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

Pull s390 updates from Martin Schwidefsky: "The most notable change for this pull request is the ftrace rework from Heiko. It brings a small performance improvement and the ground work to support a new gcc option to replace the mcount blocks with a single nop. Two new s390 specific system calls are added to emulate user space mmio for PCI, an artifact of the how PCI memory is accessed. Two patches for the memory management with changes to common code. For KVM mm_forbids_zeropage is added which disables the empty zero page for an mm that is used by a KVM process. And an optimization, pmdp_get_and_clear_full is added analog to ptep_get_and_clear_full. Some micro optimization for the cmpxchg and the spinlock code. And as usual bug fixes and cleanups" * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (46 commits) s390/cputime: fix 31-bit compile s390/scm_block: make the number of reqs per HW req configurable s390/scm_block: handle multiple requests in one HW request s390/scm_block: allocate aidaw pages only when necessary s390/scm_block: use mempool to manage aidaw requests s390/eadm: change timeout value s390/mm: fix memory leak of ptlock in pmd_free_tlb s390: use local symbol names in entry[64].S s390/ptrace: always include vector registers in core files s390/simd: clear vector register pointer on fork/clone s390: translate cputime magic constants to macros s390/idle: convert open coded idle time seqcount s390/idle: add missing irq off lockdep annotation s390/debug: avoid function call for debug_sprintf_* s390/kprobes: fix instruction copy for out of line execution s390: remove diag 44 calls from cpu_relax() s390/dasd: retry partition detection s390/dasd: fix list corruption for sleep_on requests s390/dasd: fix infinite term I/O loop s390/dasd: remove unused code ...
author: Linus Torvalds 2014-12-11 17:30:55 -0800
committer: Linus Torvalds 2014-12-11 17:30:55 -0800
commit: 27afc5dbda52ee3dbcd0bda7375c917c6936b470 (patch)
tree: 47591400f85590d48fa71bbfa50e0707e20e4bd0 /arch/s390/kernel
parent: 70e71ca0af244f48a5dcf56dc435243792e3a495 (diff)
parent: 351997810131565fe62aec2c366deccbf6bda3f4 (diff)
23 files changed, 739 insertions, 590 deletions
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index ef279a136801..e07e91605353 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -17,8 +17,8 @@
  * Make sure that the compiler is new enough. We want a compiler that
  * is known to work with the "Q" assembler constraint.
  */
-#if __GNUC__ < 3 || (__GNUC__ == 3 && __GNUC_MINOR__ < 3)
-#error Your compiler is too old; please use version 3.3.3 or newer
+#if __GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 3)
+#error Your compiler is too old; please use version 4.3 or newer
 #endif
 
 int main(void)
@@ -156,7 +156,6 @@ int main(void)
 	DEFINE(__LC_INT_CLOCK, offsetof(struct _lowcore, int_clock));
 	DEFINE(__LC_MCCK_CLOCK, offsetof(struct _lowcore, mcck_clock));
 	DEFINE(__LC_MACHINE_FLAGS, offsetof(struct _lowcore, machine_flags));
-	DEFINE(__LC_FTRACE_FUNC, offsetof(struct _lowcore, ftrace_func));
 	DEFINE(__LC_DUMP_REIPL, offsetof(struct _lowcore, ipib));
 	BLANK();
 	DEFINE(__LC_CPU_TIMER_SAVE_AREA, offsetof(struct _lowcore, cpu_timer_save_area));
diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c
index 009f5eb11125..34d5fa7b01b5 100644
--- a/arch/s390/kernel/compat_signal.c
+++ b/arch/s390/kernel/compat_signal.c
@@ -434,7 +434,7 @@ static int setup_frame32(struct ksignal *ksig, sigset_t *set,
 			ksig->ka.sa.sa_restorer | PSW32_ADDR_AMODE;
 	} else {
 		/* Signal frames without vectors registers are short ! */
-		__u16 __user *svc = (void *) frame + frame_size - 2;
+		__u16 __user *svc = (void __user *) frame + frame_size - 2;
 		if (__put_user(S390_SYSCALL_OPCODE | __NR_sigreturn, svc))
 			return -EFAULT;
 		restorer = (unsigned long __force) svc | PSW32_ADDR_AMODE;
diff --git a/arch/s390/kernel/compat_wrapper.c b/arch/s390/kernel/compat_wrapper.c
index c4f7a3d655b8..d7fa2f0f1425 100644
--- a/arch/s390/kernel/compat_wrapper.c
+++ b/arch/s390/kernel/compat_wrapper.c
@@ -218,3 +218,5 @@ COMPAT_SYSCALL_WRAP3(seccomp, unsigned int, op, unsigned int, flags, const char
 COMPAT_SYSCALL_WRAP3(getrandom, char __user *, buf, size_t, count, unsigned int, flags)
 COMPAT_SYSCALL_WRAP2(memfd_create, const char __user *, uname, unsigned int, flags)
 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);
diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index ee8390da6ea7..c1f21aca76e7 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -1019,7 +1019,7 @@ debug_count_numargs(char *string)
  */
 
 debug_entry_t*
-debug_sprintf_event(debug_info_t* id, int level,char *string,...)
+__debug_sprintf_event(debug_info_t *id, int level, char *string, ...)
 {
 	va_list   ap;
 	int numargs,idx;
@@ -1027,8 +1027,6 @@ debug_sprintf_event(debug_info_t* id, int level,char *string,...)
 	debug_sprintf_entry_t *curr_event;
 	debug_entry_t *active;
 
-	if((!id) || (level > id->level))
-		return NULL;
 	if (!debug_active || !id->areas)
 		return NULL;
 	numargs=debug_count_numargs(string);
@@ -1050,14 +1048,14 @@ debug_sprintf_event(debug_info_t* id, int level,char *string,...)
 
 	return active;
 }
-EXPORT_SYMBOL(debug_sprintf_event);
+EXPORT_SYMBOL(__debug_sprintf_event);
 
 /*
  * debug_sprintf_exception:
  */
 
 debug_entry_t*
-debug_sprintf_exception(debug_info_t* id, int level,char *string,...)
+__debug_sprintf_exception(debug_info_t *id, int level, char *string, ...)
 {
 	va_list   ap;
 	int numargs,idx;
@@ -1065,8 +1063,6 @@ debug_sprintf_exception(debug_info_t* id, int level,char *string,...)
 	debug_sprintf_entry_t *curr_event;
 	debug_entry_t *active;
 
-	if((!id) || (level > id->level))
-		return NULL;
 	if (!debug_active || !id->areas)
 		return NULL;
 
@@ -1089,7 +1085,7 @@ debug_sprintf_exception(debug_info_t* id, int level,char *string,...)
 
 	return active;
 }
-EXPORT_SYMBOL(debug_sprintf_exception);
+EXPORT_SYMBOL(__debug_sprintf_exception);
 
 /*
  * debug_register_view:
diff --git a/arch/s390/kernel/dumpstack.c b/arch/s390/kernel/dumpstack.c
index acb412442e5e..a99852e96a77 100644
--- a/arch/s390/kernel/dumpstack.c
+++ b/arch/s390/kernel/dumpstack.c
@@ -191,7 +191,8 @@ void die(struct pt_regs *regs, const char *str)
 	console_verbose();
 	spin_lock_irq(&die_lock);
 	bust_spinlocks(1);
-	printk("%s: %04x [#%d] ", str, regs->int_code & 0xffff, ++die_counter);
+	printk("%s: %04x ilc:%d [#%d] ", str, regs->int_code & 0xffff,
+	       regs->int_code >> 17, ++die_counter);
 #ifdef CONFIG_PREEMPT
 	printk("PREEMPT ");
 #endif
diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index cef2879edff3..302ac1f7f8e7 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -12,7 +12,6 @@
 #include <linux/errno.h>
 #include <linux/string.h>
 #include <linux/ctype.h>
-#include <linux/ftrace.h>
 #include <linux/lockdep.h>
 #include <linux/module.h>
 #include <linux/pfn.h>
@@ -490,8 +489,5 @@ void __init startup_init(void)
 	detect_machine_facilities();
 	setup_topology();
 	sclp_early_detect();
-#ifdef CONFIG_DYNAMIC_FTRACE
-	S390_lowcore.ftrace_func = (unsigned long)ftrace_caller;
-#endif
 	lockdep_on();
 }
diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index 70203265196f..398329b2b518 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -53,7 +53,7 @@ _PIF_WORK	= (_PIF_PER_TRAP)
 	.macro	TRACE_IRQS_ON
 #ifdef CONFIG_TRACE_IRQFLAGS
 	basr	%r2,%r0
-	l	%r1,BASED(.Lhardirqs_on)
+	l	%r1,BASED(.Lc_hardirqs_on)
 	basr	%r14,%r1		# call trace_hardirqs_on_caller
 #endif
 	.endm
@@ -61,7 +61,7 @@ _PIF_WORK	= (_PIF_PER_TRAP)
 	.macro	TRACE_IRQS_OFF
 #ifdef CONFIG_TRACE_IRQFLAGS
 	basr	%r2,%r0
-	l	%r1,BASED(.Lhardirqs_off)
+	l	%r1,BASED(.Lc_hardirqs_off)
 	basr	%r14,%r1		# call trace_hardirqs_off_caller
 #endif
 	.endm
@@ -70,7 +70,7 @@ _PIF_WORK	= (_PIF_PER_TRAP)
 #ifdef CONFIG_LOCKDEP
 	tm	__PT_PSW+1(%r11),0x01	# returning to user ?
 	jz	.+10
-	l	%r1,BASED(.Llockdep_sys_exit)
+	l	%r1,BASED(.Lc_lockdep_sys_exit)
 	basr	%r14,%r1		# call lockdep_sys_exit
 #endif
 	.endm
@@ -87,8 +87,8 @@ _PIF_WORK	= (_PIF_PER_TRAP)
 	tmh	%r8,0x0001		# interrupting from user ?
 	jnz	1f
 	lr	%r14,%r9
-	sl	%r14,BASED(.Lcritical_start)
-	cl	%r14,BASED(.Lcritical_length)
+	sl	%r14,BASED(.Lc_critical_start)
+	cl	%r14,BASED(.Lc_critical_length)
 	jhe	0f
 	la	%r11,\savearea		# inside critical section, do cleanup
 	bras	%r14,cleanup_critical
@@ -162,7 +162,7 @@ ENTRY(__switch_to)
 	lm	%r6,%r15,__SF_GPRS(%r15)	# load gprs of next task
 	br	%r14
 
-__critical_start:
+.L__critical_start:
 /*
  * SVC interrupt handler routine. System calls are synchronous events and
  * are executed with interrupts enabled.
@@ -170,145 +170,145 @@ __critical_start:
 
 ENTRY(system_call)
 	stpt	__LC_SYNC_ENTER_TIMER
-sysc_stm:
+.Lsysc_stm:
 	stm	%r8,%r15,__LC_SAVE_AREA_SYNC
 	l	%r12,__LC_THREAD_INFO
 	l	%r13,__LC_SVC_NEW_PSW+4
 	lhi	%r14,_PIF_SYSCALL
-sysc_per:
+.Lsysc_per:
 	l	%r15,__LC_KERNEL_STACK
 	la	%r11,STACK_FRAME_OVERHEAD(%r15)	# pointer to pt_regs
-sysc_vtime:
+.Lsysc_vtime:
 	UPDATE_VTIME %r8,%r9,__LC_SYNC_ENTER_TIMER
 	stm	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(32,%r11),__LC_SAVE_AREA_SYNC
 	mvc	__PT_PSW(8,%r11),__LC_SVC_OLD_PSW
 	mvc	__PT_INT_CODE(4,%r11),__LC_SVC_ILC
 	st	%r14,__PT_FLAGS(%r11)
-sysc_do_svc:
+.Lsysc_do_svc:
 	l	%r10,__TI_sysc_table(%r12)	# 31 bit system call table
 	lh	%r8,__PT_INT_CODE+2(%r11)
 	sla	%r8,2				# shift and test for svc0
-	jnz	sysc_nr_ok
+	jnz	.Lsysc_nr_ok
 	# svc 0: system call number in %r1
 	cl	%r1,BASED(.Lnr_syscalls)
-	jnl	sysc_nr_ok
+	jnl	.Lsysc_nr_ok
 	sth	%r1,__PT_INT_CODE+2(%r11)
 	lr	%r8,%r1
 	sla	%r8,2
-sysc_nr_ok:
+.Lsysc_nr_ok:
 	xc	__SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15)
 	st	%r2,__PT_ORIG_GPR2(%r11)
 	st	%r7,STACK_FRAME_OVERHEAD(%r15)
 	l	%r9,0(%r8,%r10)			# get system call addr.
 	tm	__TI_flags+3(%r12),_TIF_TRACE
-	jnz	sysc_tracesys
+	jnz	.Lsysc_tracesys
 	basr	%r14,%r9			# call sys_xxxx
 	st	%r2,__PT_R2(%r11)		# store return value
 
-sysc_return:
+.Lsysc_return:
 	LOCKDEP_SYS_EXIT
-sysc_tif:
+.Lsysc_tif:
 	tm	__PT_PSW+1(%r11),0x01		# returning to user ?
-	jno	sysc_restore
+	jno	.Lsysc_restore
 	tm	__PT_FLAGS+3(%r11),_PIF_WORK
-	jnz	sysc_work
+	jnz	.Lsysc_work
 	tm	__TI_flags+3(%r12),_TIF_WORK
-	jnz	sysc_work			# check for thread work
+	jnz	.Lsysc_work			# check for thread work
 	tm	__LC_CPU_FLAGS+3,_CIF_WORK
-	jnz	sysc_work
-sysc_restore:
+	jnz	.Lsysc_work
+.Lsysc_restore:
 	mvc	__LC_RETURN_PSW(8),__PT_PSW(%r11)
 	stpt	__LC_EXIT_TIMER
 	lm	%r0,%r15,__PT_R0(%r11)
 	lpsw	__LC_RETURN_PSW
-sysc_done:
+.Lsysc_done:
 
