From b9b2782cd55aaf5a65573df99376cf391941cb66 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Tue, 5 May 2020 11:47:50 -0400
Subject: KVM: X86: Declare KVM_CAP_SET_GUEST_DEBUG properly

KVM_CAP_SET_GUEST_DEBUG should be supported for x86 however it's not declared
as supported.  My wild guess is that userspaces like QEMU are using "#ifdef
KVM_CAP_SET_GUEST_DEBUG" to check for the capability instead, but that could be
wrong because the compilation host may not be the runtime host.

The userspace might still want to keep the old "#ifdef" though to not break the
guest debug on old kernels.

Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20200505154750.126300-1-peterx@redhat.com>
[Do the same for PPC and s390. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/powerpc/kvm/powerpc.c | 1 +
 arch/s390/kvm/kvm-s390.c   | 1 +
 arch/x86/kvm/x86.c         | 1 +
 3 files changed, 3 insertions(+)

(limited to 'arch')

diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index e15166b0a16d..ad2f172c26a6 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -521,6 +521,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_IOEVENTFD:
 	case KVM_CAP_DEVICE_CTRL:
 	case KVM_CAP_IMMEDIATE_EXIT:
+	case KVM_CAP_SET_GUEST_DEBUG:
 		r = 1;
 		break;
 	case KVM_CAP_PPC_GUEST_DEBUG_SSTEP:
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 5dcf9ff12828..d05bb040fd42 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -545,6 +545,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_S390_AIS:
 	case KVM_CAP_S390_AIS_MIGRATION:
 	case KVM_CAP_S390_VCPU_RESETS:
+	case KVM_CAP_SET_GUEST_DEBUG:
 		r = 1;
 		break;
 	case KVM_CAP_S390_HPAGE_1M:
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 8d296e3d0d56..d786c7d27ce5 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3372,6 +3372,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_GET_MSR_FEATURES:
 	case KVM_CAP_MSR_PLATFORM_INFO:
 	case KVM_CAP_EXCEPTION_PAYLOAD:
+	case KVM_CAP_SET_GUEST_DEBUG:
 		r = 1;
 		break;
 	case KVM_CAP_SYNC_REGS:
-- 
cgit v1.2.3


From 4d5523cfd5d298c58743eb31c003886cfc856709 Mon Sep 17 00:00:00 2001
From: Paolo Bonzini <pbonzini@redhat.com>
Date: Tue, 5 May 2020 07:33:20 -0400
Subject: KVM: x86: fix DR6 delivery for various cases of #DB injection

Go through kvm_queue_exception_p so that the payload is correctly delivered
through the exit qualification, and add a kvm_update_dr6 call to
kvm_deliver_exception_payload that is needed on AMD.

Reported-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/include/asm/kvm_host.h |  1 +
 arch/x86/kvm/vmx/vmx.c          |  8 ++------
 arch/x86/kvm/x86.c              | 11 ++++++-----
 3 files changed, 9 insertions(+), 11 deletions(-)

(limited to 'arch')

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 0dea9f122bb9..8c247bcb037e 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -1449,6 +1449,7 @@ bool kvm_rdpmc(struct kvm_vcpu *vcpu);
 
 void kvm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr);
 void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
+void kvm_queue_exception_p(struct kvm_vcpu *vcpu, unsigned nr, unsigned long payload);
 void kvm_requeue_exception(struct kvm_vcpu *vcpu, unsigned nr);
 void kvm_requeue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code);
 void kvm_inject_page_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault);
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index c2c6335a998c..bb5a527e49d9 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -4677,12 +4677,10 @@ static int handle_exception_nmi(struct kvm_vcpu *vcpu)
 		dr6 = vmcs_readl(EXIT_QUALIFICATION);
 		if (!(vcpu->guest_debug &
 		      (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))) {
-			vcpu->arch.dr6 &= ~DR_TRAP_BITS;
-			vcpu->arch.dr6 |= dr6 | DR6_RTM;
 			if (is_icebp(intr_info))
 				WARN_ON(!skip_emulated_instruction(vcpu));
 
-			kvm_queue_exception(vcpu, DB_VECTOR);
+			kvm_queue_exception_p(vcpu, DB_VECTOR, dr6);
 			return 1;
 		}
 		kvm_run->debug.arch.dr6 = dr6 | DR6_FIXED_1;
@@ -4936,9 +4934,7 @@ static int handle_dr(struct kvm_vcpu *vcpu)
 			vcpu->run->exit_reason = KVM_EXIT_DEBUG;
 			return 0;
 		} else {
-			vcpu->arch.dr6 &= ~DR_TRAP_BITS;
-			vcpu->arch.dr6 |= DR6_BD | DR6_RTM;
-			kvm_queue_exception(vcpu, DB_VECTOR);
+			kvm_queue_exception_p(vcpu, DB_VECTOR, DR6_BD);
 			return 1;
 		}
 	}
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index d786c7d27ce5..109115c96897 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -104,6 +104,7 @@ static u64 __read_mostly cr4_reserved_bits = CR4_RESERVED_BITS;
                                     KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK)
 
 static void update_cr8_intercept(struct kvm_vcpu *vcpu);
+static void kvm_update_dr6(struct kvm_vcpu *vcpu);
 static void process_nmi(struct kvm_vcpu *vcpu);
 static void enter_smm(struct kvm_vcpu *vcpu);
 static void __kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags);
@@ -473,6 +474,7 @@ void kvm_deliver_exception_payload(struct kvm_vcpu *vcpu)
 		 * breakpoint), it is reserved and must be zero in DR6.
 		 */
 		vcpu->arch.dr6 &= ~BIT(12);
+		kvm_update_dr6(vcpu);
 		break;
 	case PF_VECTOR:
 		vcpu->arch.cr2 = payload;
@@ -572,11 +574,12 @@ void kvm_requeue_exception(struct kvm_vcpu *vcpu, unsigned nr)
 }
 EXPORT_SYMBOL_GPL(kvm_requeue_exception);
 
-static void kvm_queue_exception_p(struct kvm_vcpu *vcpu, unsigned nr,
-				  unsigned long payload)
+void kvm_queue_exception_p(struct kvm_vcpu *vcpu, unsigned nr,
+			   unsigned long payload)
 {
 	kvm_multiple_exception(vcpu, nr, false, 0, true, payload, false);
 }
+EXPORT_SYMBOL_GPL(kvm_queue_exception_p);
 
 static void kvm_queue_exception_e_p(struct kvm_vcpu *vcpu, unsigned nr,
 				    u32 error_code, unsigned long payload)
@@ -6719,9 +6722,7 @@ static bool kvm_vcpu_check_breakpoint(struct kvm_vcpu *vcpu, int *r)
 					   vcpu->arch.db);
 
 		if (dr6 != 0) {
-			vcpu->arch.dr6 &= ~DR_TRAP_BITS;
-			vcpu->arch.dr6 |= dr6 | DR6_RTM;
-			kvm_queue_exception(vcpu, DB_VECTOR);
+			kvm_queue_exception_p(vcpu, DB_VECTOR, dr6);
 			*r = 1;
 			return true;
 		}
-- 
cgit v1.2.3


From 13196638d5de2564af90387930afa19cf27c1974 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Tue, 5 May 2020 16:49:58 -0400
Subject: KVM: X86: Set RTM for DB_VECTOR too for KVM_EXIT_DEBUG

RTM should always been set even with KVM_EXIT_DEBUG on #DB.

Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20200505205000.188252-2-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/kvm/vmx/vmx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index bb5a527e49d9..2384a2dbec44 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -4683,7 +4683,7 @@ static int handle_exception_nmi(struct kvm_vcpu *vcpu)
 			kvm_queue_exception_p(vcpu, DB_VECTOR, dr6);
 			return 1;
 		}
-		kvm_run->debug.arch.dr6 = dr6 | DR6_FIXED_1;
+		kvm_run->debug.arch.dr6 = dr6 | DR6_FIXED_1 | DR6_RTM;
 		kvm_run->debug.arch.dr7 = vmcs_readl(GUEST_DR7);
 		/* fall through */
 	case BP_VECTOR:
-- 
cgit v1.2.3


From d5d260c5ffad0e0e4e59ae2772d4a906d3287e6a Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Tue, 5 May 2020 16:49:59 -0400
Subject: KVM: X86: Fix single-step with KVM_SET_GUEST_DEBUG

When single-step triggered with KVM_SET_GUEST_DEBUG, we should fill in the pc
value with current linear RIP rather than the cached singlestep address.

Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20200505205000.188252-3-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/kvm/x86.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 109115c96897..f7628555f036 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -6662,7 +6662,7 @@ static int kvm_vcpu_do_singlestep(struct kvm_vcpu *vcpu)
 
 	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) {
 		kvm_run->debug.arch.dr6 = DR6_BS | DR6_FIXED_1 | DR6_RTM;
-		kvm_run->debug.arch.pc = vcpu->arch.singlestep_rip;
+		kvm_run->debug.arch.pc = kvm_get_linear_rip(vcpu);
 		kvm_run->debug.arch.exception = DB_VECTOR;
 		kvm_run->exit_reason = KVM_EXIT_DEBUG;
 		return 0;
-- 
cgit v1.2.3


From 2c19dba6803bfeb694da0b0b50e788b77a37fe75 Mon Sep 17 00:00:00 2001
From: Paolo Bonzini <pbonzini@redhat.com>
Date: Thu, 7 May 2020 07:20:27 -0400
Subject: KVM: nSVM: trap #DB and #BP to userspace if guest debugging is on

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/kvm/svm/nested.c | 8 ++++++++
 1 file changed, 8 insertions(+)

(limited to 'arch')

diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c
index 90a1ca939627..adab5b1c5fe1 100644
--- a/arch/x86/kvm/svm/nested.c
+++ b/arch/x86/kvm/svm/nested.c
@@ -608,6 +608,11 @@ static int nested_svm_intercept_db(struct vcpu_svm *svm)
 {
 	unsigned long dr6;
 
+	/* Always catch it and pass it to userspace if debugging.  */
+	if (svm->vcpu.guest_debug &
+	    (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))
+		return NESTED_EXIT_HOST;
+
 	/* if we're not singlestepping, it's not ours */
 	if (!svm->nmi_singlestep)
 		return NESTED_EXIT_DONE;
@@ -682,6 +687,9 @@ static int nested_svm_intercept(struct vcpu_svm *svm)
 		if (svm->nested.intercept_exceptions & excp_bits) {
 			if (exit_code == SVM_EXIT_EXCP_BASE + DB_VECTOR)
 				vmexit = nested_svm_intercept_db(svm);
+			else if (exit_code == SVM_EXIT_EXCP_BASE + BP_VECTOR &&
+				 svm->vcpu.guest_debug & KVM_GUESTDBG_USE_SW_BP)
+				vmexit = NESTED_EXIT_HOST;
 			else
 				vmexit = NESTED_EXIT_DONE;
 		}
-- 
cgit v1.2.3


From 5679b803e44ed8947e8c2a7f44cdef1d93ea24d5 Mon Sep 17 00:00:00 2001
From: Paolo Bonzini <pbonzini@redhat.com>
Date: Mon, 4 May 2020 11:28:25 -0400
Subject: KVM: SVM: keep DR6 synchronized with vcpu->arch.dr6

kvm_x86_ops.set_dr6 is only ever called with vcpu->arch.dr6 as the
second argument.  Ensure that the VMCB value is synchronized to
vcpu->arch.dr6 on #DB (both "normal" and nested) and nested vmentry, so
that the current value of DR6 is always available in vcpu->arch.dr6.
The get_dr6 callback can just access vcpu->arch.dr6 and becomes redundant.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/include/asm/kvm_host.h |  2 +-
 arch/x86/kvm/svm/nested.c       | 24 ++++++++++++++++--------
 arch/x86/kvm/svm/svm.c          |  9 ++-------
 arch/x86/kvm/vmx/vmx.c          |  6 ------
 arch/x86/kvm/x86.c              |  8 ++------
 5 files changed, 21 insertions(+), 28 deletions(-)

(limited to 'arch')

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 8c247bcb037e..de0c28814348 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -1093,7 +1093,6 @@ struct kvm_x86_ops {
 	void (*set_idt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
 	void (*get_gdt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
 	void (*set_gdt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
-	u64 (*get_dr6)(struct kvm_vcpu *vcpu);
 	void (*set_dr6)(struct kvm_vcpu *vcpu, unsigned long value);
 	void (*sync_dirty_debug_regs)(struct kvm_vcpu *vcpu);
 	void (*set_dr7)(struct kvm_vcpu *vcpu, unsigned long value);
@@ -1624,6 +1623,7 @@ int kvm_pv_send_ipi(struct kvm *kvm, unsigned long ipi_bitmap_low,
 
 void kvm_define_shared_msr(unsigned index, u32 msr);
 int kvm_set_shared_msr(unsigned index, u64 val, u64 mask);
+void kvm_update_dr6(struct kvm_vcpu *vcpu);
 
 u64 kvm_scale_tsc(struct kvm_vcpu *vcpu, u64 tsc);
 u64 kvm_read_l1_tsc(struct kvm_vcpu *vcpu, u64 host_tsc);
diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c
index adab5b1c5fe1..9cfa8098995e 100644
--- a/arch/x86/kvm/svm/nested.c
+++ b/arch/x86/kvm/svm/nested.c
@@ -19,6 +19,7 @@
 #include <linux/kernel.h>
 
