Merge commit 'v3.2-rc6' into perf/core

Merge reason: Update with the latest fixes.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

22 files changed, 187 insertions, 50 deletions

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index a9fbd43395f7..373679b3744a 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -572,6 +572,7 @@ config KEXEC
 config CRASH_DUMP
 	bool "kernel crash dumps"
 	depends on 64BIT
+	select KEXEC
 	help
 	  Generate crash dump after being started by kexec.
 	  Crash dump kernels are loaded in the main kernel with kexec-tools
diff --git a/arch/s390/crypto/crypt_s390.h b/arch/s390/crypto/crypt_s390.h
index 49676771bd66..ffd1ac255f19 100644
--- a/arch/s390/crypto/crypt_s390.h
+++ b/arch/s390/crypto/crypt_s390.h
@@ -368,9 +368,12 @@ static inline int crypt_s390_func_available(int func,
 
 	if (facility_mask & CRYPT_S390_MSA && !test_facility(17))
 		return 0;
-	if (facility_mask & CRYPT_S390_MSA3 && !test_facility(76))
+
+	if (facility_mask & CRYPT_S390_MSA3 &&
+	    (!test_facility(2) || !test_facility(76)))
 		return 0;
-	if (facility_mask & CRYPT_S390_MSA4 && !test_facility(77))
+	if (facility_mask & CRYPT_S390_MSA4 &&
+	    (!test_facility(2) || !test_facility(77)))
 		return 0;
 
 	switch (func & CRYPT_S390_OP_MASK) {
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index 24e18473d926..b0c235cb6ad5 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -47,7 +47,7 @@ struct sca_block {
 #define KVM_HPAGE_MASK(x)	(~(KVM_HPAGE_SIZE(x) - 1))
 #define KVM_PAGES_PER_HPAGE(x)	(KVM_HPAGE_SIZE(x) / PAGE_SIZE)
 
-#define CPUSTAT_HOST       0x80000000
+#define CPUSTAT_STOPPED    0x80000000
 #define CPUSTAT_WAIT       0x10000000
 #define CPUSTAT_ECALL_PEND 0x08000000
 #define CPUSTAT_STOP_INT   0x04000000
@@ -139,6 +139,7 @@ struct kvm_vcpu_stat {
 	u32 instruction_stfl;
 	u32 instruction_tprot;
 	u32 instruction_sigp_sense;
+	u32 instruction_sigp_sense_running;
 	u32 instruction_sigp_external_call;
 	u32 instruction_sigp_emergency;
 	u32 instruction_sigp_stop;
diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
index 34ede0ea85a9..4f289ff0b7fe 100644
--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h
@@ -593,14 +593,16 @@ static inline pgste_t pgste_update_all(pte_t *ptep, pgste_t pgste)
 	unsigned long address, bits;
 	unsigned char skey;
 
+	if (!pte_present(*ptep))
+		return pgste;
 	address = pte_val(*ptep) & PAGE_MASK;
 	skey = page_get_storage_key(address);
 	bits = skey & (_PAGE_CHANGED | _PAGE_REFERENCED);
 	/* Clear page changed & referenced bit in the storage key */
-	if (bits) {
-		skey ^= bits;
-		page_set_storage_key(address, skey, 1);
-	}
+	if (bits & _PAGE_CHANGED)
+		page_set_storage_key(address, skey ^ bits, 1);
+	else if (bits)
+		page_reset_referenced(address);
 	/* Transfer page changed & referenced bit to guest bits in pgste */
 	pgste_val(pgste) |= bits << 48;		/* RCP_GR_BIT & RCP_GC_BIT */
 	/* Get host changed & referenced bits from pgste */
@@ -625,6 +627,8 @@ static inline pgste_t pgste_update_young(pte_t *ptep, pgste_t pgste)
 #ifdef CONFIG_PGSTE
 	int young;
 