 #
 # One of the work bits is on. Find out which one.
 #
-sysc_work:
+.Lsysc_work:
 	tm	__LC_CPU_FLAGS+3,_CIF_MCCK_PENDING
-	jo	sysc_mcck_pending
+	jo	.Lsysc_mcck_pending
 	tm	__TI_flags+3(%r12),_TIF_NEED_RESCHED
-	jo	sysc_reschedule
+	jo	.Lsysc_reschedule
 	tm	__PT_FLAGS+3(%r11),_PIF_PER_TRAP
-	jo	sysc_singlestep
+	jo	.Lsysc_singlestep
 	tm	__TI_flags+3(%r12),_TIF_SIGPENDING
-	jo	sysc_sigpending
+	jo	.Lsysc_sigpending
 	tm	__TI_flags+3(%r12),_TIF_NOTIFY_RESUME
-	jo	sysc_notify_resume
+	jo	.Lsysc_notify_resume
 	tm	__LC_CPU_FLAGS+3,_CIF_ASCE
-	jo	sysc_uaccess
-	j	sysc_return		# beware of critical section cleanup
+	jo	.Lsysc_uaccess
+	j	.Lsysc_return		# beware of critical section cleanup
 
 #
 # _TIF_NEED_RESCHED is set, call schedule
 #
-sysc_reschedule:
-	l	%r1,BASED(.Lschedule)
-	la	%r14,BASED(sysc_return)
+.Lsysc_reschedule:
+	l	%r1,BASED(.Lc_schedule)
+	la	%r14,BASED(.Lsysc_return)
 	br	%r1			# call schedule
 
 #
 # _CIF_MCCK_PENDING is set, call handler
 #
-sysc_mcck_pending:
-	l	%r1,BASED(.Lhandle_mcck)
-	la	%r14,BASED(sysc_return)
+.Lsysc_mcck_pending:
+	l	%r1,BASED(.Lc_handle_mcck)
+	la	%r14,BASED(.Lsysc_return)
 	br	%r1			# TIF bit will be cleared by handler
 
 #
 # _CIF_ASCE is set, load user space asce
 #
-sysc_uaccess:
+.Lsysc_uaccess:
 	ni	__LC_CPU_FLAGS+3,255-_CIF_ASCE
 	lctl	%c1,%c1,__LC_USER_ASCE	# load primary asce
-	j	sysc_return
+	j	.Lsysc_return
 
 #
 # _TIF_SIGPENDING is set, call do_signal
 #
-sysc_sigpending:
+.Lsysc_sigpending:
 	lr	%r2,%r11		# pass pointer to pt_regs
-	l	%r1,BASED(.Ldo_signal)
+	l	%r1,BASED(.Lc_do_signal)
 	basr	%r14,%r1		# call do_signal
 	tm	__PT_FLAGS+3(%r11),_PIF_SYSCALL
-	jno	sysc_return
+	jno	.Lsysc_return
 	lm	%r2,%r7,__PT_R2(%r11)	# load svc arguments
 	l	%r10,__TI_sysc_table(%r12)	# 31 bit system call table
 	xr	%r8,%r8			# svc 0 returns -ENOSYS
 	clc	__PT_INT_CODE+2(2,%r11),BASED(.Lnr_syscalls+2)
-	jnl	sysc_nr_ok		# invalid svc number -> do svc 0
+	jnl	.Lsysc_nr_ok		# invalid svc number -> do svc 0
 	lh	%r8,__PT_INT_CODE+2(%r11)	# load new svc number
 	sla	%r8,2
-	j	sysc_nr_ok		# restart svc
+	j	.Lsysc_nr_ok		# restart svc
 
 #
 # _TIF_NOTIFY_RESUME is set, call do_notify_resume
 #
-sysc_notify_resume:
+.Lsysc_notify_resume:
 	lr	%r2,%r11		# pass pointer to pt_regs
-	l	%r1,BASED(.Ldo_notify_resume)
-	la	%r14,BASED(sysc_return)
+	l	%r1,BASED(.Lc_do_notify_resume)
+	la	%r14,BASED(.Lsysc_return)
 	br	%r1			# call do_notify_resume
 
 #
 # _PIF_PER_TRAP is set, call do_per_trap
 #
-sysc_singlestep:
+.Lsysc_singlestep:
 	ni	__PT_FLAGS+3(%r11),255-_PIF_PER_TRAP
 	lr	%r2,%r11		# pass pointer to pt_regs
-	l	%r1,BASED(.Ldo_per_trap)
-	la	%r14,BASED(sysc_return)
+	l	%r1,BASED(.Lc_do_per_trap)
+	la	%r14,BASED(.Lsysc_return)
 	br	%r1			# call do_per_trap
 
 #
 # call tracehook_report_syscall_entry/tracehook_report_syscall_exit before
 # and after the system call
 #
-sysc_tracesys:
-	l	%r1,BASED(.Ltrace_enter)
+.Lsysc_tracesys:
+	l	%r1,BASED(.Lc_trace_enter)
 	lr	%r2,%r11		# pass pointer to pt_regs
 	la	%r3,0
 	xr	%r0,%r0
@@ -316,22 +316,22 @@ sysc_tracesys:
 	st	%r0,__PT_R2(%r11)
 	basr	%r14,%r1		# call do_syscall_trace_enter
 	cl	%r2,BASED(.Lnr_syscalls)
-	jnl	sysc_tracenogo
+	jnl	.Lsysc_tracenogo
 	lr	%r8,%r2
 	sll	%r8,2
 	l	%r9,0(%r8,%r10)
-sysc_tracego:
+.Lsysc_tracego:
 	lm	%r3,%r7,__PT_R3(%r11)
 	st	%r7,STACK_FRAME_OVERHEAD(%r15)
 	l	%r2,__PT_ORIG_GPR2(%r11)
 	basr	%r14,%r9		# call sys_xxx
 	st	%r2,__PT_R2(%r11)	# store return value
-sysc_tracenogo:
+.Lsysc_tracenogo:
 	tm	__TI_flags+3(%r12),_TIF_TRACE
-	jz	sysc_return
-	l	%r1,BASED(.Ltrace_exit)
+	jz	.Lsysc_return
+	l	%r1,BASED(.Lc_trace_exit)
 	lr	%r2,%r11		# pass pointer to pt_regs
-	la	%r14,BASED(sysc_return)
+	la	%r14,BASED(.Lsysc_return)
 	br	%r1			# call do_syscall_trace_exit
 
 #
@@ -341,18 +341,18 @@ ENTRY(ret_from_fork)
 	la	%r11,STACK_FRAME_OVERHEAD(%r15)
 	l	%r12,__LC_THREAD_INFO
 	l	%r13,__LC_SVC_NEW_PSW+4
-	l	%r1,BASED(.Lschedule_tail)
+	l	%r1,BASED(.Lc_schedule_tail)
 	basr	%r14,%r1		# call schedule_tail
 	TRACE_IRQS_ON
 	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
 	tm	__PT_PSW+1(%r11),0x01	# forking a kernel thread ?
-	jne	sysc_tracenogo
+	jne	.Lsysc_tracenogo
 	# it's a kernel thread
 	lm	%r9,%r10,__PT_R9(%r11)	# load gprs
 ENTRY(kernel_thread_starter)
 	la	%r2,0(%r10)
 	basr	%r14,%r9
-	j	sysc_tracenogo
+	j	.Lsysc_tracenogo
 
 /*
  * Program check handler routine
@@ -369,7 +369,7 @@ ENTRY(pgm_check_handler)
 	tmh	%r8,0x4000		# PER bit set in old PSW ?
 	jnz	0f			# -> enabled, can't be a double fault
 	tm	__LC_PGM_ILC+3,0x80	# check for per exception
-	jnz	pgm_svcper		# -> single stepped svc
+	jnz	.Lpgm_svcper		# -> single stepped svc
 0:	CHECK_STACK STACK_SIZE,__LC_SAVE_AREA_SYNC
 	ahi	%r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
 	j	2f
@@ -386,42 +386,42 @@ ENTRY(pgm_check_handler)
 	jz	0f
 	l	%r1,__TI_task(%r12)
 	tmh	%r8,0x0001		# kernel per event ?
-	jz	pgm_kprobe
+	jz	.Lpgm_kprobe
 	oi	__PT_FLAGS+3(%r11),_PIF_PER_TRAP
 	mvc	__THREAD_per_address(4,%r1),__LC_PER_ADDRESS
 	mvc	__THREAD_per_cause(2,%r1),__LC_PER_CODE
 	mvc	__THREAD_per_paid(1,%r1),__LC_PER_ACCESS_ID
 0:	REENABLE_IRQS
 	xc	__SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15)
-	l	%r1,BASED(.Ljump_table)
+	l	%r1,BASED(.Lc_jump_table)
 	la	%r10,0x7f
 	n	%r10,__PT_INT_CODE(%r11)
-	je	sysc_return
+	je	.Lsysc_return
 	sll	%r10,2
 	l	%r1,0(%r10,%r1)		# load address of handler routine
 	lr	%r2,%r11		# pass pointer to pt_regs
 	basr	%r14,%r1		# branch to interrupt-handler
-	j	sysc_return
+	j	.Lsysc_return
 
 #
 # PER event in supervisor state, must be kprobes
 #
-pgm_kprobe:
+.Lpgm_kprobe:
 	REENABLE_IRQS
 	xc	__SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15)
-	l	%r1,BASED(.Ldo_per_trap)
+	l	%r1,BASED(.Lc_do_per_trap)
 	lr	%r2,%r11		# pass pointer to pt_regs
 	basr	%r14,%r1		# call do_per_trap
-	j	sysc_return
+	j	.Lsysc_return
 
 #
 # single stepped system call
 #
-pgm_svcper:
+.Lpgm_svcper:
 	mvc	__LC_RETURN_PSW(4),__LC_SVC_NEW_PSW
-	mvc	__LC_RETURN_PSW+4(4),BASED(.Lsysc_per)
+	mvc	__LC_RETURN_PSW+4(4),BASED(.Lc_sysc_per)
 	lhi	%r14,_PIF_SYSCALL | _PIF_PER_TRAP
-	lpsw	__LC_RETURN_PSW		# branch to sysc_per and enable irqs
+	lpsw	__LC_RETURN_PSW		# branch to .Lsysc_per and enable irqs
 
 /*
  * IO interrupt handler routine
@@ -435,9 +435,9 @@ ENTRY(io_int_handler)
 	l	%r13,__LC_SVC_NEW_PSW+4
 	lm	%r8,%r9,__LC_IO_OLD_PSW
 	tmh	%r8,0x0001		# interrupting from user ?
-	jz	io_skip
+	jz	.Lio_skip
 	UPDATE_VTIME %r14,%r15,__LC_ASYNC_ENTER_TIMER
-io_skip:
+.Lio_skip:
 	SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_STACK,STACK_SHIFT
 	stm	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(32,%r11),__LC_SAVE_AREA_ASYNC
@@ -446,35 +446,35 @@ io_skip:
 	xc	__PT_FLAGS(4,%r11),__PT_FLAGS(%r11)
 	TRACE_IRQS_OFF
 	xc	__SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15)
-io_loop:
-	l	%r1,BASED(.Ldo_IRQ)
+.Lio_loop:
+	l	%r1,BASED(.Lc_do_IRQ)
 	lr	%r2,%r11		# pass pointer to pt_regs
 	lhi	%r3,IO_INTERRUPT
 	tm	__PT_INT_CODE+8(%r11),0x80	# adapter interrupt ?
-	jz	io_call
+	jz	.Lio_call
 	lhi	%r3,THIN_INTERRUPT
-io_call:
+.Lio_call:
 	basr	%r14,%r1		# call do_IRQ
 	tm	__LC_MACHINE_FLAGS+2,0x10	# MACHINE_FLAG_LPAR
-	jz	io_return
+	jz	.Lio_return
 	tpi	0
-	jz	io_return
+	jz	.Lio_return
 	mvc	__PT_INT_CODE(12,%r11),__LC_SUBCHANNEL_ID
-	j	io_loop
-io_return:
+	j	.Lio_loop
+.Lio_return:
 	LOCKDEP_SYS_EXIT
 	TRACE_IRQS_ON
-io_tif:
+.Lio_tif:
 	tm	__TI_flags+3(%r12),_TIF_WORK
-	jnz	io_work			# there is work to do (signals etc.)
+	jnz	.Lio_work		# there is work to do (signals etc.)
 	tm	__LC_CPU_FLAGS+3,_CIF_WORK
-	jnz	io_work
-io_restore:
+	jnz	.Lio_work
+.Lio_restore:
 	mvc	__LC_RETURN_PSW(8),__PT_PSW(%r11)
 	stpt	__LC_EXIT_TIMER
 	lm	%r0,%r15,__PT_R0(%r11)
 	lpsw	__LC_RETURN_PSW
-io_done:
+.Lio_done:
 
 #
 # There is work todo, find out in which context we have been interrupted:
@@ -483,15 +483,15 @@ io_done:
 #    the preemption counter and if it is zero call preempt_schedule_irq
 # Before any work can be done, a switch to the kernel stack is required.
 #
-io_work:
+.Lio_work:
 	tm	__PT_PSW+1(%r11),0x01	# returning to user ?
-	jo	io_work_user		# yes -> do resched & signal
+	jo	.Lio_work_user		# yes -> do resched & signal
 #ifdef CONFIG_PREEMPT
 	# check for preemptive scheduling
 	icm	%r0,15,__TI_precount(%r12)
-	jnz	io_restore		# preemption disabled
+	jnz	.Lio_restore		# preemption disabled
 	tm	__TI_flags+3(%r12),_TIF_NEED_RESCHED
-	jno	io_restore
+	jno	.Lio_restore
 	# switch to kernel stack
 	l	%r1,__PT_R15(%r11)
 	ahi	%r1,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
@@ -499,20 +499,20 @@ io_work:
 	xc	__SF_BACKCHAIN(4,%r1),__SF_BACKCHAIN(%r1)
 	la	%r11,STACK_FRAME_OVERHEAD(%r1)
 	lr	%r15,%r1
-	# TRACE_IRQS_ON already done at io_return, call
+	# TRACE_IRQS_ON already done at .Lio_return, call
 	# TRACE_IRQS_OFF to keep things symmetrical
 	TRACE_IRQS_OFF
-	l	%r1,BASED(.Lpreempt_irq)
+	l	%r1,BASED(.Lc_preempt_irq)
 	basr	%r14,%r1		# call preempt_schedule_irq
-	j	io_return
+	j	.Lio_return
 #else
-	j	io_restore
+	j	.Lio_restore
 #endif
 