 #include <asm/msr-index.h>
+#include <asm/debugreg.h>
 
 #include "kvm_emulate.h"
 #include "trace.h"
@@ -267,7 +268,8 @@ void enter_svm_guest_mode(struct vcpu_svm *svm, u64 vmcb_gpa,
 	svm->vmcb->save.rsp = nested_vmcb->save.rsp;
 	svm->vmcb->save.rip = nested_vmcb->save.rip;
 	svm->vmcb->save.dr7 = nested_vmcb->save.dr7;
-	svm->vmcb->save.dr6 = nested_vmcb->save.dr6;
+	svm->vcpu.arch.dr6  = nested_vmcb->save.dr6;
+	kvm_update_dr6(&svm->vcpu);
 	svm->vmcb->save.cpl = nested_vmcb->save.cpl;
 
 	svm->nested.vmcb_msrpm = nested_vmcb->control.msrpm_base_pa & ~0x0fffULL;
@@ -482,7 +484,7 @@ int nested_svm_vmexit(struct vcpu_svm *svm)
 	nested_vmcb->save.rsp    = vmcb->save.rsp;
 	nested_vmcb->save.rax    = vmcb->save.rax;
 	nested_vmcb->save.dr7    = vmcb->save.dr7;
-	nested_vmcb->save.dr6    = vmcb->save.dr6;
+	nested_vmcb->save.dr6    = svm->vcpu.arch.dr6;
 	nested_vmcb->save.cpl    = vmcb->save.cpl;
 
 	nested_vmcb->control.int_ctl           = vmcb->control.int_ctl;
@@ -606,7 +608,7 @@ static int nested_svm_exit_handled_msr(struct vcpu_svm *svm)
 /* DB exceptions for our internal use must not cause vmexit */
 static int nested_svm_intercept_db(struct vcpu_svm *svm)
 {
-	unsigned long dr6;
+	unsigned long dr6 = svm->vmcb->save.dr6;
 
 	/* Always catch it and pass it to userspace if debugging.  */
 	if (svm->vcpu.guest_debug &
@@ -615,22 +617,28 @@ static int nested_svm_intercept_db(struct vcpu_svm *svm)
 
 	/* if we're not singlestepping, it's not ours */
 	if (!svm->nmi_singlestep)
-		return NESTED_EXIT_DONE;
+		goto reflected_db;
 
 	/* if it's not a singlestep exception, it's not ours */
-	if (kvm_get_dr(&svm->vcpu, 6, &dr6))
-		return NESTED_EXIT_DONE;
 	if (!(dr6 & DR6_BS))
-		return NESTED_EXIT_DONE;
+		goto reflected_db;
 
 	/* if the guest is singlestepping, it should get the vmexit */
 	if (svm->nmi_singlestep_guest_rflags & X86_EFLAGS_TF) {
 		disable_nmi_singlestep(svm);
-		return NESTED_EXIT_DONE;
+		goto reflected_db;
 	}
 
 	/* it's ours, the nested hypervisor must not see this one */
 	return NESTED_EXIT_HOST;
+
+reflected_db:
+	/*
+	 * Synchronize guest DR6 here just like in db_interception; it will
+	 * be moved into the nested VMCB by nested_svm_vmexit.
+	 */
+	svm->vcpu.arch.dr6 = dr6;
+	return NESTED_EXIT_DONE;
 }
 
 static int nested_svm_intercept_ioio(struct vcpu_svm *svm)
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index 38f6aeefeb55..f03bffafd9e6 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -1672,11 +1672,6 @@ static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *sd)
 	mark_dirty(svm->vmcb, VMCB_ASID);
 }
 
-static u64 svm_get_dr6(struct kvm_vcpu *vcpu)
-{
-	return to_svm(vcpu)->vmcb->save.dr6;
-}
-
 static void svm_set_dr6(struct kvm_vcpu *vcpu, unsigned long value)
 {
 	struct vcpu_svm *svm = to_svm(vcpu);
@@ -1693,7 +1688,7 @@ static void svm_sync_dirty_debug_regs(struct kvm_vcpu *vcpu)
 	get_debugreg(vcpu->arch.db[1], 1);
 	get_debugreg(vcpu->arch.db[2], 2);
 	get_debugreg(vcpu->arch.db[3], 3);
-	vcpu->arch.dr6 = svm_get_dr6(vcpu);
+	vcpu->arch.dr6 = svm->vmcb->save.dr6;
 	vcpu->arch.dr7 = svm->vmcb->save.dr7;
 
 	vcpu->arch.switch_db_regs &= ~KVM_DEBUGREG_WONT_EXIT;
@@ -1739,6 +1734,7 @@ static int db_interception(struct vcpu_svm *svm)
 	if (!(svm->vcpu.guest_debug &
 	      (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP)) &&
 		!svm->nmi_singlestep) {
+		vcpu->arch.dr6 = svm->vmcb->save.dr6;
 		kvm_queue_exception(&svm->vcpu, DB_VECTOR);
 		return 1;
 	}
@@ -3931,7 +3927,6 @@ static struct kvm_x86_ops svm_x86_ops __initdata = {
 	.set_idt = svm_set_idt,
 	.get_gdt = svm_get_gdt,
 	.set_gdt = svm_set_gdt,
-	.get_dr6 = svm_get_dr6,
 	.set_dr6 = svm_set_dr6,
 	.set_dr7 = svm_set_dr7,
 	.sync_dirty_debug_regs = svm_sync_dirty_debug_regs,
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index 2384a2dbec44..6153a47109d3 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -4965,11 +4965,6 @@ static int handle_dr(struct kvm_vcpu *vcpu)
 	return kvm_skip_emulated_instruction(vcpu);
 }
 
