Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull more x86 kvm fixes from Paolo Bonzini:

 - Cache coherency fix for SEV live migration

 - Fix for instruction emulation with PKU

 - fixes for rare delaying of interrupt delivery

 - fix for SEV-ES buffer overflow

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: SEV-ES: go over the sev_pio_data buffer in multiple passes if needed
  KVM: SEV-ES: keep INS functions together
  KVM: x86: remove unnecessary arguments from complete_emulator_pio_in
  KVM: x86: split the two parts of emulator_pio_in
  KVM: SEV-ES: clean up kvm_sev_es_ins/outs
  KVM: x86: leave vcpu->arch.pio.count alone in emulator_pio_in_out
  KVM: SEV-ES: rename guest_ins_data to sev_pio_data
  KVM: SEV: Flush cache on non-coherent systems before RECEIVE_UPDATE_DATA
  KVM: MMU: Reset mmu->pkru_mask to avoid stale data
  KVM: nVMX: promptly process interrupts delivered while in guest mode
  KVM: x86: check for interrupts before deciding whether to exit the fast path

Author: torvalds

arch/x86/include/asm/kvm_host.h

@@ -702,7 +702,8 @@ struct kvm_vcpu_arch {
 
 	struct kvm_pio_request pio;
 	void *pio_data;
-	void *guest_ins_data;
+	void *sev_pio_data;
+	unsigned sev_pio_count;
 
 	u8 event_exit_inst_len;
 

arch/x86/kvm/mmu/mmu.c

@@ -4596,10 +4596,10 @@ static void update_pkru_bitmask(struct kvm_mmu *mmu)
 	unsigned bit;
 	bool wp;
 
-	if (!is_cr4_pke(mmu)) {
-		mmu->pkru_mask = 0;
+	mmu->pkru_mask = 0;
+
+	if (!is_cr4_pke(mmu))
 		return;
-	}
 
 	wp = is_cr0_wp(mmu);
 

arch/x86/kvm/svm/sev.c

@@ -1484,6 +1484,13 @@ static int sev_receive_update_data(struct kvm *kvm, struct kvm_sev_cmd *argp)
 		goto e_free_trans;
 	}
 
+	/*
+	 * Flush (on non-coherent CPUs) before RECEIVE_UPDATE_DATA, the PSP
+	 * encrypts the written data with the guest's key, and the cache may
+	 * contain dirty, unencrypted data.
+	 */
+	sev_clflush_pages(guest_page, n);
+
 	/* The RECEIVE_UPDATE_DATA command requires C-bit to be always set. */
 	data.guest_address = (page_to_pfn(guest_page[0]) << PAGE_SHIFT) + offset;
 	data.guest_address |= sev_me_mask;

arch/x86/kvm/vmx/vmx.c

@@ -6305,18 +6305,13 @@ static int vmx_sync_pir_to_irr(struct kvm_vcpu *vcpu)
 
 		/*
 		 * If we are running L2 and L1 has a new pending interrupt
-		 * which can be injected, we should re-evaluate
-		 * what should be done with this new L1 interrupt.
-		 * If L1 intercepts external-interrupts, we should
-		 * exit from L2 to L1. Otherwise, interrupt should be
-		 * delivered directly to L2.
+		 * which can be injected, this may cause a vmexit or it may
+		 * be injected into L2.  Either way, this interrupt will be
+		 * processed via KVM_REQ_EVENT, not RVI, because we do not use
+		 * virtual interrupt delivery to inject L1 interrupts into L2.
 		 */
-		if (is_guest_mode(vcpu) && max_irr_updated) {
-			if (nested_exit_on_intr(vcpu))
-				kvm_vcpu_exiting_guest_mode(vcpu);
-			else
-				kvm_make_request(KVM_REQ_EVENT, vcpu);
-		}
+		if (is_guest_mode(vcpu) && max_irr_updated)
+			kvm_make_request(KVM_REQ_EVENT, vcpu);
 	} else {
 		max_irr = kvm_lapic_find_highest_irr(vcpu);
 	}

arch/x86/kvm/x86.c

@@ -6906,18 +6906,16 @@ static int kernel_pio(struct kvm_vcpu *vcpu, void *pd)
 }
 
 static int emulator_pio_in_out(struct kvm_vcpu *vcpu, int size,
-			       unsigned short port, void *val,
+			       unsigned short port,
 			       unsigned int count, bool in)
 {
 	vcpu->arch.pio.port = port;
 	vcpu->arch.pio.in = in;
 	vcpu->arch.pio.count  = count;
 	vcpu->arch.pio.size = size;
 
