Merge branch 'x86/urgent' into core/efi, to pick up a pending EFI fix

Signed-off-by: Ingo Molnar <mingo@kernel.org>
author: Ingo Molnar <mingo@kernel.org> 2015-10-14 16:05:18 +0200
committer: Ingo Molnar <mingo@kernel.org> 2015-10-14 16:05:18 +0200
commit: c7d77a7980e434c3af17de19e3348157f9b9ccce (patch)
tree: b32c5988ce8239b80c83e94c22d68f5eb0fb84da /arch/s390/kernel
parent: efi: Use the generic efi.memmap instead of 'memmap' (diff)
parent: x86/efi: Fix multiple GOP device support (diff)
download: linux-c7d77a7980e434c3af17de19e3348157f9b9ccce.tar.xz
linux-c7d77a7980e434c3af17de19e3348157f9b9ccce.zip
27 files changed, 1057 insertions, 954 deletions
diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile
index ffb87617a36c..b756c6348ac6 100644
--- a/arch/s390/kernel/Makefile
+++ b/arch/s390/kernel/Makefile
@@ -28,6 +28,17 @@ CFLAGS_ptrace.o		+= -DUTS_MACHINE='"$(UTS_MACHINE)"'
 
 CFLAGS_sysinfo.o += -w
 
+#
+# Use -march=z900 for sclp.c to be able to print an error message if
+# the kernel is started on a machine which is too old
+#
+CFLAGS_REMOVE_sclp.o = $(CC_FLAGS_FTRACE)
+ifneq ($(CC_FLAGS_MARCH),-march=z900)
+CFLAGS_REMOVE_sclp.o += $(CC_FLAGS_MARCH)
+CFLAGS_sclp.o	+= -march=z900
+endif
+GCOV_PROFILE_sclp.o := n
+
 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 sclp.o vdso.o
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index a2da259d9327..3aeeb1b562c0 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -28,6 +28,9 @@ int main(void)
 	DEFINE(__TASK_pid, offsetof(struct task_struct, pid));
 	BLANK();
 	DEFINE(__THREAD_ksp, offsetof(struct thread_struct, ksp));
+	DEFINE(__THREAD_FPU_fpc, offsetof(struct thread_struct, fpu.fpc));
+	DEFINE(__THREAD_FPU_flags, offsetof(struct thread_struct, fpu.flags));
+	DEFINE(__THREAD_FPU_regs, offsetof(struct thread_struct, fpu.regs));
 	DEFINE(__THREAD_per_cause, offsetof(struct thread_struct, per_event.cause));
 	DEFINE(__THREAD_per_address, offsetof(struct thread_struct, per_event.address));
 	DEFINE(__THREAD_per_paid, offsetof(struct thread_struct, per_event.paid));
@@ -173,6 +176,7 @@ int main(void)
 	DEFINE(__LC_PASTE, offsetof(struct _lowcore, paste));
 	DEFINE(__LC_FP_CREG_SAVE_AREA, offsetof(struct _lowcore, fpt_creg_save_area));
 	DEFINE(__LC_LAST_BREAK, offsetof(struct _lowcore, breaking_event_addr));
+	DEFINE(__LC_PERCPU_OFFSET, offsetof(struct _lowcore, percpu_offset));
 	DEFINE(__LC_VDSO_PER_CPU, offsetof(struct _lowcore, vdso_per_cpu_data));
 	DEFINE(__LC_GMAP, offsetof(struct _lowcore, gmap));
 	DEFINE(__LC_PGM_TDB, offsetof(struct _lowcore, pgm_tdb));
diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c
index fe8d6924efaa..e0f9d270b30f 100644
--- a/arch/s390/kernel/compat_signal.c
+++ b/arch/s390/kernel/compat_signal.c
@@ -48,6 +48,19 @@ typedef struct
 	struct ucontext32 uc;
 } rt_sigframe32;
 
+static inline void sigset_to_sigset32(unsigned long *set64,
+				      compat_sigset_word *set32)
+{
+	set32[0] = (compat_sigset_word) set64[0];
+	set32[1] = (compat_sigset_word)(set64[0] >> 32);
+}
+
+static inline void sigset32_to_sigset(compat_sigset_word *set32,
+				      unsigned long *set64)
+{
+	set64[0] = (unsigned long) set32[0] | ((unsigned long) set32[1] << 32);
+}
+
 int copy_siginfo_to_user32(compat_siginfo_t __user *to, const siginfo_t *from)
 {
 	int err;
@@ -153,33 +166,14 @@ int copy_siginfo_from_user32(siginfo_t *to, compat_siginfo_t __user *from)
 /* Store registers needed to create the signal frame */
 static void store_sigregs(void)
 {
-	int i;
-
 	save_access_regs(current->thread.acrs);
-	save_fp_ctl(&current->thread.fp_regs.fpc);
-	if (current->thread.vxrs) {
-		save_vx_regs(current->thread.vxrs);
-		for (i = 0; i < __NUM_FPRS; i++)
-			current->thread.fp_regs.fprs[i] =
-				*(freg_t *)(current->thread.vxrs + i);
-	} else
-		save_fp_regs(current->thread.fp_regs.fprs);
+	save_fpu_regs();
 }
 
 /* Load registers after signal return */
 static void load_sigregs(void)
 {
-	int i;
-
 	restore_access_regs(current->thread.acrs);
-	/* restore_fp_ctl is done in restore_sigregs */
-	if (current->thread.vxrs) {
-		for (i = 0; i < __NUM_FPRS; i++)
-			*(freg_t *)(current->thread.vxrs + i) =
-				current->thread.fp_regs.fprs[i];
-		restore_vx_regs(current->thread.vxrs);
-	} else
-		restore_fp_regs(current->thread.fp_regs.fprs);
 }
 
 static int save_sigregs32(struct pt_regs *regs, _sigregs32 __user *sregs)
@@ -196,8 +190,7 @@ static int save_sigregs32(struct pt_regs *regs, _sigregs32 __user *sregs)
 		user_sregs.regs.gprs[i] = (__u32) regs->gprs[i];
 	memcpy(&user_sregs.regs.acrs, current->thread.acrs,
 	       sizeof(user_sregs.regs.acrs));
-	memcpy(&user_sregs.fpregs, &current->thread.fp_regs,
-	       sizeof(user_sregs.fpregs));
+	fpregs_store((_s390_fp_regs *) &user_sregs.fpregs, &current->thread.fpu);
 	if (__copy_to_user(sregs, &user_sregs, sizeof(_sigregs32)))
 		return -EFAULT;
 	return 0;
@@ -217,8 +210,8 @@ static int restore_sigregs32(struct pt_regs *regs,_sigregs32 __user *sregs)
 	if (!is_ri_task(current) && (user_sregs.regs.psw.mask & PSW32_MASK_RI))
 		return -EINVAL;
 
-	/* Loading the floating-point-control word can fail. Do that first. */
-	if (restore_fp_ctl(&user_sregs.fpregs.fpc))
+	/* Test the floating-point-control word. */
+	if (test_fp_ctl(user_sregs.fpregs.fpc))
 		return -EINVAL;
 
 	/* Use regs->psw.mask instead of PSW_USER_BITS to preserve PER bit. */
@@ -235,9 +228,7 @@ static int restore_sigregs32(struct pt_regs *regs,_sigregs32 __user *sregs)
 		regs->gprs[i] = (__u64) user_sregs.regs.gprs[i];
 	memcpy(&current->thread.acrs, &user_sregs.regs.acrs,
 	       sizeof(current->thread.acrs));
-
-	memcpy(&current->thread.fp_regs, &user_sregs.fpregs,
-	       sizeof(current->thread.fp_regs));
+	fpregs_load((_s390_fp_regs *) &user_sregs.fpregs, &current->thread.fpu);
 
 	clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */
 	return 0;
@@ -258,13 +249,13 @@ static int save_sigregs_ext32(struct pt_regs *regs,
 		return -EFAULT;
 
 	/* Save vector registers to signal stack */
-	if (current->thread.vxrs) {
+	if (is_vx_task(current)) {
 		for (i = 0; i < __NUM_VXRS_LOW; i++)
-			vxrs[i] = *((__u64 *)(current->thread.vxrs + i) + 1);
+			vxrs[i] = *((__u64 *)(current->thread.fpu.vxrs + i) + 1);
 		if (__copy_to_user(&sregs_ext->vxrs_low, vxrs,
 				   sizeof(sregs_ext->vxrs_low)) ||
 		    __copy_to_user(&sregs_ext->vxrs_high,
-				   current->thread.vxrs + __NUM_VXRS_LOW,
+				   current->thread.fpu.vxrs + __NUM_VXRS_LOW,
 				   sizeof(sregs_ext->vxrs_high)))
 			return -EFAULT;
 	}
@@ -286,15 +277,15 @@ static int restore_sigregs_ext32(struct pt_regs *regs,
 		*(__u32 *)&regs->gprs[i] = gprs_high[i];
 
 	/* Restore vector registers from signal stack */
-	if (current->thread.vxrs) {
+	if (is_vx_task(current)) {
 		if (__copy_from_user(vxrs, &sregs_ext->vxrs_low,
 				     sizeof(sregs_ext->vxrs_low)) ||
-		    __copy_from_user(current->thread.vxrs + __NUM_VXRS_LOW,
+		    __copy_from_user(current->thread.fpu.vxrs + __NUM_VXRS_LOW,
 				     &sregs_ext->vxrs_high,
 				     sizeof(sregs_ext->vxrs_high)))
 			return -EFAULT;
 		for (i = 0; i < __NUM_VXRS_LOW; i++)
-			*((__u64 *)(current->thread.vxrs + i) + 1) = vxrs[i];
+			*((__u64 *)(current->thread.fpu.vxrs + i) + 1) = vxrs[i];
 	}
 	return 0;
 }
@@ -303,11 +294,14 @@ COMPAT_SYSCALL_DEFINE0(sigreturn)
 {
 	struct pt_regs *regs = task_pt_regs(current);
 	sigframe32 __user *frame = (sigframe32 __user *)regs->gprs[15];
+	compat_sigset_t cset;
 	sigset_t set;
 
-	if (__copy_from_user(&set.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE32))
+	if (__copy_from_user(&cset.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE32))
 		goto badframe;
+	sigset32_to_sigset(cset.sig, set.sig);
 	set_current_blocked(&set);
+	save_fpu_regs();
 	if (restore_sigregs32(regs, &frame->sregs))
 		goto badframe;
 	if (restore_sigregs_ext32(regs, &frame->sregs_ext))
@@ -323,13 +317,16 @@ COMPAT_SYSCALL_DEFINE0(rt_sigreturn)
 {
 	struct pt_regs *regs = task_pt_regs(current);
 	rt_sigframe32 __user *frame = (rt_sigframe32 __user *)regs->gprs[15];
+	compat_sigset_t cset;
 	sigset_t set;
 
-	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
+	if (__copy_from_user(&cset, &frame->uc.uc_sigmask, sizeof(cset)))
 		goto badframe;
+	sigset32_to_sigset(cset.sig, set.sig);
 	set_current_blocked(&set);
 	if (compat_restore_altstack(&frame->uc.uc_stack))
 		goto badframe;
+	save_fpu_regs();
 	if (restore_sigregs32(regs, &frame->uc.uc_mcontext))
 		goto badframe;
 	if (restore_sigregs_ext32(regs, &frame->uc.uc_mcontext_ext))
@@ -397,7 +394,7 @@ static int setup_frame32(struct ksignal *ksig, sigset_t *set,
 		return -EFAULT;
 
 	/* Create struct sigcontext32 on the signal stack */
-	memcpy(&sc.oldmask, &set->sig, _SIGMASK_COPY_SIZE32);
+	sigset_to_sigset32(set->sig, sc.oldmask);
 	sc.sregs = (__u32)(unsigned long __force) &frame->sregs;
 	if (__copy_to_user(&frame->sc, &sc, sizeof(frame->sc)))
 		return -EFAULT;
@@ -458,6 +455,7 @@ static int setup_frame32(struct ksignal *ksig, sigset_t *set,
 static int setup_rt_frame32(struct ksignal *ksig, sigset_t *set,
 			    struct pt_regs *regs)
 {
+	compat_sigset_t cset;
 	rt_sigframe32 __user *frame;
 	unsigned long restorer;
 	size_t frame_size;
@@ -472,7 +470,7 @@ static int setup_rt_frame32(struct ksignal *ksig, sigset_t *set,
 	 */
 	uc_flags = UC_GPRS_HIGH;
 	if (MACHINE_HAS_VX) {
-		if (current->thread.vxrs)
+		if (is_vx_task(current))
 			uc_flags |= UC_VXRS;
 	} else
 		frame_size -= sizeof(frame->uc.uc_mcontext_ext.vxrs_low) +
@@ -505,11 +503,12 @@ static int setup_rt_frame32(struct ksignal *ksig, sigset_t *set,
 	store_sigregs();
 
 	/* Create ucontext on the signal stack. */
+	sigset_to_sigset32(set->sig, cset.sig);
 	if (__put_user(uc_flags, &frame->uc.uc_flags) ||
 	    __put_user(0, &frame->uc.uc_link) ||
 	    __compat_save_altstack(&frame->uc.uc_stack, regs->gprs[15]) ||
 	    save_sigregs32(regs, &frame->uc.uc_mcontext) ||
-	    __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)) ||
+	    __copy_to_user(&frame->uc.uc_sigmask, &cset, sizeof(cset)) ||
 	    save_sigregs_ext32(regs, &frame->uc.uc_mcontext_ext))
 		return -EFAULT;
 
diff --git a/arch/s390/kernel/compat_wrapper.c b/arch/s390/kernel/compat_wrapper.c
index f8498dde67b1..09f194052df3 100644
--- a/arch/s390/kernel/compat_wrapper.c
+++ b/arch/s390/kernel/compat_wrapper.c
@@ -52,15 +52,13 @@
  * the regular system call wrappers.
  */
 #define COMPAT_SYSCALL_WRAPx(x, name, ...)					\
-	asmlinkage long sys##name(__MAP(x,__SC_DECL,__VA_ARGS__));		\
-	asmlinkage long compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__));\
-	asmlinkage long compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__))	\
-	{									\
-		return sys##name(__MAP(x,__SC_COMPAT_CAST,__VA_ARGS__));	\
-	}
+asmlinkage long sys##name(__MAP(x,__SC_DECL,__VA_ARGS__));			\
+asmlinkage long notrace compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__));\
+asmlinkage long notrace compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__))	\
+{										\
+	return sys##name(__MAP(x,__SC_COMPAT_CAST,__VA_ARGS__));		\
+}
 
