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

Pull KVM fixes from Radim Krčmář:
 "MIPS:
   - Fix build with KVM, DYNAMIC_DEBUG and JUMP_LABEL.

  PPC:
   - Fix host crashes/hangs on POWER9.
   - Properly restore userspace state after KVM_RUN ioctl.

  s390:
   - Fix address translation in odd-ball cases (real-space designation
     ASCEs).

  x86:
   - Fix privilege escalation in 64-bit Windows guests

  All patches are for stable and the x86 also has a CVE"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: x86: fix singlestepping over syscall
  KVM: s390: gaccess: fix real-space designation asce handling for gmap shadows
  KVM: MIPS: Fix maybe-uninitialized build failure
  KVM: PPC: Book3S HV: Ignore timebase offset on POWER9 DD1
  KVM: PPC: Book3S HV: Save/restore host values of debug registers
  KVM: PPC: Book3S HV: Preserve userspace HTM state properly
  KVM: PPC: Book3S HV: Restore critical SPRs to host values on guest exit
  KVM: PPC: Book3S HV: Context-switch EBB registers properly
  KVM: PPC: Book3S HV: Cope with host using large decrementer mode

arch/mips/kvm/tlb.c

@@ -166,7 +166,11 @@ static int _kvm_mips_host_tlb_inv(unsigned long entryhi)
 int kvm_mips_host_tlb_inv(struct kvm_vcpu *vcpu, unsigned long va,
 			  bool user, bool kernel)
 {
-	int idx_user, idx_kernel;
+	/*
+	 * Initialize idx_user and idx_kernel to workaround bogus
+	 * maybe-initialized warning when using GCC 6.
+	 */
+	int idx_user = 0, idx_kernel = 0;
 	unsigned long flags, old_entryhi;
 
 	local_irq_save(flags);

arch/powerpc/kvm/book3s_hv.c

@@ -1486,6 +1486,14 @@ static int kvmppc_set_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 		r = set_vpa(vcpu, &vcpu->arch.dtl, addr, len);
 		break;
 	case KVM_REG_PPC_TB_OFFSET:
+		/*
+		 * POWER9 DD1 has an erratum where writing TBU40 causes
+		 * the timebase to lose ticks.  So we don't let the
+		 * timebase offset be changed on P9 DD1.  (It is
+		 * initialized to zero.)
+		 */
+		if (cpu_has_feature(CPU_FTR_POWER9_DD1))
+			break;
 		/* round up to multiple of 2^24 */
 		vcpu->arch.vcore->tb_offset =
 			ALIGN(set_reg_val(id, *val), 1UL << 24);
@@ -2907,12 +2915,36 @@ static int kvmppc_vcpu_run_hv(struct kvm_run *run, struct kvm_vcpu *vcpu)
 {
 	int r;
 	int srcu_idx;
+	unsigned long ebb_regs[3] = {};	/* shut up GCC */
+	unsigned long user_tar = 0;
+	unsigned int user_vrsave;
 
 	if (!vcpu->arch.sane) {
 		run->exit_reason = KVM_EXIT_INTERNAL_ERROR;
 		return -EINVAL;
 	}
 
+	/*
+	 * Don't allow entry with a suspended transaction, because
+	 * the guest entry/exit code will lose it.
+	 * If the guest has TM enabled, save away their TM-related SPRs
+	 * (they will get restored by the TM unavailable interrupt).
+	 */
+#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
+	if (cpu_has_feature(CPU_FTR_TM) && current->thread.regs &&
+	    (current->thread.regs->msr & MSR_TM)) {
+		if (MSR_TM_ACTIVE(current->thread.regs->msr)) {
+			run->exit_reason = KVM_EXIT_FAIL_ENTRY;
+			run->fail_entry.hardware_entry_failure_reason = 0;
+			return -EINVAL;
+		}
+		current->thread.tm_tfhar = mfspr(SPRN_TFHAR);
+		current->thread.tm_tfiar = mfspr(SPRN_TFIAR);
+		current->thread.tm_texasr = mfspr(SPRN_TEXASR);
+		current->thread.regs->msr &= ~MSR_TM;
+	}
+#endif
+
 	kvmppc_core_prepare_to_enter(vcpu);
 
 	/* No need to go into the guest when all we'll do is come back out */
@@ -2934,6 +2966,15 @@ static int kvmppc_vcpu_run_hv(struct kvm_run *run, struct kvm_vcpu *vcpu)
 
 	flush_all_to_thread(current);
 
