Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull s390 updates from Martin Schwidefsky:

 - three merges for KVM/s390 with changes for vfio-ccw and cpacf. The
   patches are included in the KVM tree as well, let git sort it out.

 - add the new 'trng' random number generator

 - provide the secure key verification API for the pkey interface

 - introduce the z13 cpu counters to perf

 - add a new system call to set up the guarded storage facility

 - simplify TASK_SIZE and arch_get_unmapped_area

 - export the raw STSI data related to CPU topology to user space

 - ... and the usual churn of bug-fixes and cleanups.

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (74 commits)
  s390/crypt: use the correct module alias for paes_s390.
  s390/cpacf: Introduce kma instruction
  s390/cpacf: query instructions use unique parameters for compatibility with KMA
  s390/trng: Introduce s390 TRNG device driver.
  s390/crypto: Provide s390 specific arch random functionality.
  s390/crypto: Add new subfunctions to the cpacf PRNO function.
  s390/crypto: Renaming PPNO to PRNO.
  s390/pageattr: avoid unnecessary page table splitting
  s390/mm: simplify arch_get_unmapped_area[_topdown]
  s390/mm: make TASK_SIZE independent from the number of page table levels
  s390/gs: add regset for the guarded storage broadcast control block
  s390/kvm: Add use_cmma field to mm_context_t
  s390/kvm: Add PGSTE manipulation functions
  vfio: ccw: improve error handling for vfio_ccw_mdev_remove
  vfio: ccw: remove unnecessary NULL checks of a pointer
  s390/spinlock: remove compare and delay instruction
  s390/spinlock: use atomic primitives for spinlocks
  s390/cpumf: simplify detection of guest samples
  s390/pci: remove forward declaration
  s390/pci: increase the PCI_NR_FUNCTIONS default
  ...

24 files changed, 761 insertions, 142 deletions

diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile
index 060ce54..adb3fe2 100644
--- a/arch/s390/kernel/Makefile
+++ b/arch/s390/kernel/Makefile
@@ -51,14 +51,12 @@ CFLAGS_dumpstack.o	+= -fno-optimize-sibling-calls
 #
 CFLAGS_ptrace.o		+= -DUTS_MACHINE='"$(UTS_MACHINE)"'
 
-CFLAGS_sysinfo.o	+= -w
-
 obj-y	:= traps.o time.o process.o base.o early.o setup.o idle.o vtime.o
 obj-y	+= processor.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o nmi.o
 obj-y	+= debug.o irq.o ipl.o dis.o diag.o vdso.o als.o
 obj-y	+= sysinfo.o jump_label.o lgr.o os_info.o machine_kexec.o pgm_check.o
-obj-y	+= runtime_instr.o cache.o fpu.o dumpstack.o
-obj-y	+= entry.o reipl.o relocate_kernel.o
+obj-y	+= runtime_instr.o cache.o fpu.o dumpstack.o guarded_storage.o
+obj-y	+= entry.o reipl.o relocate_kernel.o kdebugfs.o
 
 extra-y				+= head.o head64.o vmlinux.lds
 
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index c4b3570..6bb2963 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -175,7 +175,7 @@ int main(void)
 	/* software defined ABI-relevant lowcore locations 0xe00 - 0xe20 */
 	OFFSET(__LC_DUMP_REIPL, lowcore, ipib);
 	/* hardware defined lowcore locations 0x1000 - 0x18ff */
-	OFFSET(__LC_VX_SAVE_AREA_ADDR, lowcore, vector_save_area_addr);
+	OFFSET(__LC_MCESAD, lowcore, mcesad);
 	OFFSET(__LC_EXT_PARAMS2, lowcore, ext_params2);
 	OFFSET(__LC_FPREGS_SAVE_AREA, lowcore, floating_pt_save_area);
 	OFFSET(__LC_GPREGS_SAVE_AREA, lowcore, gpregs_save_area);
diff --git a/arch/s390/kernel/compat_wrapper.c b/arch/s390/kernel/compat_wrapper.c
index e89cc2e..986642a 100644
--- a/arch/s390/kernel/compat_wrapper.c
+++ b/arch/s390/kernel/compat_wrapper.c
@@ -178,4 +178,5 @@ COMPAT_SYSCALL_WRAP3(getpeername, int, fd, struct sockaddr __user *, usockaddr,
 COMPAT_SYSCALL_WRAP6(sendto, int, fd, void __user *, buff, size_t, len, unsigned int, flags, struct sockaddr __user *, addr, int, addr_len);
 COMPAT_SYSCALL_WRAP3(mlock2, unsigned long, start, size_t, len, int, flags);
 COMPAT_SYSCALL_WRAP6(copy_file_range, int, fd_in, loff_t __user *, off_in, int, fd_out, loff_t __user *, off_out, size_t, len, unsigned int, flags);
+COMPAT_SYSCALL_WRAP2(s390_guarded_storage, int, command, struct gs_cb *, gs_cb);
 COMPAT_SYSCALL_WRAP5(statx, int, dfd, const char __user *, path, unsigned, flags, unsigned, mask, struct statx __user *, buffer);
diff --git a/arch/s390/kernel/crash_dump.c b/arch/s390/kernel/crash_dump.c
index dd1d5c6..d628afc 100644
--- a/arch/s390/kernel/crash_dump.c
+++ b/arch/s390/kernel/crash_dump.c
@@ -429,6 +429,20 @@ static void *nt_vmcoreinfo(void *ptr)
 }
 
 /*
+ * Initialize final note (needed for /proc/vmcore code)
+ */
+static void *nt_final(void *ptr)
+{
+	Elf64_Nhdr *note;
+
+	note = (Elf64_Nhdr *) ptr;
+	note->n_namesz = 0;
+	note->n_descsz = 0;
+	note->n_type = 0;
+	return PTR_ADD(ptr, sizeof(Elf64_Nhdr));
+}
+
+/*
  * Initialize ELF header (new kernel)
  */
 static void *ehdr_init(Elf64_Ehdr *ehdr, int mem_chunk_cnt)
@@ -515,6 +529,7 @@ static void *notes_init(Elf64_Phdr *phdr, void *ptr, u64 notes_offset)
 		if (sa->prefix != 0)
 			ptr = fill_cpu_elf_notes(ptr, cpu++, sa);
 	ptr = nt_vmcoreinfo(ptr);
+	ptr = nt_final(ptr);
 	memset(phdr, 0, sizeof(*phdr));
 	phdr->p_type = PT_NOTE;
 	phdr->p_offset = notes_offset;
diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index 4e65c79..5d20182 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -231,9 +231,29 @@ static noinline __init void detect_machine_type(void)
 		S390_lowcore.machine_flags |= MACHINE_FLAG_VM;
 }
 
+/* Remove leading, trailing and double whitespace. */
+static inline void strim_all(char *str)
+{
+	char *s;
+
+	s = strim(str);
+	if (s != str)
+		memmove(str, s, strlen(s));
+	while (*str) {
+		if (!isspace(*str++))
+			continue;
+		if (isspace(*str)) {
+			s = skip_spaces(str);
+			memmove(str, s, strlen(s) + 1);
+		}
+	}
+}
+
 static noinline __init void setup_arch_string(void)
 {
 	struct sysinfo_1_1_1 *mach = (struct sysinfo_1_1_1 *)&sysinfo_page;
+	struct sysinfo_3_2_2 *vm = (struct sysinfo_3_2_2 *)&sysinfo_page;
+	char mstr[80], hvstr[17];
 
 	if (stsi(mach, 1, 1, 1))
 		return;
@@ -241,14 +261,21 @@ static noinline __init void setup_arch_string(void)
 	EBCASC(mach->type, sizeof(mach->type));
 	EBCASC(mach->model, sizeof(mach->model));
 	EBCASC(mach->model_capacity, sizeof(mach->model_capacity));
-	dump_stack_set_arch_desc("%-16.16s %-4.4s %-16.16s %-16.16s (%s)",
-				 mach->manufacturer,
-				 mach->type,
-				 mach->model,
-				 mach->model_capacity,
-				 MACHINE_IS_LPAR ? "LPAR" :
-				 MACHINE_IS_VM ? "z/VM" :
-				 MACHINE_IS_KVM ? "KVM" : "unknown");
+	sprintf(mstr, "%-16.16s %-4.4s %-16.16s %-16.16s",
+		mach->manufacturer, mach->type,
+		mach->model, mach->model_capacity);
+	strim_all(mstr);
+	if (stsi(vm, 3, 2, 2) == 0 && vm->count) {
+		EBCASC(vm->vm[0].cpi, sizeof(vm->vm[0].cpi));
+		sprintf(hvstr, "%-16.16s", vm->vm[0].cpi);
+		strim_all(hvstr);
+	} else {
+		sprintf(hvstr, "%s",
+			MACHINE_IS_LPAR ? "LPAR" :
+			MACHINE_IS_VM ? "z/VM" :
+			MACHINE_IS_KVM ? "KVM" : "unknown");
+	}
+	dump_stack_set_arch_desc("%s (%s)", mstr, hvstr);
 }
 
 static __init void setup_topology(void)
@@ -358,6 +385,8 @@ static __init void detect_machine_facilities(void)
 		S390_lowcore.machine_flags |= MACHINE_FLAG_NX;
 		__ctl_set_bit(0, 20);
 	}
+	if (test_facility(133))
+		S390_lowcore.machine_flags |= MACHINE_FLAG_GS;
 }
 
 static inline void save_vector_registers(void)
@@ -375,7 +404,7 @@ static int __init topology_setup(char *str)
 
 	rc = kstrtobool(str, &enabled);
 	if (!rc && !enabled)
-		S390_lowcore.machine_flags &= ~MACHINE_HAS_TOPOLOGY;
+		S390_lowcore.machine_flags &= ~MACHINE_FLAG_TOPOLOGY;
 	return rc;
 }
 early_param("topology", topology_setup);
@@ -405,23 +434,16 @@ early_param("noexec", noexec_setup);
 
 static int __init cad_setup(char *str)
 {
-	int val;
-
-	get_option(&str, &val);
-	if (val && test_facility(128))
-		S390_lowcore.machine_flags |= MACHINE_FLAG_CAD;
-	return 0;
-}
-early_param("cad", cad_setup);
+	bool enabled;
+	int rc;
 