-	if (!kernel_pio(vcpu, vcpu->arch.pio_data)) {
-		vcpu->arch.pio.count = 0;
+	if (!kernel_pio(vcpu, vcpu->arch.pio_data))
 		return 1;
-	}
 
 	vcpu->run->exit_reason = KVM_EXIT_IO;
 	vcpu->run->io.direction = in ? KVM_EXIT_IO_IN : KVM_EXIT_IO_OUT;
@@ -6929,26 +6927,39 @@ static int emulator_pio_in_out(struct kvm_vcpu *vcpu, int size,
 	return 0;
 }
 
-static int emulator_pio_in(struct kvm_vcpu *vcpu, int size,
-			   unsigned short port, void *val, unsigned int count)
+static int __emulator_pio_in(struct kvm_vcpu *vcpu, int size,
+			     unsigned short port, unsigned int count)
 {
-	int ret;
+	WARN_ON(vcpu->arch.pio.count);
+	memset(vcpu->arch.pio_data, 0, size * count);
+	return emulator_pio_in_out(vcpu, size, port, count, true);
+}
 
-	if (vcpu->arch.pio.count)
-		goto data_avail;
+static void complete_emulator_pio_in(struct kvm_vcpu *vcpu, void *val)
+{
+	int size = vcpu->arch.pio.size;
+	unsigned count = vcpu->arch.pio.count;
+	memcpy(val, vcpu->arch.pio_data, size * count);
+	trace_kvm_pio(KVM_PIO_IN, vcpu->arch.pio.port, size, count, vcpu->arch.pio_data);
+	vcpu->arch.pio.count = 0;
+}
 
-	memset(vcpu->arch.pio_data, 0, size * count);
+static int emulator_pio_in(struct kvm_vcpu *vcpu, int size,
+			   unsigned short port, void *val, unsigned int count)
+{
+	if (vcpu->arch.pio.count) {
+		/* Complete previous iteration.  */
+	} else {
+		int r = __emulator_pio_in(vcpu, size, port, count);
+		if (!r)
+			return r;
 
-	ret = emulator_pio_in_out(vcpu, size, port, val, count, true);
-	if (ret) {
-data_avail:
-		memcpy(val, vcpu->arch.pio_data, size * count);
-		trace_kvm_pio(KVM_PIO_IN, port, size, count, vcpu->arch.pio_data);
-		vcpu->arch.pio.count = 0;
-		return 1;
+		/* Results already available, fall through.  */
 	}
 
-	return 0;
+	WARN_ON(count != vcpu->arch.pio.count);
+	complete_emulator_pio_in(vcpu, val);
+	return 1;
 }
 
 static int emulator_pio_in_emulated(struct x86_emulate_ctxt *ctxt,
@@ -6963,9 +6974,15 @@ static int emulator_pio_out(struct kvm_vcpu *vcpu, int size,
 			    unsigned short port, const void *val,
 			    unsigned int count)
 {
+	int ret;
+
 	memcpy(vcpu->arch.pio_data, val, size * count);
 	trace_kvm_pio(KVM_PIO_OUT, port, size, count, vcpu->arch.pio_data);
-	return emulator_pio_in_out(vcpu, size, port, (void *)val, count, false);
+	ret = emulator_pio_in_out(vcpu, size, port, count, false);
+	if (ret)
+                vcpu->arch.pio.count = 0;
+
+        return ret;
 }
 
 static int emulator_pio_out_emulated(struct x86_emulate_ctxt *ctxt,
@@ -9643,14 +9660,14 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 		if (likely(exit_fastpath != EXIT_FASTPATH_REENTER_GUEST))
 			break;
 
-                if (unlikely(kvm_vcpu_exit_request(vcpu))) {
+		if (vcpu->arch.apicv_active)
+			static_call(kvm_x86_sync_pir_to_irr)(vcpu);
+
+		if (unlikely(kvm_vcpu_exit_request(vcpu))) {
 			exit_fastpath = EXIT_FASTPATH_EXIT_HANDLED;
 			break;
 		}
-
-		if (vcpu->arch.apicv_active)
-			static_call(kvm_x86_sync_pir_to_irr)(vcpu);
-        }
+	}
 
 	/*
 	 * Do this here before restoring debug registers on the host.  And
@@ -12368,53 +12385,92 @@ int kvm_sev_es_mmio_read(struct kvm_vcpu *vcpu, gpa_t gpa, unsigned int bytes,
 }
 EXPORT_SYMBOL_GPL(kvm_sev_es_mmio_read);
 