+	/* Save userspace EBB and other register values */
+	if (cpu_has_feature(CPU_FTR_ARCH_207S)) {
+		ebb_regs[0] = mfspr(SPRN_EBBHR);
+		ebb_regs[1] = mfspr(SPRN_EBBRR);
+		ebb_regs[2] = mfspr(SPRN_BESCR);
+		user_tar = mfspr(SPRN_TAR);
+	}
+	user_vrsave = mfspr(SPRN_VRSAVE);
+
 	vcpu->arch.wqp = &vcpu->arch.vcore->wq;
 	vcpu->arch.pgdir = current->mm->pgd;
 	vcpu->arch.state = KVMPPC_VCPU_BUSY_IN_HOST;
@@ -2960,6 +3001,16 @@ static int kvmppc_vcpu_run_hv(struct kvm_run *run, struct kvm_vcpu *vcpu)
 		}
 	} while (is_kvmppc_resume_guest(r));
 
+	/* Restore userspace EBB and other register values */
+	if (cpu_has_feature(CPU_FTR_ARCH_207S)) {
+		mtspr(SPRN_EBBHR, ebb_regs[0]);
+		mtspr(SPRN_EBBRR, ebb_regs[1]);
+		mtspr(SPRN_BESCR, ebb_regs[2]);
+		mtspr(SPRN_TAR, user_tar);
+		mtspr(SPRN_FSCR, current->thread.fscr);
+	}
+	mtspr(SPRN_VRSAVE, user_vrsave);
+
  out:
 	vcpu->arch.state = KVMPPC_VCPU_NOTREADY;
 	atomic_dec(&vcpu->kvm->arch.vcpus_running);

arch/powerpc/kvm/book3s_hv_interrupts.S

@@ -121,10 +121,20 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 	 * Put whatever is in the decrementer into the
 	 * hypervisor decrementer.
 	 */
+BEGIN_FTR_SECTION
+	ld	r5, HSTATE_KVM_VCORE(r13)
+	ld	r6, VCORE_KVM(r5)
+	ld	r9, KVM_HOST_LPCR(r6)
+	andis.	r9, r9, LPCR_LD@h
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
 	mfspr	r8,SPRN_DEC
 	mftb	r7
-	mtspr	SPRN_HDEC,r8
+BEGIN_FTR_SECTION
+	/* On POWER9, don't sign-extend if host LPCR[LD] bit is set */
+	bne	32f
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
 	extsw	r8,r8
+32:	mtspr	SPRN_HDEC,r8
 	add	r8,r8,r7
 	std	r8,HSTATE_DECEXP(r13)
 

arch/powerpc/kvm/book3s_hv_rmhandlers.S

@@ -32,12 +32,29 @@
 #include <asm/opal.h>
 #include <asm/xive-regs.h>
 
+/* Sign-extend HDEC if not on POWER9 */
+#define EXTEND_HDEC(reg)			\
+BEGIN_FTR_SECTION;				\
+	extsw	reg, reg;			\
+END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
+
 #define VCPU_GPRS_TM(reg) (((reg) * ULONG_SIZE) + VCPU_GPR_TM)
 
 /* Values in HSTATE_NAPPING(r13) */
 #define NAPPING_CEDE	1
 #define NAPPING_NOVCPU	2
 
+/* Stack frame offsets for kvmppc_hv_entry */
+#define SFS			144
+#define STACK_SLOT_TRAP		(SFS-4)
+#define STACK_SLOT_TID		(SFS-16)
+#define STACK_SLOT_PSSCR	(SFS-24)
+#define STACK_SLOT_PID		(SFS-32)
+#define STACK_SLOT_IAMR		(SFS-40)
+#define STACK_SLOT_CIABR	(SFS-48)
+#define STACK_SLOT_DAWR		(SFS-56)
+#define STACK_SLOT_DAWRX	(SFS-64)
+
 /*
  * Call kvmppc_hv_entry in real mode.
  * Must be called with interrupts hard-disabled.
@@ -214,6 +231,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 kvmppc_primary_no_guest:
 	/* We handle this much like a ceded vcpu */
 	/* put the HDEC into the DEC, since HDEC interrupts don't wake us */
+	/* HDEC may be larger than DEC for arch >= v3.00, but since the */
+	/* HDEC value came from DEC in the first place, it will fit */
 	mfspr	r3, SPRN_HDEC
 	mtspr	SPRN_DEC, r3
 	/*
@@ -295,8 +314,9 @@ kvm_novcpu_wakeup:
 
 	/* See if our timeslice has expired (HDEC is negative) */
 	mfspr	r0, SPRN_HDEC
+	EXTEND_HDEC(r0)
 	li	r12, BOOK3S_INTERRUPT_HV_DECREMENTER
-	cmpwi	r0, 0
+	cmpdi	r0, 0
 	blt	kvm_novcpu_exit
 