+	if (!pte_present(*ptep))
+		return pgste;
 	young = page_reset_referenced(pte_val(*ptep) & PAGE_MASK);
 	/* Transfer page referenced bit to pte software bit (host view) */
 	if (young || (pgste_val(pgste) & RCP_HR_BIT))
@@ -638,13 +642,15 @@ static inline pgste_t pgste_update_young(pte_t *ptep, pgste_t pgste)
 
 }
 
-static inline void pgste_set_pte(pte_t *ptep, pgste_t pgste)
+static inline void pgste_set_pte(pte_t *ptep, pgste_t pgste, pte_t entry)
 {
 #ifdef CONFIG_PGSTE
 	unsigned long address;
 	unsigned long okey, nkey;
 
-	address = pte_val(*ptep) & PAGE_MASK;
+	if (!pte_present(entry))
+		return;
+	address = pte_val(entry) & PAGE_MASK;
 	okey = nkey = page_get_storage_key(address);
 	nkey &= ~(_PAGE_ACC_BITS | _PAGE_FP_BIT);
 	/* Set page access key and fetch protection bit from pgste */
@@ -712,7 +718,7 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
 
 	if (mm_has_pgste(mm)) {
 		pgste = pgste_get_lock(ptep);
-		pgste_set_pte(ptep, pgste);
+		pgste_set_pte(ptep, pgste, entry);
 		*ptep = entry;
 		pgste_set_unlock(ptep, pgste);
 	} else
diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h
index 5a099714df04..097183c70407 100644
--- a/arch/s390/include/asm/setup.h
+++ b/arch/s390/include/asm/setup.h
@@ -82,6 +82,7 @@ extern unsigned int user_mode;
 #define MACHINE_FLAG_LPAR	(1UL << 12)
 #define MACHINE_FLAG_SPP	(1UL << 13)
 #define MACHINE_FLAG_TOPOLOGY	(1UL << 14)
+#define MACHINE_FLAG_STCKF	(1UL << 15)
 
 #define MACHINE_IS_VM		(S390_lowcore.machine_flags & MACHINE_FLAG_VM)
 #define MACHINE_IS_KVM		(S390_lowcore.machine_flags & MACHINE_FLAG_KVM)
@@ -100,6 +101,7 @@ extern unsigned int user_mode;
 #define MACHINE_HAS_PFMF	(0)
 #define MACHINE_HAS_SPP		(0)
 #define MACHINE_HAS_TOPOLOGY	(0)
+#define MACHINE_HAS_STCKF	(0)
 #else /* __s390x__ */
 #define MACHINE_HAS_IEEE	(1)
 #define MACHINE_HAS_CSP		(1)
@@ -111,6 +113,7 @@ extern unsigned int user_mode;
 #define MACHINE_HAS_PFMF	(S390_lowcore.machine_flags & MACHINE_FLAG_PFMF)
 #define MACHINE_HAS_SPP		(S390_lowcore.machine_flags & MACHINE_FLAG_SPP)
 #define MACHINE_HAS_TOPOLOGY	(S390_lowcore.machine_flags & MACHINE_FLAG_TOPOLOGY)
+#define MACHINE_HAS_STCKF	(S390_lowcore.machine_flags & MACHINE_FLAG_STCKF)
 #endif /* __s390x__ */
 
 #define ZFCPDUMP_HSA_SIZE	(32UL<<20)
diff --git a/arch/s390/include/asm/timex.h b/arch/s390/include/asm/timex.h
index d610bef9c5e9..c447a27a7fdb 100644
--- a/arch/s390/include/asm/timex.h
+++ b/arch/s390/include/asm/timex.h
@@ -90,7 +90,7 @@ static inline unsigned long long get_clock_fast(void)
 {
 	unsigned long long clk;
 