-COMPAT_SYSCALL_WRAP1(exit, int, error_code);
-COMPAT_SYSCALL_WRAP1(close, unsigned int, fd);
 COMPAT_SYSCALL_WRAP2(creat, const char __user *, pathname, umode_t, mode);
 COMPAT_SYSCALL_WRAP2(link, const char __user *, oldname, const char __user *, newname);
 COMPAT_SYSCALL_WRAP1(unlink, const char __user *, pathname);
@@ -68,23 +66,16 @@ COMPAT_SYSCALL_WRAP1(chdir, const char __user *, filename);
 COMPAT_SYSCALL_WRAP3(mknod, const char __user *, filename, umode_t, mode, unsigned, dev);
 COMPAT_SYSCALL_WRAP2(chmod, const char __user *, filename, umode_t, mode);
 COMPAT_SYSCALL_WRAP1(oldumount, char __user *, name);
-COMPAT_SYSCALL_WRAP1(alarm, unsigned int, seconds);
 COMPAT_SYSCALL_WRAP2(access, const char __user *, filename, int, mode);
-COMPAT_SYSCALL_WRAP1(nice, int, increment);
-COMPAT_SYSCALL_WRAP2(kill, int, pid, int, sig);
 COMPAT_SYSCALL_WRAP2(rename, const char __user *, oldname, const char __user *, newname);
 COMPAT_SYSCALL_WRAP2(mkdir, const char __user *, pathname, umode_t, mode);
 COMPAT_SYSCALL_WRAP1(rmdir, const char __user *, pathname);
-COMPAT_SYSCALL_WRAP1(dup, unsigned int, fildes);
 COMPAT_SYSCALL_WRAP1(pipe, int __user *, fildes);
 COMPAT_SYSCALL_WRAP1(brk, unsigned long, brk);
 COMPAT_SYSCALL_WRAP2(signal, int, sig, __sighandler_t, handler);
 COMPAT_SYSCALL_WRAP1(acct, const char __user *, name);
 COMPAT_SYSCALL_WRAP2(umount, char __user *, name, int, flags);
-COMPAT_SYSCALL_WRAP2(setpgid, pid_t, pid, pid_t, pgid);
-COMPAT_SYSCALL_WRAP1(umask, int, mask);
 COMPAT_SYSCALL_WRAP1(chroot, const char __user *, filename);
-COMPAT_SYSCALL_WRAP2(dup2, unsigned int, oldfd, unsigned int, newfd);
 COMPAT_SYSCALL_WRAP3(sigsuspend, int, unused1, int, unused2, old_sigset_t, mask);
 COMPAT_SYSCALL_WRAP2(sethostname, char __user *, name, int, len);
 COMPAT_SYSCALL_WRAP2(symlink, const char __user *, old, const char __user *, new);
@@ -93,37 +84,23 @@ COMPAT_SYSCALL_WRAP1(uselib, const char __user *, library);
 COMPAT_SYSCALL_WRAP2(swapon, const char __user *, specialfile, int, swap_flags);
 COMPAT_SYSCALL_WRAP4(reboot, int, magic1, int, magic2, unsigned int, cmd, void __user *, arg);
 COMPAT_SYSCALL_WRAP2(munmap, unsigned long, addr, size_t, len);
-COMPAT_SYSCALL_WRAP2(fchmod, unsigned int, fd, umode_t, mode);
-COMPAT_SYSCALL_WRAP2(getpriority, int, which, int, who);
-COMPAT_SYSCALL_WRAP3(setpriority, int, which, int, who, int, niceval);
 COMPAT_SYSCALL_WRAP3(syslog, int, type, char __user *, buf, int, len);
 COMPAT_SYSCALL_WRAP1(swapoff, const char __user *, specialfile);
-COMPAT_SYSCALL_WRAP1(fsync, unsigned int, fd);
 COMPAT_SYSCALL_WRAP2(setdomainname, char __user *, name, int, len);
 COMPAT_SYSCALL_WRAP1(newuname, struct new_utsname __user *, name);
 COMPAT_SYSCALL_WRAP3(mprotect, unsigned long, start, size_t, len, unsigned long, prot);
 COMPAT_SYSCALL_WRAP3(init_module, void __user *, umod, unsigned long, len, const char __user *, uargs);
 COMPAT_SYSCALL_WRAP2(delete_module, const char __user *, name_user, unsigned int, flags);
 COMPAT_SYSCALL_WRAP4(quotactl, unsigned int, cmd, const char __user *, special, qid_t, id, void __user *, addr);
-COMPAT_SYSCALL_WRAP1(getpgid, pid_t, pid);
-COMPAT_SYSCALL_WRAP1(fchdir, unsigned int, fd);
 COMPAT_SYSCALL_WRAP2(bdflush, int, func, long, data);
 COMPAT_SYSCALL_WRAP3(sysfs, int, option, unsigned long, arg1, unsigned long, arg2);
-COMPAT_SYSCALL_WRAP1(s390_personality, unsigned int, personality);
 COMPAT_SYSCALL_WRAP5(llseek, unsigned int, fd, unsigned long, high, unsigned long, low, loff_t __user *, result, unsigned int, whence);
-COMPAT_SYSCALL_WRAP2(flock, unsigned int, fd, unsigned int, cmd);
 COMPAT_SYSCALL_WRAP3(msync, unsigned long, start, size_t, len, int, flags);
-COMPAT_SYSCALL_WRAP1(getsid, pid_t, pid);
-COMPAT_SYSCALL_WRAP1(fdatasync, unsigned int, fd);
 COMPAT_SYSCALL_WRAP2(mlock, unsigned long, start, size_t, len);
 COMPAT_SYSCALL_WRAP2(munlock, unsigned long, start, size_t, len);
-COMPAT_SYSCALL_WRAP1(mlockall, int, flags);
 COMPAT_SYSCALL_WRAP2(sched_setparam, pid_t, pid, struct sched_param __user *, param);
 COMPAT_SYSCALL_WRAP2(sched_getparam, pid_t, pid, struct sched_param __user *, param);
 COMPAT_SYSCALL_WRAP3(sched_setscheduler, pid_t, pid, int, policy, struct sched_param __user *, param);
-COMPAT_SYSCALL_WRAP1(sched_getscheduler, pid_t, pid);
-COMPAT_SYSCALL_WRAP1(sched_get_priority_max, int, policy);
-COMPAT_SYSCALL_WRAP1(sched_get_priority_min, int, policy);
 COMPAT_SYSCALL_WRAP5(mremap, unsigned long, addr, unsigned long, old_len, unsigned long, new_len, unsigned long, flags, unsigned long, new_addr);
 COMPAT_SYSCALL_WRAP3(poll, struct pollfd __user *, ufds, unsigned int, nfds, int, timeout);
 COMPAT_SYSCALL_WRAP5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, unsigned long, arg4, unsigned long, arg5);
@@ -131,20 +108,11 @@ COMPAT_SYSCALL_WRAP2(getcwd, char __user *, buf, unsigned long, size);
 COMPAT_SYSCALL_WRAP2(capget, cap_user_header_t, header, cap_user_data_t, dataptr);
 COMPAT_SYSCALL_WRAP2(capset, cap_user_header_t, header, const cap_user_data_t, data);
 COMPAT_SYSCALL_WRAP3(lchown, const char __user *, filename, uid_t, user, gid_t, group);
-COMPAT_SYSCALL_WRAP2(setreuid, uid_t, ruid, uid_t, euid);
-COMPAT_SYSCALL_WRAP2(setregid, gid_t, rgid, gid_t, egid);
 COMPAT_SYSCALL_WRAP2(getgroups, int, gidsetsize, gid_t __user *, grouplist);
 COMPAT_SYSCALL_WRAP2(setgroups, int, gidsetsize, gid_t __user *, grouplist);
-COMPAT_SYSCALL_WRAP3(fchown, unsigned int, fd, uid_t, user, gid_t, group);
-COMPAT_SYSCALL_WRAP3(setresuid, uid_t, ruid, uid_t, euid, uid_t, suid);
 COMPAT_SYSCALL_WRAP3(getresuid, uid_t __user *, ruid, uid_t __user *, euid, uid_t __user *, suid);
-COMPAT_SYSCALL_WRAP3(setresgid, gid_t, rgid, gid_t, egid, gid_t, sgid);
 COMPAT_SYSCALL_WRAP3(getresgid, gid_t __user *, rgid, gid_t __user *, egid, gid_t __user *, sgid);
 COMPAT_SYSCALL_WRAP3(chown, const char __user *, filename, uid_t, user, gid_t, group);
-COMPAT_SYSCALL_WRAP1(setuid, uid_t, uid);
-COMPAT_SYSCALL_WRAP1(setgid, gid_t, gid);
-COMPAT_SYSCALL_WRAP1(setfsuid, uid_t, uid);
-COMPAT_SYSCALL_WRAP1(setfsgid, gid_t, gid);
 COMPAT_SYSCALL_WRAP2(pivot_root, const char __user *, new_root, const char __user *, put_old);
 COMPAT_SYSCALL_WRAP3(mincore, unsigned long, start, size_t, len, unsigned char __user *, vec);
 COMPAT_SYSCALL_WRAP3(madvise, unsigned long, start, size_t, len, int, behavior);
@@ -161,23 +129,16 @@ COMPAT_SYSCALL_WRAP3(flistxattr, int, fd, char __user *, list, size_t, size);
 COMPAT_SYSCALL_WRAP2(removexattr, const char __user *, path, const char __user *, name);
 COMPAT_SYSCALL_WRAP2(lremovexattr, const char __user *, path, const char __user *, name);
 COMPAT_SYSCALL_WRAP2(fremovexattr, int, fd, const char __user *, name);
-COMPAT_SYSCALL_WRAP1(exit_group, int, error_code);
 COMPAT_SYSCALL_WRAP1(set_tid_address, int __user *, tidptr);
-COMPAT_SYSCALL_WRAP1(epoll_create, int, size);
 COMPAT_SYSCALL_WRAP4(epoll_ctl, int, epfd, int, op, int, fd, struct epoll_event __user *, event);
 COMPAT_SYSCALL_WRAP4(epoll_wait, int, epfd, struct epoll_event __user *, events, int, maxevents, int, timeout);
-COMPAT_SYSCALL_WRAP1(timer_getoverrun, timer_t, timer_id);
-COMPAT_SYSCALL_WRAP1(timer_delete, compat_timer_t, compat_timer_id);
 COMPAT_SYSCALL_WRAP1(io_destroy, aio_context_t, ctx);
 COMPAT_SYSCALL_WRAP3(io_cancel, aio_context_t, ctx_id, struct iocb __user *, iocb, struct io_event __user *, result);
 COMPAT_SYSCALL_WRAP1(mq_unlink, const char __user *, name);
 COMPAT_SYSCALL_WRAP5(add_key, const char __user *, tp, const char __user *, dsc, const void __user *, pld, size_t, len, key_serial_t, id);
 COMPAT_SYSCALL_WRAP4(request_key, const char __user *, tp, const char __user *, dsc, const char __user *, info, key_serial_t, id);
 COMPAT_SYSCALL_WRAP5(remap_file_pages, unsigned long, start, unsigned long, size, unsigned long, prot, unsigned long, pgoff, unsigned long, flags);
-COMPAT_SYSCALL_WRAP3(ioprio_set, int, which, int, who, int, ioprio);
-COMPAT_SYSCALL_WRAP2(ioprio_get, int, which, int, who);
 COMPAT_SYSCALL_WRAP3(inotify_add_watch, int, fd, const char __user *, path, u32, mask);
-COMPAT_SYSCALL_WRAP2(inotify_rm_watch, int, fd, __s32, wd);
 COMPAT_SYSCALL_WRAP3(mkdirat, int, dfd, const char __user *, pathname, umode_t, mode);
 COMPAT_SYSCALL_WRAP4(mknodat, int, dfd, const char __user *, filename, umode_t, mode, unsigned, dev);
 COMPAT_SYSCALL_WRAP5(fchownat, int, dfd, const char __user *, filename, uid_t, user, gid_t, group, int, flag);
@@ -192,23 +153,11 @@ COMPAT_SYSCALL_WRAP1(unshare, unsigned long, unshare_flags);
 COMPAT_SYSCALL_WRAP6(splice, int, fd_in, loff_t __user *, off_in, int, fd_out, loff_t __user *, off_out, size_t, len, unsigned int, flags);
 COMPAT_SYSCALL_WRAP4(tee, int, fdin, int, fdout, size_t, len, unsigned int, flags);
 COMPAT_SYSCALL_WRAP3(getcpu, unsigned __user *, cpu, unsigned __user *, node, struct getcpu_cache __user *, cache);
-COMPAT_SYSCALL_WRAP1(eventfd, unsigned int, count);
-COMPAT_SYSCALL_WRAP2(timerfd_create, int, clockid, int, flags);
-COMPAT_SYSCALL_WRAP2(eventfd2, unsigned int, count, int, flags);
-COMPAT_SYSCALL_WRAP1(inotify_init1, int, flags);
 COMPAT_SYSCALL_WRAP2(pipe2, int __user *, fildes, int, flags);
-COMPAT_SYSCALL_WRAP3(dup3, unsigned int, oldfd, unsigned int, newfd, int, flags);
-COMPAT_SYSCALL_WRAP1(epoll_create1, int, flags);
-COMPAT_SYSCALL_WRAP2(tkill, int, pid, int, sig);
-COMPAT_SYSCALL_WRAP3(tgkill, int, tgid, int, pid, int, sig);
 COMPAT_SYSCALL_WRAP5(perf_event_open, struct perf_event_attr __user *, attr_uptr, pid_t, pid, int, cpu, int, group_fd, unsigned long, flags);
 COMPAT_SYSCALL_WRAP5(clone, unsigned long, newsp, unsigned long, clone_flags, int __user *, parent_tidptr, int __user *, child_tidptr, unsigned long, tls);
-COMPAT_SYSCALL_WRAP2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags);
 COMPAT_SYSCALL_WRAP4(prlimit64, pid_t, pid, unsigned int, resource, const struct rlimit64 __user *, new_rlim, struct rlimit64 __user *, old_rlim);
 COMPAT_SYSCALL_WRAP5(name_to_handle_at, int, dfd, const char __user *, name, struct file_handle __user *, handle, int __user *, mnt_id, int, flag);
-COMPAT_SYSCALL_WRAP1(syncfs, int, fd);
-COMPAT_SYSCALL_WRAP2(setns, int, fd, int, nstype);
-COMPAT_SYSCALL_WRAP2(s390_runtime_instr, int, command, int, signum);
 COMPAT_SYSCALL_WRAP5(kcmp, pid_t, pid1, pid_t, pid2, int, type, unsigned long, idx1, unsigned long, idx2);
 COMPAT_SYSCALL_WRAP3(finit_module, int, fd, const char __user *, uargs, int, flags);
 COMPAT_SYSCALL_WRAP3(sched_setattr, pid_t, pid, struct sched_attr __user *, attr, unsigned int, flags);
@@ -220,3 +169,10 @@ COMPAT_SYSCALL_WRAP2(memfd_create, const char __user *, uname, unsigned int, fla
 COMPAT_SYSCALL_WRAP3(bpf, int, cmd, union bpf_attr *, attr, unsigned int, size);
 COMPAT_SYSCALL_WRAP3(s390_pci_mmio_write, const unsigned long, mmio_addr, const void __user *, user_buffer, const size_t, length);
 COMPAT_SYSCALL_WRAP3(s390_pci_mmio_read, const unsigned long, mmio_addr, void __user *, user_buffer, const size_t, length);
+COMPAT_SYSCALL_WRAP4(socketpair, int, family, int, type, int, protocol, int __user *, usockvec);
+COMPAT_SYSCALL_WRAP3(bind, int, fd, struct sockaddr __user *, umyaddr, int, addrlen);
+COMPAT_SYSCALL_WRAP3(connect, int, fd, struct sockaddr __user *, uservaddr, int, addrlen);
+COMPAT_SYSCALL_WRAP4(accept4, int, fd, struct sockaddr __user *, upeer_sockaddr, int __user *, upeer_addrlen, int, flags);
+COMPAT_SYSCALL_WRAP3(getsockname, int, fd, struct sockaddr __user *, usockaddr, int __user *, usockaddr_len);
+COMPAT_SYSCALL_WRAP3(getpeername, int, fd, struct sockaddr __user *, usockaddr, int __user *, usockaddr_len);
+COMPAT_SYSCALL_WRAP6(sendto, int, fd, void __user *, buff, size_t, len, unsigned int, flags, struct sockaddr __user *, addr, int, addr_len);
diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index 84062e7a77da..582fe44ab07c 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -20,6 +20,8 @@
 #include <asm/page.h>
 #include <asm/sigp.h>
 #include <asm/irq.h>
+#include <asm/fpu-internal.h>
+#include <asm/vx-insn.h>
 
 __PT_R0      =	__PT_GPRS
 __PT_R1      =	__PT_GPRS + 8
@@ -46,10 +48,10 @@ _TIF_WORK	= (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_NEED_RESCHED | \
 		   _TIF_UPROBE)
 _TIF_TRACE	= (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | \
 		   _TIF_SYSCALL_TRACEPOINT)
-_CIF_WORK	= (_CIF_MCCK_PENDING | _CIF_ASCE)
+_CIF_WORK	= (_CIF_MCCK_PENDING | _CIF_ASCE | _CIF_FPU)
 _PIF_WORK	= (_PIF_PER_TRAP)
 
-#define BASED(name) name-system_call(%r13)
+#define BASED(name) name-cleanup_critical(%r13)
 
 	.macro	TRACE_IRQS_ON
 #ifdef CONFIG_TRACE_IRQFLAGS
@@ -73,38 +75,6 @@ _PIF_WORK	= (_PIF_PER_TRAP)
 #endif
 	.endm
 
-	.macro LPP newpp
-#if IS_ENABLED(CONFIG_KVM)
-	tm	__LC_MACHINE_FLAGS+6,0x20	# MACHINE_FLAG_LPP
-	jz	.+8
-	.insn	s,0xb2800000,\newpp
-#endif
-	.endm
-
-	.macro	HANDLE_SIE_INTERCEPT scratch,reason
-#if IS_ENABLED(CONFIG_KVM)
-	tmhh	%r8,0x0001		# interrupting from user ?
-	jnz	.+62
-	lgr	\scratch,%r9
-	slg	\scratch,BASED(.Lsie_critical)
-	clg	\scratch,BASED(.Lsie_critical_length)
-	.if	\reason==1
-	# Some program interrupts are suppressing (e.g. protection).
-	# We must also check the instruction after SIE in that case.
-	# do_protection_exception will rewind to .Lrewind_pad
-	jh	.+42
-	.else
-	jhe	.+42
-	.endif
-	lg	%r14,__SF_EMPTY(%r15)		# get control block pointer
-	LPP	__SF_EMPTY+16(%r15)		# set host id
-	ni	__SIE_PROG0C+3(%r14),0xfe	# no longer in SIE
-	lctlg	%c1,%c1,__LC_USER_ASCE		# load primary asce
-	larl	%r9,sie_exit			# skip forward to sie_exit
-	mvi	__SF_EMPTY+31(%r15),\reason	# set exit reason
-#endif
-	.endm
-
 	.macro	CHECK_STACK stacksize,savearea
 #ifdef CONFIG_CHECK_STACK
 	tml	%r15,\stacksize - CONFIG_STACK_GUARD
@@ -113,7 +83,7 @@ _PIF_WORK	= (_PIF_PER_TRAP)
 #endif
 	.endm
 
-	.macro	SWITCH_ASYNC savearea,stack,shift
+	.macro	SWITCH_ASYNC savearea,timer
 	tmhh	%r8,0x0001		# interrupting from user ?
 	jnz	1f
 	lgr	%r14,%r9
@@ -124,26 +94,28 @@ _PIF_WORK	= (_PIF_PER_TRAP)
 	brasl	%r14,cleanup_critical
 	tmhh	%r8,0x0001		# retest problem state after cleanup
 	jnz	1f
-0:	lg	%r14,\stack		# are we already on the target stack?
+0:	lg	%r14,__LC_ASYNC_STACK	# are we already on the async stack?
 	slgr	%r14,%r15
-	srag	%r14,%r14,\shift
-	jnz	1f
-	CHECK_STACK 1<<\shift,\savearea
+	srag	%r14,%r14,STACK_SHIFT
+	jnz	2f
+	CHECK_STACK 1<<STACK_SHIFT,\savearea
 	aghi	%r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
-	j	2f
-1:	lg	%r15,\stack		# load target stack
-2:	la	%r11,STACK_FRAME_OVERHEAD(%r15)
+	j	3f
+1:	LAST_BREAK %r14
+	UPDATE_VTIME %r14,%r15,\timer
+2:	lg	%r15,__LC_ASYNC_STACK	# load async stack
+3:	la	%r11,STACK_FRAME_OVERHEAD(%r15)
 	.endm
 
-	.macro UPDATE_VTIME scratch,enter_timer
-	lg	\scratch,__LC_EXIT_TIMER
-	slg	\scratch,\enter_timer
-	alg	\scratch,__LC_USER_TIMER
-	stg	\scratch,__LC_USER_TIMER
-	lg	\scratch,__LC_LAST_UPDATE_TIMER
-	slg	\scratch,__LC_EXIT_TIMER
-	alg	\scratch,__LC_SYSTEM_TIMER
-	stg	\scratch,__LC_SYSTEM_TIMER
+	.macro UPDATE_VTIME w1,w2,enter_timer
+	lg	\w1,__LC_EXIT_TIMER
+	lg	\w2,__LC_LAST_UPDATE_TIMER
+	slg	\w1,\enter_timer
+	slg	\w2,__LC_EXIT_TIMER
+	alg	\w1,__LC_USER_TIMER
+	alg	\w2,__LC_SYSTEM_TIMER
+	stg	\w1,__LC_USER_TIMER
+	stg	\w2,__LC_SYSTEM_TIMER
 	mvc	__LC_LAST_UPDATE_TIMER(8),\enter_timer
 	.endm
 
@@ -197,6 +169,69 @@ ENTRY(__switch_to)
 	br	%r14
 
 .L__critical_start:
+
+#if IS_ENABLED(CONFIG_KVM)
+/*
+ * sie64a calling convention:
+ * %r2 pointer to sie control block
+ * %r3 guest register save area
+ */
+ENTRY(sie64a)
+	stmg	%r6,%r14,__SF_GPRS(%r15)	# save kernel registers
+	stg	%r2,__SF_EMPTY(%r15)		# save control block pointer
+	stg	%r3,__SF_EMPTY+8(%r15)		# save guest register save area
+	xc	__SF_EMPTY+16(16,%r15),__SF_EMPTY+16(%r15) # host id & reason
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU	# load guest fp/vx registers ?
+	jno	.Lsie_load_guest_gprs
+	brasl	%r14,load_fpu_regs		# load guest fp/vx regs
+.Lsie_load_guest_gprs:
+	lmg	%r0,%r13,0(%r3)			# load guest gprs 0-13
+	lg	%r14,__LC_GMAP			# get gmap pointer
+	ltgr	%r14,%r14
+	jz	.Lsie_gmap
+	lctlg	%c1,%c1,__GMAP_ASCE(%r14)	# load primary asce
+.Lsie_gmap:
+	lg	%r14,__SF_EMPTY(%r15)		# get control block pointer
+	oi	__SIE_PROG0C+3(%r14),1		# we are going into SIE now
+	tm	__SIE_PROG20+3(%r14),3		# last exit...
+	jnz	.Lsie_skip
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	jo	.Lsie_skip			# exit if fp/vx regs changed
+	tm	__LC_MACHINE_FLAGS+6,0x20	# MACHINE_FLAG_LPP
+	jz	.Lsie_enter
+	.insn	s,0xb2800000,__LC_CURRENT_PID	# set guest id to pid
+.Lsie_enter:
+	sie	0(%r14)
+	tm	__LC_MACHINE_FLAGS+6,0x20	# MACHINE_FLAG_LPP
+	jz	.Lsie_skip
+	.insn	s,0xb2800000,__SF_EMPTY+16(%r15)# set host id
+.Lsie_skip:
+	ni	__SIE_PROG0C+3(%r14),0xfe	# no longer in SIE
+	lctlg	%c1,%c1,__LC_USER_ASCE		# load primary asce
+.Lsie_done:
+# some program checks are suppressing. C code (e.g. do_protection_exception)
+# will rewind the PSW by the ILC, which is 4 bytes in case of SIE. Other
+# instructions between sie64a and .Lsie_done should not cause program
+# interrupts. So lets use a nop (47 00 00 00) as a landing pad.
+# See also .Lcleanup_sie
+.Lrewind_pad:
+	nop	0
+	.globl sie_exit
+sie_exit:
+	lg	%r14,__SF_EMPTY+8(%r15)		# load guest register save area
+	stmg	%r0,%r13,0(%r14)		# save guest gprs 0-13
+	lmg	%r6,%r14,__SF_GPRS(%r15)	# restore kernel registers
+	lg	%r2,__SF_EMPTY+24(%r15)		# return exit reason code
+	br	%r14
+.Lsie_fault:
+	lghi	%r14,-EFAULT
+	stg	%r14,__SF_EMPTY+24(%r15)	# set exit reason code
+	j	sie_exit
+
+	EX_TABLE(.Lrewind_pad,.Lsie_fault)
+	EX_TABLE(sie_exit,.Lsie_fault)
+#endif
+
 /*
  * SVC interrupt handler routine. System calls are synchronous events and
  * are executed with interrupts enabled.
@@ -212,9 +247,9 @@ ENTRY(system_call)
 .Lsysc_per:
 	lg	%r15,__LC_KERNEL_STACK
 	la	%r11,STACK_FRAME_OVERHEAD(%r15)	# pointer to pt_regs
-.Lsysc_vtime:
-	UPDATE_VTIME %r13,__LC_SYNC_ENTER_TIMER
 	LAST_BREAK %r13
+.Lsysc_vtime:
+	UPDATE_VTIME %r10,%r13,__LC_SYNC_ENTER_TIMER
 	stmg	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(64,%r11),__LC_SAVE_AREA_SYNC
 	mvc	__PT_PSW(16,%r11),__LC_SVC_OLD_PSW
@@ -244,8 +279,6 @@ ENTRY(system_call)
 .Lsysc_return:
 	LOCKDEP_SYS_EXIT
 .Lsysc_tif:
-	tm	__PT_PSW+1(%r11),0x01		# returning to user ?
-	jno	.Lsysc_restore
 	tm	__PT_FLAGS+7(%r11),_PIF_WORK
 	jnz	.Lsysc_work
 	tm	__TI_flags+7(%r12),_TIF_WORK
@@ -280,6 +313,8 @@ ENTRY(system_call)
 	jo	.Lsysc_sigpending
 	tm	__TI_flags+7(%r12),_TIF_NOTIFY_RESUME
 	jo	.Lsysc_notify_resume
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	jo	.Lsysc_vxrs
 	tm	__LC_CPU_FLAGS+7,_CIF_ASCE
 	jo	.Lsysc_uaccess
 	j	.Lsysc_return		# beware of critical section cleanup
@@ -307,6 +342,13 @@ ENTRY(system_call)
 	j	.Lsysc_return
 