 	/* Got an IPI but other vcpus aren't yet exiting, must be a latecomer */
@@ -319,10 +339,10 @@ kvm_novcpu_exit:
 	bl	kvmhv_accumulate_time
 #endif
 13:	mr	r3, r12
-	stw	r12, 112-4(r1)
+	stw	r12, STACK_SLOT_TRAP(r1)
 	bl	kvmhv_commence_exit
 	nop
-	lwz	r12, 112-4(r1)
+	lwz	r12, STACK_SLOT_TRAP(r1)
 	b	kvmhv_switch_to_host
 
 /*
@@ -390,8 +410,8 @@ kvm_secondary_got_guest:
 	lbz	r4, HSTATE_PTID(r13)
 	cmpwi	r4, 0
 	bne	63f
-	lis	r6, 0x7fff
-	ori	r6, r6, 0xffff
+	LOAD_REG_ADDR(r6, decrementer_max)
+	ld	r6, 0(r6)
 	mtspr	SPRN_HDEC, r6
 	/* and set per-LPAR registers, if doing dynamic micro-threading */
 	ld	r6, HSTATE_SPLIT_MODE(r13)
@@ -545,11 +565,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
  *                                                                            *
  *****************************************************************************/
 
-/* Stack frame offsets */
-#define STACK_SLOT_TID		(112-16)
-#define STACK_SLOT_PSSCR	(112-24)
-#define STACK_SLOT_PID		(112-32)
-
 .global kvmppc_hv_entry
 kvmppc_hv_entry:
 
@@ -565,7 +580,7 @@ kvmppc_hv_entry:
 	 */
 	mflr	r0
 	std	r0, PPC_LR_STKOFF(r1)
-	stdu	r1, -112(r1)
+	stdu	r1, -SFS(r1)
 
 	/* Save R1 in the PACA */
 	std	r1, HSTATE_HOST_R1(r13)
@@ -749,10 +764,20 @@ BEGIN_FTR_SECTION
 	mfspr	r5, SPRN_TIDR
 	mfspr	r6, SPRN_PSSCR
 	mfspr	r7, SPRN_PID
+	mfspr	r8, SPRN_IAMR
 	std	r5, STACK_SLOT_TID(r1)
 	std	r6, STACK_SLOT_PSSCR(r1)
 	std	r7, STACK_SLOT_PID(r1)
+	std	r8, STACK_SLOT_IAMR(r1)
 END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
+BEGIN_FTR_SECTION
+	mfspr	r5, SPRN_CIABR
+	mfspr	r6, SPRN_DAWR
+	mfspr	r7, SPRN_DAWRX
+	std	r5, STACK_SLOT_CIABR(r1)
+	std	r6, STACK_SLOT_DAWR(r1)
+	std	r7, STACK_SLOT_DAWRX(r1)
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 
 BEGIN_FTR_SECTION
 	/* Set partition DABR */
@@ -968,7 +993,8 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
 
 	/* Check if HDEC expires soon */
 	mfspr	r3, SPRN_HDEC
-	cmpwi	r3, 512		/* 1 microsecond */
+	EXTEND_HDEC(r3)
+	cmpdi	r3, 512		/* 1 microsecond */
 	blt	hdec_soon
 
 #ifdef CONFIG_KVM_XICS
@@ -1505,11 +1531,10 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300)
 	 * set by the guest could disrupt the host.
 	 */
 	li	r0, 0
-	mtspr	SPRN_IAMR, r0
-	mtspr	SPRN_CIABR, r0
-	mtspr	SPRN_DAWRX, r0
+	mtspr	SPRN_PSPB, r0
 	mtspr	SPRN_WORT, r0
 BEGIN_FTR_SECTION
+	mtspr	SPRN_IAMR, r0
 	mtspr	SPRN_TCSCR, r0
 	/* Set MMCRS to 1<<31 to freeze and disable the SPMC counters */
 	li	r0, 1
@@ -1525,6 +1550,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300)
 	std	r6,VCPU_UAMOR(r9)
 	li	r6,0
 	mtspr	SPRN_AMR,r6
+	mtspr	SPRN_UAMOR, r6
 
 	/* Switch DSCR back to host value */
 	mfspr	r8, SPRN_DSCR
@@ -1669,13 +1695,23 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 	ptesync
 