 #
 # Need to do work before returning to userspace, switch to kernel stack
 #
-io_work_user:
+.Lio_work_user:
 	l	%r1,__LC_KERNEL_STACK
 	mvc	STACK_FRAME_OVERHEAD(__PT_SIZE,%r1),0(%r11)
 	xc	__SF_BACKCHAIN(4,%r1),__SF_BACKCHAIN(%r1)
@@ -522,74 +522,74 @@ io_work_user:
 #
 # One of the work bits is on. Find out which one.
 #
-io_work_tif:
+.Lio_work_tif:
 	tm	__LC_CPU_FLAGS+3(%r12),_CIF_MCCK_PENDING
-	jo	io_mcck_pending
+	jo	.Lio_mcck_pending
 	tm	__TI_flags+3(%r12),_TIF_NEED_RESCHED
-	jo	io_reschedule
+	jo	.Lio_reschedule
 	tm	__TI_flags+3(%r12),_TIF_SIGPENDING
-	jo	io_sigpending
+	jo	.Lio_sigpending
 	tm	__TI_flags+3(%r12),_TIF_NOTIFY_RESUME
-	jo	io_notify_resume
+	jo	.Lio_notify_resume
 	tm	__LC_CPU_FLAGS+3,_CIF_ASCE
-	jo	io_uaccess
-	j	io_return		# beware of critical section cleanup
+	jo	.Lio_uaccess
+	j	.Lio_return		# beware of critical section cleanup
 
 #
 # _CIF_MCCK_PENDING is set, call handler
 #
-io_mcck_pending:
-	# TRACE_IRQS_ON already done at io_return
-	l	%r1,BASED(.Lhandle_mcck)
+.Lio_mcck_pending:
+	# TRACE_IRQS_ON already done at .Lio_return
+	l	%r1,BASED(.Lc_handle_mcck)
 	basr	%r14,%r1		# TIF bit will be cleared by handler
 	TRACE_IRQS_OFF
-	j	io_return
+	j	.Lio_return
 
 #
 # _CIF_ASCE is set, load user space asce
 #
-io_uaccess:
+.Lio_uaccess:
 	ni	__LC_CPU_FLAGS+3,255-_CIF_ASCE
 	lctl	%c1,%c1,__LC_USER_ASCE	# load primary asce
-	j	io_return
+	j	.Lio_return
 
 #
 # _TIF_NEED_RESCHED is set, call schedule
 #
-io_reschedule:
-	# TRACE_IRQS_ON already done at io_return
-	l	%r1,BASED(.Lschedule)
+.Lio_reschedule:
+	# TRACE_IRQS_ON already done at .Lio_return
+	l	%r1,BASED(.Lc_schedule)
 	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
 	basr	%r14,%r1		# call scheduler
 	ssm	__LC_PGM_NEW_PSW	# disable I/O and ext. interrupts
 	TRACE_IRQS_OFF
-	j	io_return
+	j	.Lio_return
 
 #
 # _TIF_SIGPENDING is set, call do_signal
 #
-io_sigpending:
-	# TRACE_IRQS_ON already done at io_return
-	l	%r1,BASED(.Ldo_signal)
+.Lio_sigpending:
+	# TRACE_IRQS_ON already done at .Lio_return
+	l	%r1,BASED(.Lc_do_signal)
 	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
 	lr	%r2,%r11		# pass pointer to pt_regs
 	basr	%r14,%r1		# call do_signal
 	ssm	__LC_PGM_NEW_PSW	# disable I/O and ext. interrupts
 	TRACE_IRQS_OFF
-	j	io_return
+	j	.Lio_return
 
 #
 # _TIF_SIGPENDING is set, call do_signal
 #
-io_notify_resume:
-	# TRACE_IRQS_ON already done at io_return
-	l	%r1,BASED(.Ldo_notify_resume)
+.Lio_notify_resume:
+	# TRACE_IRQS_ON already done at .Lio_return
+	l	%r1,BASED(.Lc_do_notify_resume)
 	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
 	lr	%r2,%r11		# pass pointer to pt_regs
 	basr	%r14,%r1		# call do_notify_resume
 	ssm	__LC_PGM_NEW_PSW	# disable I/O and ext. interrupts
 	TRACE_IRQS_OFF
-	j	io_return
+	j	.Lio_return
 
 /*
  * External interrupt handler routine
@@ -603,9 +603,9 @@ ENTRY(ext_int_handler)
 	l	%r13,__LC_SVC_NEW_PSW+4
 	lm	%r8,%r9,__LC_EXT_OLD_PSW
 	tmh	%r8,0x0001		# interrupting from user ?
-	jz	ext_skip
+	jz	.Lext_skip
 	UPDATE_VTIME %r14,%r15,__LC_ASYNC_ENTER_TIMER
-ext_skip:
+.Lext_skip:
 	SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_STACK,STACK_SHIFT
 	stm	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(32,%r11),__LC_SAVE_AREA_ASYNC
@@ -614,29 +614,29 @@ ext_skip:
 	mvc	__PT_INT_PARM(4,%r11),__LC_EXT_PARAMS
 	xc	__PT_FLAGS(4,%r11),__PT_FLAGS(%r11)
 	TRACE_IRQS_OFF
-	l	%r1,BASED(.Ldo_IRQ)
+	l	%r1,BASED(.Lc_do_IRQ)
 	lr	%r2,%r11		# pass pointer to pt_regs
 	lhi	%r3,EXT_INTERRUPT
 	basr	%r14,%r1		# call do_IRQ
-	j	io_return
+	j	.Lio_return
 
 /*
- * Load idle PSW. The second "half" of this function is in cleanup_idle.
+ * Load idle PSW. The second "half" of this function is in .Lcleanup_idle.
  */
 ENTRY(psw_idle)
 	st	%r3,__SF_EMPTY(%r15)
 	basr	%r1,0
-	la	%r1,psw_idle_lpsw+4-.(%r1)
+	la	%r1,.Lpsw_idle_lpsw+4-.(%r1)
 	st	%r1,__SF_EMPTY+4(%r15)
 	oi	__SF_EMPTY+4(%r15),0x80
 	stck	__CLOCK_IDLE_ENTER(%r2)
 	stpt	__TIMER_IDLE_ENTER(%r2)
-psw_idle_lpsw:
+.Lpsw_idle_lpsw:
 	lpsw	__SF_EMPTY(%r15)
 	br	%r14
-psw_idle_end:
+.Lpsw_idle_end:
 
-__critical_end:
+.L__critical_end:
 
 /*
  * Machine check handler routines
@@ -650,7 +650,7 @@ ENTRY(mcck_int_handler)
 	l	%r13,__LC_SVC_NEW_PSW+4
 	lm	%r8,%r9,__LC_MCK_OLD_PSW
 	tm	__LC_MCCK_CODE,0x80	# system damage?
-	jo	mcck_panic		# yes -> rest of mcck code invalid
+	jo	.Lmcck_panic		# yes -> rest of mcck code invalid
 	la	%r14,__LC_CPU_TIMER_SAVE_AREA
 	mvc	__LC_MCCK_ENTER_TIMER(8),0(%r14)
 	tm	__LC_MCCK_CODE+5,0x02	# stored cpu timer value valid?
@@ -668,22 +668,22 @@ ENTRY(mcck_int_handler)
 2:	spt	0(%r14)
 	mvc	__LC_MCCK_ENTER_TIMER(8),0(%r14)
 3:	tm	__LC_MCCK_CODE+2,0x09	# mwp + ia of old psw valid?
-	jno	mcck_panic		# no -> skip cleanup critical
+	jno	.Lmcck_panic		# no -> skip cleanup critical
 	tm	%r8,0x0001		# interrupting from user ?
-	jz	mcck_skip
+	jz	.Lmcck_skip
 	UPDATE_VTIME %r14,%r15,__LC_MCCK_ENTER_TIMER
-mcck_skip:
+.Lmcck_skip:
 	SWITCH_ASYNC __LC_GPREGS_SAVE_AREA+32,__LC_PANIC_STACK,PAGE_SHIFT
 	stm	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(32,%r11),__LC_GPREGS_SAVE_AREA+32
 	stm	%r8,%r9,__PT_PSW(%r11)
 	xc	__PT_FLAGS(4,%r11),__PT_FLAGS(%r11)
 	xc	__SF_BACKCHAIN(4,%r15),__SF_BACKCHAIN(%r15)
-	l	%r1,BASED(.Ldo_machine_check)
+	l	%r1,BASED(.Lc_do_machine_check)
 	lr	%r2,%r11		# pass pointer to pt_regs
 	basr	%r14,%r1		# call s390_do_machine_check
 	tm	__PT_PSW+1(%r11),0x01	# returning to user ?
-	jno	mcck_return
+	jno	.Lmcck_return
 	l	%r1,__LC_KERNEL_STACK	# switch to kernel stack
 	mvc	STACK_FRAME_OVERHEAD(__PT_SIZE,%r1),0(%r11)
 	xc	__SF_BACKCHAIN(4,%r1),__SF_BACKCHAIN(%r1)
@@ -691,12 +691,12 @@ mcck_skip:
 	lr	%r15,%r1
 	ssm	__LC_PGM_NEW_PSW	# turn dat on, keep irqs off
 	tm	__LC_CPU_FLAGS+3,_CIF_MCCK_PENDING
-	jno	mcck_return
+	jno	.Lmcck_return
 	TRACE_IRQS_OFF
-	l	%r1,BASED(.Lhandle_mcck)
+	l	%r1,BASED(.Lc_handle_mcck)
 	basr	%r14,%r1		# call s390_handle_mcck
 	TRACE_IRQS_ON
-mcck_return:
+.Lmcck_return:
 	mvc	__LC_RETURN_MCCK_PSW(8),__PT_PSW(%r11) # move return PSW
 	tm	__LC_RETURN_MCCK_PSW+1,0x01 # returning to user ?
 	jno	0f
@@ -706,15 +706,15 @@ mcck_return:
 0:	lm	%r0,%r15,__PT_R0(%r11)
 	lpsw	__LC_RETURN_MCCK_PSW
 
-mcck_panic:
+.Lmcck_panic:
 	l	%r14,__LC_PANIC_STACK
 	slr	%r14,%r15
 	sra	%r14,PAGE_SHIFT
 	jz	0f
 	l	%r15,__LC_PANIC_STACK
-	j	mcck_skip
+	j	.Lmcck_skip
 0:	ahi	%r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
-	j	mcck_skip
+	j	.Lmcck_skip
 
 #
 # PSW restart interrupt handler
@@ -764,58 +764,58 @@ stack_overflow:
 1:	.long	kernel_stack_overflow
 #endif
 
-cleanup_table:
+.Lcleanup_table:
 	.long	system_call + 0x80000000
-	.long	sysc_do_svc + 0x80000000
-	.long	sysc_tif + 0x80000000
-	.long	sysc_restore + 0x80000000
-	.long	sysc_done + 0x80000000
-	.long	io_tif + 0x80000000
-	.long	io_restore + 0x80000000
-	.long	io_done + 0x80000000
+	.long	.Lsysc_do_svc + 0x80000000
+	.long	.Lsysc_tif + 0x80000000
+	.long	.Lsysc_restore + 0x80000000
+	.long	.Lsysc_done + 0x80000000
+	.long	.Lio_tif + 0x80000000
+	.long	.Lio_restore + 0x80000000
+	.long	.Lio_done + 0x80000000
 	.long	psw_idle + 0x80000000
-	.long	psw_idle_end + 0x80000000
+	.long	.Lpsw_idle_end + 0x80000000
 
 cleanup_critical:
-	cl	%r9,BASED(cleanup_table)	# system_call
+	cl	%r9,BASED(.Lcleanup_table)	# system_call
 	jl	0f
-	cl	%r9,BASED(cleanup_table+4)	# sysc_do_svc
-	jl	cleanup_system_call
-	cl	%r9,BASED(cleanup_table+8)	# sysc_tif
+	cl	%r9,BASED(.Lcleanup_table+4)	# .Lsysc_do_svc
+	jl	.Lcleanup_system_call
+	cl	%r9,BASED(.Lcleanup_table+8)	# .Lsysc_tif
 	jl	0f
-	cl	%r9,BASED(cleanup_table+12)	# sysc_restore
-	jl	cleanup_sysc_tif
-	cl	%r9,BASED(cleanup_table+16)	# sysc_done
-	jl	cleanup_sysc_restore
-	cl	%r9,BASED(cleanup_table+20)	# io_tif
+	cl	%r9,BASED(.Lcleanup_table+12)	# .Lsysc_restore
+	jl	.Lcleanup_sysc_tif
+	cl	%r9,BASED(.Lcleanup_table+16)	# .Lsysc_done
+	jl	.Lcleanup_sysc_restore
+	cl	%r9,BASED(.Lcleanup_table+20)	# .Lio_tif
 	jl	0f
-	cl	%r9,BASED(cleanup_table+24)	# io_restore
-	jl	cleanup_io_tif
-	cl	%r9,BASED(cleanup_table+28)	# io_done
-	jl	cleanup_io_restore
-	cl	%r9,BASED(cleanup_table+32)	# psw_idle
+	cl	%r9,BASED(.Lcleanup_table+24)	# .Lio_restore
+	jl	.Lcleanup_io_tif
+	cl	%r9,BASED(.Lcleanup_table+28)	# .Lio_done
+	jl	.Lcleanup_io_restore
+	cl	%r9,BASED(.Lcleanup_table+32)	# psw_idle
 	jl	0f
-	cl	%r9,BASED(cleanup_table+36)	# psw_idle_end
-	jl	cleanup_idle
+	cl	%r9,BASED(.Lcleanup_table+36)	# .Lpsw_idle_end
+	jl	.Lcleanup_idle
 0:	br	%r14
 