 #
+# CIF_FPU is set, restore floating-point controls and floating-point registers.
+#
+.Lsysc_vxrs:
+	larl	%r14,.Lsysc_return
+	jg	load_fpu_regs
+
+#
 # _TIF_SIGPENDING is set, call do_signal
 #
 .Lsysc_sigpending:
@@ -405,28 +447,35 @@ ENTRY(pgm_check_handler)
 	stmg	%r8,%r15,__LC_SAVE_AREA_SYNC
 	lg	%r10,__LC_LAST_BREAK
 	lg	%r12,__LC_THREAD_INFO
-	larl	%r13,system_call
+	larl	%r13,cleanup_critical
 	lmg	%r8,%r9,__LC_PGM_OLD_PSW
-	HANDLE_SIE_INTERCEPT %r14,1
 	tmhh	%r8,0x0001		# test problem state bit
-	jnz	1f			# -> fault in user space
-	tmhh	%r8,0x4000		# PER bit set in old PSW ?
-	jnz	0f			# -> enabled, can't be a double fault
+	jnz	2f			# -> fault in user space
+#if IS_ENABLED(CONFIG_KVM)
+	# cleanup critical section for sie64a
+	lgr	%r14,%r9
+	slg	%r14,BASED(.Lsie_critical_start)
+	clg	%r14,BASED(.Lsie_critical_length)
+	jhe	0f
+	brasl	%r14,.Lcleanup_sie
+#endif
+0:	tmhh	%r8,0x4000		# PER bit set in old PSW ?
+	jnz	1f			# -> enabled, can't be a double fault
 	tm	__LC_PGM_ILC+3,0x80	# check for per exception
 	jnz	.Lpgm_svcper		# -> single stepped svc
-0:	CHECK_STACK STACK_SIZE,__LC_SAVE_AREA_SYNC
+1:	CHECK_STACK STACK_SIZE,__LC_SAVE_AREA_SYNC
 	aghi	%r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
-	j	2f
-1:	UPDATE_VTIME %r14,__LC_SYNC_ENTER_TIMER
-	LAST_BREAK %r14
+	j	3f
+2:	LAST_BREAK %r14
+	UPDATE_VTIME %r14,%r15,__LC_SYNC_ENTER_TIMER
 	lg	%r15,__LC_KERNEL_STACK
 	lg	%r14,__TI_task(%r12)
 	aghi	%r14,__TASK_thread	# pointer to thread_struct
 	lghi	%r13,__LC_PGM_TDB
 	tm	__LC_PGM_ILC+2,0x02	# check for transaction abort
-	jz	2f
+	jz	3f
 	mvc	__THREAD_trap_tdb(256,%r14),0(%r13)
-2:	la	%r11,STACK_FRAME_OVERHEAD(%r15)
+3:	la	%r11,STACK_FRAME_OVERHEAD(%r15)
 	stmg	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(64,%r11),__LC_SAVE_AREA_SYNC
 	stmg	%r8,%r9,__PT_PSW(%r11)
@@ -435,24 +484,28 @@ ENTRY(pgm_check_handler)
 	xc	__PT_FLAGS(8,%r11),__PT_FLAGS(%r11)
 	stg	%r10,__PT_ARGS(%r11)
 	tm	__LC_PGM_ILC+3,0x80	# check for per exception
-	jz	0f
+	jz	4f
 	tmhh	%r8,0x0001		# kernel per event ?
 	jz	.Lpgm_kprobe
 	oi	__PT_FLAGS+7(%r11),_PIF_PER_TRAP
 	mvc	__THREAD_per_address(8,%r14),__LC_PER_ADDRESS
 	mvc	__THREAD_per_cause(2,%r14),__LC_PER_CODE
 	mvc	__THREAD_per_paid(1,%r14),__LC_PER_ACCESS_ID
-0:	REENABLE_IRQS
+4:	REENABLE_IRQS
 	xc	__SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
 	larl	%r1,pgm_check_table
 	llgh	%r10,__PT_INT_CODE+2(%r11)
 	nill	%r10,0x007f
 	sll	%r10,2
-	je	.Lsysc_return
+	je	.Lpgm_return
 	lgf	%r1,0(%r10,%r1)		# load address of handler routine
 	lgr	%r2,%r11		# pass pointer to pt_regs
 	basr	%r14,%r1		# branch to interrupt-handler
-	j	.Lsysc_return
+.Lpgm_return:
+	LOCKDEP_SYS_EXIT
+	tm	__PT_PSW+1(%r11),0x01	# returning to user ?
+	jno	.Lsysc_restore
+	j	.Lsysc_tif
 
 #
 # PER event in supervisor state, must be kprobes
@@ -462,7 +515,7 @@ ENTRY(pgm_check_handler)
 	xc	__SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
 	lgr	%r2,%r11		# pass pointer to pt_regs
 	brasl	%r14,do_per_trap
-	j	.Lsysc_return
+	j	.Lpgm_return
 
 #
 # single stepped system call
@@ -483,15 +536,9 @@ ENTRY(io_int_handler)
 	stmg	%r8,%r15,__LC_SAVE_AREA_ASYNC
 	lg	%r10,__LC_LAST_BREAK
 	lg	%r12,__LC_THREAD_INFO
-	larl	%r13,system_call
+	larl	%r13,cleanup_critical
 	lmg	%r8,%r9,__LC_IO_OLD_PSW
-	HANDLE_SIE_INTERCEPT %r14,2
-	SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_STACK,STACK_SHIFT
-	tmhh	%r8,0x0001		# interrupting from user?
-	jz	.Lio_skip
-	UPDATE_VTIME %r14,__LC_ASYNC_ENTER_TIMER
-	LAST_BREAK %r14
-.Lio_skip:
+	SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER
 	stmg	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC
 	stmg	%r8,%r9,__PT_PSW(%r11)
@@ -587,6 +634,8 @@ ENTRY(io_int_handler)
 	jo	.Lio_sigpending
 	tm	__TI_flags+7(%r12),_TIF_NOTIFY_RESUME
 	jo	.Lio_notify_resume
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	jo	.Lio_vxrs
 	tm	__LC_CPU_FLAGS+7,_CIF_ASCE
 	jo	.Lio_uaccess
 	j	.Lio_return		# beware of critical section cleanup
@@ -609,6 +658,13 @@ ENTRY(io_int_handler)
 	j	.Lio_return
 
 #
+# CIF_FPU is set, restore floating-point controls and floating-point registers.
+#
+.Lio_vxrs:
+	larl	%r14,.Lio_return
+	jg	load_fpu_regs
+
+#
 # _TIF_NEED_RESCHED is set, call schedule
 #
 .Lio_reschedule:
@@ -652,15 +708,9 @@ ENTRY(ext_int_handler)
 	stmg	%r8,%r15,__LC_SAVE_AREA_ASYNC
 	lg	%r10,__LC_LAST_BREAK
 	lg	%r12,__LC_THREAD_INFO
-	larl	%r13,system_call
+	larl	%r13,cleanup_critical
 	lmg	%r8,%r9,__LC_EXT_OLD_PSW
-	HANDLE_SIE_INTERCEPT %r14,3
-	SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_STACK,STACK_SHIFT
-	tmhh	%r8,0x0001		# interrupting from user ?
-	jz	.Lext_skip
-	UPDATE_VTIME %r14,__LC_ASYNC_ENTER_TIMER
-	LAST_BREAK %r14
-.Lext_skip:
+	SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_ENTER_TIMER
 	stmg	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC
 	stmg	%r8,%r9,__PT_PSW(%r11)
@@ -683,6 +733,14 @@ ENTRY(psw_idle)
 	stg	%r3,__SF_EMPTY(%r15)
 	larl	%r1,.Lpsw_idle_lpsw+4
 	stg	%r1,__SF_EMPTY+8(%r15)
+#ifdef CONFIG_SMP
+	larl	%r1,smp_cpu_mtid
+	llgf	%r1,0(%r1)
+	ltgr	%r1,%r1
+	jz	.Lpsw_idle_stcctm
+	.insn	rsy,0xeb0000000017,%r1,5,__SF_EMPTY+16(%r15)
+.Lpsw_idle_stcctm:
+#endif
 	STCK	__CLOCK_IDLE_ENTER(%r2)
 	stpt	__TIMER_IDLE_ENTER(%r2)
 .Lpsw_idle_lpsw:
@@ -690,6 +748,122 @@ ENTRY(psw_idle)
 	br	%r14
 .Lpsw_idle_end:
 
+/* Store floating-point controls and floating-point or vector extension
+ * registers instead.  A critical section cleanup assures that the registers
+ * are stored even if interrupted for some other work.	The register %r2
+ * designates a struct fpu to store register contents.	If the specified
+ * structure does not contain a register save area, the register store is
+ * omitted (see also comments in arch_dup_task_struct()).
+ *
+ * The CIF_FPU flag is set in any case.  The CIF_FPU triggers a lazy restore
+ * of the register contents at system call or io return.
+ */
+ENTRY(save_fpu_regs)
+	lg	%r2,__LC_CURRENT
+	aghi	%r2,__TASK_thread
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	bor	%r14
+	stfpc	__THREAD_FPU_fpc(%r2)
+.Lsave_fpu_regs_fpc_end:
+	lg	%r3,__THREAD_FPU_regs(%r2)
+	ltgr	%r3,%r3
+	jz	.Lsave_fpu_regs_done	  # no save area -> set CIF_FPU
+	tm	__THREAD_FPU_flags+3(%r2),FPU_USE_VX
+	jz	.Lsave_fpu_regs_fp	  # no -> store FP regs
+.Lsave_fpu_regs_vx_low:
+	VSTM	%v0,%v15,0,%r3		  # vstm 0,15,0(3)
+.Lsave_fpu_regs_vx_high:
+	VSTM	%v16,%v31,256,%r3	  # vstm 16,31,256(3)
+	j	.Lsave_fpu_regs_done	  # -> set CIF_FPU flag
+.Lsave_fpu_regs_fp:
+	std	0,0(%r3)
+	std	1,8(%r3)
+	std	2,16(%r3)
+	std	3,24(%r3)
+	std	4,32(%r3)
+	std	5,40(%r3)
+	std	6,48(%r3)
+	std	7,56(%r3)
+	std	8,64(%r3)
+	std	9,72(%r3)
+	std	10,80(%r3)
+	std	11,88(%r3)
+	std	12,96(%r3)
+	std	13,104(%r3)
+	std	14,112(%r3)
+	std	15,120(%r3)
+.Lsave_fpu_regs_done:
+	oi	__LC_CPU_FLAGS+7,_CIF_FPU
+	br	%r14
+.Lsave_fpu_regs_end:
+
+/* Load floating-point controls and floating-point or vector extension
+ * registers.  A critical section cleanup assures that the register contents
+ * are loaded even if interrupted for some other work.	Depending on the saved
+ * FP/VX state, the vector-enablement control, CR0.46, is either set or cleared.
+ *
+ * There are special calling conventions to fit into sysc and io return work:
+ *	%r15:	<kernel stack>
+ * The function requires:
+ *	%r4 and __SF_EMPTY+32(%r15)
+ */
+load_fpu_regs:
+	lg	%r4,__LC_CURRENT
+	aghi	%r4,__TASK_thread
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	bnor	%r14
+	lfpc	__THREAD_FPU_fpc(%r4)
+	stctg	%c0,%c0,__SF_EMPTY+32(%r15)	# store CR0
+	tm	__THREAD_FPU_flags+3(%r4),FPU_USE_VX	# VX-enabled task ?
+	lg	%r4,__THREAD_FPU_regs(%r4)	# %r4 <- reg save area
+	jz	.Lload_fpu_regs_fp_ctl		# -> no VX, load FP regs
+.Lload_fpu_regs_vx_ctl:
+	tm	__SF_EMPTY+32+5(%r15),2		# test VX control
+	jo	.Lload_fpu_regs_vx
+	oi	__SF_EMPTY+32+5(%r15),2		# set VX control
+	lctlg	%c0,%c0,__SF_EMPTY+32(%r15)
+.Lload_fpu_regs_vx:
+	VLM	%v0,%v15,0,%r4
+.Lload_fpu_regs_vx_high:
+	VLM	%v16,%v31,256,%r4
+	j	.Lload_fpu_regs_done
+.Lload_fpu_regs_fp_ctl:
+	tm	__SF_EMPTY+32+5(%r15),2		# test VX control
+	jz	.Lload_fpu_regs_fp
+	ni	__SF_EMPTY+32+5(%r15),253	# clear VX control
+	lctlg	%c0,%c0,__SF_EMPTY+32(%r15)
+.Lload_fpu_regs_fp:
+	ld	0,0(%r4)
+	ld	1,8(%r4)
+	ld	2,16(%r4)
+	ld	3,24(%r4)
+	ld	4,32(%r4)
+	ld	5,40(%r4)
+	ld	6,48(%r4)
+	ld	7,56(%r4)
+	ld	8,64(%r4)
+	ld	9,72(%r4)
+	ld	10,80(%r4)
+	ld	11,88(%r4)
+	ld	12,96(%r4)
+	ld	13,104(%r4)
+	ld	14,112(%r4)
+	ld	15,120(%r4)
+.Lload_fpu_regs_done:
+	ni	__LC_CPU_FLAGS+7,255-_CIF_FPU
+	br	%r14
+.Lload_fpu_regs_end:
+
+/* Test and set the vector enablement control in CR0.46 */
+ENTRY(__ctl_set_vx)
+	stctg	%c0,%c0,__SF_EMPTY(%r15)
+	tm	__SF_EMPTY+5(%r15),2
+	bor	%r14
+	oi	__SF_EMPTY+5(%r15),2
+	lctlg	%c0,%c0,__SF_EMPTY(%r15)
+	br	%r14
+.L__ctl_set_vx_end:
+
 .L__critical_end:
 
 /*
@@ -702,9 +876,8 @@ ENTRY(mcck_int_handler)
 	lmg	%r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r1)# revalidate gprs
 	lg	%r10,__LC_LAST_BREAK
 	lg	%r12,__LC_THREAD_INFO
-	larl	%r13,system_call
+	larl	%r13,cleanup_critical
 	lmg	%r8,%r9,__LC_MCK_OLD_PSW
-	HANDLE_SIE_INTERCEPT %r14,4
 	tm	__LC_MCCK_CODE,0x80	# system damage?
 	jo	.Lmcck_panic		# yes -> rest of mcck code invalid
 	lghi	%r14,__LC_CPU_TIMER_SAVE_AREA
@@ -725,11 +898,7 @@ ENTRY(mcck_int_handler)
 	mvc	__LC_MCCK_ENTER_TIMER(8),0(%r14)
 3:	tm	__LC_MCCK_CODE+2,0x09	# mwp + ia of old psw valid?
 	jno	.Lmcck_panic		# no -> skip cleanup critical
-	SWITCH_ASYNC __LC_GPREGS_SAVE_AREA+64,__LC_PANIC_STACK,PAGE_SHIFT
-	tm	%r8,0x0001		# interrupting from user ?
-	jz	.Lmcck_skip
-	UPDATE_VTIME %r14,__LC_MCCK_ENTER_TIMER
-	LAST_BREAK %r14
+	SWITCH_ASYNC __LC_GPREGS_SAVE_AREA+64,__LC_MCCK_ENTER_TIMER
 .Lmcck_skip:
 	lghi	%r14,__LC_GPREGS_SAVE_AREA+64
 	stmg	%r0,%r7,__PT_R0(%r11)
@@ -764,12 +933,8 @@ ENTRY(mcck_int_handler)
 	lpswe	__LC_RETURN_MCCK_PSW
 
 .Lmcck_panic:
-	lg	%r14,__LC_PANIC_STACK
-	slgr	%r14,%r15
-	srag	%r14,%r14,PAGE_SHIFT
-	jz	0f
 	lg	%r15,__LC_PANIC_STACK
-0:	aghi	%r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
+	aghi	%r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
 	j	.Lmcck_skip
 
 #
@@ -819,20 +984,13 @@ stack_overflow:
 	jg	kernel_stack_overflow
 #endif
 
-	.align	8
-.Lcleanup_table:
-	.quad	system_call
-	.quad	.Lsysc_do_svc
-	.quad	.Lsysc_tif
-	.quad	.Lsysc_restore
-	.quad	.Lsysc_done
-	.quad	.Lio_tif
-	.quad	.Lio_restore
-	.quad	.Lio_done
-	.quad	psw_idle
-	.quad	.Lpsw_idle_end
-
 cleanup_critical:
+#if IS_ENABLED(CONFIG_KVM)
+	clg	%r9,BASED(.Lcleanup_table_sie)	# .Lsie_gmap
+	jl	0f
+	clg	%r9,BASED(.Lcleanup_table_sie+8)# .Lsie_done
+	jl	.Lcleanup_sie
+#endif
 	clg	%r9,BASED(.Lcleanup_table)	# system_call
 	jl	0f
 	clg	%r9,BASED(.Lcleanup_table+8)	# .Lsysc_do_svc
@@ -853,8 +1011,54 @@ cleanup_critical:
 	jl	0f
 	clg	%r9,BASED(.Lcleanup_table+72)	# .Lpsw_idle_end
 	jl	.Lcleanup_idle
+	clg	%r9,BASED(.Lcleanup_table+80)	# save_fpu_regs
+	jl	0f
+	clg	%r9,BASED(.Lcleanup_table+88)	# .Lsave_fpu_regs_end
+	jl	.Lcleanup_save_fpu_regs
+	clg	%r9,BASED(.Lcleanup_table+96)	# load_fpu_regs
+	jl	0f
+	clg	%r9,BASED(.Lcleanup_table+104)	# .Lload_fpu_regs_end
+	jl	.Lcleanup_load_fpu_regs
+	clg	%r9,BASED(.Lcleanup_table+112)	# __ctl_set_vx
+	jl	0f
+	clg	%r9,BASED(.Lcleanup_table+120)	# .L__ctl_set_vx_end
+	jl	.Lcleanup___ctl_set_vx
 0:	br	%r14
 
+	.align	8
+.Lcleanup_table:
+	.quad	system_call
+	.quad	.Lsysc_do_svc
+	.quad	.Lsysc_tif
+	.quad	.Lsysc_restore
+	.quad	.Lsysc_done
+	.quad	.Lio_tif
+	.quad	.Lio_restore
+	.quad	.Lio_done
+	.quad	psw_idle
+	.quad	.Lpsw_idle_end
+	.quad	save_fpu_regs
+	.quad	.Lsave_fpu_regs_end
+	.quad	load_fpu_regs
+	.quad	.Lload_fpu_regs_end
+	.quad	__ctl_set_vx
+	.quad	.L__ctl_set_vx_end
+
+#if IS_ENABLED(CONFIG_KVM)
+.Lcleanup_table_sie:
+	.quad	.Lsie_gmap
+	.quad	.Lsie_done
+
+.Lcleanup_sie:
+	lg	%r9,__SF_EMPTY(%r15)		# get control block pointer
+	tm	__LC_MACHINE_FLAGS+6,0x20	# MACHINE_FLAG_LPP
+	jz	0f
+	.insn	s,0xb2800000,__SF_EMPTY+16(%r15)# set host id
+0:	ni	__SIE_PROG0C+3(%r9),0xfe	# no longer in SIE
+	lctlg	%c1,%c1,__LC_USER_ASCE		# load primary asce
+	larl	%r9,sie_exit			# skip forward to sie_exit
+	br	%r14
+#endif
 
 .Lcleanup_system_call:
 	# check if stpt has been executed
@@ -915,7 +1119,7 @@ cleanup_critical:
 	.quad	system_call
 	.quad	.Lsysc_stmg
 	.quad	.Lsysc_per
-	.quad	.Lsysc_vtime+18
+	.quad	.Lsysc_vtime+36
 	.quad	.Lsysc_vtime+42
 
 .Lcleanup_sysc_tif:
@@ -963,7 +1167,27 @@ cleanup_critical:
 	jhe	1f
 	mvc	__CLOCK_IDLE_ENTER(8,%r2),__CLOCK_IDLE_EXIT(%r2)
 	mvc	__TIMER_IDLE_ENTER(8,%r2),__TIMER_IDLE_EXIT(%r2)
-1:	# account system time going idle
+1:	# calculate idle cycles
+#ifdef CONFIG_SMP
+	clg	%r9,BASED(.Lcleanup_idle_insn)
+	jl	3f
+	larl	%r1,smp_cpu_mtid
+	llgf	%r1,0(%r1)
+	ltgr	%r1,%r1
+	jz	3f
+	.insn	rsy,0xeb0000000017,%r1,5,__SF_EMPTY+80(%r15)
+	larl	%r3,mt_cycles
+	ag	%r3,__LC_PERCPU_OFFSET
+	la	%r4,__SF_EMPTY+16(%r15)
+2:	lg	%r0,0(%r3)
+	slg	%r0,0(%r4)
+	alg	%r0,64(%r4)
+	stg	%r0,0(%r3)
+	la	%r3,8(%r3)
+	la	%r4,8(%r4)
+	brct	%r1,2b
+#endif
+3:	# account system time going idle
 	lg	%r9,__LC_STEAL_TIMER
 	alg	%r9,__CLOCK_IDLE_ENTER(%r2)
 	slg	%r9,__LC_LAST_UPDATE_CLOCK
@@ -981,6 +1205,147 @@ cleanup_critical:
 .Lcleanup_idle_insn:
 	.quad	.Lpsw_idle_lpsw
 