-static int complete_sev_es_emulated_ins(struct kvm_vcpu *vcpu)
+static int kvm_sev_es_outs(struct kvm_vcpu *vcpu, unsigned int size,
+			   unsigned int port);
+
+static int complete_sev_es_emulated_outs(struct kvm_vcpu *vcpu)
 {
-	memcpy(vcpu->arch.guest_ins_data, vcpu->arch.pio_data,
-	       vcpu->arch.pio.count * vcpu->arch.pio.size);
-	vcpu->arch.pio.count = 0;
+	int size = vcpu->arch.pio.size;
+	int port = vcpu->arch.pio.port;
 
+	vcpu->arch.pio.count = 0;
+	if (vcpu->arch.sev_pio_count)
+		return kvm_sev_es_outs(vcpu, size, port);
 	return 1;
 }
 
 static int kvm_sev_es_outs(struct kvm_vcpu *vcpu, unsigned int size,
-			   unsigned int port, void *data,  unsigned int count)
+			   unsigned int port)
 {
-	int ret;
-
-	ret = emulator_pio_out_emulated(vcpu->arch.emulate_ctxt, size, port,
-					data, count);
-	if (ret)
-		return ret;
+	for (;;) {
+		unsigned int count =
+			min_t(unsigned int, PAGE_SIZE / size, vcpu->arch.sev_pio_count);
+		int ret = emulator_pio_out(vcpu, size, port, vcpu->arch.sev_pio_data, count);
+
+		/* memcpy done already by emulator_pio_out.  */
+		vcpu->arch.sev_pio_count -= count;
+		vcpu->arch.sev_pio_data += count * vcpu->arch.pio.size;
+		if (!ret)
+			break;
 
-	vcpu->arch.pio.count = 0;
+		/* Emulation done by the kernel.  */
+		if (!vcpu->arch.sev_pio_count)
+			return 1;
+	}
 
+	vcpu->arch.complete_userspace_io = complete_sev_es_emulated_outs;
 	return 0;
 }
 
 static int kvm_sev_es_ins(struct kvm_vcpu *vcpu, unsigned int size,
-			  unsigned int port, void *data, unsigned int count)
+			  unsigned int port);
+
+static void advance_sev_es_emulated_ins(struct kvm_vcpu *vcpu)
 {
-	int ret;
+	unsigned count = vcpu->arch.pio.count;
+	complete_emulator_pio_in(vcpu, vcpu->arch.sev_pio_data);
+	vcpu->arch.sev_pio_count -= count;
+	vcpu->arch.sev_pio_data += count * vcpu->arch.pio.size;
+}
 
-	ret = emulator_pio_in_emulated(vcpu->arch.emulate_ctxt, size, port,
-				       data, count);
-	if (ret) {
-		vcpu->arch.pio.count = 0;
-	} else {
-		vcpu->arch.guest_ins_data = data;
-		vcpu->arch.complete_userspace_io = complete_sev_es_emulated_ins;
+static int complete_sev_es_emulated_ins(struct kvm_vcpu *vcpu)
+{
+	int size = vcpu->arch.pio.size;
+	int port = vcpu->arch.pio.port;
+
+	advance_sev_es_emulated_ins(vcpu);
+	if (vcpu->arch.sev_pio_count)
+		return kvm_sev_es_ins(vcpu, size, port);
+	return 1;
+}
+
+static int kvm_sev_es_ins(struct kvm_vcpu *vcpu, unsigned int size,
+			  unsigned int port)
+{
+	for (;;) {
+		unsigned int count =
+			min_t(unsigned int, PAGE_SIZE / size, vcpu->arch.sev_pio_count);
+		if (!__emulator_pio_in(vcpu, size, port, count))
+			break;
+
+		/* Emulation done by the kernel.  */
+		advance_sev_es_emulated_ins(vcpu);
+		if (!vcpu->arch.sev_pio_count)
+			return 1;
 	}
 
+	vcpu->arch.complete_userspace_io = complete_sev_es_emulated_ins;
 	return 0;
 }
 
 int kvm_sev_es_string_io(struct kvm_vcpu *vcpu, unsigned int size,
 			 unsigned int port, void *data,  unsigned int count,
 			 int in)
 {
-	return in ? kvm_sev_es_ins(vcpu, size, port, data, count)
-		  : kvm_sev_es_outs(vcpu, size, port, data, count);
+	vcpu->arch.sev_pio_data = data;
+	vcpu->arch.sev_pio_count = count;
+	return in ? kvm_sev_es_ins(vcpu, size, port)
+		  : kvm_sev_es_outs(vcpu, size, port);
 }
 EXPORT_SYMBOL_GPL(kvm_sev_es_string_io);