-cleanup_system_call:
+.Lcleanup_system_call:
 	# check if stpt has been executed
-	cl	%r9,BASED(cleanup_system_call_insn)
+	cl	%r9,BASED(.Lcleanup_system_call_insn)
 	jh	0f
 	mvc	__LC_SYNC_ENTER_TIMER(8),__LC_ASYNC_ENTER_TIMER
 	chi	%r11,__LC_SAVE_AREA_ASYNC
 	je	0f
 	mvc	__LC_SYNC_ENTER_TIMER(8),__LC_MCCK_ENTER_TIMER
 0:	# check if stm has been executed
-	cl	%r9,BASED(cleanup_system_call_insn+4)
+	cl	%r9,BASED(.Lcleanup_system_call_insn+4)
 	jh	0f
 	mvc	__LC_SAVE_AREA_SYNC(32),0(%r11)
 0:	# set up saved registers r12, and r13
 	st	%r12,16(%r11)		# r12 thread-info pointer
 	st	%r13,20(%r11)		# r13 literal-pool pointer
 	# check if the user time calculation has been done
-	cl	%r9,BASED(cleanup_system_call_insn+8)
+	cl	%r9,BASED(.Lcleanup_system_call_insn+8)
 	jh	0f
 	l	%r10,__LC_EXIT_TIMER
 	l	%r15,__LC_EXIT_TIMER+4
@@ -824,7 +824,7 @@ cleanup_system_call:
 	st	%r10,__LC_USER_TIMER
 	st	%r15,__LC_USER_TIMER+4
 0:	# check if the system time calculation has been done
-	cl	%r9,BASED(cleanup_system_call_insn+12)
+	cl	%r9,BASED(.Lcleanup_system_call_insn+12)
 	jh	0f
 	l	%r10,__LC_LAST_UPDATE_TIMER
 	l	%r15,__LC_LAST_UPDATE_TIMER+4
@@ -848,20 +848,20 @@ cleanup_system_call:
 	# setup saved register 15
 	st	%r15,28(%r11)		# r15 stack pointer
 	# set new psw address and exit
-	l	%r9,BASED(cleanup_table+4)	# sysc_do_svc + 0x80000000
+	l	%r9,BASED(.Lcleanup_table+4)	# .Lsysc_do_svc + 0x80000000
 	br	%r14
-cleanup_system_call_insn:
+.Lcleanup_system_call_insn:
 	.long	system_call + 0x80000000
-	.long	sysc_stm + 0x80000000
-	.long	sysc_vtime + 0x80000000 + 36
-	.long	sysc_vtime + 0x80000000 + 76
+	.long	.Lsysc_stm + 0x80000000
+	.long	.Lsysc_vtime + 0x80000000 + 36
+	.long	.Lsysc_vtime + 0x80000000 + 76
 
-cleanup_sysc_tif:
-	l	%r9,BASED(cleanup_table+8)	# sysc_tif + 0x80000000
+.Lcleanup_sysc_tif:
+	l	%r9,BASED(.Lcleanup_table+8)	# .Lsysc_tif + 0x80000000
 	br	%r14
 
-cleanup_sysc_restore:
-	cl	%r9,BASED(cleanup_sysc_restore_insn)
+.Lcleanup_sysc_restore:
+	cl	%r9,BASED(.Lcleanup_sysc_restore_insn)
 	jhe	0f
 	l	%r9,12(%r11)		# get saved pointer to pt_regs
 	mvc	__LC_RETURN_PSW(8),__PT_PSW(%r9)
@@ -869,15 +869,15 @@ cleanup_sysc_restore:
 	lm	%r0,%r7,__PT_R0(%r9)
 0:	lm	%r8,%r9,__LC_RETURN_PSW
 	br	%r14
-cleanup_sysc_restore_insn:
-	.long	sysc_done - 4 + 0x80000000
+.Lcleanup_sysc_restore_insn:
+	.long	.Lsysc_done - 4 + 0x80000000
 
-cleanup_io_tif:
-	l	%r9,BASED(cleanup_table+20)	# io_tif + 0x80000000
+.Lcleanup_io_tif:
+	l	%r9,BASED(.Lcleanup_table+20)	# .Lio_tif + 0x80000000
 	br	%r14
 
-cleanup_io_restore:
-	cl	%r9,BASED(cleanup_io_restore_insn)
+.Lcleanup_io_restore:
+	cl	%r9,BASED(.Lcleanup_io_restore_insn)
 	jhe	0f
 	l	%r9,12(%r11)		# get saved r11 pointer to pt_regs
 	mvc	__LC_RETURN_PSW(8),__PT_PSW(%r9)
@@ -885,10 +885,10 @@ cleanup_io_restore:
 	lm	%r0,%r7,__PT_R0(%r9)
 0:	lm	%r8,%r9,__LC_RETURN_PSW
 	br	%r14
-cleanup_io_restore_insn:
-	.long	io_done - 4 + 0x80000000
+.Lcleanup_io_restore_insn:
+	.long	.Lio_done - 4 + 0x80000000
 
-cleanup_idle:
+.Lcleanup_idle:
 	# copy interrupt clock & cpu timer
 	mvc	__CLOCK_IDLE_EXIT(8,%r2),__LC_INT_CLOCK
 	mvc	__TIMER_IDLE_EXIT(8,%r2),__LC_ASYNC_ENTER_TIMER
@@ -897,7 +897,7 @@ cleanup_idle:
 	mvc	__CLOCK_IDLE_EXIT(8,%r2),__LC_MCCK_CLOCK
 	mvc	__TIMER_IDLE_EXIT(8,%r2),__LC_MCCK_ENTER_TIMER
 0:	# check if stck has been executed
-	cl	%r9,BASED(cleanup_idle_insn)
+	cl	%r9,BASED(.Lcleanup_idle_insn)
 	jhe	1f
 	mvc	__CLOCK_IDLE_ENTER(8,%r2),__CLOCK_IDLE_EXIT(%r2)
 	mvc	__TIMER_IDLE_ENTER(8,%r2),__TIMER_IDLE_EXIT(%r3)
@@ -913,12 +913,12 @@ cleanup_idle:
 	stm	%r9,%r10,__LC_SYSTEM_TIMER
 	mvc	__LC_LAST_UPDATE_TIMER(8),__TIMER_IDLE_EXIT(%r2)
 	# prepare return psw
-	n	%r8,BASED(cleanup_idle_wait)	# clear irq & wait state bits
+	n	%r8,BASED(.Lcleanup_idle_wait)	# clear irq & wait state bits
 	l	%r9,24(%r11)			# return from psw_idle
 	br	%r14
-cleanup_idle_insn:
-	.long	psw_idle_lpsw + 0x80000000
-cleanup_idle_wait:
+.Lcleanup_idle_insn:
+	.long	.Lpsw_idle_lpsw + 0x80000000
+.Lcleanup_idle_wait:
 	.long	0xfcfdffff
 
 /*
@@ -933,30 +933,30 @@ cleanup_idle_wait:
 /*
  * Symbol constants
  */
-.Ldo_machine_check:	.long	s390_do_machine_check
-.Lhandle_mcck:		.long	s390_handle_mcck
-.Ldo_IRQ:		.long	do_IRQ
-.Ldo_signal:		.long	do_signal
-.Ldo_notify_resume:	.long	do_notify_resume
-.Ldo_per_trap:		.long	do_per_trap
-.Ljump_table:		.long	pgm_check_table
-.Lschedule:		.long	schedule
+.Lc_do_machine_check:	.long	s390_do_machine_check
+.Lc_handle_mcck:	.long	s390_handle_mcck
+.Lc_do_IRQ:		.long	do_IRQ
+.Lc_do_signal:		.long	do_signal
+.Lc_do_notify_resume:	.long	do_notify_resume
+.Lc_do_per_trap:	.long	do_per_trap
+.Lc_jump_table:		.long	pgm_check_table
+.Lc_schedule:		.long	schedule
 #ifdef CONFIG_PREEMPT
-.Lpreempt_irq:		.long	preempt_schedule_irq
+.Lc_preempt_irq:	.long	preempt_schedule_irq
 #endif
-.Ltrace_enter:		.long	do_syscall_trace_enter
-.Ltrace_exit:		.long	do_syscall_trace_exit
-.Lschedule_tail:	.long	schedule_tail
-.Lsysc_per:		.long	sysc_per + 0x80000000
+.Lc_trace_enter:	.long	do_syscall_trace_enter
+.Lc_trace_exit:		.long	do_syscall_trace_exit
+.Lc_schedule_tail:	.long	schedule_tail
+.Lc_sysc_per:		.long	.Lsysc_per + 0x80000000
 #ifdef CONFIG_TRACE_IRQFLAGS
-.Lhardirqs_on:		.long	trace_hardirqs_on_caller
-.Lhardirqs_off:		.long	trace_hardirqs_off_caller
+.Lc_hardirqs_on:	.long	trace_hardirqs_on_caller
+.Lc_hardirqs_off:	.long	trace_hardirqs_off_caller
 #endif
 #ifdef CONFIG_LOCKDEP
-.Llockdep_sys_exit:	.long	lockdep_sys_exit
+.Lc_lockdep_sys_exit:	.long	lockdep_sys_exit
 #endif
-.Lcritical_start:	.long	__critical_start + 0x80000000
-.Lcritical_length:	.long	__critical_end - __critical_start
+.Lc_critical_start:	.long	.L__critical_start + 0x80000000
+.Lc_critical_length:	.long	.L__critical_end - .L__critical_start
 
 		.section .rodata, "a"
 #define SYSCALL(esa,esame,emu)	.long esa
diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h
index 0554b9771c9f..8e61393c8275 100644
--- a/arch/s390/kernel/entry.h
+++ b/arch/s390/kernel/entry.h
@@ -74,4 +74,6 @@ struct old_sigaction;
 long sys_s390_personality(unsigned int personality);
 long sys_s390_runtime_instr(int command, int signum);
 
+long sys_s390_pci_mmio_write(unsigned long, const void __user *, size_t);
+long sys_s390_pci_mmio_read(unsigned long, void __user *, size_t);
 #endif /* _ENTRY_H */
diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S
index 7b2e03afd017..c329446a951d 100644
--- a/arch/s390/kernel/entry64.S
+++ b/arch/s390/kernel/entry64.S
@@ -91,7 +91,7 @@ _PIF_WORK	= (_PIF_PER_TRAP)
 	.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 rewind_pad
+	# do_protection_exception will rewind to .Lrewind_pad
 	jh	.+42
 	.else
 	jhe	.+42
@@ -192,7 +192,7 @@ ENTRY(__switch_to)
 	lmg	%r6,%r15,__SF_GPRS(%r15)	# load gprs of next task
 	br	%r14
 
-__critical_start:
+.L__critical_start:
 /*
  * SVC interrupt handler routine. System calls are synchronous events and
  * are executed with interrupts enabled.
@@ -200,15 +200,15 @@ __critical_start:
 
 ENTRY(system_call)
 	stpt	__LC_SYNC_ENTER_TIMER
-sysc_stmg:
+.Lsysc_stmg:
 	stmg	%r8,%r15,__LC_SAVE_AREA_SYNC
 	lg	%r10,__LC_LAST_BREAK
 	lg	%r12,__LC_THREAD_INFO
 	lghi	%r14,_PIF_SYSCALL
-sysc_per:
+.Lsysc_per:
 	lg	%r15,__LC_KERNEL_STACK
 	la	%r11,STACK_FRAME_OVERHEAD(%r15)	# pointer to pt_regs
-sysc_vtime:
+.Lsysc_vtime:
 	UPDATE_VTIME %r13,__LC_SYNC_ENTER_TIMER
 	LAST_BREAK %r13
 	stmg	%r0,%r7,__PT_R0(%r11)
@@ -216,39 +216,39 @@ sysc_vtime:
 	mvc	__PT_PSW(16,%r11),__LC_SVC_OLD_PSW
 	mvc	__PT_INT_CODE(4,%r11),__LC_SVC_ILC
 	stg	%r14,__PT_FLAGS(%r11)
-sysc_do_svc:
+.Lsysc_do_svc:
 	lg	%r10,__TI_sysc_table(%r12)	# address of system call table
 	llgh	%r8,__PT_INT_CODE+2(%r11)
 	slag	%r8,%r8,2			# shift and test for svc 0
-	jnz	sysc_nr_ok
+	jnz	.Lsysc_nr_ok
 	# svc 0: system call number in %r1
 	llgfr	%r1,%r1				# clear high word in r1
 	cghi	%r1,NR_syscalls
-	jnl	sysc_nr_ok
+	jnl	.Lsysc_nr_ok
 	sth	%r1,__PT_INT_CODE+2(%r11)
 	slag	%r8,%r1,2
-sysc_nr_ok:
+.Lsysc_nr_ok:
 	xc	__SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
 	stg	%r2,__PT_ORIG_GPR2(%r11)
 	stg	%r7,STACK_FRAME_OVERHEAD(%r15)
 	lgf	%r9,0(%r8,%r10)			# get system call add.
 	tm	__TI_flags+7(%r12),_TIF_TRACE
-	jnz	sysc_tracesys
+	jnz	.Lsysc_tracesys
 	basr	%r14,%r9			# call sys_xxxx
 	stg	%r2,__PT_R2(%r11)		# store return value
 
-sysc_return:
+.Lsysc_return:
 	LOCKDEP_SYS_EXIT
-sysc_tif:
+.Lsysc_tif:
 	tm	__PT_PSW+1(%r11),0x01		# returning to user ?
-	jno	sysc_restore
+	jno	.Lsysc_restore
 	tm	__PT_FLAGS+7(%r11),_PIF_WORK
-	jnz	sysc_work
+	jnz	.Lsysc_work
 	tm	__TI_flags+7(%r12),_TIF_WORK
-	jnz	sysc_work			# check for work
+	jnz	.Lsysc_work			# check for work
 	tm	__LC_CPU_FLAGS+7,_CIF_WORK
-	jnz	sysc_work
-sysc_restore:
+	jnz	.Lsysc_work
+.Lsysc_restore:
 	lg	%r14,__LC_VDSO_PER_CPU
 	lmg	%r0,%r10,__PT_R0(%r11)
 	mvc	__LC_RETURN_PSW(16),__PT_PSW(%r11)
@@ -256,101 +256,101 @@ sysc_restore:
 	mvc	__VDSO_ECTG_BASE(16,%r14),__LC_EXIT_TIMER
 	lmg	%r11,%r15,__PT_R11(%r11)
 	lpswe	__LC_RETURN_PSW
-sysc_done:
+.Lsysc_done:
 