+.Lcleanup_save_fpu_regs:
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	bor	%r14
+	clg	%r9,BASED(.Lcleanup_save_fpu_regs_done)
+	jhe	5f
+	clg	%r9,BASED(.Lcleanup_save_fpu_regs_fp)
+	jhe	4f
+	clg	%r9,BASED(.Lcleanup_save_fpu_regs_vx_high)
+	jhe	3f
+	clg	%r9,BASED(.Lcleanup_save_fpu_regs_vx_low)
+	jhe	2f
+	clg	%r9,BASED(.Lcleanup_save_fpu_fpc_end)
+	jhe	1f
+	lg	%r2,__LC_CURRENT
+	aghi	%r2,__TASK_thread
+0:	# Store floating-point controls
+	stfpc	__THREAD_FPU_fpc(%r2)
+1:	# Load register save area and check if VX is active
+	lg	%r3,__THREAD_FPU_regs(%r2)
+	ltgr	%r3,%r3
+	jz	5f			  # no save area -> set CIF_FPU
+	tm	__THREAD_FPU_flags+3(%r2),FPU_USE_VX
+	jz	4f			  # no VX -> store FP regs
+2:	# Store vector registers (V0-V15)
+	VSTM	%v0,%v15,0,%r3		  # vstm 0,15,0(3)
+3:	# Store vector registers (V16-V31)
+	VSTM	%v16,%v31,256,%r3	  # vstm 16,31,256(3)
+	j	5f			  # -> done, set CIF_FPU flag
+4:	# Store floating-point registers
+	std	0,0(%r3)
+	std	1,8(%r3)
+	std	2,16(%r3)
+	std	3,24(%r3)
+	std	4,32(%r3)
+	std	5,40(%r3)
+	std	6,48(%r3)
+	std	7,56(%r3)
+	std	8,64(%r3)
+	std	9,72(%r3)
+	std	10,80(%r3)
+	std	11,88(%r3)
+	std	12,96(%r3)
+	std	13,104(%r3)
+	std	14,112(%r3)
+	std	15,120(%r3)
+5:	# Set CIF_FPU flag
+	oi	__LC_CPU_FLAGS+7,_CIF_FPU
+	lg	%r9,48(%r11)		# return from save_fpu_regs
+	br	%r14
+.Lcleanup_save_fpu_fpc_end:
+	.quad	.Lsave_fpu_regs_fpc_end
+.Lcleanup_save_fpu_regs_vx_low:
+	.quad	.Lsave_fpu_regs_vx_low
+.Lcleanup_save_fpu_regs_vx_high:
+	.quad	.Lsave_fpu_regs_vx_high
+.Lcleanup_save_fpu_regs_fp:
+	.quad	.Lsave_fpu_regs_fp
+.Lcleanup_save_fpu_regs_done:
+	.quad	.Lsave_fpu_regs_done
+
+.Lcleanup_load_fpu_regs:
+	tm	__LC_CPU_FLAGS+7,_CIF_FPU
+	bnor	%r14
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_done)
+	jhe	1f
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_fp)
+	jhe	2f
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_fp_ctl)
+	jhe	3f
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_vx_high)
+	jhe	4f
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_vx)
+	jhe	5f
+	clg	%r9,BASED(.Lcleanup_load_fpu_regs_vx_ctl)
+	jhe	6f
+	lg	%r4,__LC_CURRENT
+	aghi	%r4,__TASK_thread
+	lfpc	__THREAD_FPU_fpc(%r4)
+	tm	__THREAD_FPU_flags+3(%r4),FPU_USE_VX	# VX-enabled task ?
+	lg	%r4,__THREAD_FPU_regs(%r4)	# %r4 <- reg save area
+	jz	3f				# -> no VX, load FP regs
+6:	# Set VX-enablement control
+	stctg	%c0,%c0,__SF_EMPTY+32(%r15)	# store CR0
+	tm	__SF_EMPTY+32+5(%r15),2		# test VX control
+	jo	5f
+	oi	__SF_EMPTY+32+5(%r15),2		# set VX control
+	lctlg	%c0,%c0,__SF_EMPTY+32(%r15)
+5:	# Load V0 ..V15 registers
+	VLM	%v0,%v15,0,%r4
+4:	# Load V16..V31 registers
+	VLM	%v16,%v31,256,%r4
+	j	1f
+3:	# Clear VX-enablement control for FP
+	stctg	%c0,%c0,__SF_EMPTY+32(%r15)	# store CR0
+	tm	__SF_EMPTY+32+5(%r15),2		# test VX control
+	jz	2f
+	ni	__SF_EMPTY+32+5(%r15),253	# clear VX control
+	lctlg	%c0,%c0,__SF_EMPTY+32(%r15)
+2:	# Load floating-point registers
+	ld	0,0(%r4)
+	ld	1,8(%r4)
+	ld	2,16(%r4)
+	ld	3,24(%r4)
+	ld	4,32(%r4)
+	ld	5,40(%r4)
+	ld	6,48(%r4)
+	ld	7,56(%r4)
+	ld	8,64(%r4)
+	ld	9,72(%r4)
+	ld	10,80(%r4)
+	ld	11,88(%r4)
+	ld	12,96(%r4)
+	ld	13,104(%r4)
+	ld	14,112(%r4)
+	ld	15,120(%r4)
+1:	# Clear CIF_FPU bit
+	ni	__LC_CPU_FLAGS+7,255-_CIF_FPU
+	lg	%r9,48(%r11)		# return from load_fpu_regs
+	br	%r14
+.Lcleanup_load_fpu_regs_vx_ctl:
+	.quad	.Lload_fpu_regs_vx_ctl
+.Lcleanup_load_fpu_regs_vx:
+	.quad	.Lload_fpu_regs_vx
+.Lcleanup_load_fpu_regs_vx_high:
+	.quad	.Lload_fpu_regs_vx_high
+.Lcleanup_load_fpu_regs_fp_ctl:
+	.quad	.Lload_fpu_regs_fp_ctl
+.Lcleanup_load_fpu_regs_fp:
+	.quad	.Lload_fpu_regs_fp
+.Lcleanup_load_fpu_regs_done:
+	.quad	.Lload_fpu_regs_done
+
+.Lcleanup___ctl_set_vx:
+	stctg	%c0,%c0,__SF_EMPTY(%r15)
+	tm	__SF_EMPTY+5(%r15),2
+	bor	%r14
+	oi	__SF_EMPTY+5(%r15),2
+	lctlg	%c0,%c0,__SF_EMPTY(%r15)
+	lg	%r9,48(%r11)		# return from __ctl_set_vx
+	br	%r14
+
 /*
  * Integer constants
  */
@@ -989,62 +1354,11 @@ cleanup_critical:
 	.quad	.L__critical_start
 .Lcritical_length:
 	.quad	.L__critical_end - .L__critical_start
-
-
 #if IS_ENABLED(CONFIG_KVM)
-/*
- * sie64a calling convention:
- * %r2 pointer to sie control block
- * %r3 guest register save area
- */
-ENTRY(sie64a)
-	stmg	%r6,%r14,__SF_GPRS(%r15)	# save kernel registers
-	stg	%r2,__SF_EMPTY(%r15)		# save control block pointer
-	stg	%r3,__SF_EMPTY+8(%r15)		# save guest register save area
-	xc	__SF_EMPTY+16(16,%r15),__SF_EMPTY+16(%r15) # host id & reason
-	lmg	%r0,%r13,0(%r3)			# load guest gprs 0-13
-	lg	%r14,__LC_GMAP			# get gmap pointer
-	ltgr	%r14,%r14
-	jz	.Lsie_gmap
-	lctlg	%c1,%c1,__GMAP_ASCE(%r14)	# load primary asce
-.Lsie_gmap:
-	lg	%r14,__SF_EMPTY(%r15)		# get control block pointer
-	oi	__SIE_PROG0C+3(%r14),1		# we are going into SIE now
-	tm	__SIE_PROG20+3(%r14),3		# last exit...
-	jnz	.Lsie_done
-	LPP	__SF_EMPTY(%r15)		# set guest id
-	sie	0(%r14)
-.Lsie_done:
-	LPP	__SF_EMPTY+16(%r15)		# set host id
-	ni	__SIE_PROG0C+3(%r14),0xfe	# no longer in SIE
-	lctlg	%c1,%c1,__LC_USER_ASCE		# load primary asce
-# some program checks are suppressing. C code (e.g. do_protection_exception)
-# will rewind the PSW by the ILC, which is 4 bytes in case of SIE. Other
-# instructions between sie64a and .Lsie_done should not cause program
-# interrupts. So lets use a nop (47 00 00 00) as a landing pad.
-# See also HANDLE_SIE_INTERCEPT
-.Lrewind_pad:
-	nop	0
-	.globl sie_exit
-sie_exit:
-	lg	%r14,__SF_EMPTY+8(%r15)		# load guest register save area
-	stmg	%r0,%r13,0(%r14)		# save guest gprs 0-13
-	lmg	%r6,%r14,__SF_GPRS(%r15)	# restore kernel registers
-	lg	%r2,__SF_EMPTY+24(%r15)		# return exit reason code
-	br	%r14
-.Lsie_fault:
-	lghi	%r14,-EFAULT
-	stg	%r14,__SF_EMPTY+24(%r15)	# set exit reason code
-	j	sie_exit
-
-	.align	8
-.Lsie_critical:
+.Lsie_critical_start:
 	.quad	.Lsie_gmap
 .Lsie_critical_length:
 	.quad	.Lsie_done - .Lsie_gmap
-
-	EX_TABLE(.Lrewind_pad,.Lsie_fault)
-	EX_TABLE(sie_exit,.Lsie_fault)
 #endif
 
 	.section .rodata, "a"
diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S
index 59b7c6470567..1255c6c5353e 100644
--- a/arch/s390/kernel/head.S
+++ b/arch/s390/kernel/head.S
@@ -370,6 +370,7 @@ ENTRY(startup_kdump)
 	xc	0x200(256),0x200	# partially clear lowcore
 	xc	0x300(256),0x300
 	xc	0xe00(256),0xe00
+	lctlg	%c0,%c15,0x200(%r0)	# initialize control registers
 	stck	__LC_LAST_UPDATE_CLOCK
 	spt	6f-.LPG0(%r13)
 	mvc	__LC_LAST_UPDATE_TIMER(8),6f-.LPG0(%r13)
@@ -413,9 +414,9 @@ ENTRY(startup_kdump)
 # followed by the facility words.
 
 #if defined(CONFIG_MARCH_Z13)
-	.long 3, 0xc100eff2, 0xf46ce800, 0x00400000
+	.long 2, 0xc100eff2, 0xf46cc800
 #elif defined(CONFIG_MARCH_ZEC12)
-	.long 3, 0xc100eff2, 0xf46ce800, 0x00400000
+	.long 2, 0xc100eff2, 0xf46cc800
 #elif defined(CONFIG_MARCH_Z196)
 	.long 2, 0xc100eff2, 0xf46c0000
 #elif defined(CONFIG_MARCH_Z10)
diff --git a/arch/s390/kernel/jump_label.c b/arch/s390/kernel/jump_label.c
index a90299600483..083b05f5f5ab 100644
--- a/arch/s390/kernel/jump_label.c
+++ b/arch/s390/kernel/jump_label.c
@@ -44,12 +44,9 @@ static void jump_label_bug(struct jump_entry *entry, struct insn *expected,
 	unsigned char *ipn = (unsigned char *)new;
 
 	pr_emerg("Jump label code mismatch at %pS [%p]\n", ipc, ipc);
-	pr_emerg("Found:    %02x %02x %02x %02x %02x %02x\n",
-		 ipc[0], ipc[1], ipc[2], ipc[3], ipc[4], ipc[5]);
-	pr_emerg("Expected: %02x %02x %02x %02x %02x %02x\n",
-		 ipe[0], ipe[1], ipe[2], ipe[3], ipe[4], ipe[5]);
-	pr_emerg("New:      %02x %02x %02x %02x %02x %02x\n",
-		 ipn[0], ipn[1], ipn[2], ipn[3], ipn[4], ipn[5]);
+	pr_emerg("Found:    %6ph\n", ipc);
+	pr_emerg("Expected: %6ph\n", ipe);
+	pr_emerg("New:      %6ph\n", ipn);
 	panic("Corrupted kernel text");
 }
 
@@ -64,7 +61,7 @@ static void __jump_label_transform(struct jump_entry *entry,
 {
 	struct insn old, new;
 
-	if (type == JUMP_LABEL_ENABLE) {
+	if (type == JUMP_LABEL_JMP) {
 		jump_label_make_nop(entry, &old);
 		jump_label_make_branch(entry, &new);
 	} else {
diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c
index 56b550893593..0ae6f8e74840 100644
--- a/arch/s390/kernel/nmi.c
+++ b/arch/s390/kernel/nmi.c
@@ -21,6 +21,7 @@
 #include <asm/nmi.h>
 #include <asm/crw.h>
 #include <asm/switch_to.h>
+#include <asm/fpu-internal.h>
 #include <asm/ctl_reg.h>
 
 struct mcck_struct {
@@ -164,8 +165,12 @@ static int notrace s390_revalidate_registers(struct mci *mci)
 		cr0.val = S390_lowcore.cregs_save_area[0];
 		cr0.afp = cr0.vx = 1;
 		__ctl_load(cr0.val, 0, 0);
-		restore_vx_regs((__vector128 *)
-				&S390_lowcore.vector_save_area);
+		asm volatile(
+			"	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");
 		__ctl_load(S390_lowcore.cregs_save_area[0], 0, 0);
 	}
 	/* Revalidate access registers */
@@ -358,4 +363,4 @@ static int __init machine_check_init(void)
 	ctl_set_bit(14, 24);	/* enable warning MCH */
 	return 0;
 }
-arch_initcall(machine_check_init);
+early_initcall(machine_check_init);
diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c
index 56fdad479115..a9563409c36e 100644
--- a/arch/s390/kernel/perf_cpum_cf.c
+++ b/arch/s390/kernel/perf_cpum_cf.c
@@ -157,10 +157,14 @@ static int validate_ctr_auth(const struct hw_perf_event *hwc)
 
 	cpuhw = &get_cpu_var(cpu_hw_events);
 
-	/* check authorization for cpu counter sets */
+	/* Check authorization for cpu counter sets.
+	 * If the particular CPU counter set is not authorized,
+	 * return with -ENOENT in order to fall back to other
+	 * PMUs that might suffice the event request.
+	 */
 	ctrs_state = cpumf_state_ctl[hwc->config_base];
 	if (!(ctrs_state & cpuhw->info.auth_ctl))
-		err = -EPERM;
+		err = -ENOENT;
 
 	put_cpu_var(cpu_hw_events);
 	return err;
@@ -536,7 +540,7 @@ static int cpumf_pmu_add(struct perf_event *event, int flags)
 	 */
 	if (!(cpuhw->flags & PERF_EVENT_TXN))
 		if (validate_ctr_auth(&event->hw))
-			return -EPERM;
+			return -ENOENT;
 
 	ctr_set_enable(&cpuhw->state, event->hw.config_base);
 	event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
@@ -611,7 +615,7 @@ static int cpumf_pmu_commit_txn(struct pmu *pmu)
 	state = cpuhw->state & ~((1 << CPUMF_LCCTL_ENABLE_SHIFT) - 1);
 	state >>= CPUMF_LCCTL_ENABLE_SHIFT;
 	if ((state & cpuhw->info.auth_ctl) != state)
-		return -EPERM;
+		return -ENOENT;
 
 	cpuhw->flags &= ~PERF_EVENT_TXN;
 	perf_pmu_enable(pmu);
diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c
index afe05bfb7e00..b973972f6ba5 100644
--- a/arch/s390/kernel/perf_cpum_sf.c
+++ b/arch/s390/kernel/perf_cpum_sf.c
@@ -1019,12 +1019,9 @@ static int perf_push_sample(struct perf_event *event, struct sf_raw_sample *sfr)
 		break;
 	}
 
-	/* The host-program-parameter (hpp) contains the sie control
-	 * block that is set by sie64a() in entry64.S.	Check if hpp
-	 * refers to a valid control block and set sde_regs flags
-	 * accordingly.  This would allow to use hpp values for other
-	 * purposes too.
-	 * For now, simply use a non-zero value as guest indicator.
+	/* The host-program-parameter (hpp) contains the pid of
+	 * the CPU thread as set by sie64a() in entry.S.
+	 * If non-zero assume a guest sample.
 	 */
 	if (sfr->basic.hpp)
 		sde_regs->in_guest = 1;
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index 8f587d871b9f..f2dac9f0799d 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -81,8 +81,38 @@ void release_thread(struct task_struct *dead_task)
 
 void arch_release_task_struct(struct task_struct *tsk)
 {
-	if (tsk->thread.vxrs)
-		kfree(tsk->thread.vxrs);
+	/* Free either the floating-point or the vector register save area */
+	kfree(tsk->thread.fpu.regs);
+}
+
+int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
+{
+	*dst = *src;
+
+	/* Set up a new floating-point register save area */
+	dst->thread.fpu.fpc = 0;
+	dst->thread.fpu.flags = 0;	/* Always start with VX disabled */
+	dst->thread.fpu.fprs = kzalloc(sizeof(freg_t) * __NUM_FPRS,
+				       GFP_KERNEL|__GFP_REPEAT);
+	if (!dst->thread.fpu.fprs)
+		return -ENOMEM;
+
+	/*
+	 * Save the floating-point or vector register state of the current
+	 * task.  The state is not saved for early kernel threads, for example,
+	 * the init_task, which do not have an allocated save area.
+	 * The CIF_FPU flag is set in any case to lazy clear or restore a saved
+	 * state when switching to a different task or returning to user space.
+	 */
+	save_fpu_regs();
+	dst->thread.fpu.fpc = current->thread.fpu.fpc;
+	if (is_vx_task(current))
+		convert_vx_to_fp(dst->thread.fpu.fprs,
+				 current->thread.fpu.vxrs);
+	else
+		memcpy(dst->thread.fpu.fprs, current->thread.fpu.fprs,
+		       sizeof(freg_t) * __NUM_FPRS);
+	return 0;
 }
 
 int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
@@ -142,11 +172,6 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
 	p->thread.ri_signum = 0;
 	frame->childregs.psw.mask &= ~PSW_MASK_RI;
 
-	/* Save the fpu registers to new thread structure. */
-	save_fp_ctl(&p->thread.fp_regs.fpc);
-	save_fp_regs(p->thread.fp_regs.fprs);
-	p->thread.fp_regs.pad = 0;
-	p->thread.vxrs = NULL;
 	/* Set a new TLS ?  */
 	if (clone_flags & CLONE_SETTLS) {
 		unsigned long tls = frame->childregs.gprs[6];
@@ -162,7 +187,7 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
 
 asmlinkage void execve_tail(void)
 {
-	current->thread.fp_regs.fpc = 0;
+	current->thread.fpu.fpc = 0;
 	asm volatile("sfpc %0" : : "d" (0));
 }
 
@@ -171,8 +196,15 @@ asmlinkage void execve_tail(void)
  */
 int dump_fpu (struct pt_regs * regs, s390_fp_regs *fpregs)
 {
-	save_fp_ctl(&fpregs->fpc);
-	save_fp_regs(fpregs->fprs);
+	save_fpu_regs();
+	fpregs->fpc = current->thread.fpu.fpc;
+	fpregs->pad = 0;
+	if (is_vx_task(current))
+		convert_vx_to_fp((freg_t *)&fpregs->fprs,
+				 current->thread.fpu.vxrs);
+	else
+		memcpy(&fpregs->fprs, current->thread.fpu.fprs,
+		       sizeof(fpregs->fprs));
 	return 1;
 }
 EXPORT_SYMBOL(dump_fpu);
diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c
index dc488e13b7e3..e6e077ae3990 100644
--- a/arch/s390/kernel/processor.c
+++ b/arch/s390/kernel/processor.c
@@ -41,6 +41,15 @@ void cpu_init(void)
 }
 
 /*
+ * cpu_have_feature - Test CPU features on module initialization
+ */
+int cpu_have_feature(unsigned int num)
+{
+	return elf_hwcap & (1UL << num);
+}
+EXPORT_SYMBOL(cpu_have_feature);
+
+/*
  * show_cpuinfo - Get information on one CPU for use by procfs.
  */
 static int show_cpuinfo(struct seq_file *m, void *v)
diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index d363c9c322a1..8b1c8e33f184 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -45,39 +45,27 @@ void update_cr_regs(struct task_struct *task)
 	struct per_regs old, new;
 
 	/* Take care of the enable/disable of transactional execution. */
-	if (MACHINE_HAS_TE || MACHINE_HAS_VX) {
+	if (MACHINE_HAS_TE) {
 		unsigned long cr, cr_new;
 
 		__ctl_store(cr, 0, 0);
-		cr_new = cr;
-		if (MACHINE_HAS_TE) {
-			/* Set or clear transaction execution TXC bit 8. */
-			cr_new |= (1UL << 55);
-			if (task->thread.per_flags & PER_FLAG_NO_TE)
-				cr_new &= ~(1UL << 55);
-		}
-		if (MACHINE_HAS_VX) {
-			/* Enable/disable of vector extension */
-			cr_new &= ~(1UL << 17);
-			if (task->thread.vxrs)
-				cr_new |= (1UL << 17);
-		}
+		/* Set or clear transaction execution TXC bit 8. */
+		cr_new = cr | (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);
-		if (MACHINE_HAS_TE) {
-			/* Set/clear transaction execution TDC bits 62/63. */
-			__ctl_store(cr, 2, 2);
-			cr_new = cr & ~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;
-				else
-					cr_new |= 2UL;
-			}
-			if (cr_new != cr)
-				__ctl_load(cr_new, 2, 2);
+		/* Set or clear transaction execution TDC bits 62 and 63. */
+		__ctl_store(cr, 2, 2);
+		cr_new = cr & ~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;
+			else
+				cr_new |= 2UL;
 		}
+		if (cr_new != cr)
+			__ctl_load(cr_new, 2, 2);
 	}
 	/* Copy user specified PER registers */
 	new.control = thread->per_user.control;
@@ -242,21 +230,21 @@ static unsigned long __peek_user(struct task_struct *child, addr_t addr)
 		/*
 		 * floating point control reg. is in the thread structure
 		 */
-		tmp = child->thread.fp_regs.fpc;
+		tmp = child->thread.fpu.fpc;
 		tmp <<= BITS_PER_LONG - 32;
 