-static u64 vmx_get_dr6(struct kvm_vcpu *vcpu)
-{
-	return vcpu->arch.dr6;
-}
-
 static void vmx_set_dr6(struct kvm_vcpu *vcpu, unsigned long val)
 {
 }
@@ -7736,7 +7731,6 @@ static struct kvm_x86_ops vmx_x86_ops __initdata = {
 	.set_idt = vmx_set_idt,
 	.get_gdt = vmx_get_gdt,
 	.set_gdt = vmx_set_gdt,
-	.get_dr6 = vmx_get_dr6,
 	.set_dr6 = vmx_set_dr6,
 	.set_dr7 = vmx_set_dr7,
 	.sync_dirty_debug_regs = vmx_sync_dirty_debug_regs,
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index f7628555f036..4ea644827b8a 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -104,7 +104,6 @@ static u64 __read_mostly cr4_reserved_bits = CR4_RESERVED_BITS;
                                     KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK)
 
 static void update_cr8_intercept(struct kvm_vcpu *vcpu);
-static void kvm_update_dr6(struct kvm_vcpu *vcpu);
 static void process_nmi(struct kvm_vcpu *vcpu);
 static void enter_smm(struct kvm_vcpu *vcpu);
 static void __kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags);
@@ -1048,7 +1047,7 @@ static void kvm_update_dr0123(struct kvm_vcpu *vcpu)
 	}
 }
 
-static void kvm_update_dr6(struct kvm_vcpu *vcpu)
+void kvm_update_dr6(struct kvm_vcpu *vcpu)
 {
 	if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP))
 		kvm_x86_ops.set_dr6(vcpu, vcpu->arch.dr6);
@@ -1129,10 +1128,7 @@ int kvm_get_dr(struct kvm_vcpu *vcpu, int dr, unsigned long *val)
 	case 4:
 		/* fall through */
 	case 6:
-		if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)
-			*val = vcpu->arch.dr6;
-		else
-			*val = kvm_x86_ops.get_dr6(vcpu);
+		*val = vcpu->arch.dr6;
 		break;
 	case 5:
 		/* fall through */
-- 
cgit v1.2.3


From d67668e9dd76d98136048935723947156737932b Mon Sep 17 00:00:00 2001
From: Paolo Bonzini <pbonzini@redhat.com>
Date: Wed, 6 May 2020 06:40:04 -0400
Subject: KVM: x86, SVM: isolate vcpu->arch.dr6 from vmcb->save.dr6

There are two issues with KVM_EXIT_DEBUG on AMD, whose root cause is the
different handling of DR6 on intercepted #DB exceptions on Intel and AMD.

On Intel, #DB exceptions transmit the DR6 value via the exit qualification
field of the VMCS, and the exit qualification only contains the description
of the precise event that caused a vmexit.

On AMD, instead the DR6 field of the VMCB is filled in as if the #DB exception
was to be injected into the guest.  This has two effects when guest debugging
is in use:

* the guest DR6 is clobbered

* the kvm_run->debug.arch.dr6 field can accumulate more debug events, rather
than just the last one that happened (the testcase in the next patch covers
this issue).

This patch fixes both issues by emulating, so to speak, the Intel behavior
on AMD processors.  The important observation is that (after the previous
patches) the VMCB value of DR6 is only ever observable from the guest is
KVM_DEBUGREG_WONT_EXIT is set.  Therefore we can actually set vmcb->save.dr6
to any value we want as long as KVM_DEBUGREG_WONT_EXIT is clear, which it
will be if guest debugging is enabled.

Therefore it is possible to enter the guest with an all-zero DR6,
reconstruct the #DB payload from the DR6 we get at exit time, and let
kvm_deliver_exception_payload move the newly set bits into vcpu->arch.dr6.
Some extra bits may be included in the payload if KVM_DEBUGREG_WONT_EXIT
is set, but this is harmless.

This may not be the most optimized way to deal with this, but it is
simple and, being confined within SVM code, it gets rid of the set_dr6
callback and kvm_update_dr6.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/include/asm/kvm_host.h |  2 --
 arch/x86/kvm/svm/nested.c       | 15 +++++++++++----
 arch/x86/kvm/svm/svm.c          | 29 +++++++++++++++++++++--------
 arch/x86/kvm/vmx/vmx.c          |  5 -----
 arch/x86/kvm/x86.c              | 11 -----------
 5 files changed, 32 insertions(+), 30 deletions(-)

(limited to 'arch')

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index de0c28814348..9e8263b1e6fe 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -1093,7 +1093,6 @@ struct kvm_x86_ops {
 	void (*set_idt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
 	void (*get_gdt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
 	void (*set_gdt)(struct kvm_vcpu *vcpu, struct desc_ptr *dt);
-	void (*set_dr6)(struct kvm_vcpu *vcpu, unsigned long value);
 	void (*sync_dirty_debug_regs)(struct kvm_vcpu *vcpu);
 	void (*set_dr7)(struct kvm_vcpu *vcpu, unsigned long value);
 	void (*cache_reg)(struct kvm_vcpu *vcpu, enum kvm_reg reg);
@@ -1623,7 +1622,6 @@ int kvm_pv_send_ipi(struct kvm *kvm, unsigned long ipi_bitmap_low,
 
 void kvm_define_shared_msr(unsigned index, u32 msr);
 int kvm_set_shared_msr(unsigned index, u64 val, u64 mask);
-void kvm_update_dr6(struct kvm_vcpu *vcpu);
 
 u64 kvm_scale_tsc(struct kvm_vcpu *vcpu, u64 tsc);
 u64 kvm_read_l1_tsc(struct kvm_vcpu *vcpu, u64 host_tsc);
diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c
index 9cfa8098995e..9a2a62e5afeb 100644
--- a/arch/x86/kvm/svm/nested.c
+++ b/arch/x86/kvm/svm/nested.c
@@ -269,7 +269,6 @@ void enter_svm_guest_mode(struct vcpu_svm *svm, u64 vmcb_gpa,
 	svm->vmcb->save.rip = nested_vmcb->save.rip;
 	svm->vmcb->save.dr7 = nested_vmcb->save.dr7;
 	svm->vcpu.arch.dr6  = nested_vmcb->save.dr6;
-	kvm_update_dr6(&svm->vcpu);
 	svm->vmcb->save.cpl = nested_vmcb->save.cpl;
 
 	svm->nested.vmcb_msrpm = nested_vmcb->control.msrpm_base_pa & ~0x0fffULL;
@@ -634,10 +633,18 @@ static int nested_svm_intercept_db(struct vcpu_svm *svm)
 
 reflected_db:
 	/*
-	 * Synchronize guest DR6 here just like in db_interception; it will
-	 * be moved into the nested VMCB by nested_svm_vmexit.
+	 * Synchronize guest DR6 here just like in kvm_deliver_exception_payload;
+	 * it will be moved into the nested VMCB by nested_svm_vmexit.  Once
+	 * exceptions will be moved to svm_check_nested_events, all this stuff
+	 * will just go away and we could just return NESTED_EXIT_HOST
+	 * unconditionally.  db_interception will queue the exception, which
+	 * will be processed by svm_check_nested_events if a nested vmexit is
+	 * required, and we will just use kvm_deliver_exception_payload to copy
+	 * the payload to DR6 before vmexit.
 	 */
-	svm->vcpu.arch.dr6 = dr6;
+	WARN_ON(svm->vcpu.arch.switch_db_regs & KVM_DEBUGREG_WONT_EXIT);
+	svm->vcpu.arch.dr6 &= ~(DR_TRAP_BITS | DR6_RTM);
+	svm->vcpu.arch.dr6 |= dr6 & ~DR6_FIXED_1;
 	return NESTED_EXIT_DONE;
 }
 