 #
 # One of the work bits is on. Find out which one.
 #
-sysc_work:
+.Lsysc_work:
 	tm	__LC_CPU_FLAGS+7,_CIF_MCCK_PENDING
-	jo	sysc_mcck_pending
+	jo	.Lsysc_mcck_pending
 	tm	__TI_flags+7(%r12),_TIF_NEED_RESCHED
-	jo	sysc_reschedule
+	jo	.Lsysc_reschedule
 #ifdef CONFIG_UPROBES
 	tm	__TI_flags+7(%r12),_TIF_UPROBE
-	jo	sysc_uprobe_notify
+	jo	.Lsysc_uprobe_notify
 #endif
 	tm	__PT_FLAGS+7(%r11),_PIF_PER_TRAP
-	jo	sysc_singlestep
+	jo	.Lsysc_singlestep
 	tm	__TI_flags+7(%r12),_TIF_SIGPENDING
-	jo	sysc_sigpending
+	jo	.Lsysc_sigpending
 	tm	__TI_flags+7(%r12),_TIF_NOTIFY_RESUME
-	jo	sysc_notify_resume
+	jo	.Lsysc_notify_resume
 	tm	__LC_CPU_FLAGS+7,_CIF_ASCE
-	jo	sysc_uaccess
-	j	sysc_return		# beware of critical section cleanup
+	jo	.Lsysc_uaccess
+	j	.Lsysc_return		# beware of critical section cleanup
 
 #
 # _TIF_NEED_RESCHED is set, call schedule
 #
-sysc_reschedule:
-	larl	%r14,sysc_return
+.Lsysc_reschedule:
+	larl	%r14,.Lsysc_return
 	jg	schedule
 
 #
 # _CIF_MCCK_PENDING is set, call handler
 #
-sysc_mcck_pending:
-	larl	%r14,sysc_return
+.Lsysc_mcck_pending:
+	larl	%r14,.Lsysc_return
 	jg	s390_handle_mcck	# TIF bit will be cleared by handler
 
 #
 # _CIF_ASCE is set, load user space asce
 #
-sysc_uaccess:
+.Lsysc_uaccess:
 	ni	__LC_CPU_FLAGS+7,255-_CIF_ASCE
 	lctlg	%c1,%c1,__LC_USER_ASCE		# load primary asce
-	j	sysc_return
+	j	.Lsysc_return
 
 #
 # _TIF_SIGPENDING is set, call do_signal
 #
-sysc_sigpending:
+.Lsysc_sigpending:
 	lgr	%r2,%r11		# pass pointer to pt_regs
 	brasl	%r14,do_signal
 	tm	__PT_FLAGS+7(%r11),_PIF_SYSCALL
-	jno	sysc_return
+	jno	.Lsysc_return
 	lmg	%r2,%r7,__PT_R2(%r11)	# load svc arguments
 	lg	%r10,__TI_sysc_table(%r12)	# address of system call table
 	lghi	%r8,0			# svc 0 returns -ENOSYS
 	llgh	%r1,__PT_INT_CODE+2(%r11)	# load new svc number
 	cghi	%r1,NR_syscalls
-	jnl	sysc_nr_ok		# invalid svc number -> do svc 0
+	jnl	.Lsysc_nr_ok		# invalid svc number -> do svc 0
 	slag	%r8,%r1,2
-	j	sysc_nr_ok		# restart svc
+	j	.Lsysc_nr_ok		# restart svc
 
 #
 # _TIF_NOTIFY_RESUME is set, call do_notify_resume
 #
-sysc_notify_resume:
+.Lsysc_notify_resume:
 	lgr	%r2,%r11		# pass pointer to pt_regs
-	larl	%r14,sysc_return
+	larl	%r14,.Lsysc_return
 	jg	do_notify_resume
 
 #
 # _TIF_UPROBE is set, call uprobe_notify_resume
 #
 #ifdef CONFIG_UPROBES
-sysc_uprobe_notify:
+.Lsysc_uprobe_notify:
 	lgr	%r2,%r11		# pass pointer to pt_regs
-	larl	%r14,sysc_return
+	larl	%r14,.Lsysc_return
 	jg	uprobe_notify_resume
 #endif
 
 #
 # _PIF_PER_TRAP is set, call do_per_trap
 #
-sysc_singlestep:
+.Lsysc_singlestep:
 	ni	__PT_FLAGS+7(%r11),255-_PIF_PER_TRAP
 	lgr	%r2,%r11		# pass pointer to pt_regs
-	larl	%r14,sysc_return
+	larl	%r14,.Lsysc_return
 	jg	do_per_trap
 
 #
 # call tracehook_report_syscall_entry/tracehook_report_syscall_exit before
 # and after the system call
 #
-sysc_tracesys:
+.Lsysc_tracesys:
 	lgr	%r2,%r11		# pass pointer to pt_regs
 	la	%r3,0
 	llgh	%r0,__PT_INT_CODE+2(%r11)
@@ -358,20 +358,20 @@ sysc_tracesys:
 	brasl	%r14,do_syscall_trace_enter
 	lghi	%r0,NR_syscalls
 	clgr	%r0,%r2
-	jnh	sysc_tracenogo
+	jnh	.Lsysc_tracenogo
 	sllg	%r8,%r2,2
 	lgf	%r9,0(%r8,%r10)
-sysc_tracego:
+.Lsysc_tracego:
 	lmg	%r3,%r7,__PT_R3(%r11)
 	stg	%r7,STACK_FRAME_OVERHEAD(%r15)
 	lg	%r2,__PT_ORIG_GPR2(%r11)
 	basr	%r14,%r9		# call sys_xxx
 	stg	%r2,__PT_R2(%r11)	# store return value
-sysc_tracenogo:
+.Lsysc_tracenogo:
 	tm	__TI_flags+7(%r12),_TIF_TRACE
-	jz	sysc_return
+	jz	.Lsysc_return
 	lgr	%r2,%r11		# pass pointer to pt_regs
-	larl	%r14,sysc_return
+	larl	%r14,.Lsysc_return
 	jg	do_syscall_trace_exit
 
 #
@@ -384,13 +384,13 @@ ENTRY(ret_from_fork)
 	TRACE_IRQS_ON
 	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
 	tm	__PT_PSW+1(%r11),0x01	# forking a kernel thread ?
-	jne	sysc_tracenogo
+	jne	.Lsysc_tracenogo
 	# it's a kernel thread
 	lmg	%r9,%r10,__PT_R9(%r11)	# load gprs
 ENTRY(kernel_thread_starter)
 	la	%r2,0(%r10)
 	basr	%r14,%r9
-	j	sysc_tracenogo
+	j	.Lsysc_tracenogo
 
 /*
  * Program check handler routine
@@ -409,7 +409,7 @@ ENTRY(pgm_check_handler)
 	tmhh	%r8,0x4000		# PER bit set in old PSW ?
 	jnz	0f			# -> enabled, can't be a double fault
 	tm	__LC_PGM_ILC+3,0x80	# check for per exception
-	jnz	pgm_svcper		# -> single stepped svc
+	jnz	.Lpgm_svcper		# -> single stepped svc
 0:	CHECK_STACK STACK_SIZE,__LC_SAVE_AREA_SYNC
 	aghi	%r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
 	j	2f
@@ -432,7 +432,7 @@ ENTRY(pgm_check_handler)
 	tm	__LC_PGM_ILC+3,0x80	# check for per exception
 	jz	0f
 	tmhh	%r8,0x0001		# kernel per event ?
-	jz	pgm_kprobe
+	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
@@ -443,31 +443,31 @@ ENTRY(pgm_check_handler)
 	llgh	%r10,__PT_INT_CODE+2(%r11)
 	nill	%r10,0x007f
 	sll	%r10,2
-	je	sysc_return
+	je	.Lsysc_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	sysc_return
+	j	.Lsysc_return
 
 #
 # PER event in supervisor state, must be kprobes
 #
-pgm_kprobe:
+.Lpgm_kprobe:
 	REENABLE_IRQS
 	xc	__SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
 	lgr	%r2,%r11		# pass pointer to pt_regs
 	brasl	%r14,do_per_trap
-	j	sysc_return
+	j	.Lsysc_return
 
 #
 # single stepped system call
 #
-pgm_svcper:
+.Lpgm_svcper:
 	mvc	__LC_RETURN_PSW(8),__LC_SVC_NEW_PSW
-	larl	%r14,sysc_per
+	larl	%r14,.Lsysc_per
 	stg	%r14,__LC_RETURN_PSW+8
 	lghi	%r14,_PIF_SYSCALL | _PIF_PER_TRAP
-	lpswe	__LC_RETURN_PSW		# branch to sysc_per and enable irqs
+	lpswe	__LC_RETURN_PSW		# branch to .Lsysc_per and enable irqs
 
 /*
  * IO interrupt handler routine
@@ -483,10 +483,10 @@ ENTRY(io_int_handler)
 	HANDLE_SIE_INTERCEPT %r14,2
 	SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_STACK,STACK_SHIFT
 	tmhh	%r8,0x0001		# interrupting from user?
-	jz	io_skip
+	jz	.Lio_skip
 	UPDATE_VTIME %r14,__LC_ASYNC_ENTER_TIMER
 	LAST_BREAK %r14
-io_skip:
+.Lio_skip:
 	stmg	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC
 	stmg	%r8,%r9,__PT_PSW(%r11)
@@ -494,29 +494,29 @@ io_skip:
 	xc	__PT_FLAGS(8,%r11),__PT_FLAGS(%r11)
 	TRACE_IRQS_OFF
 	xc	__SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
-io_loop:
+.Lio_loop:
 	lgr	%r2,%r11		# pass pointer to pt_regs
 	lghi	%r3,IO_INTERRUPT
 	tm	__PT_INT_CODE+8(%r11),0x80	# adapter interrupt ?
-	jz	io_call
+	jz	.Lio_call
 	lghi	%r3,THIN_INTERRUPT
-io_call:
+.Lio_call:
 	brasl	%r14,do_IRQ
 	tm	__LC_MACHINE_FLAGS+6,0x10	# MACHINE_FLAG_LPAR
-	jz	io_return
+	jz	.Lio_return
 	tpi	0
-	jz	io_return
+	jz	.Lio_return
 	mvc	__PT_INT_CODE(12,%r11),__LC_SUBCHANNEL_ID
-	j	io_loop
-io_return:
+	j	.Lio_loop
+.Lio_return:
 	LOCKDEP_SYS_EXIT
 	TRACE_IRQS_ON
-io_tif:
+.Lio_tif:
 	tm	__TI_flags+7(%r12),_TIF_WORK
-	jnz	io_work 		# there is work to do (signals etc.)
+	jnz	.Lio_work		# there is work to do (signals etc.)
 	tm	__LC_CPU_FLAGS+7,_CIF_WORK
-	jnz	io_work
-io_restore:
+	jnz	.Lio_work
+.Lio_restore:
 	lg	%r14,__LC_VDSO_PER_CPU
 	lmg	%r0,%r10,__PT_R0(%r11)
 	mvc	__LC_RETURN_PSW(16),__PT_PSW(%r11)
@@ -524,7 +524,7 @@ io_restore:
 	mvc	__VDSO_ECTG_BASE(16,%r14),__LC_EXIT_TIMER
 	lmg	%r11,%r15,__PT_R11(%r11)
 	lpswe	__LC_RETURN_PSW
-io_done:
+.Lio_done:
 
 #
 # There is work todo, find out in which context we have been interrupted:
@@ -535,15 +535,15 @@ io_done:
 #    the preemption counter and if it is zero call preempt_schedule_irq
 # Before any work can be done, a switch to the kernel stack is required.
 #
-io_work:
+.Lio_work:
 	tm	__PT_PSW+1(%r11),0x01	# returning to user ?
-	jo	io_work_user		# yes -> do resched & signal
+	jo	.Lio_work_user		# yes -> do resched & signal
 #ifdef CONFIG_PREEMPT
 	# check for preemptive scheduling
 	icm	%r0,15,__TI_precount(%r12)
-	jnz	io_restore		# preemption is disabled
+	jnz	.Lio_restore		# preemption is disabled
 	tm	__TI_flags+7(%r12),_TIF_NEED_RESCHED
-	jno	io_restore
+	jno	.Lio_restore
 	# switch to kernel stack
 	lg	%r1,__PT_R15(%r11)
 	aghi	%r1,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
@@ -551,19 +551,19 @@ io_work:
 	xc	__SF_BACKCHAIN(8,%r1),__SF_BACKCHAIN(%r1)
 	la	%r11,STACK_FRAME_OVERHEAD(%r1)
 	lgr	%r15,%r1
-	# TRACE_IRQS_ON already done at io_return, call
+	# TRACE_IRQS_ON already done at .Lio_return, call
 	# TRACE_IRQS_OFF to keep things symmetrical
 	TRACE_IRQS_OFF
 	brasl	%r14,preempt_schedule_irq
-	j	io_return
+	j	.Lio_return
 #else
-	j	io_restore
+	j	.Lio_restore
 #endif
 
 #
 # Need to do work before returning to userspace, switch to kernel stack
 #
-io_work_user:
+.Lio_work_user:
 	lg	%r1,__LC_KERNEL_STACK
 	mvc	STACK_FRAME_OVERHEAD(__PT_SIZE,%r1),0(%r11)
 	xc	__SF_BACKCHAIN(8,%r1),__SF_BACKCHAIN(%r1)
@@ -573,70 +573,70 @@ io_work_user:
 #
 # One of the work bits is on. Find out which one.
 #
-io_work_tif:
+.Lio_work_tif:
 	tm	__LC_CPU_FLAGS+7,_CIF_MCCK_PENDING
-	jo	io_mcck_pending
+	jo	.Lio_mcck_pending
 	tm	__TI_flags+7(%r12),_TIF_NEED_RESCHED
-	jo	io_reschedule
+	jo	.Lio_reschedule
 	tm	__TI_flags+7(%r12),_TIF_SIGPENDING
-	jo	io_sigpending
+	jo	.Lio_sigpending
 	tm	__TI_flags+7(%r12),_TIF_NOTIFY_RESUME
-	jo	io_notify_resume
+	jo	.Lio_notify_resume
 	tm	__LC_CPU_FLAGS+7,_CIF_ASCE
-	jo	io_uaccess
-	j	io_return		# beware of critical section cleanup
+	jo	.Lio_uaccess
+	j	.Lio_return		# beware of critical section cleanup
 