-static int __init cad_init(void)
-{
-	if (MACHINE_HAS_CAD)
+	rc = kstrtobool(str, &enabled);
+	if (!rc && enabled && test_facility(128))
 		/* Enable problem state CAD. */
 		__ctl_set_bit(2, 3);
-	return 0;
+	return rc;
 }
-early_initcall(cad_init);
+early_param("cad", cad_setup);
 
 static __init void memmove_early(void *dst, const void *src, size_t n)
 {
diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index 6a7d737..c6cf338 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -47,7 +47,7 @@ STACK_SIZE  = 1 << STACK_SHIFT
 STACK_INIT = STACK_SIZE - STACK_FRAME_OVERHEAD - __PT_SIZE
 
 _TIF_WORK	= (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \
-		   _TIF_UPROBE)
+		   _TIF_UPROBE | _TIF_GUARDED_STORAGE)
 _TIF_TRACE	= (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | \
 		   _TIF_SYSCALL_TRACEPOINT)
 _CIF_WORK	= (_CIF_MCCK_PENDING | _CIF_ASCE_PRIMARY | \
@@ -189,8 +189,6 @@ ENTRY(__switch_to)
 	stg	%r3,__LC_CURRENT		# store task struct of next
 	stg	%r15,__LC_KERNEL_STACK		# store end of kernel stack
 	lg	%r15,__THREAD_ksp(%r1)		# load kernel stack of next
-	/* c4 is used in guest detection: arch/s390/kernel/perf_cpum_sf.c */
-	lctl	%c4,%c4,__TASK_pid(%r3)		# load pid to control reg. 4
 	mvc	__LC_CURRENT_PID(4,%r0),__TASK_pid(%r3) # store pid of next
 	lmg	%r6,%r15,__SF_GPRS(%r15)	# load gprs of next task
 	TSTMSK	__LC_MACHINE_FLAGS,MACHINE_FLAG_LPP
@@ -332,6 +330,8 @@ ENTRY(system_call)
 	TSTMSK	__TI_flags(%r12),_TIF_UPROBE
 	jo	.Lsysc_uprobe_notify
 #endif
+	TSTMSK	__TI_flags(%r12),_TIF_GUARDED_STORAGE
+	jo	.Lsysc_guarded_storage
 	TSTMSK	__PT_FLAGS(%r11),_PIF_PER_TRAP
 	jo	.Lsysc_singlestep
 	TSTMSK	__TI_flags(%r12),_TIF_SIGPENDING
@@ -409,6 +409,14 @@ ENTRY(system_call)
 #endif
 
 #
+# _TIF_GUARDED_STORAGE is set, call guarded_storage_load
+#
+.Lsysc_guarded_storage:
+	lgr	%r2,%r11		# pass pointer to pt_regs
+	larl	%r14,.Lsysc_return
+	jg	gs_load_bc_cb
+
+#
 # _PIF_PER_TRAP is set, call do_per_trap
 #
 .Lsysc_singlestep:
@@ -663,6 +671,8 @@ ENTRY(io_int_handler)
 	jo	.Lio_sigpending
 	TSTMSK	__TI_flags(%r12),_TIF_NOTIFY_RESUME
 	jo	.Lio_notify_resume
+	TSTMSK	__TI_flags(%r12),_TIF_GUARDED_STORAGE
+	jo	.Lio_guarded_storage
 	TSTMSK	__LC_CPU_FLAGS,_CIF_FPU
 	jo	.Lio_vxrs
 	TSTMSK	__LC_CPU_FLAGS,(_CIF_ASCE_PRIMARY|_CIF_ASCE_SECONDARY)
@@ -697,6 +707,18 @@ ENTRY(io_int_handler)
 	jg	load_fpu_regs
 
 #
+# _TIF_GUARDED_STORAGE is set, call guarded_storage_load
+#
+.Lio_guarded_storage:
+	# TRACE_IRQS_ON already done at .Lio_return
+	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
+	lgr	%r2,%r11		# pass pointer to pt_regs
+	brasl	%r14,gs_load_bc_cb
+	ssm	__LC_PGM_NEW_PSW	# disable I/O and ext. interrupts
+	TRACE_IRQS_OFF
+	j	.Lio_return
+
+#
 # _TIF_NEED_RESCHED is set, call schedule
 #
 .Lio_reschedule:
diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h
index 33f9018..dbf5f7e 100644
--- a/arch/s390/kernel/entry.h
+++ b/arch/s390/kernel/entry.h
@@ -74,12 +74,14 @@ long sys_sigreturn(void);
 
 long sys_s390_personality(unsigned int personality);
 long sys_s390_runtime_instr(int command, int signum);
+long sys_s390_guarded_storage(int command, struct gs_cb __user *);
 long sys_s390_pci_mmio_write(unsigned long, const void __user *, size_t);
 long sys_s390_pci_mmio_read(unsigned long, void __user *, size_t);
 
 DECLARE_PER_CPU(u64, mt_cycles[8]);
 
 void verify_facilities(void);
+void gs_load_bc_cb(struct pt_regs *regs);
 void set_fs_fixup(void);
 
 #endif /* _ENTRY_H */
diff --git a/arch/s390/kernel/guarded_storage.c b/arch/s390/kernel/guarded_storage.c
new file mode 100644
index 0000000..6f06474
--- /dev/null
+++ b/arch/s390/kernel/guarded_storage.c
@@ -0,0 +1,128 @@
+/*
+ * Copyright IBM Corp. 2016
+ * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/syscalls.h>
+#include <linux/signal.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <asm/guarded_storage.h>
+#include "entry.h"
+
+void exit_thread_gs(void)
+{
+	kfree(current->thread.gs_cb);
+	kfree(current->thread.gs_bc_cb);
+	current->thread.gs_cb = current->thread.gs_bc_cb = NULL;
+}
+
+static int gs_enable(void)
+{
+	struct gs_cb *gs_cb;
+
+	if (!current->thread.gs_cb) {
+		gs_cb = kzalloc(sizeof(*gs_cb), GFP_KERNEL);
+		if (!gs_cb)
+			return -ENOMEM;
+		gs_cb->gsd = 25;
+		preempt_disable();
+		__ctl_set_bit(2, 4);
+		load_gs_cb(gs_cb);
+		current->thread.gs_cb = gs_cb;
+		preempt_enable();
+	}
+	return 0;
+}
+
+static int gs_disable(void)
+{
+	if (current->thread.gs_cb) {
+		preempt_disable();
+		kfree(current->thread.gs_cb);
+		current->thread.gs_cb = NULL;
+		__ctl_clear_bit(2, 4);
+		preempt_enable();
+	}
+	return 0;
+}
+
+static int gs_set_bc_cb(struct gs_cb __user *u_gs_cb)
+{
+	struct gs_cb *gs_cb;
+
+	gs_cb = current->thread.gs_bc_cb;
+	if (!gs_cb) {
+		gs_cb = kzalloc(sizeof(*gs_cb), GFP_KERNEL);
+		if (!gs_cb)
+			return -ENOMEM;
+		current->thread.gs_bc_cb = gs_cb;
+	}
+	if (copy_from_user(gs_cb, u_gs_cb, sizeof(*gs_cb)))
+		return -EFAULT;
+	return 0;
+}
+
+static int gs_clear_bc_cb(void)
+{
+	struct gs_cb *gs_cb;
+
+	gs_cb = current->thread.gs_bc_cb;
+	current->thread.gs_bc_cb = NULL;
+	kfree(gs_cb);
+	return 0;
+}
+
+void gs_load_bc_cb(struct pt_regs *regs)
+{
+	struct gs_cb *gs_cb;
+
+	preempt_disable();
+	clear_thread_flag(TIF_GUARDED_STORAGE);
+	gs_cb = current->thread.gs_bc_cb;
+	if (gs_cb) {
+		kfree(current->thread.gs_cb);
+		current->thread.gs_bc_cb = NULL;
+		__ctl_set_bit(2, 4);
+		load_gs_cb(gs_cb);
+		current->thread.gs_cb = gs_cb;
+	}
+	preempt_enable();
+}
+
+static int gs_broadcast(void)
+{
+	struct task_struct *sibling;
+
+	read_lock(&tasklist_lock);
+	for_each_thread(current, sibling) {
+		if (!sibling->thread.gs_bc_cb)
+			continue;
+		if (test_and_set_tsk_thread_flag(sibling, TIF_GUARDED_STORAGE))
+			kick_process(sibling);
+	}
+	read_unlock(&tasklist_lock);
+	return 0;
+}
+
+SYSCALL_DEFINE2(s390_guarded_storage, int, command,
+		struct gs_cb __user *, gs_cb)
+{
+	if (!MACHINE_HAS_GS)
+		return -EOPNOTSUPP;
+	switch (command) {
+	case GS_ENABLE:
+		return gs_enable();
+	case GS_DISABLE:
+		return gs_disable();
+	case GS_SET_BC_CB:
+		return gs_set_bc_cb(gs_cb);
+	case GS_CLEAR_BC_CB:
+		return gs_clear_bc_cb();
+	case GS_BROADCAST:
+		return gs_broadcast();
+	default:
+		return -EINVAL;
+	}
+}
diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S
index 0b5ebf8..eff5b31 100644
--- a/arch/s390/kernel/head.S
+++ b/arch/s390/kernel/head.S
@@ -25,7 +25,6 @@
 #include <linux/linkage.h>
 #include <asm/asm-offsets.h>
 #include <asm/thread_info.h>
-#include <asm/facility.h>
 #include <asm/page.h>
 #include <asm/ptrace.h>
 
diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index 482d352..31c91f2 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -52,7 +52,7 @@ ENTRY(startup_continue)
 	.quad	0			# cr1: primary space segment table
 	.quad	.Lduct			# cr2: dispatchable unit control table
 	.quad	0			# cr3: instruction authorization
-	.quad	0			# cr4: instruction authorization
+	.quad	0xffff			# cr4: instruction authorization
 	.quad	.Lduct			# cr5: primary-aste origin
 	.quad	0			# cr6:	I/O interrupts
 	.quad	0			# cr7:	secondary space segment table
diff --git a/arch/s390/kernel/kdebugfs.c b/arch/s390/kernel/kdebugfs.c
new file mode 100644
index 0000000..ee85e17
--- /dev/null
+++ b/arch/s390/kernel/kdebugfs.c
@@ -0,0 +1,15 @@
+#include <linux/debugfs.h>
+#include <linux/export.h>
+#include <linux/init.h>
+
+struct dentry *arch_debugfs_dir;
+EXPORT_SYMBOL(arch_debugfs_dir);
+
+static int __init arch_kdebugfs_init(void)
+{
+	arch_debugfs_dir = debugfs_create_dir("s390", NULL);
+	if (IS_ERR(arch_debugfs_dir))
+		arch_debugfs_dir = NULL;
+	return 0;
+}
+postcore_initcall(arch_kdebugfs_init);
diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c
index 3074c1d..db5658d 100644
--- a/arch/s390/kernel/machine_kexec.c
+++ b/arch/s390/kernel/machine_kexec.c
@@ -27,6 +27,7 @@
 #include <asm/cacheflush.h>
 #include <asm/os_info.h>
 #include <asm/switch_to.h>
+#include <asm/nmi.h>
 
 typedef void (*relocate_kernel_t)(kimage_entry_t *, unsigned long);
 
@@ -102,6 +103,8 @@ static void __do_machine_kdump(void *image)
  */
 static noinline void __machine_kdump(void *image)
 {
+	struct mcesa *mcesa;
+	unsigned long cr2_old, cr2_new;
 	int this_cpu, cpu;
 
 	lgr_info_log();
@@ -114,8 +117,16 @@ static noinline void __machine_kdump(void *image)
 			continue;
 	}
 	/* Store status of the boot CPU */
+	mcesa = (struct mcesa *)(S390_lowcore.mcesad & MCESA_ORIGIN_MASK);
 	if (MACHINE_HAS_VX)
-		save_vx_regs((void *) &S390_lowcore.vector_save_area);
+		save_vx_regs((__vector128 *) mcesa->vector_save_area);
+	if (MACHINE_HAS_GS) {
+		__ctl_store(cr2_old, 2, 2);
+		cr2_new = cr2_old | (1UL << 4);
+		__ctl_load(cr2_new, 2, 2);
+		save_gs_cb((struct gs_cb *) mcesa->guarded_storage_save_area);
+		__ctl_load(cr2_old, 2, 2);
+	}
 	/*
 	 * To create a good backchain for this CPU in the dump store_status
 	 * is passed the address of a function. The address is saved into
diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c
index 9bf8327..9855895 100644
--- a/arch/s390/kernel/nmi.c
+++ b/arch/s390/kernel/nmi.c
@@ -106,6 +106,7 @@ static int notrace s390_validate_registers(union mci mci, int umode)
 	int kill_task;
 	u64 zero;
 	void *fpt_save_area;
+	struct mcesa *mcesa;
 
 	kill_task = 0;
 	zero = 0;
@@ -165,6 +166,7 @@ static int notrace s390_validate_registers(union mci mci, int umode)
 			     : : "Q" (S390_lowcore.fpt_creg_save_area));
 	}
 
+	mcesa = (struct mcesa *)(S390_lowcore.mcesad & MCESA_ORIGIN_MASK);
 	if (!MACHINE_HAS_VX) {
 		/* Validate floating point registers */
 		asm volatile(
@@ -209,8 +211,8 @@ static int notrace s390_validate_registers(union mci mci, int umode)
 			"	la	1,%0\n"
 			"	.word	0xe70f,0x1000,0x0036\n"	/* vlm 0,15,0(1) */
 			"	.word	0xe70f,0x1100,0x0c36\n"	/* vlm 16,31,256(1) */
-			: : "Q" (*(struct vx_array *)
-				 &S390_lowcore.vector_save_area) : "1");
+			: : "Q" (*(struct vx_array *) mcesa->vector_save_area)
+			: "1");
 		__ctl_load(S390_lowcore.cregs_save_area[0], 0, 0);
 	}
 	/* Validate access registers */
@@ -224,6 +226,19 @@ static int notrace s390_validate_registers(union mci mci, int umode)
 		 */
 		kill_task = 1;
 	}
+	/* Validate guarded storage registers */
+	if (MACHINE_HAS_GS && (S390_lowcore.cregs_save_area[2] & (1UL << 4))) {
+		if (!mci.gs)
+			/*
+			 * Guarded storage register can't be restored and
+			 * the current processes uses guarded storage.
+			 * It has to be terminated.
+			 */
+			kill_task = 1;
+		else
+			load_gs_cb((struct gs_cb *)
+				   mcesa->guarded_storage_save_area);
+	}
 	/*
 	 * We don't even try to validate the TOD register, since we simply
 	 * can't write something sensible into that register.
diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c
index 1aba10e..746d034 100644
--- a/arch/s390/kernel/perf_cpum_cf.c
+++ b/arch/s390/kernel/perf_cpum_cf.c
@@ -1,7 +1,7 @@
 /*
  * Performance event support for s390x - CPU-measurement Counter Facility
  *
- *  Copyright IBM Corp. 2012
+ *  Copyright IBM Corp. 2012, 2017
  *  Author(s): Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
  *
  * This program is free software; you can redistribute it and/or modify
@@ -22,19 +22,12 @@
 #include <asm/irq.h>
 #include <asm/cpu_mf.h>
 
-/* CPU-measurement counter facility supports these CPU counter sets:
- * For CPU counter sets:
- *    Basic counter set:	     0-31
- *    Problem-state counter set:    32-63
- *    Crypto-activity counter set:  64-127
- *    Extented counter set:	   128-159
- */
 enum cpumf_ctr_set {
-	/* CPU counter sets */
-	CPUMF_CTR_SET_BASIC   = 0,
-	CPUMF_CTR_SET_USER    = 1,
-	CPUMF_CTR_SET_CRYPTO  = 2,
-	CPUMF_CTR_SET_EXT     = 3,
+	CPUMF_CTR_SET_BASIC   = 0,    /* Basic Counter Set */
+	CPUMF_CTR_SET_USER    = 1,    /* Problem-State Counter Set */
+	CPUMF_CTR_SET_CRYPTO  = 2,    /* Crypto-Activity Counter Set */
+	CPUMF_CTR_SET_EXT     = 3,    /* Extended Counter Set */
+	CPUMF_CTR_SET_MT_DIAG = 4,    /* MT-diagnostic Counter Set */
 
 	/* Maximum number of counter sets */
 	CPUMF_CTR_SET_MAX,
@@ -47,6 +40,7 @@ static const u64 cpumf_state_ctl[CPUMF_CTR_SET_MAX] = {
 	[CPUMF_CTR_SET_USER]	= 0x04,
 	[CPUMF_CTR_SET_CRYPTO]	= 0x08,
 	[CPUMF_CTR_SET_EXT]	= 0x01,
+	[CPUMF_CTR_SET_MT_DIAG] = 0x20,
 };
 
 static void ctr_set_enable(u64 *state, int ctr_set)
@@ -76,19 +70,20 @@ struct cpu_hw_events {
 };
 static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
 	.ctr_set = {
-		[CPUMF_CTR_SET_BASIC]  = ATOMIC_INIT(0),
-		[CPUMF_CTR_SET_USER]   = ATOMIC_INIT(0),
-		[CPUMF_CTR_SET_CRYPTO] = ATOMIC_INIT(0),
-		[CPUMF_CTR_SET_EXT]    = ATOMIC_INIT(0),
+		[CPUMF_CTR_SET_BASIC]	= ATOMIC_INIT(0),
+		[CPUMF_CTR_SET_USER]	= ATOMIC_INIT(0),
+		[CPUMF_CTR_SET_CRYPTO]	= ATOMIC_INIT(0),
+		[CPUMF_CTR_SET_EXT]	= ATOMIC_INIT(0),
+		[CPUMF_CTR_SET_MT_DIAG] = ATOMIC_INIT(0),
 	},
 	.state = 0,
 	.flags = 0,
 	.txn_flags = 0,
 };
 
