM_AX
+.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.
+	 */
+	mov %_ASM_ARG1, %_ASM_DI
+.endif
+
+	/* Clobbers RAX, RCX, RDX.  */
+	RESTORE_GUEST_SPEC_CTRL
+
+	/*
+	 * Use a single vmcb (vmcb01 because it's always valid) for
+	 * context switching guest state via VMLOAD/VMSAVE, that way
+	 * the state doesn't need to be copied between vmcb01 and
+	 * vmcb02 when switching vmcbs for nested virtualization.
+	 */
+	mov SVM_vmcb01_pa(%_ASM_DI), %_ASM_AX
+1:	vmload %_ASM_AX
+2:
+
+	/* Get svm->current_vmcb->pa into RAX. */
+	mov SVM_current_vmcb(%_ASM_DI), %_ASM_AX
+	mov KVM_VMCB_pa(%_ASM_AX), %_ASM_AX
 
 	/* Load guest registers. */
-	mov VCPU_RCX(%_ASM_AX), %_ASM_CX
-	mov VCPU_RDX(%_ASM_AX), %_ASM_DX
-	mov VCPU_RBX(%_ASM_AX), %_ASM_BX
-	mov VCPU_RBP(%_ASM_AX), %_ASM_BP
-	mov VCPU_RSI(%_ASM_AX), %_ASM_SI
-	mov VCPU_RDI(%_ASM_AX), %_ASM_DI
+	mov VCPU_RCX(%_ASM_DI), %_ASM_CX
+	mov VCPU_RDX(%_ASM_DI), %_ASM_DX
+	mov VCPU_RBX(%_ASM_DI), %_ASM_BX
+	mov VCPU_RBP(%_ASM_DI), %_ASM_BP
+	mov VCPU_RSI(%_ASM_DI), %_ASM_SI
 #ifdef CONFIG_X86_64
-	mov VCPU_R8 (%_ASM_AX),  %r8
-	mov VCPU_R9 (%_ASM_AX),  %r9
-	mov VCPU_R10(%_ASM_AX), %r10
-	mov VCPU_R11(%_ASM_AX), %r11
-	mov VCPU_R12(%_ASM_AX), %r12
-	mov VCPU_R13(%_ASM_AX), %r13
-	mov VCPU_R14(%_ASM_AX), %r14
-	mov VCPU_R15(%_ASM_AX), %r15
+	mov VCPU_R8 (%_ASM_DI),  %r8
+	mov VCPU_R9 (%_ASM_DI),  %r9
+	mov VCPU_R10(%_ASM_DI), %r10
+	mov VCPU_R11(%_ASM_DI), %r11
+	mov VCPU_R12(%_ASM_DI), %r12
+	mov VCPU_R13(%_ASM_DI), %r13
+	mov VCPU_R14(%_ASM_DI), %r14
+	mov VCPU_R15(%_ASM_DI), %r15
 #endif
-
-	/* "POP" @vmcb to RAX. */
-	pop %_ASM_AX
+	mov VCPU_RDI(%_ASM_DI), %_ASM_DI
 
 	/* Enter guest mode */
 	sti
 
-1:	vmrun %_ASM_AX
-
-2:	cli
-
-#ifdef CONFIG_RETPOLINE
-	/* IMPORTANT: Stuff the RSB immediately after VM-Exit, before RET! */
-	FILL_RETURN_BUFFER %_ASM_AX, RSB_CLEAR_LOOPS, X86_FEATURE_RETPOLINE
-#endif
+3:	vmrun %_ASM_AX
+4:
+	cli
 
-	/* "POP" @regs to RAX. */
+	/* Pop @svm to RAX while it's the only available register. */
 	pop %_ASM_AX
 
 	/* Save all guest registers.  */
@@ -110,6 +194,26 @@ SYM_FUNC_START(__svm_vcpu_run)
 	mov %r15, VCPU_R15(%_ASM_AX)
 #endif
 