 	} else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) {
 		/*
-		 * floating point regs. are either in child->thread.fp_regs
-		 * or the child->thread.vxrs array
+		 * floating point regs. are either in child->thread.fpu
+		 * or the child->thread.fpu.vxrs array
 		 */
 		offset = addr - (addr_t) &dummy->regs.fp_regs.fprs;
-		if (child->thread.vxrs)
+		if (is_vx_task(child))
 			tmp = *(addr_t *)
-			       ((addr_t) child->thread.vxrs + 2*offset);
+			       ((addr_t) child->thread.fpu.vxrs + 2*offset);
 		else
 			tmp = *(addr_t *)
-			       ((addr_t) &child->thread.fp_regs.fprs + offset);
+			       ((addr_t) &child->thread.fpu.fprs + offset);
 
 	} else if (addr < (addr_t) (&dummy->regs.per_info + 1)) {
 		/*
@@ -387,20 +375,20 @@ static int __poke_user(struct task_struct *child, addr_t addr, addr_t data)
 		if ((unsigned int) data != 0 ||
 		    test_fp_ctl(data >> (BITS_PER_LONG - 32)))
 			return -EINVAL;
-		child->thread.fp_regs.fpc = data >> (BITS_PER_LONG - 32);
+		child->thread.fpu.fpc = data >> (BITS_PER_LONG - 32);
 
 	} else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) {
 		/*
-		 * floating point regs. are either in child->thread.fp_regs
-		 * or the child->thread.vxrs array
+		 * floating point regs. are either in child->thread.fpu
+		 * or the child->thread.fpu.vxrs array
 		 */
 		offset = addr - (addr_t) &dummy->regs.fp_regs.fprs;
-		if (child->thread.vxrs)
+		if (is_vx_task(child))
 			*(addr_t *)((addr_t)
-				child->thread.vxrs + 2*offset) = data;
+				child->thread.fpu.vxrs + 2*offset) = data;
 		else
 			*(addr_t *)((addr_t)
-				&child->thread.fp_regs.fprs + offset) = data;
+				&child->thread.fpu.fprs + offset) = data;
 
 	} else if (addr < (addr_t) (&dummy->regs.per_info + 1)) {
 		/*
@@ -621,20 +609,20 @@ static u32 __peek_user_compat(struct task_struct *child, addr_t addr)
 		/*
 		 * floating point control reg. is in the thread structure
 		 */
-		tmp = child->thread.fp_regs.fpc;
+		tmp = child->thread.fpu.fpc;
 
 	} else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) {
 		/*
-		 * floating point regs. are either in child->thread.fp_regs
-		 * or the child->thread.vxrs array
+		 * floating point regs. are either in child->thread.fpu
+		 * or the child->thread.fpu.vxrs array
 		 */
 		offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs;
-		if (child->thread.vxrs)
+		if (is_vx_task(child))
 			tmp = *(__u32 *)
-			       ((addr_t) child->thread.vxrs + 2*offset);
+			       ((addr_t) child->thread.fpu.vxrs + 2*offset);
 		else
 			tmp = *(__u32 *)
-			       ((addr_t) &child->thread.fp_regs.fprs + offset);
+			       ((addr_t) &child->thread.fpu.fprs + offset);
 
 	} else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) {
 		/*
@@ -746,20 +734,20 @@ static int __poke_user_compat(struct task_struct *child,
 		 */
 		if (test_fp_ctl(tmp))
 			return -EINVAL;
-		child->thread.fp_regs.fpc = data;
+		child->thread.fpu.fpc = data;
 
 	} else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) {
 		/*
-		 * floating point regs. are either in child->thread.fp_regs
-		 * or the child->thread.vxrs array
+		 * floating point regs. are either in child->thread.fpu
+		 * or the child->thread.fpu.vxrs array
 		 */
 		offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs;
-		if (child->thread.vxrs)
+		if (is_vx_task(child))
 			*(__u32 *)((addr_t)
-				child->thread.vxrs + 2*offset) = tmp;
+				child->thread.fpu.vxrs + 2*offset) = tmp;
 		else
 			*(__u32 *)((addr_t)
-				&child->thread.fp_regs.fprs + offset) = tmp;
+				&child->thread.fpu.fprs + offset) = tmp;
 
 	} else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) {
 		/*
@@ -952,18 +940,16 @@ static int s390_fpregs_get(struct task_struct *target,
 			   const struct user_regset *regset, unsigned int pos,
 			   unsigned int count, void *kbuf, void __user *ubuf)
 {
-	if (target == current) {
-		save_fp_ctl(&target->thread.fp_regs.fpc);
-		save_fp_regs(target->thread.fp_regs.fprs);
-	} else if (target->thread.vxrs) {
-		int i;
+	_s390_fp_regs fp_regs;
+
+	if (target == current)
+		save_fpu_regs();
+
+	fp_regs.fpc = target->thread.fpu.fpc;
+	fpregs_store(&fp_regs, &target->thread.fpu);
 
-		for (i = 0; i < __NUM_VXRS_LOW; i++)
-			target->thread.fp_regs.fprs[i] =
-				*(freg_t *)(target->thread.vxrs + i);
-	}
 	return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
-				   &target->thread.fp_regs, 0, -1);
+				   &fp_regs, 0, -1);
 }
 
 static int s390_fpregs_set(struct task_struct *target,
@@ -972,41 +958,33 @@ static int s390_fpregs_set(struct task_struct *target,
 			   const void __user *ubuf)
 {
 	int rc = 0;
+	freg_t fprs[__NUM_FPRS];
 
-	if (target == current) {
-		save_fp_ctl(&target->thread.fp_regs.fpc);
-		save_fp_regs(target->thread.fp_regs.fprs);
-	}
+	if (target == current)
+		save_fpu_regs();
 
 	/* If setting FPC, must validate it first. */
 	if (count > 0 && pos < offsetof(s390_fp_regs, fprs)) {
-		u32 ufpc[2] = { target->thread.fp_regs.fpc, 0 };
+		u32 ufpc[2] = { target->thread.fpu.fpc, 0 };
 		rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &ufpc,
 					0, offsetof(s390_fp_regs, fprs));
 		if (rc)
 			return rc;
 		if (ufpc[1] != 0 || test_fp_ctl(ufpc[0]))
 			return -EINVAL;
-		target->thread.fp_regs.fpc = ufpc[0];
+		target->thread.fpu.fpc = ufpc[0];
 	}
 
 	if (rc == 0 && count > 0)
 		rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
-					target->thread.fp_regs.fprs,
-					offsetof(s390_fp_regs, fprs), -1);
-
-	if (rc == 0) {
-		if (target == current) {
-			restore_fp_ctl(&target->thread.fp_regs.fpc);
-			restore_fp_regs(target->thread.fp_regs.fprs);
-		} else if (target->thread.vxrs) {
-			int i;
-
-			for (i = 0; i < __NUM_VXRS_LOW; i++)
-				*(freg_t *)(target->thread.vxrs + i) =
-					target->thread.fp_regs.fprs[i];
-		}
-	}
+					fprs, offsetof(s390_fp_regs, fprs), -1);
+	if (rc)
+		return rc;
+
+	if (is_vx_task(target))
+		convert_fp_to_vx(target->thread.fpu.vxrs, fprs);
+	else
+		memcpy(target->thread.fpu.fprs, &fprs, sizeof(fprs));
 
 	return rc;
 }
@@ -1069,11 +1047,11 @@ static int s390_vxrs_low_get(struct task_struct *target,
 
 	if (!MACHINE_HAS_VX)
 		return -ENODEV;
-	if (target->thread.vxrs) {
+	if (is_vx_task(target)) {
 		if (target == current)
-			save_vx_regs(target->thread.vxrs);
+			save_fpu_regs();
 		for (i = 0; i < __NUM_VXRS_LOW; i++)
-			vxrs[i] = *((__u64 *)(target->thread.vxrs + i) + 1);
+			vxrs[i] = *((__u64 *)(target->thread.fpu.vxrs + i) + 1);
 	} else
 		memset(vxrs, 0, sizeof(vxrs));
 	return user_regset_copyout(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1);
@@ -1089,20 +1067,17 @@ static int s390_vxrs_low_set(struct task_struct *target,
 
 	if (!MACHINE_HAS_VX)
 		return -ENODEV;
-	if (!target->thread.vxrs) {
+	if (!is_vx_task(target)) {
 		rc = alloc_vector_registers(target);
 		if (rc)
 			return rc;
 	} else if (target == current)
-		save_vx_regs(target->thread.vxrs);
+		save_fpu_regs();
 
 	rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1);
-	if (rc == 0) {
+	if (rc == 0)
 		for (i = 0; i < __NUM_VXRS_LOW; i++)
-			*((__u64 *)(target->thread.vxrs + i) + 1) = vxrs[i];
-		if (target == current)
-			restore_vx_regs(target->thread.vxrs);
-	}
+			*((__u64 *)(target->thread.fpu.vxrs + i) + 1) = vxrs[i];
 
 	return rc;
 }
@@ -1116,10 +1091,10 @@ static int s390_vxrs_high_get(struct task_struct *target,
 
 	if (!MACHINE_HAS_VX)
 		return -ENODEV;
-	if (target->thread.vxrs) {
+	if (is_vx_task(target)) {
 		if (target == current)
-			save_vx_regs(target->thread.vxrs);
-		memcpy(vxrs, target->thread.vxrs + __NUM_VXRS_LOW,
+			save_fpu_regs();
+		memcpy(vxrs, target->thread.fpu.vxrs + __NUM_VXRS_LOW,
 		       sizeof(vxrs));
 	} else
 		memset(vxrs, 0, sizeof(vxrs));
@@ -1135,18 +1110,15 @@ static int s390_vxrs_high_set(struct task_struct *target,
 
 	if (!MACHINE_HAS_VX)
 		return -ENODEV;
-	if (!target->thread.vxrs) {
+	if (!is_vx_task(target)) {
 		rc = alloc_vector_registers(target);
 		if (rc)
 			return rc;
 	} else if (target == current)
-		save_vx_regs(target->thread.vxrs);
+		save_fpu_regs();
 
 	rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
-				target->thread.vxrs + __NUM_VXRS_LOW, 0, -1);
-	if (rc == 0 && target == current)
-		restore_vx_regs(target->thread.vxrs);
-
+				target->thread.fpu.vxrs + __NUM_VXRS_LOW, 0, -1);
 	return rc;
 }
 
diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c
index 9f60467938d1..5090d3dad10b 100644
--- a/arch/s390/kernel/s390_ksyms.c
+++ b/arch/s390/kernel/s390_ksyms.c
@@ -1,5 +1,6 @@
 #include <linux/module.h>
 #include <linux/kvm_host.h>
+#include <asm/fpu-internal.h>
 #include <asm/ftrace.h>
 