-	if (test_facility(25))
+	if (MACHINE_HAS_STCKF)
 		asm volatile(".insn	s,0xb27c0000,%0" : "=Q" (clk) : : "cc");
 	else
 		clk = get_clock();
diff --git a/arch/s390/include/asm/unistd.h b/arch/s390/include/asm/unistd.h
index 404bdb9671b4..58de4c91c333 100644
--- a/arch/s390/include/asm/unistd.h
+++ b/arch/s390/include/asm/unistd.h
@@ -277,7 +277,9 @@
 #define __NR_clock_adjtime	337
 #define __NR_syncfs		338
 #define __NR_setns		339
-#define NR_syscalls 340
+#define __NR_process_vm_readv	340
+#define __NR_process_vm_writev	341
+#define NR_syscalls 342
 
 /* 
  * There are some system calls that are not present on 64 bit, some
diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S
index 5006a1d9f5d0..18c51df9fe06 100644
--- a/arch/s390/kernel/compat_wrapper.S
+++ b/arch/s390/kernel/compat_wrapper.S
@@ -1627,3 +1627,23 @@ ENTRY(sys_setns_wrapper)
 	lgfr	%r2,%r2			# int
 	lgfr	%r3,%r3			# int
 	jg	sys_setns
+
+ENTRY(compat_sys_process_vm_readv_wrapper)
+	lgfr	%r2,%r2			# compat_pid_t
+	llgtr	%r3,%r3			# struct compat_iovec __user *
+	llgfr	%r4,%r4			# unsigned long
+	llgtr	%r5,%r5			# struct compat_iovec __user *
+	llgfr	%r6,%r6			# unsigned long
+	llgf	%r0,164(%r15)		# unsigned long
+	stg	%r0,160(%r15)
+	jg	sys_process_vm_readv
+
+ENTRY(compat_sys_process_vm_writev_wrapper)
+	lgfr	%r2,%r2			# compat_pid_t
+	llgtr	%r3,%r3			# struct compat_iovec __user *
+	llgfr	%r4,%r4			# unsigned long
+	llgtr	%r5,%r5			# struct compat_iovec __user *
+	llgfr	%r6,%r6			# unsigned long
+	llgf	%r0,164(%r15)		# unsigned long
+	stg	%r0,160(%r15)
+	jg	sys_process_vm_writev
diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index 37394b3413e2..c9ffe0025197 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -390,6 +390,8 @@ static __init void detect_machine_facilities(void)
 		S390_lowcore.machine_flags |= MACHINE_FLAG_MVCOS;
 	if (test_facility(40))
 		S390_lowcore.machine_flags |= MACHINE_FLAG_SPP;
+	if (test_facility(25))
+		S390_lowcore.machine_flags |= MACHINE_FLAG_STCKF;
 #endif
 }
 
diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index 450931a45b68..573bc29551ef 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -296,13 +296,6 @@ static int __poke_user(struct task_struct *child, addr_t addr, addr_t data)
 		     ((data & PSW_MASK_EA) && !(data & PSW_MASK_BA))))
 			/* Invalid psw mask. */
 			return -EINVAL;
-		if (addr == (addr_t) &dummy->regs.psw.addr)
-			/*
-			 * The debugger changed the instruction address,
-			 * reset system call restart, see signal.c:do_signal
-			 */
-			task_thread_info(child)->system_call = 0;
-
 		*(addr_t *)((addr_t) &task_pt_regs(child)->psw + addr) = data;
 
 	} else if (addr < (addr_t) (&dummy->regs.orig_gpr2)) {
@@ -614,11 +607,6 @@ static int __poke_user_compat(struct task_struct *child,
 			/* Transfer 31 bit amode bit to psw mask. */
 			regs->psw.mask = (regs->psw.mask & ~PSW_MASK_BA) |
 				(__u64)(tmp & PSW32_ADDR_AMODE);
-			/*
-			 * The debugger changed the instruction address,
-			 * reset system call restart, see signal.c:do_signal
-			 */
-			task_thread_info(child)->system_call = 0;
 		} else {
 			/* gpr 0-15 */
 			*(__u32*)((addr_t) &regs->psw + addr*2 + 4) = tmp;
@@ -905,6 +893,14 @@ static int s390_last_break_get(struct task_struct *target,
 	return 0;
 }
 