-static int get_counter_set(u64 event)
+static enum cpumf_ctr_set get_counter_set(u64 event)
 {
-	int set = -1;
+	int set = CPUMF_CTR_SET_MAX;
 
 	if (event < 32)
 		set = CPUMF_CTR_SET_BASIC;
@@ -98,34 +93,17 @@ static int get_counter_set(u64 event)
 		set = CPUMF_CTR_SET_CRYPTO;
 	else if (event < 256)
 		set = CPUMF_CTR_SET_EXT;
+	else if (event >= 448 && event < 496)
+		set = CPUMF_CTR_SET_MT_DIAG;
 
 	return set;
 }
 
-static int validate_event(const struct hw_perf_event *hwc)
-{
-	switch (hwc->config_base) {
-	case CPUMF_CTR_SET_BASIC:
-	case CPUMF_CTR_SET_USER:
-	case CPUMF_CTR_SET_CRYPTO:
-	case CPUMF_CTR_SET_EXT:
-		/* check for reserved counters */
-		if ((hwc->config >=  6 && hwc->config <=  31) ||
-		    (hwc->config >= 38 && hwc->config <=  63) ||
-		    (hwc->config >= 80 && hwc->config <= 127))
-			return -EOPNOTSUPP;
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
 static int validate_ctr_version(const struct hw_perf_event *hwc)
 {
 	struct cpu_hw_events *cpuhw;
 	int err = 0;
+	u16 mtdiag_ctl;
 
 	cpuhw = &get_cpu_var(cpu_hw_events);
 
@@ -145,6 +123,27 @@ static int validate_ctr_version(const struct hw_perf_event *hwc)
 		    (cpuhw->info.csvn  > 2 && hwc->config > 255))
 			err = -EOPNOTSUPP;
 		break;
+	case CPUMF_CTR_SET_MT_DIAG:
+		if (cpuhw->info.csvn <= 3)
+			err = -EOPNOTSUPP;
+		/*
+		 * MT-diagnostic counters are read-only.  The counter set
+		 * is automatically enabled and activated on all CPUs with
+		 * multithreading (SMT).  Deactivation of multithreading
+		 * also disables the counter set.  State changes are ignored
+		 * by lcctl().	Because Linux controls SMT enablement through
+		 * a kernel parameter only, the counter set is either disabled
+		 * or enabled and active.
+		 *
+		 * Thus, the counters can only be used if SMT is on and the
+		 * counter set is enabled and active.
+		 */
+		mtdiag_ctl = cpumf_state_ctl[CPUMF_CTR_SET_MT_DIAG];
+		if (!((cpuhw->info.auth_ctl & mtdiag_ctl) &&
+		      (cpuhw->info.enable_ctl & mtdiag_ctl) &&
+		      (cpuhw->info.act_ctl & mtdiag_ctl)))
+			err = -EOPNOTSUPP;
+		break;
 	}
 
 	put_cpu_var(cpu_hw_events);
@@ -250,6 +249,11 @@ static void cpumf_measurement_alert(struct ext_code ext_code,
 	/* loss of counter data alert */
 	if (alert & CPU_MF_INT_CF_LCDA)
 		pr_err("CPU[%i] Counter data was lost\n", smp_processor_id());
+
+	/* loss of MT counter data alert */
+	if (alert & CPU_MF_INT_CF_MTDA)
+		pr_warn("CPU[%i] MT counter data was lost\n",
+			smp_processor_id());
 }
 
 #define PMC_INIT      0
@@ -330,6 +334,7 @@ static int __hw_perf_event_init(struct perf_event *event)
 {
 	struct perf_event_attr *attr = &event->attr;
 	struct hw_perf_event *hwc = &event->hw;
+	enum cpumf_ctr_set set;
 	int err;
 	u64 ev;
 
@@ -370,25 +375,30 @@ static int __hw_perf_event_init(struct perf_event *event)
 	if (ev == -1)
 		return -ENOENT;
 
-	if (ev >= PERF_CPUM_CF_MAX_CTR)
+	if (ev > PERF_CPUM_CF_MAX_CTR)
 		return -EINVAL;
 
-	/* Use the hardware perf event structure to store the counter number
-	 * in 'config' member and the counter set to which the counter belongs
-	 * in the 'config_base'.  The counter set (config_base) is then used
-	 * to enable/disable the counters.
-	 */
-	hwc->config = ev;
-	hwc->config_base = get_counter_set(ev);
-
-	/* Validate the counter that is assigned to this event.
-	 * Because the counter facility can use numerous counters at the
-	 * same time without constraints, it is not necessary to explicitly
-	 * validate event groups (event->group_leader != event).
-	 */
-	err = validate_event(hwc);
-	if (err)
-		return err;
+	/* Obtain the counter set to which the specified counter belongs */
+	set = get_counter_set(ev);
+	switch (set) {
+	case CPUMF_CTR_SET_BASIC:
+	case CPUMF_CTR_SET_USER:
+	case CPUMF_CTR_SET_CRYPTO:
+	case CPUMF_CTR_SET_EXT:
+	case CPUMF_CTR_SET_MT_DIAG:
+		/*
+		 * Use the hardware perf event structure to store the
+		 * counter number in the 'config' member and the counter
+		 * set number in the 'config_base'.  The counter set number
+		 * is then later used to enable/disable the counter(s).
+		 */
+		hwc->config = ev;
+		hwc->config_base = set;
+		break;
+	case CPUMF_CTR_SET_MAX:
+		/* The counter could not be associated to a counter set */
+		return -EINVAL;
+	};
 
 	/* Initialize for using the CPU-measurement counter facility */
 	if (!atomic_inc_not_zero(&num_events)) {
@@ -452,7 +462,7 @@ static int hw_perf_event_reset(struct perf_event *event)
 	return err;
 }
 
-static int hw_perf_event_update(struct perf_event *event)
+static void hw_perf_event_update(struct perf_event *event)
 {
 	u64 prev, new, delta;
 	int err;
@@ -461,14 +471,12 @@ static int hw_perf_event_update(struct perf_event *event)
 		prev = local64_read(&event->hw.prev_count);
 		err = ecctr(event->hw.config, &new);
 		if (err)
-			goto out;
+			return;
 	} while (local64_cmpxchg(&event->hw.prev_count, prev, new) != prev);
 
 	delta = (prev <= new) ? new - prev
 			      : (-1ULL - prev) + new + 1;	 /* overflow */
 	local64_add(delta, &event->count);
-out:
-	return err;
 }
 