 	/* Restore host values of some registers */
+BEGIN_FTR_SECTION
+	ld	r5, STACK_SLOT_CIABR(r1)
+	ld	r6, STACK_SLOT_DAWR(r1)
+	ld	r7, STACK_SLOT_DAWRX(r1)
+	mtspr	SPRN_CIABR, r5
+	mtspr	SPRN_DAWR, r6
+	mtspr	SPRN_DAWRX, r7
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S)
 BEGIN_FTR_SECTION
 	ld	r5, STACK_SLOT_TID(r1)
 	ld	r6, STACK_SLOT_PSSCR(r1)
 	ld	r7, STACK_SLOT_PID(r1)
+	ld	r8, STACK_SLOT_IAMR(r1)
 	mtspr	SPRN_TIDR, r5
 	mtspr	SPRN_PSSCR, r6
 	mtspr	SPRN_PID, r7
+	mtspr	SPRN_IAMR, r8
 END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
 BEGIN_FTR_SECTION
 	PPC_INVALIDATE_ERAT
@@ -1819,8 +1855,8 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_TYPE_RADIX)
 	li	r0, KVM_GUEST_MODE_NONE
 	stb	r0, HSTATE_IN_GUEST(r13)
 
-	ld	r0, 112+PPC_LR_STKOFF(r1)
-	addi	r1, r1, 112
+	ld	r0, SFS+PPC_LR_STKOFF(r1)
+	addi	r1, r1, SFS
 	mtlr	r0
 	blr
 
@@ -2366,12 +2402,13 @@ END_FTR_SECTION_IFSET(CPU_FTR_TM)
 	mfspr	r3, SPRN_DEC
 	mfspr	r4, SPRN_HDEC
 	mftb	r5
-	cmpw	r3, r4
+	extsw	r3, r3
+	EXTEND_HDEC(r4)
+	cmpd	r3, r4
 	ble	67f
 	mtspr	SPRN_DEC, r4
 67:
 	/* save expiry time of guest decrementer */
-	extsw	r3, r3
 	add	r3, r3, r5
 	ld	r4, HSTATE_KVM_VCPU(r13)
 	ld	r5, HSTATE_KVM_VCORE(r13)

arch/s390/kvm/gaccess.c

@@ -977,11 +977,12 @@ static int kvm_s390_shadow_tables(struct gmap *sg, unsigned long saddr,
 	ptr = asce.origin * 4096;
 	if (asce.r) {
 		*fake = 1;
+		ptr = 0;
 		asce.dt = ASCE_TYPE_REGION1;
 	}
 	switch (asce.dt) {
 	case ASCE_TYPE_REGION1:
-		if (vaddr.rfx01 > asce.tl && !asce.r)
+		if (vaddr.rfx01 > asce.tl && !*fake)
 			return PGM_REGION_FIRST_TRANS;
 		break;
 	case ASCE_TYPE_REGION2:
@@ -1009,8 +1010,7 @@ static int kvm_s390_shadow_tables(struct gmap *sg, unsigned long saddr,
 		union region1_table_entry rfte;
 
 		if (*fake) {
-			/* offset in 16EB guest memory block */
-			ptr = ptr + ((unsigned long) vaddr.rsx << 53UL);
+			ptr += (unsigned long) vaddr.rfx << 53;
 			rfte.val = ptr;
 			goto shadow_r2t;
 		}
@@ -1036,8 +1036,7 @@ static int kvm_s390_shadow_tables(struct gmap *sg, unsigned long saddr,
 		union region2_table_entry rste;
 
 		if (*fake) {
-			/* offset in 8PB guest memory block */
-			ptr = ptr + ((unsigned long) vaddr.rtx << 42UL);
+			ptr += (unsigned long) vaddr.rsx << 42;
 			rste.val = ptr;
 			goto shadow_r3t;
 		}
@@ -1064,8 +1063,7 @@ static int kvm_s390_shadow_tables(struct gmap *sg, unsigned long saddr,
 		union region3_table_entry rtte;
 
 		if (*fake) {
-			/* offset in 4TB guest memory block */
-			ptr = ptr + ((unsigned long) vaddr.sx << 31UL);
+			ptr += (unsigned long) vaddr.rtx << 31;
 			rtte.val = ptr;
 			goto shadow_sgt;
 		}
@@ -1101,8 +1099,7 @@ static int kvm_s390_shadow_tables(struct gmap *sg, unsigned long saddr,
 		union segment_table_entry ste;
 
 		if (*fake) {
-			/* offset in 2G guest memory block */
-			ptr = ptr + ((unsigned long) vaddr.sx << 20UL);
+			ptr += (unsigned long) vaddr.sx << 20;
 			ste.val = ptr;
 			goto shadow_pgt;
 		}