+static int s390_last_break_set(struct task_struct *target,
+			       const struct user_regset *regset,
+			       unsigned int pos, unsigned int count,
+			       const void *kbuf, const void __user *ubuf)
+{
+	return 0;
+}
+
 #endif
 
 static int s390_system_call_get(struct task_struct *target,
@@ -951,6 +947,7 @@ static const struct user_regset s390_regsets[] = {
 		.size = sizeof(long),
 		.align = sizeof(long),
 		.get = s390_last_break_get,
+		.set = s390_last_break_set,
 	},
 #endif
 	[REGSET_SYSTEM_CALL] = {
@@ -1116,6 +1113,14 @@ static int s390_compat_last_break_get(struct task_struct *target,
 	return 0;
 }
 
+static int s390_compat_last_break_set(struct task_struct *target,
+				      const struct user_regset *regset,
+				      unsigned int pos, unsigned int count,
+				      const void *kbuf, const void __user *ubuf)
+{
+	return 0;
+}
+
 static const struct user_regset s390_compat_regsets[] = {
 	[REGSET_GENERAL] = {
 		.core_note_type = NT_PRSTATUS,
@@ -1139,6 +1144,7 @@ static const struct user_regset s390_compat_regsets[] = {
 		.size = sizeof(long),
 		.align = sizeof(long),
 		.get = s390_compat_last_break_get,
+		.set = s390_compat_last_break_set,
 	},
 	[REGSET_SYSTEM_CALL] = {
 		.core_note_type = NT_S390_SYSTEM_CALL,
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 8ac6bfa2786c..e54c4ff8abaa 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -211,6 +211,8 @@ static void __init setup_zfcpdump(unsigned int console_devno)
 
 	if (ipl_info.type != IPL_TYPE_FCP_DUMP)
 		return;
+	if (OLDMEM_BASE)
+		return;
 	if (console_devno != -1)
 		sprintf(str, " cio_ignore=all,!0.0.%04x,!0.0.%04x",
 			ipl_info.data.fcp.dev_id.devno, console_devno);
@@ -482,7 +484,7 @@ static void __init setup_memory_end(void)
 
 
 #ifdef CONFIG_ZFCPDUMP
-	if (ipl_info.type == IPL_TYPE_FCP_DUMP) {
+	if (ipl_info.type == IPL_TYPE_FCP_DUMP && !OLDMEM_BASE) {
 		memory_end = ZFCPDUMP_HSA_SIZE;
 		memory_end_set = 1;
 	}
@@ -577,7 +579,7 @@ static unsigned long __init find_crash_base(unsigned long crash_size,
 		*msg = "first memory chunk must be at least crashkernel size";
 		return 0;
 	}
-	if (is_kdump_kernel() && (crash_size == OLDMEM_SIZE))
+	if (OLDMEM_BASE && crash_size == OLDMEM_SIZE)
 		return OLDMEM_BASE;
 
 	for (i = MEMORY_CHUNKS - 1; i >= 0; i--) {
diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c
index 05a85bc14c98..7f6f9f354545 100644
--- a/arch/s390/kernel/signal.c
+++ b/arch/s390/kernel/signal.c
@@ -460,9 +460,9 @@ void do_signal(struct pt_regs *regs)
 						     regs->svc_code >> 16);
 				break;
 			}
-			/* No longer in a system call */
-			clear_thread_flag(TIF_SYSCALL);
 		}
+		/* No longer in a system call */
+		clear_thread_flag(TIF_SYSCALL);
 
 		if ((is_compat_task() ?
 		     handle_signal32(signr, &ka, &info, oldset, regs) :
@@ -486,6 +486,7 @@ void do_signal(struct pt_regs *regs)
 	}
 