 #
 # _CIF_MCCK_PENDING is set, call handler
 #
-io_mcck_pending:
-	# TRACE_IRQS_ON already done at io_return
+.Lio_mcck_pending:
+	# TRACE_IRQS_ON already done at .Lio_return
 	brasl	%r14,s390_handle_mcck	# TIF bit will be cleared by handler
 	TRACE_IRQS_OFF
-	j	io_return
+	j	.Lio_return
 
 #
 # _CIF_ASCE is set, load user space asce
 #
-io_uaccess:
+.Lio_uaccess:
 	ni	__LC_CPU_FLAGS+7,255-_CIF_ASCE
 	lctlg	%c1,%c1,__LC_USER_ASCE		# load primary asce
-	j	io_return
+	j	.Lio_return
 
 #
 # _TIF_NEED_RESCHED is set, call schedule
 #
-io_reschedule:
-	# TRACE_IRQS_ON already done at io_return
+.Lio_reschedule:
+	# TRACE_IRQS_ON already done at .Lio_return
 	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
 	brasl	%r14,schedule		# call scheduler
 	ssm	__LC_PGM_NEW_PSW	# disable I/O and ext. interrupts
 	TRACE_IRQS_OFF
-	j	io_return
+	j	.Lio_return
 
 #
 # _TIF_SIGPENDING or is set, call do_signal
 #
-io_sigpending:
-	# TRACE_IRQS_ON already done at io_return
+.Lio_sigpending:
+	# TRACE_IRQS_ON already done at .Lio_return
 	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
 	lgr	%r2,%r11		# pass pointer to pt_regs
 	brasl	%r14,do_signal
 	ssm	__LC_PGM_NEW_PSW	# disable I/O and ext. interrupts
 	TRACE_IRQS_OFF
-	j	io_return
+	j	.Lio_return
 
 #
 # _TIF_NOTIFY_RESUME or is set, call do_notify_resume
 #
-io_notify_resume:
-	# TRACE_IRQS_ON already done at io_return
+.Lio_notify_resume:
+	# TRACE_IRQS_ON already done at .Lio_return
 	ssm	__LC_SVC_NEW_PSW	# reenable interrupts
 	lgr	%r2,%r11		# pass pointer to pt_regs
 	brasl	%r14,do_notify_resume
 	ssm	__LC_PGM_NEW_PSW	# disable I/O and ext. interrupts
 	TRACE_IRQS_OFF
-	j	io_return
+	j	.Lio_return
 
 /*
  * External interrupt handler routine
@@ -652,10 +652,10 @@ ENTRY(ext_int_handler)
 	HANDLE_SIE_INTERCEPT %r14,3
 	SWITCH_ASYNC __LC_SAVE_AREA_ASYNC,__LC_ASYNC_STACK,STACK_SHIFT
 	tmhh	%r8,0x0001		# interrupting from user ?
-	jz	ext_skip
+	jz	.Lext_skip
 	UPDATE_VTIME %r14,__LC_ASYNC_ENTER_TIMER
 	LAST_BREAK %r14
-ext_skip:
+.Lext_skip:
 	stmg	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(64,%r11),__LC_SAVE_AREA_ASYNC
 	stmg	%r8,%r9,__PT_PSW(%r11)
@@ -669,23 +669,23 @@ ext_skip:
 	lgr	%r2,%r11		# pass pointer to pt_regs
 	lghi	%r3,EXT_INTERRUPT
 	brasl	%r14,do_IRQ
-	j	io_return
+	j	.Lio_return
 
 /*
- * Load idle PSW. The second "half" of this function is in cleanup_idle.
+ * Load idle PSW. The second "half" of this function is in .Lcleanup_idle.
  */
 ENTRY(psw_idle)
 	stg	%r3,__SF_EMPTY(%r15)
-	larl	%r1,psw_idle_lpsw+4
+	larl	%r1,.Lpsw_idle_lpsw+4
 	stg	%r1,__SF_EMPTY+8(%r15)
 	STCK	__CLOCK_IDLE_ENTER(%r2)
 	stpt	__TIMER_IDLE_ENTER(%r2)
-psw_idle_lpsw:
+.Lpsw_idle_lpsw:
 	lpswe	__SF_EMPTY(%r15)
 	br	%r14
-psw_idle_end:
+.Lpsw_idle_end:
 
-__critical_end:
+.L__critical_end:
 
 /*
  * Machine check handler routines
@@ -701,7 +701,7 @@ ENTRY(mcck_int_handler)
 	lmg	%r8,%r9,__LC_MCK_OLD_PSW
 	HANDLE_SIE_INTERCEPT %r14,4
 	tm	__LC_MCCK_CODE,0x80	# system damage?
-	jo	mcck_panic		# yes -> rest of mcck code invalid
+	jo	.Lmcck_panic		# yes -> rest of mcck code invalid
 	lghi	%r14,__LC_CPU_TIMER_SAVE_AREA
 	mvc	__LC_MCCK_ENTER_TIMER(8),0(%r14)
 	tm	__LC_MCCK_CODE+5,0x02	# stored cpu timer value valid?
@@ -719,13 +719,13 @@ ENTRY(mcck_int_handler)
 2:	spt	0(%r14)
 	mvc	__LC_MCCK_ENTER_TIMER(8),0(%r14)
 3:	tm	__LC_MCCK_CODE+2,0x09	# mwp + ia of old psw valid?
-	jno	mcck_panic		# no -> skip cleanup critical
+	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	mcck_skip
+	jz	.Lmcck_skip
 	UPDATE_VTIME %r14,__LC_MCCK_ENTER_TIMER
 	LAST_BREAK %r14
-mcck_skip:
+.Lmcck_skip:
 	lghi	%r14,__LC_GPREGS_SAVE_AREA+64
 	stmg	%r0,%r7,__PT_R0(%r11)
 	mvc	__PT_R8(64,%r11),0(%r14)
@@ -735,7 +735,7 @@ mcck_skip:
 	lgr	%r2,%r11		# pass pointer to pt_regs
 	brasl	%r14,s390_do_machine_check
 	tm	__PT_PSW+1(%r11),0x01	# returning to user ?
-	jno	mcck_return
+	jno	.Lmcck_return
 	lg	%r1,__LC_KERNEL_STACK	# switch to kernel stack
 	mvc	STACK_FRAME_OVERHEAD(__PT_SIZE,%r1),0(%r11)
 	xc	__SF_BACKCHAIN(8,%r1),__SF_BACKCHAIN(%r1)
@@ -743,11 +743,11 @@ mcck_skip:
 	lgr	%r15,%r1
 	ssm	__LC_PGM_NEW_PSW	# turn dat on, keep irqs off
 	tm	__LC_CPU_FLAGS+7,_CIF_MCCK_PENDING
-	jno	mcck_return
+	jno	.Lmcck_return
 	TRACE_IRQS_OFF
 	brasl	%r14,s390_handle_mcck
 	TRACE_IRQS_ON
-mcck_return:
+.Lmcck_return:
 	lg	%r14,__LC_VDSO_PER_CPU
 	lmg	%r0,%r10,__PT_R0(%r11)
 	mvc	__LC_RETURN_MCCK_PSW(16),__PT_PSW(%r11) # move return PSW
@@ -758,14 +758,14 @@ mcck_return:
 0:	lmg	%r11,%r15,__PT_R11(%r11)
 	lpswe	__LC_RETURN_MCCK_PSW
 
-mcck_panic:
+.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)
-	j	mcck_skip
+	j	.Lmcck_skip
 
 #
 # PSW restart interrupt handler
@@ -815,69 +815,69 @@ stack_overflow:
 #endif
 
 	.align	8
-cleanup_table:
+.Lcleanup_table:
 	.quad	system_call
-	.quad	sysc_do_svc
-	.quad	sysc_tif
-	.quad	sysc_restore
-	.quad	sysc_done
-	.quad	io_tif
-	.quad	io_restore
-	.quad	io_done
+	.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	psw_idle_end
+	.quad	.Lpsw_idle_end
 
 cleanup_critical:
-	clg	%r9,BASED(cleanup_table)	# system_call
+	clg	%r9,BASED(.Lcleanup_table)	# system_call
 	jl	0f
-	clg	%r9,BASED(cleanup_table+8)	# sysc_do_svc
-	jl	cleanup_system_call
-	clg	%r9,BASED(cleanup_table+16)	# sysc_tif
+	clg	%r9,BASED(.Lcleanup_table+8)	# .Lsysc_do_svc
+	jl	.Lcleanup_system_call
+	clg	%r9,BASED(.Lcleanup_table+16)	# .Lsysc_tif
 	jl	0f
-	clg	%r9,BASED(cleanup_table+24)	# sysc_restore
-	jl	cleanup_sysc_tif
-	clg	%r9,BASED(cleanup_table+32)	# sysc_done
-	jl	cleanup_sysc_restore
-	clg	%r9,BASED(cleanup_table+40)	# io_tif
+	clg	%r9,BASED(.Lcleanup_table+24)	# .Lsysc_restore
+	jl	.Lcleanup_sysc_tif
+	clg	%r9,BASED(.Lcleanup_table+32)	# .Lsysc_done
+	jl	.Lcleanup_sysc_restore
+	clg	%r9,BASED(.Lcleanup_table+40)	# .Lio_tif
 	jl	0f
-	clg	%r9,BASED(cleanup_table+48)	# io_restore
-	jl	cleanup_io_tif
-	clg	%r9,BASED(cleanup_table+56)	# io_done
-	jl	cleanup_io_restore
-	clg	%r9,BASED(cleanup_table+64)	# psw_idle
+	clg	%r9,BASED(.Lcleanup_table+48)	# .Lio_restore
+	jl	.Lcleanup_io_tif
+	clg	%r9,BASED(.Lcleanup_table+56)	# .Lio_done
+	jl	.Lcleanup_io_restore
+	clg	%r9,BASED(.Lcleanup_table+64)	# psw_idle
 	jl	0f
-	clg	%r9,BASED(cleanup_table+72)	# psw_idle_end
-	jl	cleanup_idle
+	clg	%r9,BASED(.Lcleanup_table+72)	# .Lpsw_idle_end
+	jl	.Lcleanup_idle
 0:	br	%r14
 
 
-cleanup_system_call:
+.Lcleanup_system_call:
 	# check if stpt has been executed
-	clg	%r9,BASED(cleanup_system_call_insn)
+	clg	%r9,BASED(.Lcleanup_system_call_insn)
 	jh	0f
 	mvc	__LC_SYNC_ENTER_TIMER(8),__LC_ASYNC_ENTER_TIMER
 	cghi	%r11,__LC_SAVE_AREA_ASYNC
 	je	0f
 	mvc	__LC_SYNC_ENTER_TIMER(8),__LC_MCCK_ENTER_TIMER
 0:	# check if stmg has been executed
-	clg	%r9,BASED(cleanup_system_call_insn+8)
+	clg	%r9,BASED(.Lcleanup_system_call_insn+8)
 	jh	0f
 	mvc	__LC_SAVE_AREA_SYNC(64),0(%r11)
 0:	# check if base register setup + TIF bit load has been done
-	clg	%r9,BASED(cleanup_system_call_insn+16)
+	clg	%r9,BASED(.Lcleanup_system_call_insn+16)
 	jhe	0f
 	# set up saved registers r10 and r12
 	stg	%r10,16(%r11)		# r10 last break
 	stg	%r12,32(%r11)		# r12 thread-info pointer
 0:	# check if the user time update has been done
-	clg	%r9,BASED(cleanup_system_call_insn+24)
+	clg	%r9,BASED(.Lcleanup_system_call_insn+24)
 	jh	0f
 	lg	%r15,__LC_EXIT_TIMER
 	slg	%r15,__LC_SYNC_ENTER_TIMER
 	alg	%r15,__LC_USER_TIMER
 	stg	%r15,__LC_USER_TIMER
 0:	# check if the system time update has been done
-	clg	%r9,BASED(cleanup_system_call_insn+32)
+	clg	%r9,BASED(.Lcleanup_system_call_insn+32)
 	jh	0f
 	lg	%r15,__LC_LAST_UPDATE_TIMER
 	slg	%r15,__LC_EXIT_TIMER
@@ -904,21 +904,21 @@ cleanup_system_call:
 	# setup saved register r15
 	stg	%r15,56(%r11)		# r15 stack pointer
 	# set new psw address and exit
-	larl	%r9,sysc_do_svc
+	larl	%r9,.Lsysc_do_svc
 	br	%r14
-cleanup_system_call_insn:
+.Lcleanup_system_call_insn:
 	.quad	system_call
-	.quad	sysc_stmg
-	.quad	sysc_per
-	.quad	sysc_vtime+18
-	.quad	sysc_vtime+42
+	.quad	.Lsysc_stmg
+	.quad	.Lsysc_per
+	.quad	.Lsysc_vtime+18
+	.quad	.Lsysc_vtime+42
 
-cleanup_sysc_tif:
-	larl	%r9,sysc_tif
+.Lcleanup_sysc_tif:
+	larl	%r9,.Lsysc_tif
 	br	%r14
 
-cleanup_sysc_restore:
-	clg	%r9,BASED(cleanup_sysc_restore_insn)
+.Lcleanup_sysc_restore:
+	clg	%r9,BASED(.Lcleanup_sysc_restore_insn)
 	je	0f
 	lg	%r9,24(%r11)		# get saved pointer to pt_regs
 	mvc	__LC_RETURN_PSW(16),__PT_PSW(%r9)
@@ -926,15 +926,15 @@ cleanup_sysc_restore:
 	lmg	%r0,%r7,__PT_R0(%r9)
 0:	lmg	%r8,%r9,__LC_RETURN_PSW
 	br	%r14
-cleanup_sysc_restore_insn:
-	.quad	sysc_done - 4
+.Lcleanup_sysc_restore_insn:
+	.quad	.Lsysc_done - 4
 
-cleanup_io_tif:
-	larl	%r9,io_tif
+.Lcleanup_io_tif:
+	larl	%r9,.Lio_tif
 	br	%r14
 