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index f03bffafd9e6..a862c768fd54 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -1672,12 +1672,14 @@ static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *sd)
 	mark_dirty(svm->vmcb, VMCB_ASID);
 }
 
-static void svm_set_dr6(struct kvm_vcpu *vcpu, unsigned long value)
+static void svm_set_dr6(struct vcpu_svm *svm, unsigned long value)
 {
-	struct vcpu_svm *svm = to_svm(vcpu);
+	struct vmcb *vmcb = svm->vmcb;
 
-	svm->vmcb->save.dr6 = value;
-	mark_dirty(svm->vmcb, VMCB_DR);
+	if (unlikely(value != vmcb->save.dr6)) {
+		vmcb->save.dr6 = value;
+		mark_dirty(vmcb, VMCB_DR);
+	}
 }
 
 static void svm_sync_dirty_debug_regs(struct kvm_vcpu *vcpu)
@@ -1688,9 +1690,12 @@ static void svm_sync_dirty_debug_regs(struct kvm_vcpu *vcpu)
 	get_debugreg(vcpu->arch.db[1], 1);
 	get_debugreg(vcpu->arch.db[2], 2);
 	get_debugreg(vcpu->arch.db[3], 3);
+	/*
+	 * We cannot reset svm->vmcb->save.dr6 to DR6_FIXED_1|DR6_RTM here,
+	 * because db_interception might need it.  We can do it before vmentry.
+	 */
 	vcpu->arch.dr6 = svm->vmcb->save.dr6;
 	vcpu->arch.dr7 = svm->vmcb->save.dr7;
-
 	vcpu->arch.switch_db_regs &= ~KVM_DEBUGREG_WONT_EXIT;
 	set_dr_intercepts(svm);
 }
@@ -1734,8 +1739,8 @@ static int db_interception(struct vcpu_svm *svm)
 	if (!(svm->vcpu.guest_debug &
 	      (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP)) &&
 		!svm->nmi_singlestep) {
-		vcpu->arch.dr6 = svm->vmcb->save.dr6;
-		kvm_queue_exception(&svm->vcpu, DB_VECTOR);
+		u32 payload = (svm->vmcb->save.dr6 ^ DR6_RTM) & ~DR6_FIXED_1;
+		kvm_queue_exception_p(&svm->vcpu, DB_VECTOR, payload);
 		return 1;
 	}
 
@@ -3313,6 +3318,15 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu)
 
 	svm->vmcb->save.cr2 = vcpu->arch.cr2;
 
+	/*
+	 * Run with all-zero DR6 unless needed, so that we can get the exact cause
+	 * of a #DB.
+	 */
+	if (unlikely(svm->vcpu.arch.switch_db_regs & KVM_DEBUGREG_WONT_EXIT))
+		svm_set_dr6(svm, vcpu->arch.dr6);
+	else
+		svm_set_dr6(svm, DR6_FIXED_1 | DR6_RTM);
+
 	clgi();
 	kvm_load_guest_xsave_state(vcpu);
 
@@ -3927,7 +3941,6 @@ static struct kvm_x86_ops svm_x86_ops __initdata = {
 	.set_idt = svm_set_idt,
 	.get_gdt = svm_get_gdt,
 	.set_gdt = svm_set_gdt,
-	.set_dr6 = svm_set_dr6,
 	.set_dr7 = svm_set_dr7,
 	.sync_dirty_debug_regs = svm_sync_dirty_debug_regs,
 	.cache_reg = svm_cache_reg,
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index 6153a47109d3..e2b71b0cdfce 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -4965,10 +4965,6 @@ static int handle_dr(struct kvm_vcpu *vcpu)
 	return kvm_skip_emulated_instruction(vcpu);
 }
 
-static void vmx_set_dr6(struct kvm_vcpu *vcpu, unsigned long val)
-{
-}
-
 static void vmx_sync_dirty_debug_regs(struct kvm_vcpu *vcpu)
 {
 	get_debugreg(vcpu->arch.db[0], 0);
@@ -7731,7 +7727,6 @@ static struct kvm_x86_ops vmx_x86_ops __initdata = {
 	.set_idt = vmx_set_idt,
 	.get_gdt = vmx_get_gdt,
 	.set_gdt = vmx_set_gdt,
-	.set_dr6 = vmx_set_dr6,
 	.set_dr7 = vmx_set_dr7,
 	.sync_dirty_debug_regs = vmx_sync_dirty_debug_regs,
 	.cache_reg = vmx_cache_reg,
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 4ea644827b8a..f780af601c5f 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -473,7 +473,6 @@ void kvm_deliver_exception_payload(struct kvm_vcpu *vcpu)
 		 * breakpoint), it is reserved and must be zero in DR6.
 		 */
 		vcpu->arch.dr6 &= ~BIT(12);
-		kvm_update_dr6(vcpu);
 		break;
 	case PF_VECTOR:
 		vcpu->arch.cr2 = payload;
@@ -1047,12 +1046,6 @@ static void kvm_update_dr0123(struct kvm_vcpu *vcpu)
 	}
 }
 
-void kvm_update_dr6(struct kvm_vcpu *vcpu)
-{
-	if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP))
-		kvm_x86_ops.set_dr6(vcpu, vcpu->arch.dr6);
-}
-
 static void kvm_update_dr7(struct kvm_vcpu *vcpu)
 {
 	unsigned long dr7;
@@ -1092,7 +1085,6 @@ static int __kvm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long val)
 		if (val & 0xffffffff00000000ULL)
 			return -1; /* #GP */
 		vcpu->arch.dr6 = (val & DR6_VOLATILE) | kvm_dr6_fixed(vcpu);
-		kvm_update_dr6(vcpu);
 		break;
 	case 5:
 		/* fall through */
@@ -4008,7 +4000,6 @@ static int kvm_vcpu_ioctl_x86_set_debugregs(struct kvm_vcpu *vcpu,
 	memcpy(vcpu->arch.db, dbgregs->db, sizeof(vcpu->arch.db));
 	kvm_update_dr0123(vcpu);
 	vcpu->arch.dr6 = dbgregs->dr6;
-	kvm_update_dr6(vcpu);
 	vcpu->arch.dr7 = dbgregs->dr7;
 	kvm_update_dr7(vcpu);
 