 	/* No handlers present - check for system call restart */
+	clear_thread_flag(TIF_SYSCALL);
 	if (current_thread_info()->system_call) {
 		regs->svc_code = current_thread_info()->system_call;
 		switch (regs->gprs[2]) {
@@ -500,9 +501,6 @@ void do_signal(struct pt_regs *regs)
 			regs->gprs[2] = regs->orig_gpr2;
 			set_thread_flag(TIF_SYSCALL);
 			break;
-		default:
-			clear_thread_flag(TIF_SYSCALL);
-			break;
 		}
 	}
 
diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
index 73eb08c874fb..bcab2f04ba58 100644
--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -348,3 +348,5 @@ SYSCALL(sys_open_by_handle_at,sys_open_by_handle_at,compat_sys_open_by_handle_at
 SYSCALL(sys_clock_adjtime,sys_clock_adjtime,compat_sys_clock_adjtime_wrapper)
 SYSCALL(sys_syncfs,sys_syncfs,sys_syncfs_wrapper)
 SYSCALL(sys_setns,sys_setns,sys_setns_wrapper)
+SYSCALL(sys_process_vm_readv,sys_process_vm_readv,compat_sys_process_vm_readv_wrapper) /* 340 */
+SYSCALL(sys_process_vm_writev,sys_process_vm_writev,compat_sys_process_vm_writev_wrapper)
diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
index 77b8942b9a15..fdb5b8cb260f 100644
--- a/arch/s390/kernel/topology.c
+++ b/arch/s390/kernel/topology.c
@@ -68,8 +68,10 @@ static cpumask_t cpu_group_map(struct mask_info *info, unsigned int cpu)
 	return mask;
 }
 
-static void add_cpus_to_mask(struct topology_cpu *tl_cpu,
-			     struct mask_info *book, struct mask_info *core)
+static struct mask_info *add_cpus_to_mask(struct topology_cpu *tl_cpu,
+					  struct mask_info *book,
+					  struct mask_info *core,
+					  int z10)
 {
 	unsigned int cpu;
 
@@ -88,10 +90,16 @@ static void add_cpus_to_mask(struct topology_cpu *tl_cpu,
 			cpu_book_id[lcpu] = book->id;
 #endif
 			cpumask_set_cpu(lcpu, &core->mask);
-			cpu_core_id[lcpu] = core->id;
+			if (z10) {
+				cpu_core_id[lcpu] = rcpu;
+				core = core->next;
+			} else {
+				cpu_core_id[lcpu] = core->id;
+			}
 			smp_cpu_polarization[lcpu] = tl_cpu->pp;
 		}
 	}
+	return core;
 }
 
 static void clear_masks(void)
@@ -123,18 +131,41 @@ static void tl_to_cores(struct sysinfo_15_1_x *info)
 {
 #ifdef CONFIG_SCHED_BOOK
 	struct mask_info *book = &book_info;
+	struct cpuid cpu_id;
 #else
 	struct mask_info *book = NULL;
 #endif
 	struct mask_info *core = &core_info;
 	union topology_entry *tle, *end;
+	int z10 = 0;
 
-
+#ifdef CONFIG_SCHED_BOOK
+	get_cpu_id(&cpu_id);
+	z10 = cpu_id.machine == 0x2097 || cpu_id.machine == 0x2098;
+#endif
 	spin_lock_irq(&topology_lock);
 	clear_masks();
 	tle = info->tle;
 	end = (union topology_entry *)((unsigned long)info + info->length);
 	while (tle < end) {
+#ifdef CONFIG_SCHED_BOOK
+		if (z10) {
+			switch (tle->nl) {
+			case 1:
+				book = book->next;
+				book->id = tle->container.id;
+				break;
+			case 0:
+				core = add_cpus_to_mask(&tle->cpu, book, core, z10);
+				break;
+			default:
+				clear_masks();
+				goto out;
+			}
+			tle = next_tle(tle);
+			continue;
+		}
+#endif
 		switch (tle->nl) {
 #ifdef CONFIG_SCHED_BOOK
 		case 2:
@@ -147,7 +178,7 @@ static void tl_to_cores(struct sysinfo_15_1_x *info)
 			core->id = tle->container.id;
 			break;
 		case 0:
-			add_cpus_to_mask(&tle->cpu, book, core);
+			add_cpus_to_mask(&tle->cpu, book, core, z10);
 			break;
 		default:
 			clear_masks();
@@ -328,8 +359,8 @@ void __init s390_init_cpu_topology(void)
 	for (i = 0; i < TOPOLOGY_NR_MAG; i++)
 		printk(" %d", info->mag[i]);
 	printk(" / %d\n", info->mnest);
-	alloc_masks(info, &core_info, 2);
+	alloc_masks(info, &core_info, 1);
 #ifdef CONFIG_SCHED_BOOK
-	alloc_masks(info, &book_info, 3);
+	alloc_masks(info, &book_info, 2);
 #endif
 }
diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S
index 56fe6bc81fee..e4c79ebb40e6 100644
--- a/arch/s390/kernel/vmlinux.lds.S
+++ b/arch/s390/kernel/vmlinux.lds.S
@@ -43,6 +43,8 @@ SECTIONS
 