-cleanup_io_restore:
-	clg	%r9,BASED(cleanup_io_restore_insn)
+.Lcleanup_io_restore:
+	clg	%r9,BASED(.Lcleanup_io_restore_insn)
 	je	0f
 	lg	%r9,24(%r11)		# get saved r11 pointer to pt_regs
 	mvc	__LC_RETURN_PSW(16),__PT_PSW(%r9)
@@ -942,10 +942,10 @@ cleanup_io_restore:
 	lmg	%r0,%r7,__PT_R0(%r9)
 0:	lmg	%r8,%r9,__LC_RETURN_PSW
 	br	%r14
-cleanup_io_restore_insn:
-	.quad	io_done - 4
+.Lcleanup_io_restore_insn:
+	.quad	.Lio_done - 4
 
-cleanup_idle:
+.Lcleanup_idle:
 	# copy interrupt clock & cpu timer
 	mvc	__CLOCK_IDLE_EXIT(8,%r2),__LC_INT_CLOCK
 	mvc	__TIMER_IDLE_EXIT(8,%r2),__LC_ASYNC_ENTER_TIMER
@@ -954,7 +954,7 @@ cleanup_idle:
 	mvc	__CLOCK_IDLE_EXIT(8,%r2),__LC_MCCK_CLOCK
 	mvc	__TIMER_IDLE_EXIT(8,%r2),__LC_MCCK_ENTER_TIMER
 0:	# check if stck & stpt have been executed
-	clg	%r9,BASED(cleanup_idle_insn)
+	clg	%r9,BASED(.Lcleanup_idle_insn)
 	jhe	1f
 	mvc	__CLOCK_IDLE_ENTER(8,%r2),__CLOCK_IDLE_EXIT(%r2)
 	mvc	__TIMER_IDLE_ENTER(8,%r2),__TIMER_IDLE_EXIT(%r2)
@@ -973,17 +973,17 @@ cleanup_idle:
 	nihh	%r8,0xfcfd		# clear irq & wait state bits
 	lg	%r9,48(%r11)		# return from psw_idle
 	br	%r14
-cleanup_idle_insn:
-	.quad	psw_idle_lpsw
+.Lcleanup_idle_insn:
+	.quad	.Lpsw_idle_lpsw
 
 /*
  * Integer constants
  */
 	.align	8
 .Lcritical_start:
-	.quad	__critical_start
+	.quad	.L__critical_start
 .Lcritical_length:
-	.quad	__critical_end - __critical_start
+	.quad	.L__critical_end - .L__critical_start
 
 
 #if IS_ENABLED(CONFIG_KVM)
@@ -1000,25 +1000,25 @@ ENTRY(sie64a)
 	lmg	%r0,%r13,0(%r3)			# load guest gprs 0-13
 	lg	%r14,__LC_GMAP			# get gmap pointer
 	ltgr	%r14,%r14
-	jz	sie_gmap
+	jz	.Lsie_gmap
 	lctlg	%c1,%c1,__GMAP_ASCE(%r14)	# load primary asce
-sie_gmap:
+.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),1		# last exit...
-	jnz	sie_done
+	jnz	.Lsie_done
 	LPP	__SF_EMPTY(%r15)		# set guest id
 	sie	0(%r14)
-sie_done:
+.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 sie_done should not cause program
+# 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
-rewind_pad:
+.Lrewind_pad:
 	nop	0
 	.globl sie_exit
 sie_exit:
@@ -1027,19 +1027,19 @@ sie_exit:
 	lmg	%r6,%r14,__SF_GPRS(%r15)	# restore kernel registers
 	lg	%r2,__SF_EMPTY+24(%r15)		# return exit reason code
 	br	%r14
-sie_fault:
+.Lsie_fault:
 	lghi	%r14,-EFAULT
 	stg	%r14,__SF_EMPTY+24(%r15)	# set exit reason code
 	j	sie_exit
 
 	.align	8
 .Lsie_critical:
-	.quad	sie_gmap
+	.quad	.Lsie_gmap
 .Lsie_critical_length:
-	.quad	sie_done - sie_gmap
+	.quad	.Lsie_done - .Lsie_gmap
 
-	EX_TABLE(rewind_pad,sie_fault)
-	EX_TABLE(sie_exit,sie_fault)
+	EX_TABLE(.Lrewind_pad,.Lsie_fault)
+	EX_TABLE(sie_exit,.Lsie_fault)
 #endif
 
 		.section .rodata, "a"
diff --git a/arch/s390/kernel/ftrace.c b/arch/s390/kernel/ftrace.c
index ca1cabb3a96c..b86bb8823f15 100644
--- a/arch/s390/kernel/ftrace.c
+++ b/arch/s390/kernel/ftrace.c
@@ -7,6 +7,7 @@
  *		Martin Schwidefsky <schwidefsky@de.ibm.com>
  */
 
+#include <linux/moduleloader.h>
 #include <linux/hardirq.h>
 #include <linux/uaccess.h>
 #include <linux/ftrace.h>
@@ -15,60 +16,39 @@
 #include <linux/kprobes.h>
 #include <trace/syscall.h>
 #include <asm/asm-offsets.h>
+#include <asm/cacheflush.h>
 #include "entry.h"
 
-void mcount_replace_code(void);
-void ftrace_disable_code(void);
-void ftrace_enable_insn(void);
-
 /*
  * The mcount code looks like this:
  *	stg	%r14,8(%r15)		# offset 0
  *	larl	%r1,<&counter>		# offset 6
  *	brasl	%r14,_mcount		# offset 12
  *	lg	%r14,8(%r15)		# offset 18
- * Total length is 24 bytes. The complete mcount block initially gets replaced
- * by ftrace_make_nop. Subsequent calls to ftrace_make_call / ftrace_make_nop
- * only patch the jg/lg instruction within the block.
- * Note: we do not patch the first instruction to an unconditional branch,
- * since that would break kprobes/jprobes. It is easier to leave the larl
- * instruction in and only modify the second instruction.
+ * Total length is 24 bytes. Only the first instruction will be patched
+ * by ftrace_make_call / ftrace_make_nop.
  * The enabled ftrace code block looks like this:
- *	larl	%r0,.+24		# offset 0
- * >	lg	%r1,__LC_FTRACE_FUNC	# offset 6
- *	br	%r1			# offset 12
- *	brcl	0,0			# offset 14
- *	brc	0,0			# offset 20
+ * >	brasl	%r0,ftrace_caller	# offset 0
+ *	larl	%r1,<&counter>		# offset 6
+ *	brasl	%r14,_mcount		# offset 12
+ *	lg	%r14,8(%r15)		# offset 18
  * The ftrace function gets called with a non-standard C function call ABI
  * where r0 contains the return address. It is also expected that the called
  * function only clobbers r0 and r1, but restores r2-r15.
+ * For module code we can't directly jump to ftrace caller, but need a
+ * trampoline (ftrace_plt), which clobbers also r1.
  * The return point of the ftrace function has offset 24, so execution
  * continues behind the mcount block.
- *	larl	%r0,.+24		# offset 0
- * >	jg	.+18			# offset 6
- *	br	%r1			# offset 12
- *	brcl	0,0			# offset 14
- *	brc	0,0			# offset 20
+ * The disabled ftrace code block looks like this:
+ * >	jg	.+24			# offset 0
+ *	larl	%r1,<&counter>		# offset 6
+ *	brasl	%r14,_mcount		# offset 12
+ *	lg	%r14,8(%r15)		# offset 18
  * The jg instruction branches to offset 24 to skip as many instructions
  * as possible.
  */
-asm(
-	"	.align	4\n"
-	"mcount_replace_code:\n"
-	"	larl	%r0,0f\n"
-	"ftrace_disable_code:\n"
-	"	jg	0f\n"
-	"	br	%r1\n"
-	"	brcl	0,0\n"
-	"	brc	0,0\n"
-	"0:\n"
-	"	.align	4\n"
-	"ftrace_enable_insn:\n"
-	"	lg	%r1,"__stringify(__LC_FTRACE_FUNC)"\n");
-
-#define MCOUNT_BLOCK_SIZE	24
-#define MCOUNT_INSN_OFFSET	6
-#define FTRACE_INSN_SIZE	6
+
+unsigned long ftrace_plt;
 
 int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr,
 		       unsigned long addr)
@@ -79,24 +59,62 @@ int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr,
 int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec,
 		    unsigned long addr)
 {
-	/* Initial replacement of the whole mcount block */
-	if (addr == MCOUNT_ADDR) {
-		if (probe_kernel_write((void *) rec->ip - MCOUNT_INSN_OFFSET,
-				       mcount_replace_code,
-				       MCOUNT_BLOCK_SIZE))
-			return -EPERM;
-		return 0;
+	struct ftrace_insn insn;
+	unsigned short op;
+	void *from, *to;
+	size_t size;
+
+	ftrace_generate_nop_insn(&insn);
+	size = sizeof(insn);
+	from = &insn;
+	to = (void *) rec->ip;
+	if (probe_kernel_read(&op, (void *) rec->ip, sizeof(op)))
+		return -EFAULT;
+	/*
+	 * If we find a breakpoint instruction, a kprobe has been placed
+	 * at the beginning of the function. We write the constant
+	 * KPROBE_ON_FTRACE_NOP into the remaining four bytes of the original
+	 * instruction so that the kprobes handler can execute a nop, if it
+	 * reaches this breakpoint.
+	 */
+	if (op == BREAKPOINT_INSTRUCTION) {
+		size -= 2;
+		from += 2;
+		to += 2;
+		insn.disp = KPROBE_ON_FTRACE_NOP;
 	}
-	if (probe_kernel_write((void *) rec->ip, ftrace_disable_code,
-			       MCOUNT_INSN_SIZE))
+	if (probe_kernel_write(to, from, size))
 		return -EPERM;
 	return 0;
 }
 
 int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr)
 {
-	if (probe_kernel_write((void *) rec->ip, ftrace_enable_insn,
-			       FTRACE_INSN_SIZE))
+	struct ftrace_insn insn;
+	unsigned short op;
+	void *from, *to;
+	size_t size;
+
+	ftrace_generate_call_insn(&insn, rec->ip);
+	size = sizeof(insn);
+	from = &insn;
+	to = (void *) rec->ip;
+	if (probe_kernel_read(&op, (void *) rec->ip, sizeof(op)))
+		return -EFAULT;
+	/*
+	 * If we find a breakpoint instruction, a kprobe has been placed
+	 * at the beginning of the function. We write the constant
+	 * KPROBE_ON_FTRACE_CALL into the remaining four bytes of the original
+	 * instruction so that the kprobes handler can execute a brasl if it
+	 * reaches this breakpoint.
+	 */
+	if (op == BREAKPOINT_INSTRUCTION) {
+		size -= 2;
+		from += 2;
+		to += 2;
+		insn.disp = KPROBE_ON_FTRACE_CALL;
+	}
+	if (probe_kernel_write(to, from, size))
 		return -EPERM;
 	return 0;
 }
@@ -111,13 +129,30 @@ int __init ftrace_dyn_arch_init(void)
 	return 0;
 }
 
+static int __init ftrace_plt_init(void)
+{
+	unsigned int *ip;
+
+	ftrace_plt = (unsigned long) module_alloc(PAGE_SIZE);
+	if (!ftrace_plt)
+		panic("cannot allocate ftrace plt\n");
+	ip = (unsigned int *) ftrace_plt;
+	ip[0] = 0x0d10e310; /* basr 1,0; lg 1,10(1); br 1 */
+	ip[1] = 0x100a0004;
+	ip[2] = 0x07f10000;
+	ip[3] = FTRACE_ADDR >> 32;
+	ip[4] = FTRACE_ADDR & 0xffffffff;
+	set_memory_ro(ftrace_plt, 1);
+	return 0;
+}
+device_initcall(ftrace_plt_init);
+
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 /*
  * Hook the return address and push it in the stack of return addresses
  * in current thread info.
  */
-unsigned long __kprobes prepare_ftrace_return(unsigned long parent,
-					      unsigned long ip)
+unsigned long prepare_ftrace_return(unsigned long parent, unsigned long ip)
 {
 	struct ftrace_graph_ent trace;
 
@@ -137,6 +172,7 @@ unsigned long __kprobes prepare_ftrace_return(unsigned long parent,
 out:
 	return parent;
 }
+NOKPROBE_SYMBOL(prepare_ftrace_return);
 
 /*
  * Patch the kernel code at ftrace_graph_caller location. The instruction
diff --git a/arch/s390/kernel/idle.c b/arch/s390/kernel/idle.c
index 7559f1beab29..7a55c29b0b33 100644
--- a/arch/s390/kernel/idle.c
+++ b/arch/s390/kernel/idle.c
@@ -19,7 +19,7 @@
 
 static DEFINE_PER_CPU(struct s390_idle_data, s390_idle);
 
-void __kprobes enabled_wait(void)
+void enabled_wait(void)
 {
 	struct s390_idle_data *idle = this_cpu_ptr(&s390_idle);
 	unsigned long long idle_time;
@@ -35,31 +35,32 @@ void __kprobes enabled_wait(void)
 	/* Call the assembler magic in entry.S */
 	psw_idle(idle, psw_mask);
 