 static void cpumf_pmu_read(struct perf_event *event)
diff --git a/arch/s390/kernel/perf_cpum_cf_events.c b/arch/s390/kernel/perf_cpum_cf_events.c
index c343ac2..d313328 100644
--- a/arch/s390/kernel/perf_cpum_cf_events.c
+++ b/arch/s390/kernel/perf_cpum_cf_events.c
@@ -114,8 +114,64 @@ CPUMF_EVENT_ATTR(cf_zec12, L1I_OFFBOOK_L3_SOURCED_WRITES_IV, 0x00a1);
 CPUMF_EVENT_ATTR(cf_zec12, TX_NC_TABORT, 0x00b1);
 CPUMF_EVENT_ATTR(cf_zec12, TX_C_TABORT_NO_SPECIAL, 0x00b2);
 CPUMF_EVENT_ATTR(cf_zec12, TX_C_TABORT_SPECIAL, 0x00b3);
+CPUMF_EVENT_ATTR(cf_z13, L1D_WRITES_RO_EXCL, 0x0080);
+CPUMF_EVENT_ATTR(cf_z13, DTLB1_WRITES, 0x0081);
+CPUMF_EVENT_ATTR(cf_z13, DTLB1_MISSES, 0x0082);
+CPUMF_EVENT_ATTR(cf_z13, DTLB1_HPAGE_WRITES, 0x0083);
+CPUMF_EVENT_ATTR(cf_z13, DTLB1_GPAGE_WRITES, 0x0084);
+CPUMF_EVENT_ATTR(cf_z13, L1D_L2D_SOURCED_WRITES, 0x0085);
+CPUMF_EVENT_ATTR(cf_z13, ITLB1_WRITES, 0x0086);
+CPUMF_EVENT_ATTR(cf_z13, ITLB1_MISSES, 0x0087);
+CPUMF_EVENT_ATTR(cf_z13, L1I_L2I_SOURCED_WRITES, 0x0088);
+CPUMF_EVENT_ATTR(cf_z13, TLB2_PTE_WRITES, 0x0089);
+CPUMF_EVENT_ATTR(cf_z13, TLB2_CRSTE_HPAGE_WRITES, 0x008a);
+CPUMF_EVENT_ATTR(cf_z13, TLB2_CRSTE_WRITES, 0x008b);
+CPUMF_EVENT_ATTR(cf_z13, TX_C_TEND, 0x008c);
+CPUMF_EVENT_ATTR(cf_z13, TX_NC_TEND, 0x008d);
+CPUMF_EVENT_ATTR(cf_z13, L1C_TLB1_MISSES, 0x008f);
+CPUMF_EVENT_ATTR(cf_z13, L1D_ONCHIP_L3_SOURCED_WRITES, 0x0090);
+CPUMF_EVENT_ATTR(cf_z13, L1D_ONCHIP_L3_SOURCED_WRITES_IV, 0x0091);
+CPUMF_EVENT_ATTR(cf_z13, L1D_ONNODE_L4_SOURCED_WRITES, 0x0092);
+CPUMF_EVENT_ATTR(cf_z13, L1D_ONNODE_L3_SOURCED_WRITES_IV, 0x0093);
+CPUMF_EVENT_ATTR(cf_z13, L1D_ONNODE_L3_SOURCED_WRITES, 0x0094);
+CPUMF_EVENT_ATTR(cf_z13, L1D_ONDRAWER_L4_SOURCED_WRITES, 0x0095);
+CPUMF_EVENT_ATTR(cf_z13, L1D_ONDRAWER_L3_SOURCED_WRITES_IV, 0x0096);
+CPUMF_EVENT_ATTR(cf_z13, L1D_ONDRAWER_L3_SOURCED_WRITES, 0x0097);
+CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_SCOL_L4_SOURCED_WRITES, 0x0098);
+CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_SCOL_L3_SOURCED_WRITES_IV, 0x0099);
+CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_SCOL_L3_SOURCED_WRITES, 0x009a);
+CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_FCOL_L4_SOURCED_WRITES, 0x009b);
+CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_FCOL_L3_SOURCED_WRITES_IV, 0x009c);
+CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_FCOL_L3_SOURCED_WRITES, 0x009d);
+CPUMF_EVENT_ATTR(cf_z13, L1D_ONNODE_MEM_SOURCED_WRITES, 0x009e);
+CPUMF_EVENT_ATTR(cf_z13, L1D_ONDRAWER_MEM_SOURCED_WRITES, 0x009f);
+CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_MEM_SOURCED_WRITES, 0x00a0);
+CPUMF_EVENT_ATTR(cf_z13, L1D_ONCHIP_MEM_SOURCED_WRITES, 0x00a1);
+CPUMF_EVENT_ATTR(cf_z13, L1I_ONCHIP_L3_SOURCED_WRITES, 0x00a2);
+CPUMF_EVENT_ATTR(cf_z13, L1I_ONCHIP_L3_SOURCED_WRITES_IV, 0x00a3);
+CPUMF_EVENT_ATTR(cf_z13, L1I_ONNODE_L4_SOURCED_WRITES, 0x00a4);
+CPUMF_EVENT_ATTR(cf_z13, L1I_ONNODE_L3_SOURCED_WRITES_IV, 0x00a5);
+CPUMF_EVENT_ATTR(cf_z13, L1I_ONNODE_L3_SOURCED_WRITES, 0x00a6);
+CPUMF_EVENT_ATTR(cf_z13, L1I_ONDRAWER_L4_SOURCED_WRITES, 0x00a7);
+CPUMF_EVENT_ATTR(cf_z13, L1I_ONDRAWER_L3_SOURCED_WRITES_IV, 0x00a8);
+CPUMF_EVENT_ATTR(cf_z13, L1I_ONDRAWER_L3_SOURCED_WRITES, 0x00a9);
+CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_SCOL_L4_SOURCED_WRITES, 0x00aa);
+CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_SCOL_L3_SOURCED_WRITES_IV, 0x00ab);
+CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_SCOL_L3_SOURCED_WRITES, 0x00ac);
+CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_FCOL_L4_SOURCED_WRITES, 0x00ad);
+CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_FCOL_L3_SOURCED_WRITES_IV, 0x00ae);
+CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_FCOL_L3_SOURCED_WRITES, 0x00af);
+CPUMF_EVENT_ATTR(cf_z13, L1I_ONNODE_MEM_SOURCED_WRITES, 0x00b0);
+CPUMF_EVENT_ATTR(cf_z13, L1I_ONDRAWER_MEM_SOURCED_WRITES, 0x00b1);
+CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_MEM_SOURCED_WRITES, 0x00b2);
+CPUMF_EVENT_ATTR(cf_z13, L1I_ONCHIP_MEM_SOURCED_WRITES, 0x00b3);
+CPUMF_EVENT_ATTR(cf_z13, TX_NC_TABORT, 0x00da);
+CPUMF_EVENT_ATTR(cf_z13, TX_C_TABORT_NO_SPECIAL, 0x00db);
+CPUMF_EVENT_ATTR(cf_z13, TX_C_TABORT_SPECIAL, 0x00dc);
+CPUMF_EVENT_ATTR(cf_z13, MT_DIAG_CYCLES_ONE_THR_ACTIVE, 0x01c0);
+CPUMF_EVENT_ATTR(cf_z13, MT_DIAG_CYCLES_TWO_THR_ACTIVE, 0x01c1);
 
-static struct attribute *cpumcf_pmu_event_attr[] = {
+static struct attribute *cpumcf_pmu_event_attr[] __initdata = {
 	CPUMF_EVENT_PTR(cf, CPU_CYCLES),
 	CPUMF_EVENT_PTR(cf, INSTRUCTIONS),
 	CPUMF_EVENT_PTR(cf, L1I_DIR_WRITES),
@@ -236,28 +292,87 @@ static struct attribute *cpumcf_zec12_pmu_event_attr[] __initdata = {
 	NULL,
 };
 
+static struct attribute *cpumcf_z13_pmu_event_attr[] __initdata = {
+	CPUMF_EVENT_PTR(cf_z13, L1D_WRITES_RO_EXCL),
+	CPUMF_EVENT_PTR(cf_z13, DTLB1_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, DTLB1_MISSES),
+	CPUMF_EVENT_PTR(cf_z13, DTLB1_HPAGE_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, DTLB1_GPAGE_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_L2D_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, ITLB1_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, ITLB1_MISSES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_L2I_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, TLB2_PTE_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, TLB2_CRSTE_HPAGE_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, TLB2_CRSTE_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, TX_C_TEND),
+	CPUMF_EVENT_PTR(cf_z13, TX_NC_TEND),
+	CPUMF_EVENT_PTR(cf_z13, L1C_TLB1_MISSES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_ONCHIP_L3_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_ONCHIP_L3_SOURCED_WRITES_IV),
+	CPUMF_EVENT_PTR(cf_z13, L1D_ONNODE_L4_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_ONNODE_L3_SOURCED_WRITES_IV),
+	CPUMF_EVENT_PTR(cf_z13, L1D_ONNODE_L3_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_ONDRAWER_L4_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_ONDRAWER_L3_SOURCED_WRITES_IV),
+	CPUMF_EVENT_PTR(cf_z13, L1D_ONDRAWER_L3_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_SCOL_L4_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_SCOL_L3_SOURCED_WRITES_IV),
+	CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_SCOL_L3_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_FCOL_L4_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_FCOL_L3_SOURCED_WRITES_IV),
+	CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_FCOL_L3_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_ONNODE_MEM_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_ONDRAWER_MEM_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_MEM_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1D_ONCHIP_MEM_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_ONCHIP_L3_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_ONCHIP_L3_SOURCED_WRITES_IV),
+	CPUMF_EVENT_PTR(cf_z13, L1I_ONNODE_L4_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_ONNODE_L3_SOURCED_WRITES_IV),
+	CPUMF_EVENT_PTR(cf_z13, L1I_ONNODE_L3_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_ONDRAWER_L4_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_ONDRAWER_L3_SOURCED_WRITES_IV),
+	CPUMF_EVENT_PTR(cf_z13, L1I_ONDRAWER_L3_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_SCOL_L4_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_SCOL_L3_SOURCED_WRITES_IV),
+	CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_SCOL_L3_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_FCOL_L4_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_FCOL_L3_SOURCED_WRITES_IV),
+	CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_FCOL_L3_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_ONNODE_MEM_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_ONDRAWER_MEM_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_MEM_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, L1I_ONCHIP_MEM_SOURCED_WRITES),
+	CPUMF_EVENT_PTR(cf_z13, TX_NC_TABORT),
+	CPUMF_EVENT_PTR(cf_z13, TX_C_TABORT_NO_SPECIAL),
+	CPUMF_EVENT_PTR(cf_z13, TX_C_TABORT_SPECIAL),
+	CPUMF_EVENT_PTR(cf_z13, MT_DIAG_CYCLES_ONE_THR_ACTIVE),
+	CPUMF_EVENT_PTR(cf_z13, MT_DIAG_CYCLES_TWO_THR_ACTIVE),
+	NULL,
+};
+
 /* END: CPUM_CF COUNTER DEFINITIONS ===================================== */
 
-static struct attribute_group cpumsf_pmu_events_group = {
+static struct attribute_group cpumcf_pmu_events_group = {
 	.name = "events",
-	.attrs = cpumcf_pmu_event_attr,
 };
 
 PMU_FORMAT_ATTR(event, "config:0-63");
 