arch/x86/include/asm/kvm_emulate.h

@@ -296,6 +296,7 @@ struct x86_emulate_ctxt {
 
 	bool perm_ok; /* do not check permissions if true */
 	bool ud;	/* inject an #UD if host doesn't support insn */
+	bool tf;	/* TF value before instruction (after for syscall/sysret) */
 
 	bool have_exception;
 	struct x86_exception exception;

arch/x86/kvm/emulate.c

@@ -2742,6 +2742,7 @@ static int em_syscall(struct x86_emulate_ctxt *ctxt)
 		ctxt->eflags &= ~(X86_EFLAGS_VM | X86_EFLAGS_IF);
 	}
 
+	ctxt->tf = (ctxt->eflags & X86_EFLAGS_TF) != 0;
 	return X86EMUL_CONTINUE;
 }
 

arch/x86/kvm/x86.c

@@ -5313,6 +5313,8 @@ static void init_emulate_ctxt(struct kvm_vcpu *vcpu)
 	kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l);
 
 	ctxt->eflags = kvm_get_rflags(vcpu);
+	ctxt->tf = (ctxt->eflags & X86_EFLAGS_TF) != 0;
+
 	ctxt->eip = kvm_rip_read(vcpu);
 	ctxt->mode = (!is_protmode(vcpu))		? X86EMUL_MODE_REAL :
 		     (ctxt->eflags & X86_EFLAGS_VM)	? X86EMUL_MODE_VM86 :
@@ -5528,22 +5530,12 @@ static int kvm_vcpu_check_hw_bp(unsigned long addr, u32 type, u32 dr7,
 	return dr6;
 }
 
-static void kvm_vcpu_check_singlestep(struct kvm_vcpu *vcpu, unsigned long rflags, int *r)
+static void kvm_vcpu_do_singlestep(struct kvm_vcpu *vcpu, int *r)
 {
 	struct kvm_run *kvm_run = vcpu->run;
 
-	/*
-	 * rflags is the old, "raw" value of the flags.  The new value has
-	 * not been saved yet.
-	 *
-	 * This is correct even for TF set by the guest, because "the
-	 * processor will not generate this exception after the instruction
-	 * that sets the TF flag".
-	 */
-	if (unlikely(rflags & X86_EFLAGS_TF)) {
 	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) {
-			kvm_run->debug.arch.dr6 = DR6_BS | DR6_FIXED_1 |
-						  DR6_RTM;
+		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.exception = DB_VECTOR;
 		kvm_run->exit_reason = KVM_EXIT_DEBUG;
@@ -5558,7 +5550,6 @@ static void kvm_vcpu_check_singlestep(struct kvm_vcpu *vcpu, unsigned long rflag
 		vcpu->arch.dr6 |= DR6_BS | DR6_RTM;
 		kvm_queue_exception(vcpu, DB_VECTOR);
 	}
-	}
 }
 
 int kvm_skip_emulated_instruction(struct kvm_vcpu *vcpu)
@@ -5567,7 +5558,17 @@ int kvm_skip_emulated_instruction(struct kvm_vcpu *vcpu)
 	int r = EMULATE_DONE;
 
 	kvm_x86_ops->skip_emulated_instruction(vcpu);
-	kvm_vcpu_check_singlestep(vcpu, rflags, &r);
+
+	/*
+	 * rflags is the old, "raw" value of the flags.  The new value has
+	 * not been saved yet.
+	 *
+	 * This is correct even for TF set by the guest, because "the
+	 * processor will not generate this exception after the instruction
+	 * that sets the TF flag".
+	 */
+	if (unlikely(rflags & X86_EFLAGS_TF))
+		kvm_vcpu_do_singlestep(vcpu, &r);
 	return r == EMULATE_DONE;
 }
 EXPORT_SYMBOL_GPL(kvm_skip_emulated_instruction);
@@ -5726,8 +5727,9 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu,
 		toggle_interruptibility(vcpu, ctxt->interruptibility);
 		vcpu->arch.emulate_regs_need_sync_to_vcpu = false;
 		kvm_rip_write(vcpu, ctxt->eip);
-		if (r == EMULATE_DONE)
-			kvm_vcpu_check_singlestep(vcpu, rflags, &r);
+		if (r == EMULATE_DONE &&
+		    (ctxt->tf || (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)))
+			kvm_vcpu_do_singlestep(vcpu, &r);
 		if (!ctxt->have_exception ||
 		    exception_type(ctxt->exception.vector) == EXCPT_TRAP)
 			__kvm_set_rflags(vcpu, ctxt->eflags);