 	NOTES :text :note
 
+	.dummy : { *(.dummy) } :data
+
 	RODATA
 
 #ifdef CONFIG_SHARED_KERNEL
diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c
index 87cedd61be04..8943e82cd4d9 100644
--- a/arch/s390/kvm/diag.c
+++ b/arch/s390/kvm/diag.c
@@ -70,7 +70,7 @@ static int __diag_ipl_functions(struct kvm_vcpu *vcpu)
 		return -EOPNOTSUPP;
 	}
 
-	atomic_clear_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
+	atomic_set_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
 	vcpu->run->s390_reset_flags |= KVM_S390_RESET_SUBSYSTEM;
 	vcpu->run->s390_reset_flags |= KVM_S390_RESET_IPL;
 	vcpu->run->s390_reset_flags |= KVM_S390_RESET_CPU_INIT;
diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c
index c7c51898984e..02434543eabb 100644
--- a/arch/s390/kvm/intercept.c
+++ b/arch/s390/kvm/intercept.c
@@ -132,7 +132,6 @@ static int handle_stop(struct kvm_vcpu *vcpu)
 	int rc = 0;
 
 	vcpu->stat.exit_stop_request++;
-	atomic_clear_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
 	spin_lock_bh(&vcpu->arch.local_int.lock);
 	if (vcpu->arch.local_int.action_bits & ACTION_STORE_ON_STOP) {
 		vcpu->arch.local_int.action_bits &= ~ACTION_STORE_ON_STOP;
@@ -149,6 +148,8 @@ static int handle_stop(struct kvm_vcpu *vcpu)
 	}
 
 	if (vcpu->arch.local_int.action_bits & ACTION_STOP_ON_STOP) {
+		atomic_set_mask(CPUSTAT_STOPPED,
+				&vcpu->arch.sie_block->cpuflags);
 		vcpu->arch.local_int.action_bits &= ~ACTION_STOP_ON_STOP;
 		VCPU_EVENT(vcpu, 3, "%s", "cpu stopped");
 		rc = -EOPNOTSUPP;
diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
index 87c16705b381..278ee009ce65 100644
--- a/arch/s390/kvm/interrupt.c
+++ b/arch/s390/kvm/interrupt.c
@@ -252,6 +252,7 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu,
 			offsetof(struct _lowcore, restart_psw), sizeof(psw_t));
 		if (rc == -EFAULT)
 			exception = 1;
+		atomic_clear_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
 		break;
 
 	case KVM_S390_PROGRAM_INT:
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 0bd3bea1e4cd..d1c445732451 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -65,6 +65,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "instruction_stfl", VCPU_STAT(instruction_stfl) },
 	{ "instruction_tprot", VCPU_STAT(instruction_tprot) },
 	{ "instruction_sigp_sense", VCPU_STAT(instruction_sigp_sense) },
+	{ "instruction_sigp_sense_running", VCPU_STAT(instruction_sigp_sense_running) },
 	{ "instruction_sigp_external_call", VCPU_STAT(instruction_sigp_external_call) },
 	{ "instruction_sigp_emergency", VCPU_STAT(instruction_sigp_emergency) },
 	{ "instruction_sigp_stop", VCPU_STAT(instruction_sigp_stop) },
@@ -127,6 +128,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	switch (ext) {
 	case KVM_CAP_S390_PSW:
 	case KVM_CAP_S390_GMAP:
+	case KVM_CAP_SYNC_MMU:
 		r = 1;
 		break;
 	default:
@@ -270,10 +272,12 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 	restore_fp_regs(&vcpu->arch.guest_fpregs);
 	restore_access_regs(vcpu->arch.guest_acrs);
 	gmap_enable(vcpu->arch.gmap);
+	atomic_set_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
 }
 