-static struct attribute *cpumsf_pmu_format_attr[] = {
+static struct attribute *cpumcf_pmu_format_attr[] = {
 	&format_attr_event.attr,
 	NULL,
 };
 
-static struct attribute_group cpumsf_pmu_format_group = {
+static struct attribute_group cpumcf_pmu_format_group = {
 	.name = "format",
-	.attrs = cpumsf_pmu_format_attr,
+	.attrs = cpumcf_pmu_format_attr,
 };
 
-static const struct attribute_group *cpumsf_pmu_attr_groups[] = {
-	&cpumsf_pmu_events_group,
-	&cpumsf_pmu_format_group,
+static const struct attribute_group *cpumcf_pmu_attr_groups[] = {
+	&cpumcf_pmu_events_group,
+	&cpumcf_pmu_format_group,
 	NULL,
 };
 
@@ -290,6 +405,7 @@ static __init struct attribute **merge_attr(struct attribute **a,
 __init const struct attribute_group **cpumf_cf_event_group(void)
 {
 	struct attribute **combined, **model;
+	struct attribute *none[] = { NULL };
 	struct cpuid cpu_id;
 
 	get_cpu_id(&cpu_id);
@@ -306,17 +422,17 @@ __init const struct attribute_group **cpumf_cf_event_group(void)
 	case 0x2828:
 		model = cpumcf_zec12_pmu_event_attr;
 		break;
+	case 0x2964:
+	case 0x2965:
+		model = cpumcf_z13_pmu_event_attr;
+		break;
 	default:
-		model = NULL;
+		model = none;
 		break;
 	}
 
-	if (!model)
-		goto out;
-
 	combined = merge_attr(cpumcf_pmu_event_attr, model);
 	if (combined)
-		cpumsf_pmu_events_group.attrs = combined;
-out:
-	return cpumsf_pmu_attr_groups;
+		cpumcf_pmu_events_group.attrs = combined;
+	return cpumcf_pmu_attr_groups;
 }
diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c
index 1c0b585..9a4f279 100644
--- a/arch/s390/kernel/perf_cpum_sf.c
+++ b/arch/s390/kernel/perf_cpum_sf.c
@@ -1009,8 +1009,8 @@ static int perf_push_sample(struct perf_event *event, struct sf_raw_sample *sfr)
 	 * sample. Some early samples or samples from guests without
 	 * lpp usage would be misaccounted to the host. We use the asn
 	 * value as an addon heuristic to detect most of these guest samples.
-	 * If the value differs from the host hpp value, we assume to be a
-	 * KVM guest.
+	 * If the value differs from 0xffff (the host value), we assume to
+	 * be a KVM guest.
 	 */
 	switch (sfr->basic.CL) {
 	case 1: /* logical partition */
@@ -1020,8 +1020,7 @@ static int perf_push_sample(struct perf_event *event, struct sf_raw_sample *sfr)
 		sde_regs->in_guest = 1;
 		break;
 	default: /* old machine, use heuristics */
-		if (sfr->basic.gpp ||
-		    sfr->basic.prim_asn != (u16)sfr->basic.hpp)
+		if (sfr->basic.gpp || sfr->basic.prim_asn != 0xffff)
 			sde_regs->in_guest = 1;
 		break;
 	}
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index f29e41c..999d715 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -73,8 +73,10 @@ extern void kernel_thread_starter(void);
  */
 void exit_thread(struct task_struct *tsk)
 {
-	if (tsk == current)
+	if (tsk == current) {
 		exit_thread_runtime_instr();
+		exit_thread_gs();
+	}
 }
 
 void flush_thread(void)
@@ -159,6 +161,9 @@ int copy_thread_tls(unsigned long clone_flags, unsigned long new_stackp,
 	/* Don't copy runtime instrumentation info */
 	p->thread.ri_cb = NULL;
 	frame->childregs.psw.mask &= ~PSW_MASK_RI;
+	/* Don't copy guarded storage control block */
+	p->thread.gs_cb = NULL;
+	p->thread.gs_bc_cb = NULL;
 
 	/* Set a new TLS ?  */
 	if (clone_flags & CLONE_SETTLS) {
diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c
index 928b929..778cd65 100644
--- a/arch/s390/kernel/processor.c
+++ b/arch/s390/kernel/processor.c
@@ -7,6 +7,7 @@
 #define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
 
 #include <linux/cpufeature.h>
+#include <linux/bitops.h>
 #include <linux/kernel.h>
 #include <linux/sched/mm.h>
 #include <linux/init.h>
@@ -91,11 +92,23 @@ int cpu_have_feature(unsigned int num)
 }
 EXPORT_SYMBOL(cpu_have_feature);
 
+static void show_facilities(struct seq_file *m)
+{
+	unsigned int bit;
+	long *facilities;
+
+	facilities = (long *)&S390_lowcore.stfle_fac_list;
+	seq_puts(m, "facilities      :");
+	for_each_set_bit_inv(bit, facilities, MAX_FACILITY_BIT)
+		seq_printf(m, " %d", bit);
+	seq_putc(m, '\n');
+}
+
 static void show_cpu_summary(struct seq_file *m, void *v)
 {
 	static const char *hwcap_str[] = {
 		"esan3", "zarch", "stfle", "msa", "ldisp", "eimm", "dfp",
-		"edat", "etf3eh", "highgprs", "te", "vx", "vxd", "vxe"
+		"edat", "etf3eh", "highgprs", "te", "vx", "vxd", "vxe", "gs"
 	};
 	static const char * const int_hwcap_str[] = {
 		"sie"
@@ -116,6 +129,7 @@ static void show_cpu_summary(struct seq_file *m, void *v)
 		if (int_hwcap_str[i] && (int_hwcap & (1UL << i)))
 			seq_printf(m, "%s ", int_hwcap_str[i]);
 	seq_puts(m, "\n");
+	show_facilities(m);
 	show_cacheinfo(m);
 	for_each_online_cpu(cpu) {
 		struct cpuid *id = &per_cpu(cpu_info.cpu_id, cpu);
diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index c14df0a..488c5bb 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -44,30 +44,42 @@ void update_cr_regs(struct task_struct *task)
 	struct pt_regs *regs = task_pt_regs(task);
 	struct thread_struct *thread = &task->thread;
 	struct per_regs old, new;
-
+	unsigned long cr0_old, cr0_new;
+	unsigned long cr2_old, cr2_new;
+	int cr0_changed, cr2_changed;
+
+	__ctl_store(cr0_old, 0, 0);
+	__ctl_store(cr2_old, 2, 2);
+	cr0_new = cr0_old;
+	cr2_new = cr2_old;
 	/* Take care of the enable/disable of transactional execution. */
 	if (MACHINE_HAS_TE) {
-		unsigned long cr, cr_new;
-
-		__ctl_store(cr, 0, 0);
 		/* Set or clear transaction execution TXC bit 8. */
-		cr_new = cr | (1UL << 55);
+		cr0_new |= (1UL << 55);
 		if (task->thread.per_flags & PER_FLAG_NO_TE)
-			cr_new &= ~(1UL << 55);
-		if (cr_new != cr)
-			__ctl_load(cr_new, 0, 0);
+			cr0_new &= ~(1UL << 55);
 		/* Set or clear transaction execution TDC bits 62 and 63. */
-		__ctl_store(cr, 2, 2);
-		cr_new = cr & ~3UL;
+		cr2_new &= ~3UL;
 		if (task->thread.per_flags & PER_FLAG_TE_ABORT_RAND) {
 			if (task->thread.per_flags & PER_FLAG_TE_ABORT_RAND_TEND)
-				cr_new |= 1UL;
+				cr2_new |= 1UL;
 			else
-				cr_new |= 2UL;
+				cr2_new |= 2UL;
 		}
-		if (cr_new != cr)
-			__ctl_load(cr_new, 2, 2);
 	}
+	/* Take care of enable/disable of guarded storage. */
+	if (MACHINE_HAS_GS) {
+		cr2_new &= ~(1UL << 4);
+		if (task->thread.gs_cb)
+			cr2_new |= (1UL << 4);
+	}
+	/* Load control register 0/2 iff changed */
+	cr0_changed = cr0_new != cr0_old;
+	cr2_changed = cr2_new != cr2_old;
+	if (cr0_changed)
+		__ctl_load(cr0_new, 0, 0);
+	if (cr2_changed)
+		__ctl_load(cr2_new, 2, 2);
 	/* Copy user specified PER registers */
 	new.control = thread->per_user.control;
 	new.start = thread->per_user.start;
@@ -1137,6 +1149,74 @@ static int s390_system_call_set(struct task_struct *target,
 				  data, 0, sizeof(unsigned int));
 }
 