@@ -8417,7 +8408,6 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 		WARN_ON(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP);
 		kvm_x86_ops.sync_dirty_debug_regs(vcpu);
 		kvm_update_dr0123(vcpu);
-		kvm_update_dr6(vcpu);
 		kvm_update_dr7(vcpu);
 		vcpu->arch.switch_db_regs &= ~KVM_DEBUGREG_RELOAD;
 	}
@@ -9478,7 +9468,6 @@ void kvm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event)
 	memset(vcpu->arch.db, 0, sizeof(vcpu->arch.db));
 	kvm_update_dr0123(vcpu);
 	vcpu->arch.dr6 = DR6_INIT;
-	kvm_update_dr6(vcpu);
 	vcpu->arch.dr7 = DR7_FIXED_1;
 	kvm_update_dr7(vcpu);
 
-- 
cgit v1.2.3


From 45981dedf555bd66d86ccb51e4049d9f7ec45bd5 Mon Sep 17 00:00:00 2001
From: Paolo Bonzini <pbonzini@redhat.com>
Date: Wed, 6 May 2020 05:59:39 -0400
Subject: KVM: VMX: pass correct DR6 for GD userspace exit

When KVM_EXIT_DEBUG is raised for the disabled-breakpoints case (DR7.GD),
DR6 was incorrectly copied from the value in the VM.  Instead,
DR6.BD should be set in order to catch this case.

On AMD this does not need any special code because the processor triggers
a #DB exception that is intercepted.  However, the testcase would fail
without the previous patch because both DR6.BS and DR6.BD would be set.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/kvm/vmx/vmx.c                          |  2 +-
 tools/testing/selftests/kvm/x86_64/debug_regs.c | 24 +++++++++++++++++++++++-
 2 files changed, 24 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index e2b71b0cdfce..e45cf89c5821 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -4927,7 +4927,7 @@ static int handle_dr(struct kvm_vcpu *vcpu)
 		 * guest debugging itself.
 		 */
 		if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) {
-			vcpu->run->debug.arch.dr6 = vcpu->arch.dr6;
+			vcpu->run->debug.arch.dr6 = DR6_BD | DR6_RTM | DR6_FIXED_1;
 			vcpu->run->debug.arch.dr7 = dr7;
 			vcpu->run->debug.arch.pc = kvm_get_linear_rip(vcpu);
 			vcpu->run->debug.arch.exception = DB_VECTOR;
diff --git a/tools/testing/selftests/kvm/x86_64/debug_regs.c b/tools/testing/selftests/kvm/x86_64/debug_regs.c
index 077f25d61d1a..8162c58a1234 100644
--- a/tools/testing/selftests/kvm/x86_64/debug_regs.c
+++ b/tools/testing/selftests/kvm/x86_64/debug_regs.c
@@ -11,10 +11,13 @@
 
 #define VCPU_ID 0
 
+#define DR6_BD		(1 << 13)
+#define DR7_GD		(1 << 13)
+
 /* For testing data access debug BP */
 uint32_t guest_value;
 
-extern unsigned char sw_bp, hw_bp, write_data, ss_start;
+extern unsigned char sw_bp, hw_bp, write_data, ss_start, bd_start;
 
 static void guest_code(void)
 {
@@ -43,6 +46,8 @@ static void guest_code(void)
 		     "rdmsr\n\t"
 		     : : : "rax", "ecx");
 
+	/* DR6.BD test */
+	asm volatile("bd_start: mov %%dr0, %%rax" : : : "rax");
 	GUEST_DONE();
 }
 
@@ -165,6 +170,23 @@ int main(void)
 			    target_dr6);
 	}
 
+	/* Finally test global disable */
+	CLEAR_DEBUG();
+	debug.control = KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_HW_BP;
+	debug.arch.debugreg[7] = 0x400 | DR7_GD;
+	APPLY_DEBUG();
+	vcpu_run(vm, VCPU_ID);
+	target_dr6 = 0xffff0ff0 | DR6_BD;
+	TEST_ASSERT(run->exit_reason == KVM_EXIT_DEBUG &&
+		    run->debug.arch.exception == DB_VECTOR &&
+		    run->debug.arch.pc == CAST_TO_RIP(bd_start) &&
+		    run->debug.arch.dr6 == target_dr6,
+			    "DR7.GD: exit %d exception %d rip 0x%llx "
+			    "(should be 0x%llx) dr6 0x%llx (should be 0x%llx)",
+			    run->exit_reason, run->debug.arch.exception,
+			    run->debug.arch.pc, target_rip, run->debug.arch.dr6,
+			    target_dr6);
+
 	/* Disable all debug controls, run to the end */
 	CLEAR_DEBUG();
 	APPLY_DEBUG();
-- 
cgit v1.2.3


From 54163a346d4a0a1b93f2ff6dc1f488419a605fa9 Mon Sep 17 00:00:00 2001
From: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Date: Wed, 6 May 2020 08:17:53 -0500
Subject: KVM: Introduce kvm_make_all_cpus_request_except()

This allows making request to all other vcpus except the one
specified in the parameter.

Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Message-Id: <1588771076-73790-2-git-send-email-suravee.suthikulpanit@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/kvm/hyperv.c    |  2 +-
 arch/x86/kvm/x86.c       |  2 +-
 include/linux/kvm_host.h |  3 +++
 virt/kvm/kvm_main.c      | 14 +++++++++++---
 4 files changed, 16 insertions(+), 5 deletions(-)

(limited to 'arch')

diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
index bcefa9d4e57e..54d4b98b49e1 100644
--- a/arch/x86/kvm/hyperv.c
+++ b/arch/x86/kvm/hyperv.c
@@ -1427,7 +1427,7 @@ static u64 kvm_hv_flush_tlb(struct kvm_vcpu *current_vcpu, u64 ingpa,
 	 */
 	kvm_make_vcpus_request_mask(kvm,
 				    KVM_REQ_TLB_FLUSH | KVM_REQUEST_NO_WAKEUP,
-				    vcpu_mask, &hv_vcpu->tlb_flush);
+				    NULL, vcpu_mask, &hv_vcpu->tlb_flush);
 
 ret_success:
 	/* We always do full TLB flush, set rep_done = rep_cnt. */
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index f780af601c5f..ba8edf3b89f6 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -8030,7 +8030,7 @@ void kvm_make_scan_ioapic_request_mask(struct kvm *kvm,
 	zalloc_cpumask_var(&cpus, GFP_ATOMIC);
 
 	kvm_make_vcpus_request_mask(kvm, KVM_REQ_SCAN_IOAPIC,
-				    vcpu_bitmap, cpus);
+				    NULL, vcpu_bitmap, cpus);
 
 	free_cpumask_var(cpus);
 }
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 01276e3d01b9..131cc1527d68 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -813,8 +813,11 @@ void kvm_flush_remote_tlbs(struct kvm *kvm);
 void kvm_reload_remote_mmus(struct kvm *kvm);
 
 bool kvm_make_vcpus_request_mask(struct kvm *kvm, unsigned int req,
+				 struct kvm_vcpu *except,
 				 unsigned long *vcpu_bitmap, cpumask_var_t tmp);
 bool kvm_make_all_cpus_request(struct kvm *kvm, unsigned int req);
+bool kvm_make_all_cpus_request_except(struct kvm *kvm, unsigned int req,
+				      struct kvm_vcpu *except);
 bool kvm_make_cpus_request_mask(struct kvm *kvm, unsigned int req,
 				unsigned long *vcpu_bitmap);
 