 #ifdef CONFIG_FUNCTION_TRACER
@@ -8,6 +9,8 @@ EXPORT_SYMBOL(_mcount);
 #if IS_ENABLED(CONFIG_KVM)
 EXPORT_SYMBOL(sie64a);
 EXPORT_SYMBOL(sie_exit);
+EXPORT_SYMBOL(save_fpu_regs);
+EXPORT_SYMBOL(__ctl_set_vx);
 #endif
 EXPORT_SYMBOL(memcpy);
 EXPORT_SYMBOL(memset);
diff --git a/arch/s390/kernel/sclp.S b/arch/s390/kernel/sclp.S
deleted file mode 100644
index ada0c07fe1a8..000000000000
--- a/arch/s390/kernel/sclp.S
+++ /dev/null
@@ -1,355 +0,0 @@
-/*
- * Mini SCLP driver.
- *
- * Copyright IBM Corp. 2004, 2009
- *
- *   Author(s):	Peter Oberparleiter <Peter.Oberparleiter@de.ibm.com>,
- *		Heiko Carstens <heiko.carstens@de.ibm.com>,
- *
- */
-
-#include <linux/linkage.h>
-#include <asm/irq.h>
-
-LC_EXT_NEW_PSW		= 0x58			# addr of ext int handler
-LC_EXT_NEW_PSW_64	= 0x1b0			# addr of ext int handler 64 bit
-LC_EXT_INT_PARAM	= 0x80			# addr of ext int parameter
-LC_EXT_INT_CODE		= 0x86			# addr of ext int code
-LC_AR_MODE_ID		= 0xa3
-
-#
-# Subroutine which waits synchronously until either an external interruption
-# or a timeout occurs.
-#
-# Parameters:
-#   R2	= 0 for no timeout, non-zero for timeout in (approximated) seconds
-#
-# Returns:
-#   R2	= 0 on interrupt, 2 on timeout
-#   R3	= external interruption parameter if R2=0
-#
-
-_sclp_wait_int:
-	stm	%r6,%r15,24(%r15)		# save registers
-	basr	%r13,0				# get base register
-.LbaseS1:
-	ahi	%r15,-96			# create stack frame
-	la	%r8,LC_EXT_NEW_PSW		# register int handler
-	la	%r9,.LextpswS1-.LbaseS1(%r13)
-	tm	LC_AR_MODE_ID,1
-	jno	.Lesa1
-	la	%r8,LC_EXT_NEW_PSW_64		# register int handler 64 bit
-	la	%r9,.LextpswS1_64-.LbaseS1(%r13)
-.Lesa1:
-	mvc	.LoldpswS1-.LbaseS1(16,%r13),0(%r8)
-	mvc	0(16,%r8),0(%r9)
-	epsw	%r6,%r7				# set current addressing mode
-	nill	%r6,0x1				# in new psw (31 or 64 bit mode)
-	nilh	%r7,0x8000
-	stm	%r6,%r7,0(%r8)
-	lhi	%r6,0x0200			# cr mask for ext int (cr0.54)
-	ltr	%r2,%r2
-	jz	.LsetctS1
-	ahi	%r6,0x0800			# cr mask for clock int (cr0.52)
-	stck	.LtimeS1-.LbaseS1(%r13)		# initiate timeout
-	al	%r2,.LtimeS1-.LbaseS1(%r13)
-	st	%r2,.LtimeS1-.LbaseS1(%r13)
-	sckc	.LtimeS1-.LbaseS1(%r13)
-
-.LsetctS1:
-	stctl	%c0,%c0,.LctlS1-.LbaseS1(%r13)	# enable required interrupts
-	l	%r0,.LctlS1-.LbaseS1(%r13)
-	lhi	%r1,~(0x200 | 0x800)		# clear old values
-	nr	%r1,%r0
-	or	%r1,%r6				# set new value
-	st	%r1,.LctlS1-.LbaseS1(%r13)
-	lctl	%c0,%c0,.LctlS1-.LbaseS1(%r13)
-	st	%r0,.LctlS1-.LbaseS1(%r13)
-	lhi	%r2,2				# return code for timeout
-.LloopS1:
-	lpsw	.LwaitpswS1-.LbaseS1(%r13)	# wait until interrupt
-.LwaitS1:
-	lh	%r7,LC_EXT_INT_CODE
-	chi	%r7,EXT_IRQ_CLK_COMP		# timeout?
-	je	.LtimeoutS1
-	chi	%r7,EXT_IRQ_SERVICE_SIG		# service int?
-	jne	.LloopS1
-	sr	%r2,%r2
-	l	%r3,LC_EXT_INT_PARAM
-.LtimeoutS1:
-	lctl	%c0,%c0,.LctlS1-.LbaseS1(%r13)	# restore interrupt setting
-	# restore old handler
-	mvc	0(16,%r8),.LoldpswS1-.LbaseS1(%r13)
-	lm	%r6,%r15,120(%r15)		# restore registers
-	br	%r14				# return to caller
-
-	.align	8
-.LoldpswS1:
-	.long	0, 0, 0, 0			# old ext int PSW
-.LextpswS1:
-	.long	0x00080000, 0x80000000+.LwaitS1	# PSW to handle ext int
-.LextpswS1_64:
-	.quad	0, .LwaitS1			# PSW to handle ext int, 64 bit
-.LwaitpswS1:
-	.long	0x010a0000, 0x00000000+.LloopS1	# PSW to wait for ext int
-.LtimeS1:
-	.quad	0				# current time
-.LctlS1:
-	.long	0				# CT0 contents
-
-#
-# Subroutine to synchronously issue a service call.
-#
-# Parameters:
-#   R2	= command word
-#   R3	= sccb address
-#
-# Returns:
-#   R2	= 0 on success, 1 on failure
-#   R3	= sccb response code if R2 = 0
-#
-
-_sclp_servc:
-	stm	%r6,%r15,24(%r15)		# save registers
-	ahi	%r15,-96			# create stack frame
-	lr	%r6,%r2				# save command word
-	lr	%r7,%r3				# save sccb address
-.LretryS2:
-	lhi	%r2,1				# error return code
-	.insn	rre,0xb2200000,%r6,%r7		# servc
-	brc	1,.LendS2			# exit if not operational
-	brc	8,.LnotbusyS2			# go on if not busy
-	sr	%r2,%r2				# wait until no longer busy
-	bras	%r14,_sclp_wait_int
-	j	.LretryS2			# retry
-.LnotbusyS2:
-	sr	%r2,%r2				# wait until result
-	bras	%r14,_sclp_wait_int
-	sr	%r2,%r2
-	lh	%r3,6(%r7)
-.LendS2:
-	lm	%r6,%r15,120(%r15)		# restore registers
-	br	%r14
-
-#
-# Subroutine to set up the SCLP interface.
-#
-# Parameters:
-#   R2	= 0 to activate, non-zero to deactivate
-#
-# Returns:
-#   R2	= 0 on success, non-zero on failure
-#
-
-_sclp_setup:
-	stm	%r6,%r15,24(%r15)		# save registers
-	ahi	%r15,-96			# create stack frame
-	basr	%r13,0				# get base register
-.LbaseS3:
-	l	%r6,.LsccbS0-.LbaseS3(%r13)	# prepare init mask sccb
-	mvc	0(.LinitendS3-.LinitsccbS3,%r6),.LinitsccbS3-.LbaseS3(%r13)
-	ltr	%r2,%r2				# initialization?
-	jz	.LdoinitS3			# go ahead
-	# clear masks
-	xc	.LinitmaskS3-.LinitsccbS3(8,%r6),.LinitmaskS3-.LinitsccbS3(%r6)
-.LdoinitS3:
-	l	%r2,.LwritemaskS3-.LbaseS3(%r13)# get command word
-	lr	%r3,%r6				# get sccb address
-	bras	%r14,_sclp_servc		# issue service call
-	ltr	%r2,%r2				# servc successful?
-	jnz	.LerrorS3
-	chi	%r3,0x20			# write mask successful?
-	jne	.LerrorS3
-	# check masks
-	la	%r2,.LinitmaskS3-.LinitsccbS3(%r6)
-	l	%r1,0(%r2)			# receive mask ok?
-	n	%r1,12(%r2)
-	cl	%r1,0(%r2)
-	jne	.LerrorS3
-	l	%r1,4(%r2)			# send mask ok?
-	n	%r1,8(%r2)
-	cl	%r1,4(%r2)
-	sr	%r2,%r2
-	je	.LendS3
-.LerrorS3:
-	lhi	%r2,1				# error return code
-.LendS3:
-	lm	%r6,%r15,120(%r15)		# restore registers
-	br	%r14
-.LwritemaskS3:
-	.long	0x00780005			# SCLP command for write mask
-.LinitsccbS3:
-	.word	.LinitendS3-.LinitsccbS3
-	.byte	0,0,0,0
-	.word	0
-	.word	0
-	.word	4
-.LinitmaskS3:
-	.long	0x80000000
-	.long	0x40000000
-	.long	0
-	.long	0
-.LinitendS3:
-
-#
-# Subroutine which prints a given text to the SCLP console.
-#
-# Parameters:
-#   R2	= address of nil-terminated ASCII text
-#
-# Returns:
-#   R2	= 0 on success, 1 on failure
-#
-
-_sclp_print:
-	stm	%r6,%r15,24(%r15)		# save registers
-	ahi	%r15,-96			# create stack frame
-	basr	%r13,0				# get base register
-.LbaseS4:
-	l	%r8,.LsccbS0-.LbaseS4(%r13)	# prepare write data sccb
-	mvc	0(.LmtoS4-.LwritesccbS4,%r8),.LwritesccbS4-.LbaseS4(%r13)
-	la	%r7,.LmtoS4-.LwritesccbS4(%r8)	# current mto addr
-	sr	%r0,%r0
-	l	%r10,.Lascebc-.LbaseS4(%r13)	# address of translation table
-.LinitmtoS4:
-	# initialize mto
-	mvc	0(.LmtoendS4-.LmtoS4,%r7),.LmtoS4-.LbaseS4(%r13)
-	lhi	%r6,.LmtoendS4-.LmtoS4		# current mto length
-.LloopS4:
-	ic	%r0,0(%r2)			# get character
-	ahi	%r2,1
-	ltr	%r0,%r0				# end of string?
-	jz	.LfinalizemtoS4
-	chi	%r0,0x0a			# end of line (NL)?
-	jz	.LfinalizemtoS4
-	stc	%r0,0(%r6,%r7)			# copy to mto
-	la	%r11,0(%r6,%r7)
-	tr	0(1,%r11),0(%r10)		# translate to EBCDIC
-	ahi	%r6,1
-	j	.LloopS4
-.LfinalizemtoS4:
-	sth	%r6,0(%r7)			# update mto length
-	lh	%r9,.LmdbS4-.LwritesccbS4(%r8)	# update mdb length
-	ar	%r9,%r6
-	sth	%r9,.LmdbS4-.LwritesccbS4(%r8)
-	lh	%r9,.LevbufS4-.LwritesccbS4(%r8)# update evbuf length
-	ar	%r9,%r6
-	sth	%r9,.LevbufS4-.LwritesccbS4(%r8)
-	lh	%r9,0(%r8)			# update sccb length
-	ar	%r9,%r6
-	sth	%r9,0(%r8)
-	ar	%r7,%r6				# update current mto address
-	ltr	%r0,%r0				# more characters?
-	jnz	.LinitmtoS4
-	l	%r2,.LwritedataS4-.LbaseS4(%r13)# write data
-	lr	%r3,%r8
-	bras	%r14,_sclp_servc
-	ltr	%r2,%r2				# servc successful?
-	jnz	.LendS4
-	chi	%r3,0x20			# write data successful?
-	je	.LendS4
-	lhi	%r2,1				# error return code
-.LendS4:
-	lm	%r6,%r15,120(%r15)		# restore registers
-	br	%r14
-
-#
-# Function which prints a given text to the SCLP console.
-#
-# Parameters:
-#   R2	= address of nil-terminated ASCII text
-#
-# Returns:
-#   R2	= 0 on success, 1 on failure
-#
-
-ENTRY(_sclp_print_early)
-	stm	%r6,%r15,24(%r15)		# save registers
-	ahi	%r15,-96			# create stack frame
-	tm	LC_AR_MODE_ID,1
-	jno	.Lesa2
-	ahi	%r15,-80
-	stmh	%r6,%r15,96(%r15)		# store upper register halves
-	basr	%r13,0
-	lmh	%r0,%r15,.Lzeroes-.(%r13)	# clear upper register halves
-.Lesa2:
-	lr	%r10,%r2			# save string pointer
-	lhi	%r2,0
-	bras	%r14,_sclp_setup		# enable console
-	ltr	%r2,%r2
-	jnz	.LendS5
-	lr	%r2,%r10
-	bras	%r14,_sclp_print		# print string
-	ltr	%r2,%r2
-	jnz	.LendS5
-	lhi	%r2,1
-	bras	%r14,_sclp_setup		# disable console
-.LendS5:
-	tm	LC_AR_MODE_ID,1
-	jno	.Lesa3
-	lgfr	%r2,%r2				# sign extend return value
-	lmh	%r6,%r15,96(%r15)		# restore upper register halves
-	ahi	%r15,80
-.Lesa3:
-	lm	%r6,%r15,120(%r15)		# restore registers
-	br	%r14
-.Lzeroes:
-	.fill	64,4,0
-
-.LwritedataS4:
-	.long	0x00760005			# SCLP command for write data
-.LwritesccbS4:
-	# sccb
-	.word	.LmtoS4-.LwritesccbS4
-	.byte	0
-	.byte	0,0,0
-	.word	0
-
-	# evbuf
-.LevbufS4:
-	.word	.LmtoS4-.LevbufS4
-	.byte	0x02
-	.byte	0
-	.word	0
-
-.LmdbS4:
-	# mdb
-	.word	.LmtoS4-.LmdbS4
-	.word	1
-	.long	0xd4c4c240
-	.long	1
-
-	# go
-.LgoS4:
-	.word	.LmtoS4-.LgoS4
-	.word	1
-	.long	0
-	.byte	0,0,0,0,0,0,0,0
-	.byte	0,0,0
-	.byte	0
-	.byte	0,0,0,0,0,0,0
-	.byte	0
-	.word	0
-	.byte	0,0,0,0,0,0,0,0,0,0
-	.byte	0,0,0,0,0,0,0,0
-	.byte	0,0,0,0,0,0,0,0
-
-.LmtoS4:
-	.word	.LmtoendS4-.LmtoS4
-	.word	4
-	.word	0x1000
-	.byte	0
-	.byte	0,0,0
-.LmtoendS4:
-
-	# Global constants
-.LsccbS0:
-	.long	_sclp_work_area
-.Lascebc:
-	.long	_ascebc
-
-.section .data,"aw",@progbits
-	.balign 4096
-_sclp_work_area:
-	.fill	4096
-.previous
diff --git a/arch/s390/kernel/sclp.c b/arch/s390/kernel/sclp.c
new file mode 100644
index 000000000000..fa0bdff1d413
--- /dev/null
+++ b/arch/s390/kernel/sclp.c
@@ -0,0 +1,160 @@
+/*
+ *    Copyright IBM Corp. 2015
+ *    Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+#include <linux/kernel.h>
+#include <asm/ebcdic.h>
+#include <asm/irq.h>
+#include <asm/lowcore.h>
+#include <asm/processor.h>
+#include <asm/sclp.h>
+
+static char _sclp_work_area[4096] __aligned(PAGE_SIZE);
+
+static void _sclp_wait_int(void)
+{
+	unsigned long cr0, cr0_new, psw_mask, addr;
+	psw_t psw_ext_save, psw_wait;
+
+	__ctl_store(cr0, 0, 0);
+	cr0_new = cr0 | 0x200;
+	__ctl_load(cr0_new, 0, 0);
+
+	psw_ext_save = S390_lowcore.external_new_psw;
+	psw_mask = __extract_psw() & (PSW_MASK_EA | PSW_MASK_BA);
+	S390_lowcore.external_new_psw.mask = psw_mask;
+	psw_wait.mask = psw_mask | PSW_MASK_EXT | PSW_MASK_WAIT;
+	S390_lowcore.ext_int_code = 0;
+
+	do {
+		asm volatile(
+			"	larl	%[addr],0f\n"
+			"	stg	%[addr],%[psw_wait_addr]\n"
+			"	stg	%[addr],%[psw_ext_addr]\n"
+			"	lpswe	%[psw_wait]\n"
+			"0:\n"
+			: [addr] "=&d" (addr),
+			  [psw_wait_addr] "=Q" (psw_wait.addr),
+			  [psw_ext_addr] "=Q" (S390_lowcore.external_new_psw.addr)
+			: [psw_wait] "Q" (psw_wait)
+			: "cc", "memory");
+	} while (S390_lowcore.ext_int_code != EXT_IRQ_SERVICE_SIG);
+
+	__ctl_load(cr0, 0, 0);
+	S390_lowcore.external_new_psw = psw_ext_save;
+}
+
+static int _sclp_servc(unsigned int cmd, char *sccb)
+{
+	unsigned int cc;
+
+	do {
+		asm volatile(
+			"	.insn	rre,0xb2200000,%1,%2\n"
+			"	ipm	%0\n"
+			: "=d" (cc) : "d" (cmd), "a" (sccb)
+			: "cc", "memory");
+		cc >>= 28;
+		if (cc == 3)
+			return -EINVAL;
+		_sclp_wait_int();
+	} while (cc != 0);
+	return (*(unsigned short *)(sccb + 6) == 0x20) ? 0 : -EIO;
+}
+
+static int _sclp_setup(int disable)
+{
+	static unsigned char init_sccb[] = {
+		0x00, 0x1c,
+		0x00, 0x00, 0x00, 0x00,	0x00, 0x00, 0x00, 0x00,
+		0x00, 0x04,
+		0x80, 0x00, 0x00, 0x00,	0x40, 0x00, 0x00, 0x00,
+		0x00, 0x00, 0x00, 0x00,	0x00, 0x00, 0x00, 0x00
+	};
+	unsigned int *masks;
+	int rc;
+
+	memcpy(_sclp_work_area, init_sccb, 28);
+	masks = (unsigned int *)(_sclp_work_area + 12);
+	if (disable)
+		memset(masks, 0, 16);
+	/* SCLP write mask */
+	rc = _sclp_servc(0x00780005, _sclp_work_area);
+	if (rc)
+		return rc;
+	if ((masks[0] & masks[3]) != masks[0] ||
+	    (masks[1] & masks[2]) != masks[1])
+		return -EIO;
+	return 0;
+}
+
+static int _sclp_print(const char *str)
+{
+	static unsigned char write_head[] = {
+		/* sccb header */
+		0x00, 0x52,					/* 0 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00,		/* 2 */
+		/* evbuf */
+		0x00, 0x4a,					/* 8 */
+		0x02, 0x00, 0x00, 0x00,				/* 10 */
+		/* mdb */
+		0x00, 0x44,					/* 14 */
+		0x00, 0x01,					/* 16 */
+		0xd4, 0xc4, 0xc2, 0x40,				/* 18 */
+		0x00, 0x00, 0x00, 0x01,				/* 22 */
+		/* go */
+		0x00, 0x38,					/* 26 */
+		0x00, 0x01,					/* 28 */
+		0x00, 0x00, 0x00, 0x00,				/* 30 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,	/* 34 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,	/* 42 */
+		0x00, 0x00, 0x00, 0x00,				/* 50 */
+		0x00, 0x00,					/* 54 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,	/* 56 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,	/* 64 */
+		0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,	/* 72 */
+		0x00, 0x00,					/* 80 */
+	};
+	static unsigned char write_mto[] = {
+		/* mto	*/
+		0x00, 0x0a,					/* 0 */
+		0x00, 0x04,					/* 2 */
+		0x10, 0x00,					/* 4 */
+		0x00, 0x00, 0x00, 0x00				/* 6 */
+	};
+	unsigned char *ptr, ch;
+	unsigned int count;
+
+	memcpy(_sclp_work_area, write_head, sizeof(write_head));
+	ptr = _sclp_work_area + sizeof(write_head);
+	do {
+		memcpy(ptr, write_mto, sizeof(write_mto));
+		for (count = sizeof(write_mto); (ch = *str++) != 0; count++) {
+			if (ch == 0x0a)
+				break;
+			ptr[count] = _ascebc[ch];
+		}
+		/* Update length fields in mto, mdb, evbuf and sccb */
+		*(unsigned short *) ptr = count;
+		*(unsigned short *)(_sclp_work_area + 14) += count;
+		*(unsigned short *)(_sclp_work_area + 8) += count;
+		*(unsigned short *)(_sclp_work_area + 0) += count;
+		ptr += count;
+	} while (ch != 0);
+
+	/* SCLP write data */
+	return _sclp_servc(0x00760005, _sclp_work_area);
+}
+
+int _sclp_print_early(const char *str)
+{
+	int rc;
+
+	rc = _sclp_setup(0);
+	if (rc)
+		return rc;
+	rc = _sclp_print(str);
+	if (rc)
+		return rc;
+	return _sclp_setup(1);
+}
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index ca070d260af2..ce0cbd6ba7ca 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -62,6 +62,7 @@
 #include <asm/os_info.h>
 #include <asm/sclp.h>
 #include <asm/sysinfo.h>
+#include <asm/numa.h>
 #include "entry.h"
 
 /*
@@ -76,7 +77,7 @@ EXPORT_SYMBOL(console_devno);
 unsigned int console_irq = -1;
 EXPORT_SYMBOL(console_irq);
 
-unsigned long elf_hwcap = 0;
+unsigned long elf_hwcap __read_mostly = 0;
 char elf_platform[ELF_PLATFORM_SIZE];
 
 int __initdata memory_end_set;
@@ -688,7 +689,7 @@ static void __init setup_memory(void)
 /*
  * Setup hardware capabilities.
  */
-static void __init setup_hwcaps(void)
+static int __init setup_hwcaps(void)
 {
 	static const int stfl_bits[6] = { 0, 2, 7, 17, 19, 21 };
 	struct cpuid cpu_id;
@@ -754,9 +755,11 @@ static void __init setup_hwcaps(void)
 		elf_hwcap |= HWCAP_S390_TE;
 
 	/*
-	 * Vector extension HWCAP_S390_VXRS is bit 11.
+	 * Vector extension HWCAP_S390_VXRS is bit 11. The Vector extension
+	 * can be disabled with the "novx" parameter. Use MACHINE_HAS_VX
+	 * instead of facility bit 129.
 	 */
-	if (test_facility(129))
+	if (MACHINE_HAS_VX)
 		elf_hwcap |= HWCAP_S390_VXRS;
 	get_cpu_id(&cpu_id);
 	add_device_randomness(&cpu_id, sizeof(cpu_id));
@@ -793,7 +796,9 @@ static void __init setup_hwcaps(void)
 		strcpy(elf_platform, "z13");
 		break;
 	}
+	return 0;
 }
+arch_initcall(setup_hwcaps);
 
 /*
  * Add system information as device randomness
@@ -879,11 +884,7 @@ void __init setup_arch(char **cmdline_p)
 	setup_lowcore();
 	smp_fill_possible_mask();
         cpu_init();
-
-	/*
-	 * Setup capabilities (ELF_HWCAP & ELF_PLATFORM).
-	 */
-	setup_hwcaps();
+	numa_setup();
 
 	/*
 	 * Create kernel page tables and switch to virtual addressing.
diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c
index c551f22ce066..9549af102d75 100644
--- a/arch/s390/kernel/signal.c
+++ b/arch/s390/kernel/signal.c
@@ -105,32 +105,13 @@ struct rt_sigframe
 static void store_sigregs(void)
 {
 	save_access_regs(current->thread.acrs);
-	save_fp_ctl(&current->thread.fp_regs.fpc);
-	if (current->thread.vxrs) {
-		int i;
-
-		save_vx_regs(current->thread.vxrs);
-		for (i = 0; i < __NUM_FPRS; i++)
-			current->thread.fp_regs.fprs[i] =
-				*(freg_t *)(current->thread.vxrs + i);
-	} else
-		save_fp_regs(current->thread.fp_regs.fprs);
+	save_fpu_regs();
 }
 
 /* Load registers after signal return */
 static void load_sigregs(void)
 {
 	restore_access_regs(current->thread.acrs);
-	/* restore_fp_ctl is done in restore_sigregs */
-	if (current->thread.vxrs) {
-		int i;
-
-		for (i = 0; i < __NUM_FPRS; i++)
-			*(freg_t *)(current->thread.vxrs + i) =
-				current->thread.fp_regs.fprs[i];
-		restore_vx_regs(current->thread.vxrs);
-	} else
-		restore_fp_regs(current->thread.fp_regs.fprs);
 }
 
 /* Returns non-zero on fault. */
@@ -146,8 +127,7 @@ static int save_sigregs(struct pt_regs *regs, _sigregs __user *sregs)
 	memcpy(&user_sregs.regs.gprs, &regs->gprs, sizeof(sregs->regs.gprs));
 	memcpy(&user_sregs.regs.acrs, current->thread.acrs,
 	       sizeof(user_sregs.regs.acrs));
-	memcpy(&user_sregs.fpregs, &current->thread.fp_regs,
-	       sizeof(user_sregs.fpregs));
+	fpregs_store(&user_sregs.fpregs, &current->thread.fpu);
 	if (__copy_to_user(sregs, &user_sregs, sizeof(_sigregs)))
 		return -EFAULT;
 	return 0;
@@ -166,8 +146,8 @@ static int restore_sigregs(struct pt_regs *regs, _sigregs __user *sregs)
 	if (!is_ri_task(current) && (user_sregs.regs.psw.mask & PSW_MASK_RI))
 		return -EINVAL;
 
-	/* Loading the floating-point-control word can fail. Do that first. */
-	if (restore_fp_ctl(&user_sregs.fpregs.fpc))
+	/* Test the floating-point-control word. */
+	if (test_fp_ctl(user_sregs.fpregs.fpc))
 		return -EINVAL;
 
 	/* Use regs->psw.mask instead of PSW_USER_BITS to preserve PER bit. */
@@ -185,8 +165,7 @@ static int restore_sigregs(struct pt_regs *regs, _sigregs __user *sregs)
 	memcpy(&current->thread.acrs, &user_sregs.regs.acrs,
 	       sizeof(current->thread.acrs));
 
-	memcpy(&current->thread.fp_regs, &user_sregs.fpregs,
-	       sizeof(current->thread.fp_regs));
+	fpregs_load(&user_sregs.fpregs, &current->thread.fpu);
 
 	clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */
 	return 0;
@@ -200,13 +179,13 @@ static int save_sigregs_ext(struct pt_regs *regs,
 	int i;
 
 	/* Save vector registers to signal stack */
-	if (current->thread.vxrs) {
+	if (is_vx_task(current)) {
 		for (i = 0; i < __NUM_VXRS_LOW; i++)
-			vxrs[i] = *((__u64 *)(current->thread.vxrs + i) + 1);
+			vxrs[i] = *((__u64 *)(current->thread.fpu.vxrs + i) + 1);
 		if (__copy_to_user(&sregs_ext->vxrs_low, vxrs,
 				   sizeof(sregs_ext->vxrs_low)) ||
 		    __copy_to_user(&sregs_ext->vxrs_high,
-				   current->thread.vxrs + __NUM_VXRS_LOW,
+				   current->thread.fpu.vxrs + __NUM_VXRS_LOW,
 				   sizeof(sregs_ext->vxrs_high)))
 			return -EFAULT;
 	}
@@ -220,15 +199,15 @@ static int restore_sigregs_ext(struct pt_regs *regs,
 	int i;
 
 	/* Restore vector registers from signal stack */
-	if (current->thread.vxrs) {
+	if (is_vx_task(current)) {
 		if (__copy_from_user(vxrs, &sregs_ext->vxrs_low,
 				     sizeof(sregs_ext->vxrs_low)) ||
-		    __copy_from_user(current->thread.vxrs + __NUM_VXRS_LOW,
+		    __copy_from_user(current->thread.fpu.vxrs + __NUM_VXRS_LOW,
 				     &sregs_ext->vxrs_high,
 				     sizeof(sregs_ext->vxrs_high)))
 			return -EFAULT;
 		for (i = 0; i < __NUM_VXRS_LOW; i++)
-			*((__u64 *)(current->thread.vxrs + i) + 1) = vxrs[i];
+			*((__u64 *)(current->thread.fpu.vxrs + i) + 1) = vxrs[i];
 	}
 	return 0;
 }
@@ -243,6 +222,7 @@ SYSCALL_DEFINE0(sigreturn)
 	if (__copy_from_user(&set.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE))
 		goto badframe;
 	set_current_blocked(&set);
+	save_fpu_regs();
 	if (restore_sigregs(regs, &frame->sregs))
 		goto badframe;
 	if (restore_sigregs_ext(regs, &frame->sregs_ext))
@@ -266,6 +246,7 @@ SYSCALL_DEFINE0(rt_sigreturn)
 	set_current_blocked(&set);
 	if (restore_altstack(&frame->uc.uc_stack))
 		goto badframe;
+	save_fpu_regs();
 	if (restore_sigregs(regs, &frame->uc.uc_mcontext))
 		goto badframe;
 	if (restore_sigregs_ext(regs, &frame->uc.uc_mcontext_ext))
@@ -400,7 +381,7 @@ static int setup_rt_frame(struct ksignal *ksig, sigset_t *set,
 	uc_flags = 0;
 	if (MACHINE_HAS_VX) {
 		frame_size += sizeof(_sigregs_ext);
-		if (current->thread.vxrs)
+		if (is_vx_task(current))
 			uc_flags |= UC_VXRS;
 	}
 	frame = get_sigframe(&ksig->ka, regs, frame_size);
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 6f54c175f5c9..c6355e6f3fcc 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -532,8 +532,8 @@ EXPORT_SYMBOL(smp_ctl_clear_bit);
 
 #ifdef CONFIG_CRASH_DUMP
 
-static void __smp_store_cpu_state(struct save_area_ext *sa_ext, u16 address,
-				  int is_boot_cpu)
+static void __init __smp_store_cpu_state(struct save_area_ext *sa_ext,
+					 u16 address, int is_boot_cpu)
 {
 	void *lc = (void *)(unsigned long) store_prefix();
 	unsigned long vx_sa;
diff --git a/arch/s390/kernel/swsusp.S b/arch/s390/kernel/swsusp.S
index ca6294645dd3..2d6b6e81f812 100644
--- a/arch/s390/kernel/swsusp.S
+++ b/arch/s390/kernel/swsusp.S
@@ -30,6 +30,9 @@ ENTRY(swsusp_arch_suspend)
 	aghi	%r15,-STACK_FRAME_OVERHEAD
 	stg	%r1,__SF_BACKCHAIN(%r15)
 
+	/* Store FPU registers */
+	brasl	%r14,save_fpu_regs
+
 	/* Deactivate DAT */
 	stnsm	__SF_EMPTY(%r15),0xfb
 
@@ -47,23 +50,6 @@ ENTRY(swsusp_arch_suspend)
 