+static int s390_gs_cb_get(struct task_struct *target,
+			  const struct user_regset *regset,
+			  unsigned int pos, unsigned int count,
+			  void *kbuf, void __user *ubuf)
+{
+	struct gs_cb *data = target->thread.gs_cb;
+
+	if (!MACHINE_HAS_GS)
+		return -ENODEV;
+	if (!data)
+		return -ENODATA;
+	return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+				   data, 0, sizeof(struct gs_cb));
+}
+
+static int s390_gs_cb_set(struct task_struct *target,
+			  const struct user_regset *regset,
+			  unsigned int pos, unsigned int count,
+			  const void *kbuf, const void __user *ubuf)
+{
+	struct gs_cb *data = target->thread.gs_cb;
+
+	if (!MACHINE_HAS_GS)
+		return -ENODEV;
+	if (!data) {
+		data = kzalloc(sizeof(*data), GFP_KERNEL);
+		if (!data)
+			return -ENOMEM;
+		target->thread.gs_cb = data;
+	}
+	return user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+				  data, 0, sizeof(struct gs_cb));
+}
+
+static int s390_gs_bc_get(struct task_struct *target,
+			  const struct user_regset *regset,
+			  unsigned int pos, unsigned int count,
+			  void *kbuf, void __user *ubuf)
+{
+	struct gs_cb *data = target->thread.gs_bc_cb;
+
+	if (!MACHINE_HAS_GS)
+		return -ENODEV;
+	if (!data)
+		return -ENODATA;
+	return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+				   data, 0, sizeof(struct gs_cb));
+}
+
+static int s390_gs_bc_set(struct task_struct *target,
+			  const struct user_regset *regset,
+			  unsigned int pos, unsigned int count,
+			  const void *kbuf, const void __user *ubuf)
+{
+	struct gs_cb *data = target->thread.gs_bc_cb;
+
+	if (!MACHINE_HAS_GS)
+		return -ENODEV;
+	if (!data) {
+		data = kzalloc(sizeof(*data), GFP_KERNEL);
+		if (!data)
+			return -ENOMEM;
+		target->thread.gs_bc_cb = data;
+	}
+	return user_regset_copyin(&pos, &count, &kbuf, &ubuf,
+				  data, 0, sizeof(struct gs_cb));
+}
+
 static const struct user_regset s390_regsets[] = {
 	{
 		.core_note_type = NT_PRSTATUS,
@@ -1194,6 +1274,22 @@ static const struct user_regset s390_regsets[] = {
 		.get = s390_vxrs_high_get,
 		.set = s390_vxrs_high_set,
 	},
+	{
+		.core_note_type = NT_S390_GS_CB,
+		.n = sizeof(struct gs_cb) / sizeof(__u64),
+		.size = sizeof(__u64),
+		.align = sizeof(__u64),
+		.get = s390_gs_cb_get,
+		.set = s390_gs_cb_set,
+	},
+	{
+		.core_note_type = NT_S390_GS_BC,
+		.n = sizeof(struct gs_cb) / sizeof(__u64),
+		.size = sizeof(__u64),
+		.align = sizeof(__u64),
+		.get = s390_gs_bc_get,
+		.set = s390_gs_bc_set,
+	},
 };
 
 static const struct user_regset_view user_s390_view = {
@@ -1422,6 +1518,14 @@ static const struct user_regset s390_compat_regsets[] = {
 		.get = s390_compat_regs_high_get,
 		.set = s390_compat_regs_high_set,
 	},
+	{
+		.core_note_type = NT_S390_GS_CB,
+		.n = sizeof(struct gs_cb) / sizeof(__u64),
+		.size = sizeof(__u64),
+		.align = sizeof(__u64),
+		.get = s390_gs_cb_get,
+		.set = s390_gs_cb_set,
+	},
 };
 
 static const struct user_regset_view user_s390_compat_view = {
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 911dc0b..3ae756c 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -339,9 +339,15 @@ static void __init setup_lowcore(void)
 	lc->stfl_fac_list = S390_lowcore.stfl_fac_list;
 	memcpy(lc->stfle_fac_list, S390_lowcore.stfle_fac_list,
 	       MAX_FACILITY_BIT/8);
-	if (MACHINE_HAS_VX)
-		lc->vector_save_area_addr =
-			(unsigned long) &lc->vector_save_area;
+	if (MACHINE_HAS_VX || MACHINE_HAS_GS) {
+		unsigned long bits, size;
+
+		bits = MACHINE_HAS_GS ? 11 : 10;
+		size = 1UL << bits;
+		lc->mcesad = (__u64) memblock_virt_alloc(size, size);
+		if (MACHINE_HAS_GS)
+			lc->mcesad |= bits;
+	}
 	lc->vdso_per_cpu_data = (unsigned long) &lc->paste[0];
 	lc->sync_enter_timer = S390_lowcore.sync_enter_timer;
 	lc->async_enter_timer = S390_lowcore.async_enter_timer;
@@ -779,6 +785,12 @@ static int __init setup_hwcaps(void)
 			elf_hwcap |= HWCAP_S390_VXRS_BCD;
 	}
 
+	/*
+	 * Guarded storage support HWCAP_S390_GS is bit 12.
+	 */
+	if (MACHINE_HAS_GS)
+		elf_hwcap |= HWCAP_S390_GS;
+
 	get_cpu_id(&cpu_id);
 	add_device_randomness(&cpu_id, sizeof(cpu_id));
 	switch (cpu_id.machine) {
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 5dab859..363000a 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -51,6 +51,7 @@
 #include <asm/os_info.h>
 #include <asm/sigp.h>
 #include <asm/idle.h>
+#include <asm/nmi.h>
 #include "entry.h"
 
 enum {
@@ -78,6 +79,8 @@ struct pcpu {
 static u8 boot_core_type;
 static struct pcpu pcpu_devices[NR_CPUS];
 
+static struct kmem_cache *pcpu_mcesa_cache;
+
 unsigned int smp_cpu_mt_shift;
 EXPORT_SYMBOL(smp_cpu_mt_shift);
 
@@ -188,8 +191,10 @@ static void pcpu_ec_call(struct pcpu *pcpu, int ec_bit)
 static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu)
 {
 	unsigned long async_stack, panic_stack;
+	unsigned long mcesa_origin, mcesa_bits;
 	struct lowcore *lc;
 
+	mcesa_origin = mcesa_bits = 0;
 	if (pcpu != &pcpu_devices[0]) {
 		pcpu->lowcore =	(struct lowcore *)
 			__get_free_pages(GFP_KERNEL | GFP_DMA, LC_ORDER);
@@ -197,20 +202,27 @@ static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu)
 		panic_stack = __get_free_page(GFP_KERNEL);
 		if (!pcpu->lowcore || !panic_stack || !async_stack)
 			goto out;
+		if (MACHINE_HAS_VX || MACHINE_HAS_GS) {
+			mcesa_origin = (unsigned long)
+				kmem_cache_alloc(pcpu_mcesa_cache, GFP_KERNEL);
+			if (!mcesa_origin)
+				goto out;
+			mcesa_bits = MACHINE_HAS_GS ? 11 : 0;
+		}
 	} else {
 		async_stack = pcpu->lowcore->async_stack - ASYNC_FRAME_OFFSET;
 		panic_stack = pcpu->lowcore->panic_stack - PANIC_FRAME_OFFSET;
+		mcesa_origin = pcpu->lowcore->mcesad & MCESA_ORIGIN_MASK;
+		mcesa_bits = pcpu->lowcore->mcesad & MCESA_LC_MASK;
 	}
 	lc = pcpu->lowcore;
 	memcpy(lc, &S390_lowcore, 512);
 	memset((char *) lc + 512, 0, sizeof(*lc) - 512);
 	lc->async_stack = async_stack + ASYNC_FRAME_OFFSET;
 	lc->panic_stack = panic_stack + PANIC_FRAME_OFFSET;
+	lc->mcesad = mcesa_origin | mcesa_bits;
 	lc->cpu_nr = cpu;
 	lc->spinlock_lockval = arch_spin_lockval(cpu);
-	if (MACHINE_HAS_VX)
-		lc->vector_save_area_addr =
-			(unsigned long) &lc->vector_save_area;
 	if (vdso_alloc_per_cpu(lc))
 		goto out;
 	lowcore_ptr[cpu] = lc;
@@ -218,6 +230,9 @@ static int pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu)
 	return 0;
 out:
 	if (pcpu != &pcpu_devices[0]) {
+		if (mcesa_origin)
+			kmem_cache_free(pcpu_mcesa_cache,
+					(void *) mcesa_origin);
 		free_page(panic_stack);
 		free_pages(async_stack, ASYNC_ORDER);
 		free_pages((unsigned long) pcpu->lowcore, LC_ORDER);
@@ -229,11 +244,17 @@ out:
 
 static void pcpu_free_lowcore(struct pcpu *pcpu)
 {
+	unsigned long mcesa_origin;
+
 	pcpu_sigp_retry(pcpu, SIGP_SET_PREFIX, 0);
 	lowcore_ptr[pcpu - pcpu_devices] = NULL;
 	vdso_free_per_cpu(pcpu->lowcore);
 	if (pcpu == &pcpu_devices[0])
 		return;
+	if (MACHINE_HAS_VX || MACHINE_HAS_GS) {
+		mcesa_origin = pcpu->lowcore->mcesad & MCESA_ORIGIN_MASK;
+		kmem_cache_free(pcpu_mcesa_cache, (void *) mcesa_origin);
+	}
 	free_page(pcpu->lowcore->panic_stack-PANIC_FRAME_OFFSET);
 	free_pages(pcpu->lowcore->async_stack-ASYNC_FRAME_OFFSET, ASYNC_ORDER);
 	free_pages((unsigned long) pcpu->lowcore, LC_ORDER);
@@ -550,9 +571,11 @@ int smp_store_status(int cpu)
 	if (__pcpu_sigp_relax(pcpu->address, SIGP_STORE_STATUS_AT_ADDRESS,
 			      pa) != SIGP_CC_ORDER_CODE_ACCEPTED)
 		return -EIO;
-	if (!MACHINE_HAS_VX)
+	if (!MACHINE_HAS_VX && !MACHINE_HAS_GS)
 		return 0;
-	pa = __pa(pcpu->lowcore->vector_save_area_addr);
+	pa = __pa(pcpu->lowcore->mcesad & MCESA_ORIGIN_MASK);
+	if (MACHINE_HAS_GS)
+		pa |= pcpu->lowcore->mcesad & MCESA_LC_MASK;
 	if (__pcpu_sigp_relax(pcpu->address, SIGP_STORE_ADDITIONAL_STATUS,
 			      pa) != SIGP_CC_ORDER_CODE_ACCEPTED)
 		return -EIO;
@@ -897,12 +920,22 @@ void __init smp_fill_possible_mask(void)
 