 void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 {
+	atomic_clear_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
 	gmap_disable(vcpu->arch.gmap);
 	save_fp_regs(&vcpu->arch.guest_fpregs);
 	save_access_regs(vcpu->arch.guest_acrs);
@@ -301,7 +305,9 @@ static void kvm_s390_vcpu_initial_reset(struct kvm_vcpu *vcpu)
 
 int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 {
-	atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH | CPUSTAT_SM);
+	atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH |
+						    CPUSTAT_SM |
+						    CPUSTAT_STOPPED);
 	vcpu->arch.sie_block->ecb   = 6;
 	vcpu->arch.sie_block->eca   = 0xC1002001U;
 	vcpu->arch.sie_block->fac   = (int) (long) facilities;
@@ -428,7 +434,7 @@ static int kvm_arch_vcpu_ioctl_set_initial_psw(struct kvm_vcpu *vcpu, psw_t psw)
 {
 	int rc = 0;
 
-	if (atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_RUNNING)
+	if (!(atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_STOPPED))
 		rc = -EBUSY;
 	else {
 		vcpu->run->psw_mask = psw.mask;
@@ -501,7 +507,7 @@ rerun_vcpu:
 	if (vcpu->sigset_active)
 		sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);
 
-	atomic_set_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
+	atomic_clear_mask(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
 
 	BUG_ON(vcpu->kvm->arch.float_int.local_int[vcpu->vcpu_id] == NULL);
 
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index 391626361084..d02638959922 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -336,6 +336,7 @@ static int handle_tprot(struct kvm_vcpu *vcpu)
 	u64 address1 = disp1 + base1 ? vcpu->arch.guest_gprs[base1] : 0;
 	u64 address2 = disp2 + base2 ? vcpu->arch.guest_gprs[base2] : 0;
 	struct vm_area_struct *vma;
+	unsigned long user_address;
 
 	vcpu->stat.instruction_tprot++;
 
@@ -349,9 +350,14 @@ static int handle_tprot(struct kvm_vcpu *vcpu)
 		return -EOPNOTSUPP;
 
 
+	/* we must resolve the address without holding the mmap semaphore.
+	 * This is ok since the userspace hypervisor is not supposed to change
+	 * the mapping while the guest queries the memory. Otherwise the guest
+	 * might crash or get wrong info anyway. */
+	user_address = (unsigned long) __guestaddr_to_user(vcpu, address1);
+
 	down_read(&current->mm->mmap_sem);
-	vma = find_vma(current->mm,
-			(unsigned long) __guestaddr_to_user(vcpu, address1));
+	vma = find_vma(current->mm, user_address);
 	if (!vma) {
 		up_read(&current->mm->mmap_sem);
 		return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
diff --git a/arch/s390/kvm/sigp.c b/arch/s390/kvm/sigp.c
index f815118835f3..0a7941d74bc6 100644
--- a/arch/s390/kvm/sigp.c
+++ b/arch/s390/kvm/sigp.c
@@ -31,9 +31,11 @@
 #define SIGP_SET_PREFIX        0x0d
 #define SIGP_STORE_STATUS_ADDR 0x0e
 #define SIGP_SET_ARCH          0x12
+#define SIGP_SENSE_RUNNING     0x15
 