 	/* Store registers */
 	mvc	0x318(4,%r1),__SF_EMPTY(%r15)	/* move prefix to lowcore */
-	stfpc	0x31c(%r1)			/* store fpu control */
-	std	0,0x200(%r1)			/* store f0 */
-	std	1,0x208(%r1)			/* store f1 */
-	std	2,0x210(%r1)			/* store f2 */
-	std	3,0x218(%r1)			/* store f3 */
-	std	4,0x220(%r1)			/* store f4 */
-	std	5,0x228(%r1)			/* store f5 */
-	std	6,0x230(%r1)			/* store f6 */
-	std	7,0x238(%r1)			/* store f7 */
-	std	8,0x240(%r1)			/* store f8 */
-	std	9,0x248(%r1)			/* store f9 */
-	std	10,0x250(%r1)			/* store f10 */
-	std	11,0x258(%r1)			/* store f11 */
-	std	12,0x260(%r1)			/* store f12 */
-	std	13,0x268(%r1)			/* store f13 */
-	std	14,0x270(%r1)			/* store f14 */
-	std	15,0x278(%r1)			/* store f15 */
 	stam	%a0,%a15,0x340(%r1)		/* store access registers */
 	stctg	%c0,%c15,0x380(%r1)		/* store control registers */
 	stmg	%r0,%r15,0x280(%r1)		/* store general registers */
@@ -249,24 +235,6 @@ restore_registers:
 	lctlg	%c0,%c15,0x380(%r13)	/* load control registers */
 	lam	%a0,%a15,0x340(%r13)	/* load access registers */
 
-	lfpc	0x31c(%r13)		/* load fpu control */
-	ld	0,0x200(%r13)		/* load f0 */
-	ld	1,0x208(%r13)		/* load f1 */
-	ld	2,0x210(%r13)		/* load f2 */
-	ld	3,0x218(%r13)		/* load f3 */
-	ld	4,0x220(%r13)		/* load f4 */
-	ld	5,0x228(%r13)		/* load f5 */
-	ld	6,0x230(%r13)		/* load f6 */
-	ld	7,0x238(%r13)		/* load f7 */
-	ld	8,0x240(%r13)		/* load f8 */
-	ld	9,0x248(%r13)		/* load f9 */
-	ld	10,0x250(%r13)		/* load f10 */
-	ld	11,0x258(%r13)		/* load f11 */
-	ld	12,0x260(%r13)		/* load f12 */
-	ld	13,0x268(%r13)		/* load f13 */
-	ld	14,0x270(%r13)		/* load f14 */
-	ld	15,0x278(%r13)		/* load f15 */
-
 	/* Load old stack */
 	lg	%r15,0x2f8(%r13)
 
diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
index 1acad02681c4..8c56929c8d82 100644
--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -9,12 +9,12 @@
 #define NI_SYSCALL SYSCALL(sys_ni_syscall,sys_ni_syscall)
 
 NI_SYSCALL						/* 0 */
-SYSCALL(sys_exit,compat_sys_exit)
+SYSCALL(sys_exit,sys_exit)
 SYSCALL(sys_fork,sys_fork)
 SYSCALL(sys_read,compat_sys_s390_read)
 SYSCALL(sys_write,compat_sys_s390_write)
 SYSCALL(sys_open,compat_sys_open)			/* 5 */
-SYSCALL(sys_close,compat_sys_close)
+SYSCALL(sys_close,sys_close)
 SYSCALL(sys_restart_syscall,sys_restart_syscall)
 SYSCALL(sys_creat,compat_sys_creat)
 SYSCALL(sys_link,compat_sys_link)
@@ -35,21 +35,21 @@ SYSCALL(sys_ni_syscall,compat_sys_s390_setuid16)	/* old setuid16 syscall*/
 SYSCALL(sys_ni_syscall,compat_sys_s390_getuid16)	/* old getuid16 syscall*/
 SYSCALL(sys_ni_syscall,compat_sys_stime)		/* 25 old stime syscall */
 SYSCALL(sys_ptrace,compat_sys_ptrace)
-SYSCALL(sys_alarm,compat_sys_alarm)
+SYSCALL(sys_alarm,sys_alarm)
 NI_SYSCALL						/* old fstat syscall */
 SYSCALL(sys_pause,sys_pause)
 SYSCALL(sys_utime,compat_sys_utime)			/* 30 */
 NI_SYSCALL						/* old stty syscall */
 NI_SYSCALL						/* old gtty syscall */
 SYSCALL(sys_access,compat_sys_access)
-SYSCALL(sys_nice,compat_sys_nice)
+SYSCALL(sys_nice,sys_nice)
 NI_SYSCALL						/* 35 old ftime syscall */
 SYSCALL(sys_sync,sys_sync)
-SYSCALL(sys_kill,compat_sys_kill)
+SYSCALL(sys_kill,sys_kill)
 SYSCALL(sys_rename,compat_sys_rename)
 SYSCALL(sys_mkdir,compat_sys_mkdir)
 SYSCALL(sys_rmdir,compat_sys_rmdir)			/* 40 */
-SYSCALL(sys_dup,compat_sys_dup)
+SYSCALL(sys_dup,sys_dup)
 SYSCALL(sys_pipe,compat_sys_pipe)
 SYSCALL(sys_times,compat_sys_times)
 NI_SYSCALL						/* old prof syscall */
@@ -65,13 +65,13 @@ NI_SYSCALL						/* old lock syscall */
 SYSCALL(sys_ioctl,compat_sys_ioctl)
 SYSCALL(sys_fcntl,compat_sys_fcntl)			/* 55 */
 NI_SYSCALL						/* intel mpx syscall */
-SYSCALL(sys_setpgid,compat_sys_setpgid)
+SYSCALL(sys_setpgid,sys_setpgid)
 NI_SYSCALL						/* old ulimit syscall */
 NI_SYSCALL						/* old uname syscall */
-SYSCALL(sys_umask,compat_sys_umask)			/* 60 */
+SYSCALL(sys_umask,sys_umask)				/* 60 */
 SYSCALL(sys_chroot,compat_sys_chroot)
 SYSCALL(sys_ustat,compat_sys_ustat)
-SYSCALL(sys_dup2,compat_sys_dup2)
+SYSCALL(sys_dup2,sys_dup2)
 SYSCALL(sys_getppid,sys_getppid)
 SYSCALL(sys_getpgrp,sys_getpgrp)			/* 65 */
 SYSCALL(sys_setsid,sys_setsid)
@@ -102,10 +102,10 @@ SYSCALL(sys_old_mmap,compat_sys_s390_old_mmap)		/* 90 */
 SYSCALL(sys_munmap,compat_sys_munmap)
 SYSCALL(sys_truncate,compat_sys_truncate)
 SYSCALL(sys_ftruncate,compat_sys_ftruncate)
-SYSCALL(sys_fchmod,compat_sys_fchmod)
+SYSCALL(sys_fchmod,sys_fchmod)
 SYSCALL(sys_ni_syscall,compat_sys_s390_fchown16)	/* 95 old fchown16 syscall*/
-SYSCALL(sys_getpriority,compat_sys_getpriority)
-SYSCALL(sys_setpriority,compat_sys_setpriority)
+SYSCALL(sys_getpriority,sys_getpriority)
+SYSCALL(sys_setpriority,sys_setpriority)
 NI_SYSCALL						/* old profil syscall */
 SYSCALL(sys_statfs,compat_sys_statfs)
 SYSCALL(sys_fstatfs,compat_sys_fstatfs)			/* 100 */
@@ -126,7 +126,7 @@ SYSCALL(sys_wait4,compat_sys_wait4)
 SYSCALL(sys_swapoff,compat_sys_swapoff)			/* 115 */
 SYSCALL(sys_sysinfo,compat_sys_sysinfo)
 SYSCALL(sys_s390_ipc,compat_sys_s390_ipc)
-SYSCALL(sys_fsync,compat_sys_fsync)
+SYSCALL(sys_fsync,sys_fsync)
 SYSCALL(sys_sigreturn,compat_sys_sigreturn)
 SYSCALL(sys_clone,compat_sys_clone)			/* 120 */
 SYSCALL(sys_setdomainname,compat_sys_setdomainname)
@@ -140,35 +140,35 @@ SYSCALL(sys_init_module,compat_sys_init_module)
 SYSCALL(sys_delete_module,compat_sys_delete_module)
 NI_SYSCALL						/* 130: old get_kernel_syms */
 SYSCALL(sys_quotactl,compat_sys_quotactl)
-SYSCALL(sys_getpgid,compat_sys_getpgid)
-SYSCALL(sys_fchdir,compat_sys_fchdir)
+SYSCALL(sys_getpgid,sys_getpgid)
+SYSCALL(sys_fchdir,sys_fchdir)
 SYSCALL(sys_bdflush,compat_sys_bdflush)
 SYSCALL(sys_sysfs,compat_sys_sysfs)			/* 135 */
-SYSCALL(sys_s390_personality,compat_sys_s390_personality)
+SYSCALL(sys_s390_personality,sys_s390_personality)
 NI_SYSCALL						/* for afs_syscall */
 SYSCALL(sys_ni_syscall,compat_sys_s390_setfsuid16)	/* old setfsuid16 syscall */
 SYSCALL(sys_ni_syscall,compat_sys_s390_setfsgid16)	/* old setfsgid16 syscall */
 SYSCALL(sys_llseek,compat_sys_llseek)			/* 140 */
 SYSCALL(sys_getdents,compat_sys_getdents)
 SYSCALL(sys_select,compat_sys_select)
-SYSCALL(sys_flock,compat_sys_flock)
+SYSCALL(sys_flock,sys_flock)
 SYSCALL(sys_msync,compat_sys_msync)
 SYSCALL(sys_readv,compat_sys_readv)			/* 145 */
 SYSCALL(sys_writev,compat_sys_writev)
-SYSCALL(sys_getsid,compat_sys_getsid)
-SYSCALL(sys_fdatasync,compat_sys_fdatasync)
+SYSCALL(sys_getsid,sys_getsid)
+SYSCALL(sys_fdatasync,sys_fdatasync)
 SYSCALL(sys_sysctl,compat_sys_sysctl)
 SYSCALL(sys_mlock,compat_sys_mlock)			/* 150 */
 SYSCALL(sys_munlock,compat_sys_munlock)
-SYSCALL(sys_mlockall,compat_sys_mlockall)
+SYSCALL(sys_mlockall,sys_mlockall)
 SYSCALL(sys_munlockall,sys_munlockall)
 SYSCALL(sys_sched_setparam,compat_sys_sched_setparam)
 SYSCALL(sys_sched_getparam,compat_sys_sched_getparam)	/* 155 */
 SYSCALL(sys_sched_setscheduler,compat_sys_sched_setscheduler)
-SYSCALL(sys_sched_getscheduler,compat_sys_sched_getscheduler)
+SYSCALL(sys_sched_getscheduler,sys_sched_getscheduler)
 SYSCALL(sys_sched_yield,sys_sched_yield)
-SYSCALL(sys_sched_get_priority_max,compat_sys_sched_get_priority_max)
-SYSCALL(sys_sched_get_priority_min,compat_sys_sched_get_priority_min)	/* 160 */
+SYSCALL(sys_sched_get_priority_max,sys_sched_get_priority_max)
+SYSCALL(sys_sched_get_priority_min,sys_sched_get_priority_min)	/* 160 */
 SYSCALL(sys_sched_rr_get_interval,compat_sys_sched_rr_get_interval)
 SYSCALL(sys_nanosleep,compat_sys_nanosleep)
 SYSCALL(sys_mremap,compat_sys_mremap)
@@ -211,20 +211,20 @@ SYSCALL(sys_getuid,sys_getuid)
 SYSCALL(sys_getgid,sys_getgid)				/* 200 */
 SYSCALL(sys_geteuid,sys_geteuid)
 SYSCALL(sys_getegid,sys_getegid)
-SYSCALL(sys_setreuid,compat_sys_setreuid)
-SYSCALL(sys_setregid,compat_sys_setregid)
+SYSCALL(sys_setreuid,sys_setreuid)
+SYSCALL(sys_setregid,sys_setregid)
 SYSCALL(sys_getgroups,compat_sys_getgroups)		/* 205 */
 SYSCALL(sys_setgroups,compat_sys_setgroups)
-SYSCALL(sys_fchown,compat_sys_fchown)
-SYSCALL(sys_setresuid,compat_sys_setresuid)
+SYSCALL(sys_fchown,sys_fchown)
+SYSCALL(sys_setresuid,sys_setresuid)
 SYSCALL(sys_getresuid,compat_sys_getresuid)
-SYSCALL(sys_setresgid,compat_sys_setresgid)		/* 210 */
+SYSCALL(sys_setresgid,sys_setresgid)			/* 210 */
 SYSCALL(sys_getresgid,compat_sys_getresgid)
 SYSCALL(sys_chown,compat_sys_chown)
-SYSCALL(sys_setuid,compat_sys_setuid)
-SYSCALL(sys_setgid,compat_sys_setgid)
-SYSCALL(sys_setfsuid,compat_sys_setfsuid)		/* 215 */
-SYSCALL(sys_setfsgid,compat_sys_setfsgid)
+SYSCALL(sys_setuid,sys_setuid)
+SYSCALL(sys_setgid,sys_setgid)
+SYSCALL(sys_setfsuid,sys_setfsuid)			/* 215 */
+SYSCALL(sys_setfsgid,sys_setfsgid)
 SYSCALL(sys_pivot_root,compat_sys_pivot_root)
 SYSCALL(sys_mincore,compat_sys_mincore)
 SYSCALL(sys_madvise,compat_sys_madvise)
@@ -245,19 +245,19 @@ SYSCALL(sys_removexattr,compat_sys_removexattr)
 SYSCALL(sys_lremovexattr,compat_sys_lremovexattr)
 SYSCALL(sys_fremovexattr,compat_sys_fremovexattr)	/* 235 */
 SYSCALL(sys_gettid,sys_gettid)
-SYSCALL(sys_tkill,compat_sys_tkill)
+SYSCALL(sys_tkill,sys_tkill)
 SYSCALL(sys_futex,compat_sys_futex)
 SYSCALL(sys_sched_setaffinity,compat_sys_sched_setaffinity)
 SYSCALL(sys_sched_getaffinity,compat_sys_sched_getaffinity)	/* 240 */
-SYSCALL(sys_tgkill,compat_sys_tgkill)
+SYSCALL(sys_tgkill,sys_tgkill)
 NI_SYSCALL						/* reserved for TUX */
 SYSCALL(sys_io_setup,compat_sys_io_setup)
 SYSCALL(sys_io_destroy,compat_sys_io_destroy)
 SYSCALL(sys_io_getevents,compat_sys_io_getevents)	/* 245 */
 SYSCALL(sys_io_submit,compat_sys_io_submit)
 SYSCALL(sys_io_cancel,compat_sys_io_cancel)
-SYSCALL(sys_exit_group,compat_sys_exit_group)
-SYSCALL(sys_epoll_create,compat_sys_epoll_create)
+SYSCALL(sys_exit_group,sys_exit_group)
+SYSCALL(sys_epoll_create,sys_epoll_create)
 SYSCALL(sys_epoll_ctl,compat_sys_epoll_ctl)		/* 250 */
 SYSCALL(sys_epoll_wait,compat_sys_epoll_wait)
 SYSCALL(sys_set_tid_address,compat_sys_set_tid_address)
@@ -265,8 +265,8 @@ SYSCALL(sys_fadvise64_64,compat_sys_s390_fadvise64)
 SYSCALL(sys_timer_create,compat_sys_timer_create)
 SYSCALL(sys_timer_settime,compat_sys_timer_settime)	/* 255 */
 SYSCALL(sys_timer_gettime,compat_sys_timer_gettime)
-SYSCALL(sys_timer_getoverrun,compat_sys_timer_getoverrun)
-SYSCALL(sys_timer_delete,compat_sys_timer_delete)
+SYSCALL(sys_timer_getoverrun,sys_timer_getoverrun)
+SYSCALL(sys_timer_delete,sys_timer_delete)
 SYSCALL(sys_clock_settime,compat_sys_clock_settime)
 SYSCALL(sys_clock_gettime,compat_sys_clock_gettime)	/* 260 */
 SYSCALL(sys_clock_getres,compat_sys_clock_getres)
@@ -276,9 +276,9 @@ SYSCALL(sys_ni_syscall,compat_sys_s390_fadvise64_64)
 SYSCALL(sys_statfs64,compat_sys_statfs64)
 SYSCALL(sys_fstatfs64,compat_sys_fstatfs64)
 SYSCALL(sys_remap_file_pages,compat_sys_remap_file_pages)
-NI_SYSCALL						/* 268 sys_mbind */
-NI_SYSCALL						/* 269 sys_get_mempolicy */
-NI_SYSCALL						/* 270 sys_set_mempolicy */
+SYSCALL(sys_mbind,compat_sys_mbind)
+SYSCALL(sys_get_mempolicy,compat_sys_get_mempolicy)
+SYSCALL(sys_set_mempolicy,compat_sys_set_mempolicy)
 SYSCALL(sys_mq_open,compat_sys_mq_open)
 SYSCALL(sys_mq_unlink,compat_sys_mq_unlink)
 SYSCALL(sys_mq_timedsend,compat_sys_mq_timedsend)
@@ -290,12 +290,12 @@ SYSCALL(sys_add_key,compat_sys_add_key)
 SYSCALL(sys_request_key,compat_sys_request_key)
 SYSCALL(sys_keyctl,compat_sys_keyctl)			/* 280 */
 SYSCALL(sys_waitid,compat_sys_waitid)
-SYSCALL(sys_ioprio_set,compat_sys_ioprio_set)
-SYSCALL(sys_ioprio_get,compat_sys_ioprio_get)
+SYSCALL(sys_ioprio_set,sys_ioprio_set)
+SYSCALL(sys_ioprio_get,sys_ioprio_get)
 SYSCALL(sys_inotify_init,sys_inotify_init)
 SYSCALL(sys_inotify_add_watch,compat_sys_inotify_add_watch)	/* 285 */
-SYSCALL(sys_inotify_rm_watch,compat_sys_inotify_rm_watch)
-NI_SYSCALL						/* 287 sys_migrate_pages */
+SYSCALL(sys_inotify_rm_watch,sys_inotify_rm_watch)
+SYSCALL(sys_migrate_pages,compat_sys_migrate_pages)
 SYSCALL(sys_openat,compat_sys_openat)
 SYSCALL(sys_mkdirat,compat_sys_mkdirat)
 SYSCALL(sys_mknodat,compat_sys_mknodat)			/* 290 */
@@ -318,7 +318,7 @@ SYSCALL(sys_splice,compat_sys_splice)
 SYSCALL(sys_sync_file_range,compat_sys_s390_sync_file_range)
 SYSCALL(sys_tee,compat_sys_tee)
 SYSCALL(sys_vmsplice,compat_sys_vmsplice)
-NI_SYSCALL						/* 310 sys_move_pages */
+SYSCALL(sys_move_pages,compat_sys_move_pages)
 SYSCALL(sys_getcpu,compat_sys_getcpu)
 SYSCALL(sys_epoll_pwait,compat_sys_epoll_pwait)
 SYSCALL(sys_utimes,compat_sys_utimes)
@@ -326,31 +326,31 @@ SYSCALL(sys_fallocate,compat_sys_s390_fallocate)
 SYSCALL(sys_utimensat,compat_sys_utimensat)		/* 315 */
 SYSCALL(sys_signalfd,compat_sys_signalfd)
 NI_SYSCALL						/* 317 old sys_timer_fd */
-SYSCALL(sys_eventfd,compat_sys_eventfd)
-SYSCALL(sys_timerfd_create,compat_sys_timerfd_create)
+SYSCALL(sys_eventfd,sys_eventfd)
+SYSCALL(sys_timerfd_create,sys_timerfd_create)
 SYSCALL(sys_timerfd_settime,compat_sys_timerfd_settime) /* 320 */
 SYSCALL(sys_timerfd_gettime,compat_sys_timerfd_gettime)
 SYSCALL(sys_signalfd4,compat_sys_signalfd4)
-SYSCALL(sys_eventfd2,compat_sys_eventfd2)
-SYSCALL(sys_inotify_init1,compat_sys_inotify_init1)
+SYSCALL(sys_eventfd2,sys_eventfd2)
+SYSCALL(sys_inotify_init1,sys_inotify_init1)
 SYSCALL(sys_pipe2,compat_sys_pipe2)			/* 325 */
-SYSCALL(sys_dup3,compat_sys_dup3)
-SYSCALL(sys_epoll_create1,compat_sys_epoll_create1)
+SYSCALL(sys_dup3,sys_dup3)
+SYSCALL(sys_epoll_create1,sys_epoll_create1)
 SYSCALL(sys_preadv,compat_sys_preadv)
 SYSCALL(sys_pwritev,compat_sys_pwritev)
 SYSCALL(sys_rt_tgsigqueueinfo,compat_sys_rt_tgsigqueueinfo) /* 330 */
 SYSCALL(sys_perf_event_open,compat_sys_perf_event_open)
-SYSCALL(sys_fanotify_init,compat_sys_fanotify_init)
+SYSCALL(sys_fanotify_init,sys_fanotify_init)
 SYSCALL(sys_fanotify_mark,compat_sys_fanotify_mark)
 SYSCALL(sys_prlimit64,compat_sys_prlimit64)
 SYSCALL(sys_name_to_handle_at,compat_sys_name_to_handle_at) /* 335 */
 SYSCALL(sys_open_by_handle_at,compat_sys_open_by_handle_at)
 SYSCALL(sys_clock_adjtime,compat_sys_clock_adjtime)
-SYSCALL(sys_syncfs,compat_sys_syncfs)
-SYSCALL(sys_setns,compat_sys_setns)
+SYSCALL(sys_syncfs,sys_syncfs)
+SYSCALL(sys_setns,sys_setns)
 SYSCALL(sys_process_vm_readv,compat_sys_process_vm_readv) /* 340 */
 SYSCALL(sys_process_vm_writev,compat_sys_process_vm_writev)
-SYSCALL(sys_s390_runtime_instr,compat_sys_s390_runtime_instr)
+SYSCALL(sys_s390_runtime_instr,sys_s390_runtime_instr)
 SYSCALL(sys_kcmp,compat_sys_kcmp)
 SYSCALL(sys_finit_module,compat_sys_finit_module)
 SYSCALL(sys_sched_setattr,compat_sys_sched_setattr)	/* 345 */
@@ -363,3 +363,22 @@ SYSCALL(sys_bpf,compat_sys_bpf)
 SYSCALL(sys_s390_pci_mmio_write,compat_sys_s390_pci_mmio_write)
 SYSCALL(sys_s390_pci_mmio_read,compat_sys_s390_pci_mmio_read)
 SYSCALL(sys_execveat,compat_sys_execveat)
+SYSCALL(sys_userfaultfd,sys_userfaultfd)		/* 355 */
+SYSCALL(sys_membarrier,sys_membarrier)
+SYSCALL(sys_recvmmsg,compat_sys_recvmmsg)
+SYSCALL(sys_sendmmsg,compat_sys_sendmmsg)
+SYSCALL(sys_socket,sys_socket)
+SYSCALL(sys_socketpair,compat_sys_socketpair)		/* 360 */
+SYSCALL(sys_bind,sys_bind)
+SYSCALL(sys_connect,sys_connect)
+SYSCALL(sys_listen,sys_listen)
+SYSCALL(sys_accept4,sys_accept4)
+SYSCALL(sys_getsockopt,compat_sys_getsockopt)		/* 365 */
+SYSCALL(sys_setsockopt,compat_sys_setsockopt)
+SYSCALL(sys_getsockname,compat_sys_getsockname)
+SYSCALL(sys_getpeername,compat_sys_getpeername)
+SYSCALL(sys_sendto,compat_sys_sendto)
+SYSCALL(sys_sendmsg,compat_sys_sendmsg)			/* 370 */
+SYSCALL(sys_recvfrom,compat_sys_recvfrom)
+SYSCALL(sys_recvmsg,compat_sys_recvmsg)
+SYSCALL(sys_shutdown,sys_shutdown)
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 9e733d965e08..017c3a9bfc28 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -58,6 +58,9 @@ EXPORT_SYMBOL_GPL(sched_clock_base_cc);
 
 static DEFINE_PER_CPU(struct clock_event_device, comparators);
 