+	/* @svm can stay in RDI from now on.  */
+	mov %_ASM_AX, %_ASM_DI
+
+	mov SVM_vmcb01_pa(%_ASM_DI), %_ASM_AX
+5:	vmsave %_ASM_AX
+6:
+
+	/* Restores GSBASE among other things, allowing access to percpu data.  */
+	pop %_ASM_AX
+7:	vmload %_ASM_AX
+8:
+
+#ifdef CONFIG_RETPOLINE
+	/* IMPORTANT: Stuff the RSB immediately after VM-Exit, before RET! */
+	FILL_RETURN_BUFFER %_ASM_AX, RSB_CLEAR_LOOPS, X86_FEATURE_RETPOLINE
+#endif
+
+	/* Clobbers RAX, RCX, RDX.  */
+	RESTORE_HOST_SPEC_CTRL
+
 	/*
 	 * Mitigate RETBleed for AMD/Hygon Zen uarch. RET should be
 	 * untrained as soon as we exit the VM and are back to the
@@ -145,6 +249,9 @@ SYM_FUNC_START(__svm_vcpu_run)
 	xor %r15d, %r15d
 #endif
 
+	/* "Pop" @spec_ctrl_intercepted.  */
+	pop %_ASM_BX
+
 	pop %_ASM_BX
 
 #ifdef CONFIG_X86_64
@@ -159,17 +266,33 @@ SYM_FUNC_START(__svm_vcpu_run)
 	pop %_ASM_BP
 	RET
 
-3:	cmpb $0, kvm_rebooting
+	RESTORE_GUEST_SPEC_CTRL_BODY
+	RESTORE_HOST_SPEC_CTRL_BODY
+
+10:	cmpb $0, kvm_rebooting
 	jne 2b
 	ud2
+30:	cmpb $0, kvm_rebooting
+	jne 4b
+	ud2
+50:	cmpb $0, kvm_rebooting
+	jne 6b
+	ud2
+70:	cmpb $0, kvm_rebooting
+	jne 8b
+	ud2
 
-	_ASM_EXTABLE(1b, 3b)
+	_ASM_EXTABLE(1b, 10b)
+	_ASM_EXTABLE(3b, 30b)
+	_ASM_EXTABLE(5b, 50b)
+	_ASM_EXTABLE(7b, 70b)
 
 SYM_FUNC_END(__svm_vcpu_run)
 
 /**
  * __svm_sev_es_vcpu_run - Run a SEV-ES vCPU via a transition to SVM guest mode
- * @vmcb_pa:	unsigned long
+ * @svm:	struct vcpu_svm *
+ * @spec_ctrl_intercepted: bool
  */
 SYM_FUNC_START(__svm_sev_es_vcpu_run)
 	push %_ASM_BP
@@ -184,8 +307,31 @@ SYM_FUNC_START(__svm_sev_es_vcpu_run)
 #endif
 	push %_ASM_BX
 
-	/* Move @vmcb to RAX. */
-	mov %_ASM_ARG1, %_ASM_AX
+	/*
+	 * Save variables needed after vmexit on the stack, in inverse
+	 * order compared to when they are needed.
+	 */
+
+	/* 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.
+	 */
+	mov %_ASM_ARG1, %_ASM_DI
+.endif
+
+	/* Clobbers RAX, RCX, RDX.  */
+	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
 
 	/* Enter guest mode */
 	sti
@@ -194,11 +340,17 @@ SYM_FUNC_START(__svm_sev_es_vcpu_run)
 
 2:	cli
 
+	/* Pop @svm to RDI, guest registers have been saved already. */
+	pop %_ASM_DI
+
 #ifdef CONFIG_RETPOLINE
 	/* IMPORTANT: Stuff the RSB immediately after VM-Exit, before RET! */
 	FILL_RETURN_BUFFER %_ASM_AX, RSB_CLEAR_LOOPS, X86_FEATURE_RETPOLINE
 #endif
 
+	/* Clobbers RAX, RCX, RDX.  */
+	RESTORE_HOST_SPEC_CTRL
+
 	/*
 	 * Mitigate RETBleed for AMD/Hygon Zen uarch. RET should be
 	 * untrained as soon as we exit the VM and are back to the
@@ -208,6 +360,9 @@ SYM_FUNC_START(__svm_sev_es_vcpu_run)
 	 */
 	UNTRAIN_RET
 
+	/* "Pop" @spec_ctrl_intercepted.  */
+	pop %_ASM_BX
+
 	pop %_ASM_BX
 
 #ifdef CONFIG_X86_64
@@ -222,6 +377,9 @@ SYM_FUNC_START(__svm_sev_es_vcpu_run)
 	pop %_ASM_BP
 	RET
 
+	RESTORE_GUEST_SPEC_CTRL_BODY
+	RESTORE_HOST_SPEC_CTRL_BODY
+
 3:	cmpb $0, kvm_rebooting
 	jne 2b
 	ud2