 /* cpu status bits */
 #define SIGP_STAT_EQUIPMENT_CHECK   0x80000000UL
+#define SIGP_STAT_NOT_RUNNING	    0x00000400UL
 #define SIGP_STAT_INCORRECT_STATE   0x00000200UL
 #define SIGP_STAT_INVALID_PARAMETER 0x00000100UL
 #define SIGP_STAT_EXT_CALL_PENDING  0x00000080UL
@@ -57,8 +59,8 @@ static int __sigp_sense(struct kvm_vcpu *vcpu, u16 cpu_addr,
 	spin_lock(&fi->lock);
 	if (fi->local_int[cpu_addr] == NULL)
 		rc = 3; /* not operational */
-	else if (atomic_read(fi->local_int[cpu_addr]->cpuflags)
-		 & CPUSTAT_RUNNING) {
+	else if (!(atomic_read(fi->local_int[cpu_addr]->cpuflags)
+		  & CPUSTAT_STOPPED)) {
 		*reg &= 0xffffffff00000000UL;
 		rc = 1; /* status stored */
 	} else {
@@ -251,7 +253,7 @@ static int __sigp_set_prefix(struct kvm_vcpu *vcpu, u16 cpu_addr, u32 address,
 
 	spin_lock_bh(&li->lock);
 	/* cpu must be in stopped state */
-	if (atomic_read(li->cpuflags) & CPUSTAT_RUNNING) {
+	if (!(atomic_read(li->cpuflags) & CPUSTAT_STOPPED)) {
 		rc = 1; /* incorrect state */
 		*reg &= SIGP_STAT_INCORRECT_STATE;
 		kfree(inti);
@@ -275,6 +277,38 @@ out_fi:
 	return rc;
 }
 
+static int __sigp_sense_running(struct kvm_vcpu *vcpu, u16 cpu_addr,
+				unsigned long *reg)
+{
+	int rc;
+	struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int;
+
+	if (cpu_addr >= KVM_MAX_VCPUS)
+		return 3; /* not operational */
+
+	spin_lock(&fi->lock);
+	if (fi->local_int[cpu_addr] == NULL)
+		rc = 3; /* not operational */
+	else {
+		if (atomic_read(fi->local_int[cpu_addr]->cpuflags)
+		    & CPUSTAT_RUNNING) {
+			/* running */
+			rc = 1;
+		} else {
+			/* not running */
+			*reg &= 0xffffffff00000000UL;
+			*reg |= SIGP_STAT_NOT_RUNNING;
+			rc = 0;
+		}
+	}
+	spin_unlock(&fi->lock);
+
+	VCPU_EVENT(vcpu, 4, "sensed running status of cpu %x rc %x", cpu_addr,
+		   rc);
+
+	return rc;
+}
+
 int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu)
 {
 	int r1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4;
@@ -331,6 +365,11 @@ int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu)
 		rc = __sigp_set_prefix(vcpu, cpu_addr, parameter,
 				       &vcpu->arch.guest_gprs[r1]);
 		break;
+	case SIGP_SENSE_RUNNING:
+		vcpu->stat.instruction_sigp_sense_running++;
+		rc = __sigp_sense_running(vcpu, cpu_addr,
+					  &vcpu->arch.guest_gprs[r1]);
+		break;
 	case SIGP_RESTART:
 		vcpu->stat.instruction_sigp_restart++;
 		/* user space must know about restart */
diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index 1766def5bc3f..a9a301866b3c 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -587,8 +587,13 @@ static void pfault_interrupt(unsigned int ext_int_code,
 		} else {
 			/* Completion interrupt was faster than initial
 			 * interrupt. Set pfault_wait to -1 so the initial
-			 * interrupt doesn't put the task to sleep. */
-			tsk->thread.pfault_wait = -1;
+			 * interrupt doesn't put the task to sleep.
+			 * If the task is not running, ignore the completion
+			 * interrupt since it must be a leftover of a PFAULT
+			 * CANCEL operation which didn't remove all pending
+			 * completion interrupts. */
+			if (tsk->state == TASK_RUNNING)
+				tsk->thread.pfault_wait = -1;
 		}
 		put_task_struct(tsk);
 	} else {