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 74bdb7bf3295..731c1e517716 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -259,6 +259,7 @@ static inline bool kvm_kick_many_cpus(const struct cpumask *cpus, bool wait)
 }
 
 bool kvm_make_vcpus_request_mask(struct kvm *kvm, unsigned int req,
+				 struct kvm_vcpu *except,
 				 unsigned long *vcpu_bitmap, cpumask_var_t tmp)
 {
 	int i, cpu, me;
@@ -268,7 +269,8 @@ bool kvm_make_vcpus_request_mask(struct kvm *kvm, unsigned int req,
 	me = get_cpu();
 
 	kvm_for_each_vcpu(i, vcpu, kvm) {
-		if (vcpu_bitmap && !test_bit(i, vcpu_bitmap))
+		if ((vcpu_bitmap && !test_bit(i, vcpu_bitmap)) ||
+		    vcpu == except)
 			continue;
 
 		kvm_make_request(req, vcpu);
@@ -288,19 +290,25 @@ bool kvm_make_vcpus_request_mask(struct kvm *kvm, unsigned int req,
 	return called;
 }
 
-bool kvm_make_all_cpus_request(struct kvm *kvm, unsigned int req)
+bool kvm_make_all_cpus_request_except(struct kvm *kvm, unsigned int req,
+				      struct kvm_vcpu *except)
 {
 	cpumask_var_t cpus;
 	bool called;
 
 	zalloc_cpumask_var(&cpus, GFP_ATOMIC);
 
-	called = kvm_make_vcpus_request_mask(kvm, req, NULL, cpus);
+	called = kvm_make_vcpus_request_mask(kvm, req, except, NULL, cpus);
 
 	free_cpumask_var(cpus);
 	return called;
 }
 
+bool kvm_make_all_cpus_request(struct kvm *kvm, unsigned int req)
+{
+	return kvm_make_all_cpus_request_except(kvm, req, NULL);
+}
+
 #ifndef CONFIG_HAVE_KVM_ARCH_TLB_FLUSH_ALL
 void kvm_flush_remote_tlbs(struct kvm *kvm)
 {
-- 
cgit v1.2.3


From 7d611233b01613c866259d6e6f44c67f7f7eb2a3 Mon Sep 17 00:00:00 2001
From: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Date: Wed, 6 May 2020 21:35:39 -0500
Subject: KVM: SVM: Disable AVIC before setting V_IRQ

The commit 64b5bd270426 ("KVM: nSVM: ignore L1 interrupt window
while running L2 with V_INTR_MASKING=1") introduced a WARN_ON,
which checks if AVIC is enabled when trying to set V_IRQ
in the VMCB for enabling irq window.

The following warning is triggered because the requesting vcpu
(to deactivate AVIC) does not get to process APICv update request
for itself until the next #vmexit.

WARNING: CPU: 0 PID: 118232 at arch/x86/kvm/svm/svm.c:1372 enable_irq_window+0x6a/0xa0 [kvm_amd]
 RIP: 0010:enable_irq_window+0x6a/0xa0 [kvm_amd]
 Call Trace:
  kvm_arch_vcpu_ioctl_run+0x6e3/0x1b50 [kvm]
  ? kvm_vm_ioctl_irq_line+0x27/0x40 [kvm]
  ? _copy_to_user+0x26/0x30
  ? kvm_vm_ioctl+0xb3e/0xd90 [kvm]
  ? set_next_entity+0x78/0xc0
  kvm_vcpu_ioctl+0x236/0x610 [kvm]
  ksys_ioctl+0x8a/0xc0
  __x64_sys_ioctl+0x1a/0x20
  do_syscall_64+0x58/0x210
  entry_SYSCALL_64_after_hwframe+0x44/0xa9

Fixes by sending APICV update request to all other vcpus, and
immediately update APIC for itself.

Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Link: https://lkml.org/lkml/2020/5/2/167
Fixes: 64b5bd270426 ("KVM: nSVM: ignore L1 interrupt window while running L2 with V_INTR_MASKING=1")
Message-Id: <1588818939-54264-1-git-send-email-suravee.suthikulpanit@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/kvm/x86.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index ba8edf3b89f6..98176b80c481 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -8060,6 +8060,7 @@ EXPORT_SYMBOL_GPL(kvm_vcpu_update_apicv);
  */
 void kvm_request_apicv_update(struct kvm *kvm, bool activate, ulong bit)
 {
+	struct kvm_vcpu *except;
 	unsigned long old, new, expected;
 
 	if (!kvm_x86_ops.check_apicv_inhibit_reasons ||
@@ -8084,7 +8085,17 @@ void kvm_request_apicv_update(struct kvm *kvm, bool activate, ulong bit)
 	trace_kvm_apicv_update_request(activate, bit);
 	if (kvm_x86_ops.pre_update_apicv_exec_ctrl)
 		kvm_x86_ops.pre_update_apicv_exec_ctrl(kvm, activate);
-	kvm_make_all_cpus_request(kvm, KVM_REQ_APICV_UPDATE);
+
+	/*
+	 * Sending request to update APICV for all other vcpus,
+	 * while update the calling vcpu immediately instead of
+	 * waiting for another #VMEXIT to handle the request.
+	 */
+	except = kvm_get_running_vcpu();
+	kvm_make_all_cpus_request_except(kvm, KVM_REQ_APICV_UPDATE,
+					 except);
+	if (except)
+		kvm_vcpu_update_apicv(except);
 }
 EXPORT_SYMBOL_GPL(kvm_request_apicv_update);
 
-- 
cgit v1.2.3


From 37486135d3a7b03acc7755b63627a130437f066a Mon Sep 17 00:00:00 2001
From: Babu Moger <babu.moger@amd.com>
Date: Tue, 12 May 2020 18:59:06 -0500
Subject: KVM: x86: Fix pkru save/restore when guest CR4.PKE=0, move it to
 x86.c

Though rdpkru and wrpkru are contingent upon CR4.PKE, the PKRU
resource isn't. It can be read with XSAVE and written with XRSTOR.
So, if we don't set the guest PKRU value here(kvm_load_guest_xsave_state),
the guest can read the host value.