 void __init smp_prepare_cpus(unsigned int max_cpus)
 {
+	unsigned long size;
+
 	/* request the 0x1201 emergency signal external interrupt */
 	if (register_external_irq(EXT_IRQ_EMERGENCY_SIG, do_ext_call_interrupt))
 		panic("Couldn't request external interrupt 0x1201");
 	/* request the 0x1202 external call external interrupt */
 	if (register_external_irq(EXT_IRQ_EXTERNAL_CALL, do_ext_call_interrupt))
 		panic("Couldn't request external interrupt 0x1202");
+	/* create slab cache for the machine-check-extended-save-areas */
+	if (MACHINE_HAS_VX || MACHINE_HAS_GS) {
+		size = 1UL << (MACHINE_HAS_GS ? 11 : 10);
+		pcpu_mcesa_cache = kmem_cache_create("nmi_save_areas",
+						     size, size, 0, NULL);
+		if (!pcpu_mcesa_cache)
+			panic("Couldn't create nmi save area cache");
+	}
 }
 
 void __init smp_prepare_boot_cpu(void)
diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
index 2659b5c..54fce7b 100644
--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -386,5 +386,5 @@ SYSCALL(sys_mlock2,compat_sys_mlock2)
 SYSCALL(sys_copy_file_range,compat_sys_copy_file_range) /* 375 */
 SYSCALL(sys_preadv2,compat_sys_preadv2)
 SYSCALL(sys_pwritev2,compat_sys_pwritev2)
-NI_SYSCALL
+SYSCALL(sys_s390_guarded_storage,compat_sys_s390_guarded_storage) /* 378 */
 SYSCALL(sys_statx,compat_sys_statx)
diff --git a/arch/s390/kernel/sysinfo.c b/arch/s390/kernel/sysinfo.c
index 12b6b13..eefcb54 100644
--- a/arch/s390/kernel/sysinfo.c
+++ b/arch/s390/kernel/sysinfo.c
@@ -4,6 +4,7 @@
  *	       Martin Schwidefsky <schwidefsky@de.ibm.com>,
  */
 
+#include <linux/debugfs.h>
 #include <linux/kernel.h>
 #include <linux/mm.h>
 #include <linux/proc_fs.h>
@@ -13,6 +14,7 @@
 #include <linux/export.h>
 #include <linux/slab.h>
 #include <asm/ebcdic.h>
+#include <asm/debug.h>
 #include <asm/sysinfo.h>
 #include <asm/cpcmd.h>
 #include <asm/topology.h>
@@ -485,3 +487,99 @@ void calibrate_delay(void)
 	       "%lu.%02lu BogoMIPS preset\n", loops_per_jiffy/(500000/HZ),
 	       (loops_per_jiffy/(5000/HZ)) % 100);
 }
+
+#ifdef CONFIG_DEBUG_FS
+
+#define STSI_FILE(fc, s1, s2)						       \
+static int stsi_open_##fc##_##s1##_##s2(struct inode *inode, struct file *file)\
+{									       \
+	file->private_data = (void *) get_zeroed_page(GFP_KERNEL);	       \
+	if (!file->private_data)					       \
+		return -ENOMEM;						       \
+	if (stsi(file->private_data, fc, s1, s2)) {			       \
+		free_page((unsigned long)file->private_data);		       \
+		file->private_data = NULL;				       \
+		return -EACCES;						       \
+	}								       \
+	return nonseekable_open(inode, file);				       \
+}									       \
+									       \
+static const struct file_operations stsi_##fc##_##s1##_##s2##_fs_ops = {       \
+	.open		= stsi_open_##fc##_##s1##_##s2,			       \
+	.release	= stsi_release,					       \
+	.read		= stsi_read,					       \
+	.llseek		= no_llseek,					       \
+};
+
+static int stsi_release(struct inode *inode, struct file *file)
+{
+	free_page((unsigned long)file->private_data);
+	return 0;
+}
+
+static ssize_t stsi_read(struct file *file, char __user *buf, size_t size, loff_t *ppos)
+{
+	return simple_read_from_buffer(buf, size, ppos, file->private_data, PAGE_SIZE);
+}
+
+STSI_FILE( 1, 1, 1);
+STSI_FILE( 1, 2, 1);
+STSI_FILE( 1, 2, 2);
+STSI_FILE( 2, 2, 1);
+STSI_FILE( 2, 2, 2);
+STSI_FILE( 3, 2, 2);
+STSI_FILE(15, 1, 2);
+STSI_FILE(15, 1, 3);
+STSI_FILE(15, 1, 4);
+STSI_FILE(15, 1, 5);
+STSI_FILE(15, 1, 6);
+
+struct stsi_file {
+	const struct file_operations *fops;
+	char *name;
+};
+
+static struct stsi_file stsi_file[] __initdata = {
+	{.fops = &stsi_1_1_1_fs_ops,  .name =  "1_1_1"},
+	{.fops = &stsi_1_2_1_fs_ops,  .name =  "1_2_1"},
+	{.fops = &stsi_1_2_2_fs_ops,  .name =  "1_2_2"},
+	{.fops = &stsi_2_2_1_fs_ops,  .name =  "2_2_1"},
+	{.fops = &stsi_2_2_2_fs_ops,  .name =  "2_2_2"},
+	{.fops = &stsi_3_2_2_fs_ops,  .name =  "3_2_2"},
+	{.fops = &stsi_15_1_2_fs_ops, .name = "15_1_2"},
+	{.fops = &stsi_15_1_3_fs_ops, .name = "15_1_3"},
+	{.fops = &stsi_15_1_4_fs_ops, .name = "15_1_4"},
+	{.fops = &stsi_15_1_5_fs_ops, .name = "15_1_5"},
+	{.fops = &stsi_15_1_6_fs_ops, .name = "15_1_6"},
+};
+
+static u8 stsi_0_0_0;
+
+static __init int stsi_init_debugfs(void)
+{
+	struct dentry *stsi_root;
+	struct stsi_file *sf;
+	int lvl, i;
+
+	stsi_root = debugfs_create_dir("stsi", arch_debugfs_dir);
+	if (IS_ERR_OR_NULL(stsi_root))
+		return 0;
+	lvl = stsi(NULL, 0, 0, 0);
+	if (lvl > 0)
+		stsi_0_0_0 = lvl;
+	debugfs_create_u8("0_0_0", 0400, stsi_root, &stsi_0_0_0);
+	for (i = 0; i < ARRAY_SIZE(stsi_file); i++) {
+		sf = &stsi_file[i];
+		debugfs_create_file(sf->name, 0400, stsi_root, NULL, sf->fops);
+	}
+	if (IS_ENABLED(CONFIG_SCHED_TOPOLOGY) && MACHINE_HAS_TOPOLOGY) {
+		char link_to[10];
+
+		sprintf(link_to, "15_1_%d", topology_mnest_limit());
+		debugfs_create_symlink("topology", stsi_root, link_to);
+	}
+	return 0;
+}
+device_initcall(stsi_init_debugfs);
+
+#endif /* CONFIG_DEBUG_FS */
diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
index 17660e8..bb47c92 100644
--- a/arch/s390/kernel/topology.c
+++ b/arch/s390/kernel/topology.c
@@ -83,6 +83,8 @@ static cpumask_t cpu_thread_map(unsigned int cpu)
 	return mask;
 }
 
+#define TOPOLOGY_CORE_BITS	64
+
 static void add_cpus_to_mask(struct topology_core *tl_core,
 			     struct mask_info *drawer,
 			     struct mask_info *book,
@@ -91,7 +93,7 @@ static void add_cpus_to_mask(struct topology_core *tl_core,
 	struct cpu_topology_s390 *topo;
 	unsigned int core;
 
-	for_each_set_bit(core, &tl_core->mask[0], TOPOLOGY_CORE_BITS) {
+	for_each_set_bit(core, &tl_core->mask, TOPOLOGY_CORE_BITS) {
 		unsigned int rcore;
 		int lcpu, i;
 
@@ -244,7 +246,7 @@ static void update_cpu_masks(void)
 
 void store_topology(struct sysinfo_15_1_x *info)
 {
-	stsi(info, 15, 1, min(topology_max_mnest, 4));
+	stsi(info, 15, 1, topology_mnest_limit());
 }
 
 static int __arch_update_cpu_topology(void)