+ATOMIC_NOTIFIER_HEAD(s390_epoch_delta_notifier);
+EXPORT_SYMBOL(s390_epoch_delta_notifier);
+
 /*
  * Scheduler clock - returns current time in nanosec units.
  */
@@ -117,11 +120,6 @@ static int s390_next_event(unsigned long delta,
 	return 0;
 }
 
-static void s390_set_mode(enum clock_event_mode mode,
-			  struct clock_event_device *evt)
-{
-}
-
 /*
  * Set up lowcore and control register of the current cpu to
  * enable TOD clock and clock comparator interrupts.
@@ -145,7 +143,6 @@ void init_cpu_timer(void)
 	cd->rating		= 400;
 	cd->cpumask		= cpumask_of(cpu);
 	cd->set_next_event	= s390_next_event;
-	cd->set_mode		= s390_set_mode;
 
 	clockevents_register_device(cd);
 
@@ -381,7 +378,7 @@ static void disable_sync_clock(void *dummy)
 	 * increase the "sequence" counter to avoid the race of an
 	 * etr event and the complete recovery against get_sync_clock.
 	 */
-	atomic_clear_mask(0x80000000, sw_ptr);
+	atomic_andnot(0x80000000, sw_ptr);
 	atomic_inc(sw_ptr);
 }
 
@@ -392,7 +389,7 @@ static void disable_sync_clock(void *dummy)
 static void enable_sync_clock(void)
 {
 	atomic_t *sw_ptr = this_cpu_ptr(&clock_sync_word);
-	atomic_set_mask(0x80000000, sw_ptr);
+	atomic_or(0x80000000, sw_ptr);
 }
 
 /*
@@ -752,7 +749,7 @@ static void clock_sync_cpu(struct clock_sync_data *sync)
 static int etr_sync_clock(void *data)
 {
 	static int first;
-	unsigned long long clock, old_clock, delay, delta;
+	unsigned long long clock, old_clock, clock_delta, delay, delta;
 	struct clock_sync_data *etr_sync;
 	struct etr_aib *sync_port, *aib;
 	int port;
@@ -789,6 +786,9 @@ static int etr_sync_clock(void *data)
 		delay = (unsigned long long)
 			(aib->edf2.etv - sync_port->edf2.etv) << 32;
 		delta = adjust_time(old_clock, clock, delay);
+		clock_delta = clock - old_clock;
+		atomic_notifier_call_chain(&s390_epoch_delta_notifier, 0,
+					   &clock_delta);
 		etr_sync->fixup_cc = delta;
 		fixup_clock_comparator(delta);
 		/* Verify that the clock is properly set. */
@@ -1526,7 +1526,7 @@ void stp_island_check(void)
 static int stp_sync_clock(void *data)
 {
 	static int first;
-	unsigned long long old_clock, delta;
+	unsigned long long old_clock, delta, new_clock, clock_delta;
 	struct clock_sync_data *stp_sync;
 	int rc;
 
@@ -1551,7 +1551,11 @@ static int stp_sync_clock(void *data)
 		old_clock = get_tod_clock();
 		rc = chsc_sstpc(stp_page, STP_OP_SYNC, 0);
 		if (rc == 0) {
-			delta = adjust_time(old_clock, get_tod_clock(), 0);
+			new_clock = get_tod_clock();
+			delta = adjust_time(old_clock, new_clock, 0);
+			clock_delta = new_clock - old_clock;
+			atomic_notifier_call_chain(&s390_epoch_delta_notifier,
+						   0, &clock_delta);
 			fixup_clock_comparator(delta);
 			rc = chsc_sstpi(stp_page, &stp_info,
 					sizeof(struct stp_sstpi));
diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
index 5728c5bd44a8..bf05e7fc3e70 100644
--- a/arch/s390/kernel/topology.c
+++ b/arch/s390/kernel/topology.c
@@ -18,7 +18,10 @@
 #include <linux/cpu.h>
 #include <linux/smp.h>
 #include <linux/mm.h>
+#include <linux/nodemask.h>
+#include <linux/node.h>
 #include <asm/sysinfo.h>
+#include <asm/numa.h>
 
 #define PTF_HORIZONTAL	(0UL)
 #define PTF_VERTICAL	(1UL)
@@ -37,8 +40,10 @@ static struct sysinfo_15_1_x *tl_info;
 static int topology_enabled = 1;
 static DECLARE_WORK(topology_work, topology_work_fn);
 
-/* topology_lock protects the socket and book linked lists */
-static DEFINE_SPINLOCK(topology_lock);
+/*
+ * Socket/Book linked lists and per_cpu(cpu_topology) updates are
+ * protected by "sched_domains_mutex".
+ */
 static struct mask_info socket_info;
 static struct mask_info book_info;
 
@@ -188,7 +193,6 @@ static void tl_to_masks(struct sysinfo_15_1_x *info)
 {
 	struct cpuid cpu_id;
 
-	spin_lock_irq(&topology_lock);
 	get_cpu_id(&cpu_id);
 	clear_masks();
 	switch (cpu_id.machine) {
@@ -199,7 +203,6 @@ static void tl_to_masks(struct sysinfo_15_1_x *info)
 	default:
 		__tl_to_masks_generic(info);
 	}
-	spin_unlock_irq(&topology_lock);
 }
 
 static void topology_update_polarization_simple(void)
@@ -244,10 +247,8 @@ int topology_set_cpu_management(int fc)
 
 static void update_cpu_masks(void)
 {
-	unsigned long flags;
 	int cpu;
 
-	spin_lock_irqsave(&topology_lock, flags);
 	for_each_possible_cpu(cpu) {
 		per_cpu(cpu_topology, cpu).thread_mask = cpu_thread_map(cpu);
 		per_cpu(cpu_topology, cpu).core_mask = cpu_group_map(&socket_info, cpu);
@@ -259,7 +260,7 @@ static void update_cpu_masks(void)
 			per_cpu(cpu_topology, cpu).book_id = cpu;
 		}
 	}
-	spin_unlock_irqrestore(&topology_lock, flags);
+	numa_update_cpu_topology();
 }
 
 void store_topology(struct sysinfo_15_1_x *info)
@@ -274,21 +275,21 @@ int arch_update_cpu_topology(void)
 {
 	struct sysinfo_15_1_x *info = tl_info;
 	struct device *dev;
-	int cpu;
+	int cpu, rc = 0;
 
-	if (!MACHINE_HAS_TOPOLOGY) {
-		update_cpu_masks();
-		topology_update_polarization_simple();
-		return 0;
+	if (MACHINE_HAS_TOPOLOGY) {
+		rc = 1;
+		store_topology(info);
+		tl_to_masks(info);
 	}
-	store_topology(info);
-	tl_to_masks(info);
 	update_cpu_masks();
+	if (!MACHINE_HAS_TOPOLOGY)
+		topology_update_polarization_simple();
 	for_each_online_cpu(cpu) {
 		dev = get_cpu_device(cpu);
 		kobject_uevent(&dev->kobj, KOBJ_CHANGE);
 	}
-	return 1;
+	return rc;
 }
 
 static void topology_work_fn(struct work_struct *work)
diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index 7bea81d8a363..9861613fb35a 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -19,7 +19,7 @@
 #include <linux/sched.h>
 #include <linux/mm.h>
 #include <linux/slab.h>
-#include <asm/switch_to.h>
+#include <asm/fpu-internal.h>
 #include "entry.h"
 
 int show_unhandled_signals = 1;
@@ -151,7 +151,7 @@ DO_ERROR_INFO(special_op_exception, SIGILL, ILL_ILLOPN,
 DO_ERROR_INFO(transaction_exception, SIGILL, ILL_ILLOPN,
 	      "transaction constraint exception")
 
-static inline void do_fp_trap(struct pt_regs *regs, int fpc)
+static inline void do_fp_trap(struct pt_regs *regs, __u32 fpc)
 {
 	int si_code = 0;
 	/* FPC[2] is Data Exception Code */
@@ -227,7 +227,7 @@ DO_ERROR_INFO(specification_exception, SIGILL, ILL_ILLOPN,
 int alloc_vector_registers(struct task_struct *tsk)
 {
 	__vector128 *vxrs;
-	int i;
+	freg_t *fprs;
 
 	/* Allocate vector register save area. */
 	vxrs = kzalloc(sizeof(__vector128) * __NUM_VXRS,
@@ -236,15 +236,13 @@ int alloc_vector_registers(struct task_struct *tsk)
 		return -ENOMEM;
 	preempt_disable();
 	if (tsk == current)
-		save_fp_regs(tsk->thread.fp_regs.fprs);
+		save_fpu_regs();
 	/* Copy the 16 floating point registers */
-	for (i = 0; i < 16; i++)
-		*(freg_t *) &vxrs[i] = tsk->thread.fp_regs.fprs[i];
-	tsk->thread.vxrs = vxrs;
-	if (tsk == current) {
-		__ctl_set_bit(0, 17);
-		restore_vx_regs(vxrs);
-	}
+	convert_fp_to_vx(vxrs, tsk->thread.fpu.fprs);
+	fprs = tsk->thread.fpu.fprs;
+	tsk->thread.fpu.vxrs = vxrs;
+	tsk->thread.fpu.flags |= FPU_USE_VX;
+	kfree(fprs);
 	preempt_enable();
 	return 0;
 }
@@ -259,8 +257,8 @@ void vector_exception(struct pt_regs *regs)
 	}
 
 	/* get vector interrupt code from fpc */
-	asm volatile("stfpc %0" : "=Q" (current->thread.fp_regs.fpc));
-	vic = (current->thread.fp_regs.fpc & 0xf00) >> 8;
+	save_fpu_regs();
+	vic = (current->thread.fpu.fpc & 0xf00) >> 8;
 	switch (vic) {
 	case 1: /* invalid vector operation */
 		si_code = FPE_FLTINV;
@@ -297,22 +295,22 @@ void data_exception(struct pt_regs *regs)
 
 	location = get_trap_ip(regs);
 
-	asm volatile("stfpc %0" : "=Q" (current->thread.fp_regs.fpc));
+	save_fpu_regs();
 	/* Check for vector register enablement */
-	if (MACHINE_HAS_VX && !current->thread.vxrs &&
-	    (current->thread.fp_regs.fpc & FPC_DXC_MASK) == 0xfe00) {
+	if (MACHINE_HAS_VX && !is_vx_task(current) &&
+	    (current->thread.fpu.fpc & FPC_DXC_MASK) == 0xfe00) {
 		alloc_vector_registers(current);
 		/* Vector data exception is suppressing, rewind psw. */
 		regs->psw.addr = __rewind_psw(regs->psw, regs->int_code >> 16);
 		clear_pt_regs_flag(regs, PIF_PER_TRAP);
 		return;
 	}
-	if (current->thread.fp_regs.fpc & FPC_DXC_MASK)
+	if (current->thread.fpu.fpc & FPC_DXC_MASK)
 		signal = SIGFPE;
 	else
 		signal = SIGILL;
 	if (signal == SIGFPE)
-		do_fp_trap(regs, current->thread.fp_regs.fpc);
+		do_fp_trap(regs, current->thread.fpu.fpc);
 	else if (signal)
 		do_trap(regs, signal, ILL_ILLOPN, "data exception");
 }
diff --git a/arch/s390/kernel/vdso32/Makefile b/arch/s390/kernel/vdso32/Makefile
index 8ad2b34ad151..ee8a18e50a25 100644
--- a/arch/s390/kernel/vdso32/Makefile
+++ b/arch/s390/kernel/vdso32/Makefile
@@ -13,7 +13,7 @@ KBUILD_AFLAGS_31 += -m31 -s
 KBUILD_CFLAGS_31 := $(filter-out -m64,$(KBUILD_CFLAGS))
 KBUILD_CFLAGS_31 += -m31 -fPIC -shared -fno-common -fno-builtin
 KBUILD_CFLAGS_31 += -nostdlib -Wl,-soname=linux-vdso32.so.1 \
-			$(call cc-ldoption, -Wl$(comma)--hash-style=sysv)
+			$(call cc-ldoption, -Wl$(comma)--hash-style=both)
 
 $(targets:%=$(obj)/%.dbg): KBUILD_CFLAGS = $(KBUILD_CFLAGS_31)
 $(targets:%=$(obj)/%.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_31)
diff --git a/arch/s390/kernel/vdso64/Makefile b/arch/s390/kernel/vdso64/Makefile
index 2a8ddfd12a5b..c4b03f9ed228 100644
--- a/arch/s390/kernel/vdso64/Makefile
+++ b/arch/s390/kernel/vdso64/Makefile
@@ -13,7 +13,7 @@ KBUILD_AFLAGS_64 += -m64 -s
 KBUILD_CFLAGS_64 := $(filter-out -m64,$(KBUILD_CFLAGS))
 KBUILD_CFLAGS_64 += -m64 -fPIC -shared -fno-common -fno-builtin
 KBUILD_CFLAGS_64 += -nostdlib -Wl,-soname=linux-vdso64.so.1 \
-			$(call cc-ldoption, -Wl$(comma)--hash-style=sysv)
+			$(call cc-ldoption, -Wl$(comma)--hash-style=both)
 
 $(targets:%=$(obj)/%.dbg): KBUILD_CFLAGS = $(KBUILD_CFLAGS_64)
 $(targets:%=$(obj)/%.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_64)
diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c
index e53d3595a7c8..dafc44f519c3 100644
--- a/arch/s390/kernel/vtime.c
+++ b/arch/s390/kernel/vtime.c
@@ -25,9 +25,10 @@ static DEFINE_SPINLOCK(virt_timer_lock);
 static atomic64_t virt_timer_current;
 static atomic64_t virt_timer_elapsed;
 
-static DEFINE_PER_CPU(u64, mt_cycles[32]);
+DEFINE_PER_CPU(u64, mt_cycles[8]);
 static DEFINE_PER_CPU(u64, mt_scaling_mult) = { 1 };
 static DEFINE_PER_CPU(u64, mt_scaling_div) = { 1 };
+static DEFINE_PER_CPU(u64, mt_scaling_jiffies);
 
 static inline u64 get_vtimer(void)
 {
@@ -59,6 +60,34 @@ static inline int virt_timer_forward(u64 elapsed)
 	return elapsed >= atomic64_read(&virt_timer_current);
 }
 
+static void update_mt_scaling(void)
+{
+	u64 cycles_new[8], *cycles_old;
+	u64 delta, fac, mult, div;
+	int i;
+
+	stcctm5(smp_cpu_mtid + 1, cycles_new);
+	cycles_old = this_cpu_ptr(mt_cycles);
+	fac = 1;
+	mult = div = 0;
+	for (i = 0; i <= smp_cpu_mtid; i++) {
+		delta = cycles_new[i] - cycles_old[i];
+		div += delta;
+		mult *= i + 1;
+		mult += delta * fac;
+		fac *= i + 1;
+	}
+	div *= fac;
+	if (div > 0) {
+		/* Update scaling factor */
+		__this_cpu_write(mt_scaling_mult, mult);
+		__this_cpu_write(mt_scaling_div, div);
+		memcpy(cycles_old, cycles_new,
+		       sizeof(u64) * (smp_cpu_mtid + 1));
+	}
+	__this_cpu_write(mt_scaling_jiffies, jiffies_64);
+}
+
 /*
  * Update process times based on virtual cpu times stored by entry.S
  * to the lowcore fields user_timer, system_timer & steal_clock.
@@ -68,7 +97,6 @@ static int do_account_vtime(struct task_struct *tsk, int hardirq_offset)
 	struct thread_info *ti = task_thread_info(tsk);
 	u64 timer, clock, user, system, steal;
 	u64 user_scaled, system_scaled;
-	int i;
 
 	timer = S390_lowcore.last_update_timer;
 	clock = S390_lowcore.last_update_clock;
@@ -84,28 +112,10 @@ static int do_account_vtime(struct task_struct *tsk, int hardirq_offset)
 	S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer;
 	S390_lowcore.steal_timer += S390_lowcore.last_update_clock - clock;
 
-	/* Do MT utilization calculation */
-	if (smp_cpu_mtid) {
-		u64 cycles_new[32], *cycles_old;
-		u64 delta, mult, div;
-
-		cycles_old = this_cpu_ptr(mt_cycles);
-		if (stcctm5(smp_cpu_mtid + 1, cycles_new) < 2) {
-			mult = div = 0;
-			for (i = 0; i <= smp_cpu_mtid; i++) {
-				delta = cycles_new[i] - cycles_old[i];
-				mult += delta;
-				div += (i + 1) * delta;
-			}
-			if (mult > 0) {
-				/* Update scaling factor */
-				__this_cpu_write(mt_scaling_mult, mult);
-				__this_cpu_write(mt_scaling_div, div);
-				memcpy(cycles_old, cycles_new,
-				       sizeof(u64) * (smp_cpu_mtid + 1));
-			}
-		}
-	}
+	/* Update MT utilization calculation */
+	if (smp_cpu_mtid &&
+	    time_after64(jiffies_64, this_cpu_read(mt_scaling_jiffies)))
+		update_mt_scaling();
 
 	user = S390_lowcore.user_timer - ti->user_timer;
 	S390_lowcore.steal_timer -= user;
@@ -174,6 +184,11 @@ void vtime_account_irq_enter(struct task_struct *tsk)
 	S390_lowcore.last_update_timer = get_vtimer();
 	S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer;
 
+	/* Update MT utilization calculation */
+	if (smp_cpu_mtid &&
+	    time_after64(jiffies_64, this_cpu_read(mt_scaling_jiffies)))
+		update_mt_scaling();
+
 	system = S390_lowcore.system_timer - ti->system_timer;
 	S390_lowcore.steal_timer -= system;
 	ti->system_timer = S390_lowcore.system_timer;
@@ -376,4 +391,11 @@ void vtime_init(void)
 {
 	/* set initial cpu timer */
 	set_vtimer(VTIMER_MAX_SLICE);
+	/* Setup initial MT scaling values */
+	if (smp_cpu_mtid) {
+		__this_cpu_write(mt_scaling_jiffies, jiffies);
+		__this_cpu_write(mt_scaling_mult, 1);
+		__this_cpu_write(mt_scaling_div, 1);
+		stcctm5(smp_cpu_mtid + 1, this_cpu_ptr(mt_cycles));
+	}
 }
author	Ingo Molnar <mingo@kernel.org>	2015-10-14 16:05:18 +0200
committer	Ingo Molnar <mingo@kernel.org>	2015-10-14 16:05:18 +0200
commit	c7d77a7980e434c3af17de19e3348157f9b9ccce (patch)
tree	b32c5988ce8239b80c83e94c22d68f5eb0fb84da /arch/s390/kernel
parent	efi: Use the generic efi.memmap instead of 'memmap' (diff)
parent	x86/efi: Fix multiple GOP device support (diff)
download	linux-c7d77a7980e434c3af17de19e3348157f9b9ccce.tar.xz linux-c7d77a7980e434c3af17de19e3348157f9b9ccce.zip