In case of kvm_load_host_xsave_state, guest with CR4.PKE clear could
potentially use XRSTOR to change the host PKRU value.

While at it, move pkru state save/restore to common code and the
host_pkru field to kvm_vcpu_arch.  This will let SVM support protection keys.

Cc: stable@vger.kernel.org
Reported-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Babu Moger <babu.moger@amd.com>
Message-Id: <158932794619.44260.14508381096663848853.stgit@naples-babu.amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/include/asm/kvm_host.h |  1 +
 arch/x86/kvm/vmx/vmx.c          | 18 ------------------
 arch/x86/kvm/x86.c              | 17 +++++++++++++++++
 3 files changed, 18 insertions(+), 18 deletions(-)

(limited to 'arch')

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 9e8263b1e6fe..0a6b35353fc7 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -578,6 +578,7 @@ struct kvm_vcpu_arch {
 	unsigned long cr4;
 	unsigned long cr4_guest_owned_bits;
 	unsigned long cr8;
+	u32 host_pkru;
 	u32 pkru;
 	u32 hflags;
 	u64 efer;
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index e45cf89c5821..89c766fad889 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -1372,7 +1372,6 @@ void vmx_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 
 	vmx_vcpu_pi_load(vcpu, cpu);
 
-	vmx->host_pkru = read_pkru();
 	vmx->host_debugctlmsr = get_debugctlmsr();
 }
 
@@ -6564,11 +6563,6 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu)
 
 	kvm_load_guest_xsave_state(vcpu);
 
-	if (static_cpu_has(X86_FEATURE_PKU) &&
-	    kvm_read_cr4_bits(vcpu, X86_CR4_PKE) &&
-	    vcpu->arch.pkru != vmx->host_pkru)
-		__write_pkru(vcpu->arch.pkru);
-
 	pt_guest_enter(vmx);
 
 	if (vcpu_to_pmu(vcpu)->version)
@@ -6658,18 +6652,6 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu)
 
 	pt_guest_exit(vmx);
 
-	/*
-	 * eager fpu is enabled if PKEY is supported and CR4 is switched
-	 * back on host, so it is safe to read guest PKRU from current
-	 * XSAVE.
-	 */
-	if (static_cpu_has(X86_FEATURE_PKU) &&
-	    kvm_read_cr4_bits(vcpu, X86_CR4_PKE)) {
-		vcpu->arch.pkru = rdpkru();
-		if (vcpu->arch.pkru != vmx->host_pkru)
-			__write_pkru(vmx->host_pkru);
-	}
-
 	kvm_load_host_xsave_state(vcpu);
 
 	vmx->nested.nested_run_pending = 0;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 98176b80c481..d11eba8b85c6 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -837,11 +837,25 @@ void kvm_load_guest_xsave_state(struct kvm_vcpu *vcpu)
 		    vcpu->arch.ia32_xss != host_xss)
 			wrmsrl(MSR_IA32_XSS, vcpu->arch.ia32_xss);
 	}
+
+	if (static_cpu_has(X86_FEATURE_PKU) &&
+	    (kvm_read_cr4_bits(vcpu, X86_CR4_PKE) ||
+	     (vcpu->arch.xcr0 & XFEATURE_MASK_PKRU)) &&
+	    vcpu->arch.pkru != vcpu->arch.host_pkru)
+		__write_pkru(vcpu->arch.pkru);
 }
 EXPORT_SYMBOL_GPL(kvm_load_guest_xsave_state);
 
 void kvm_load_host_xsave_state(struct kvm_vcpu *vcpu)
 {
+	if (static_cpu_has(X86_FEATURE_PKU) &&
+	    (kvm_read_cr4_bits(vcpu, X86_CR4_PKE) ||
+	     (vcpu->arch.xcr0 & XFEATURE_MASK_PKRU))) {
+		vcpu->arch.pkru = rdpkru();
+		if (vcpu->arch.pkru != vcpu->arch.host_pkru)
+			__write_pkru(vcpu->arch.host_pkru);
+	}
+
 	if (kvm_read_cr4_bits(vcpu, X86_CR4_OSXSAVE)) {
 
 		if (vcpu->arch.xcr0 != host_xcr0)
@@ -3549,6 +3563,9 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 
 	kvm_x86_ops.vcpu_load(vcpu, cpu);
 
+	/* Save host pkru register if supported */
+	vcpu->arch.host_pkru = read_pkru();
+
 	/* Apply any externally detected TSC adjustments (due to suspend) */
 	if (unlikely(vcpu->arch.tsc_offset_adjustment)) {
 		adjust_tsc_offset_host(vcpu, vcpu->arch.tsc_offset_adjustment);
-- 
cgit v1.2.3


From c4e0e4ab4cf3ec2b3f0b628ead108d677644ebd9 Mon Sep 17 00:00:00 2001
From: Jim Mattson <jmattson@google.com>
Date: Mon, 11 May 2020 15:56:16 -0700
Subject: KVM: x86: Fix off-by-one error in kvm_vcpu_ioctl_x86_setup_mce

Bank_num is a one-based count of banks, not a zero-based index. It
overflows the allocated space only when strictly greater than
KVM_MAX_MCE_BANKS.

Fixes: a9e38c3e01ad ("KVM: x86: Catch potential overrun in MCE setup")
Signed-off-by: Jue Wang <juew@google.com>
Signed-off-by: Jim Mattson <jmattson@google.com>
Reviewed-by: Peter Shier <pshier@google.com>
Message-Id: <20200511225616.19557-1-jmattson@google.com>
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 arch/x86/kvm/x86.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index d11eba8b85c6..c17e6eb9ad43 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3759,7 +3759,7 @@ static int kvm_vcpu_ioctl_x86_setup_mce(struct kvm_vcpu *vcpu,
 	unsigned bank_num = mcg_cap & 0xff, bank;
 
 	r = -EINVAL;
-	if (!bank_num || bank_num >= KVM_MAX_MCE_BANKS)
+	if (!bank_num || bank_num > KVM_MAX_MCE_BANKS)
 		goto out;
 	if (mcg_cap & ~(kvm_mce_cap_supported | 0xff | 0xff0000))
 		goto out;
-- 
cgit v1.2.3