+	trace_hardirqs_off();
+
 	/* Account time spent with enabled wait psw loaded as idle time. */
-	idle->sequence++;
-	smp_wmb();
+	write_seqcount_begin(&idle->seqcount);
 	idle_time = idle->clock_idle_exit - idle->clock_idle_enter;
 	idle->clock_idle_enter = idle->clock_idle_exit = 0ULL;
 	idle->idle_time += idle_time;
 	idle->idle_count++;
 	account_idle_time(idle_time);
-	smp_wmb();
-	idle->sequence++;
+	write_seqcount_end(&idle->seqcount);
 }
+NOKPROBE_SYMBOL(enabled_wait);
 
 static ssize_t show_idle_count(struct device *dev,
 				struct device_attribute *attr, char *buf)
 {
 	struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id);
 	unsigned long long idle_count;
-	unsigned int sequence;
+	unsigned int seq;
 
 	do {
-		sequence = ACCESS_ONCE(idle->sequence);
+		seq = read_seqcount_begin(&idle->seqcount);
 		idle_count = ACCESS_ONCE(idle->idle_count);
 		if (ACCESS_ONCE(idle->clock_idle_enter))
 			idle_count++;
-	} while ((sequence & 1) || (ACCESS_ONCE(idle->sequence) != sequence));
+	} while (read_seqcount_retry(&idle->seqcount, seq));
 	return sprintf(buf, "%llu\n", idle_count);
 }
 DEVICE_ATTR(idle_count, 0444, show_idle_count, NULL);
@@ -69,15 +70,15 @@ static ssize_t show_idle_time(struct device *dev,
 {
 	struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id);
 	unsigned long long now, idle_time, idle_enter, idle_exit;
-	unsigned int sequence;
+	unsigned int seq;
 
 	do {
 		now = get_tod_clock();
-		sequence = ACCESS_ONCE(idle->sequence);
+		seq = read_seqcount_begin(&idle->seqcount);
 		idle_time = ACCESS_ONCE(idle->idle_time);
 		idle_enter = ACCESS_ONCE(idle->clock_idle_enter);
 		idle_exit = ACCESS_ONCE(idle->clock_idle_exit);
-	} while ((sequence & 1) || (ACCESS_ONCE(idle->sequence) != sequence));
+	} while (read_seqcount_retry(&idle->seqcount, seq));
 	idle_time += idle_enter ? ((idle_exit ? : now) - idle_enter) : 0;
 	return sprintf(buf, "%llu\n", idle_time >> 12);
 }
@@ -87,14 +88,14 @@ cputime64_t arch_cpu_idle_time(int cpu)
 {
 	struct s390_idle_data *idle = &per_cpu(s390_idle, cpu);
 	unsigned long long now, idle_enter, idle_exit;
-	unsigned int sequence;
+	unsigned int seq;
 
 	do {
 		now = get_tod_clock();
-		sequence = ACCESS_ONCE(idle->sequence);
+		seq = read_seqcount_begin(&idle->seqcount);
 		idle_enter = ACCESS_ONCE(idle->clock_idle_enter);
 		idle_exit = ACCESS_ONCE(idle->clock_idle_exit);
-	} while ((sequence & 1) || (ACCESS_ONCE(idle->sequence) != sequence));
+	} while (read_seqcount_retry(&idle->seqcount, seq));
 	return idle_enter ? ((idle_exit ?: now) - idle_enter) : 0;
 }
 
diff --git a/arch/s390/kernel/irq.c b/arch/s390/kernel/irq.c
index 1b8a38ab7861..f238720690f3 100644
--- a/arch/s390/kernel/irq.c
+++ b/arch/s390/kernel/irq.c
@@ -127,13 +127,10 @@ int show_interrupts(struct seq_file *p, void *v)
 		for_each_online_cpu(cpu)
 			seq_printf(p, "CPU%d       ", cpu);
 		seq_putc(p, '\n');
-		goto out;
 	}
 	if (index < NR_IRQS) {
 		if (index >= NR_IRQS_BASE)
 			goto out;
-		/* Adjust index to process irqclass_main_desc array entries */
-		index--;
 		seq_printf(p, "%s: ", irqclass_main_desc[index].name);
 		irq = irqclass_main_desc[index].irq;
 		for_each_online_cpu(cpu)
@@ -158,7 +155,7 @@ out:
 
 unsigned int arch_dynirq_lower_bound(unsigned int from)
 {
-	return from < THIN_INTERRUPT ? THIN_INTERRUPT : from;
+	return from < NR_IRQS_BASE ? NR_IRQS_BASE : from;
 }
 
 /*
diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c
index 014d4729b134..1e4c710dfb92 100644
--- a/arch/s390/kernel/kprobes.c
+++ b/arch/s390/kernel/kprobes.c
@@ -29,6 +29,7 @@
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/hardirq.h>
+#include <linux/ftrace.h>
 #include <asm/cacheflush.h>
 #include <asm/sections.h>
 #include <asm/dis.h>
@@ -58,12 +59,23 @@ struct kprobe_insn_cache kprobe_dmainsn_slots = {
 	.insn_size = MAX_INSN_SIZE,
 };
 
-static void __kprobes copy_instruction(struct kprobe *p)
+static void copy_instruction(struct kprobe *p)
 {
+	unsigned long ip = (unsigned long) p->addr;
 	s64 disp, new_disp;
 	u64 addr, new_addr;
 
-	memcpy(p->ainsn.insn, p->addr, insn_length(p->opcode >> 8));
+	if (ftrace_location(ip) == ip) {
+		/*
+		 * If kprobes patches the instruction that is morphed by
+		 * ftrace make sure that kprobes always sees the branch
+		 * "jg .+24" that skips the mcount block
+		 */
+		ftrace_generate_nop_insn((struct ftrace_insn *)p->ainsn.insn);
+		p->ainsn.is_ftrace_insn = 1;
+	} else
+		memcpy(p->ainsn.insn, p->addr, insn_length(*p->addr >> 8));
+	p->opcode = p->ainsn.insn[0];
 	if (!probe_is_insn_relative_long(p->ainsn.insn))
 		return;
 	/*
@@ -79,25 +91,14 @@ static void __kprobes copy_instruction(struct kprobe *p)
 	new_disp = ((addr + (disp * 2)) - new_addr) / 2;
 	*(s32 *)&p->ainsn.insn[1] = new_disp;
 }
+NOKPROBE_SYMBOL(copy_instruction);
 
 static inline int is_kernel_addr(void *addr)
 {
 	return addr < (void *)_end;
 }
 
-static inline int is_module_addr(void *addr)
-{
-#ifdef CONFIG_64BIT
-	BUILD_BUG_ON(MODULES_LEN > (1UL << 31));
-	if (addr < (void *)MODULES_VADDR)
-		return 0;
-	if (addr > (void *)MODULES_END)
-		return 0;
-#endif
-	return 1;
-}
-
-static int __kprobes s390_get_insn_slot(struct kprobe *p)
+static int s390_get_insn_slot(struct kprobe *p)
 {
 	/*
 	 * Get an insn slot that is within the same 2GB area like the original
@@ -111,8 +112,9 @@ static int __kprobes s390_get_insn_slot(struct kprobe *p)
 		p->ainsn.insn = get_insn_slot();
 	return p->ainsn.insn ? 0 : -ENOMEM;
 }
+NOKPROBE_SYMBOL(s390_get_insn_slot);
 
-static void __kprobes s390_free_insn_slot(struct kprobe *p)
+static void s390_free_insn_slot(struct kprobe *p)
 {
 	if (!p->ainsn.insn)
 		return;
@@ -122,8 +124,9 @@ static void __kprobes s390_free_insn_slot(struct kprobe *p)
 		free_insn_slot(p->ainsn.insn, 0);
 	p->ainsn.insn = NULL;
 }
+NOKPROBE_SYMBOL(s390_free_insn_slot);
 
-int __kprobes arch_prepare_kprobe(struct kprobe *p)
+int arch_prepare_kprobe(struct kprobe *p)
 {
 	if ((unsigned long) p->addr & 0x01)
 		return -EINVAL;
@@ -132,54 +135,79 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p)
 		return -EINVAL;
 	if (s390_get_insn_slot(p))
 		return -ENOMEM;
-	p->opcode = *p->addr;
 	copy_instruction(p);
 	return 0;
 }
+NOKPROBE_SYMBOL(arch_prepare_kprobe);
 
-struct ins_replace_args {
-	kprobe_opcode_t *ptr;
-	kprobe_opcode_t opcode;
+int arch_check_ftrace_location(struct kprobe *p)
+{
+	return 0;
+}
+
+struct swap_insn_args {
+	struct kprobe *p;
+	unsigned int arm_kprobe : 1;
 };
 
-static int __kprobes swap_instruction(void *aref)
+static int swap_instruction(void *data)
 {
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 	unsigned long status = kcb->kprobe_status;
-	struct ins_replace_args *args = aref;
-
+	struct swap_insn_args *args = data;
+	struct ftrace_insn new_insn, *insn;
+	struct kprobe *p = args->p;
+	size_t len;
+
+	new_insn.opc = args->arm_kprobe ? BREAKPOINT_INSTRUCTION : p->opcode;
+	len = sizeof(new_insn.opc);
+	if (!p->ainsn.is_ftrace_insn)
+		goto skip_ftrace;
+	len = sizeof(new_insn);
+	insn = (struct ftrace_insn *) p->addr;
+	if (args->arm_kprobe) {
+		if (is_ftrace_nop(insn))
+			new_insn.disp = KPROBE_ON_FTRACE_NOP;
+		else
+			new_insn.disp = KPROBE_ON_FTRACE_CALL;
+	} else {
+		ftrace_generate_call_insn(&new_insn, (unsigned long)p->addr);
+		if (insn->disp == KPROBE_ON_FTRACE_NOP)
+			ftrace_generate_nop_insn(&new_insn);
+	}
+skip_ftrace:
 	kcb->kprobe_status = KPROBE_SWAP_INST;
-	probe_kernel_write(args->ptr, &args->opcode, sizeof(args->opcode));
+	probe_kernel_write(p->addr, &new_insn, len);
 	kcb->kprobe_status = status;
 	return 0;
 }
+NOKPROBE_SYMBOL(swap_instruction);
 
-void __kprobes arch_arm_kprobe(struct kprobe *p)
+void arch_arm_kprobe(struct kprobe *p)
 {
-	struct ins_replace_args args;
+	struct swap_insn_args args = {.p = p, .arm_kprobe = 1};
 
-	args.ptr = p->addr;
-	args.opcode = BREAKPOINT_INSTRUCTION;
 	stop_machine(swap_instruction, &args, NULL);
 }
+NOKPROBE_SYMBOL(arch_arm_kprobe);
 
-void __kprobes arch_disarm_kprobe(struct kprobe *p)
+void arch_disarm_kprobe(struct kprobe *p)
 {
-	struct ins_replace_args args;
+	struct swap_insn_args args = {.p = p, .arm_kprobe = 0};
 
-	args.ptr = p->addr;
-	args.opcode = p->opcode;
 	stop_machine(swap_instruction, &args, NULL);
 }
+NOKPROBE_SYMBOL(arch_disarm_kprobe);
 
-void __kprobes arch_remove_kprobe(struct kprobe *p)
+void arch_remove_kprobe(struct kprobe *p)
 {
 	s390_free_insn_slot(p);
 }
+NOKPROBE_SYMBOL(arch_remove_kprobe);
 
-static void __kprobes enable_singlestep(struct kprobe_ctlblk *kcb,
-					struct pt_regs *regs,
-					unsigned long ip)
+static void enable_singlestep(struct kprobe_ctlblk *kcb,
+			      struct pt_regs *regs,
+			      unsigned long ip)
 {
 	struct per_regs per_kprobe;
 
@@ -199,10 +227,11 @@ static void __kprobes enable_singlestep(struct kprobe_ctlblk *kcb,
 	regs->psw.mask &= ~(PSW_MASK_IO | PSW_MASK_EXT);
 	regs->psw.addr = ip | PSW_ADDR_AMODE;
 }
+NOKPROBE_SYMBOL(enable_singlestep);
 
-static void __kprobes disable_singlestep(struct kprobe_ctlblk *kcb,
-					 struct pt_regs *regs,
-					 unsigned long ip)
+static void disable_singlestep(struct kprobe_ctlblk *kcb,
+			       struct pt_regs *regs,
+			       unsigned long ip)
 {
 	/* Restore control regs and psw mask, set new psw address */
 	__ctl_load(kcb->kprobe_saved_ctl, 9, 11);
@@ -210,41 +239,43 @@ static void __kprobes disable_singlestep(struct kprobe_ctlblk *kcb,
 	regs->psw.mask |= kcb->kprobe_saved_imask;
 	regs->psw.addr = ip | PSW_ADDR_AMODE;
 }
+NOKPROBE_SYMBOL(disable_singlestep);
 
 /*
  * Activate a kprobe by storing its pointer to current_kprobe. The
  * previous kprobe is stored in kcb->prev_kprobe. A stack of up to
  * two kprobes can be active, see KPROBE_REENTER.
  */
-static void __kprobes push_kprobe(struct kprobe_ctlblk *kcb, struct kprobe *p)
+static void push_kprobe(struct kprobe_ctlblk *kcb, struct kprobe *p)
 {
 	kcb->prev_kprobe.kp = __this_cpu_read(current_kprobe);
 	kcb->prev_kprobe.status = kcb->kprobe_status;
 	__this_cpu_write(current_kprobe, p);
 }
+NOKPROBE_SYMBOL(push_kprobe);
 
 /*
  * Deactivate a kprobe by backing up to the previous state. If the
  * current state is KPROBE_REENTER prev_kprobe.kp will be non-NULL,
  * for any other state prev_kprobe.kp will be NULL.
  */
-static void __kprobes pop_kprobe(struct kprobe_ctlblk *kcb)
+static void pop_kprobe(struct kprobe_ctlblk *kcb)
 {
 	__this_cpu_write(current_kprobe, kcb->prev_kprobe.kp);
 	kcb->kprobe_status = kcb->prev_kprobe.status;
 }
+NOKPROBE_SYMBOL(pop_kprobe);
 
-void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
-					struct pt_regs *regs)
+void arch_prepare_kretprobe(struct kretprobe_instance *ri, struct pt_regs *regs)
 {
 	ri->ret_addr = (kprobe_opcode_t *) regs->gprs[14];
 
 	/* Replace the return addr with trampoline addr */
 	regs->gprs[14] = (unsigned long) &kretprobe_trampoline;
 }
+NOKPROBE_SYMBOL(arch_prepare_kretprobe);
 
-static void __kprobes kprobe_reenter_check(struct kprobe_ctlblk *kcb,
-					   struct kprobe *p)
+static void kprobe_reenter_check(struct kprobe_ctlblk *kcb, struct kprobe *p)
 {
 	switch (kcb->kprobe_status) {
 	case KPROBE_HIT_SSDONE:
@@ -264,8 +295,9 @@ static void __kprobes kprobe_reenter_check(struct kprobe_ctlblk *kcb,
 		BUG();
 	}
 }
+NOKPROBE_SYMBOL(kprobe_reenter_check);
 
-static int __kprobes kprobe_handler(struct pt_regs *regs)
+static int kprobe_handler(struct pt_regs *regs)
 {
 	struct kprobe_ctlblk *kcb;
 	struct kprobe *p;
@@ -339,6 +371,7 @@ static int __kprobes kprobe_handler(struct pt_regs *regs)
 	preempt_enable_no_resched();
 	return 0;
 }
+NOKPROBE_SYMBOL(kprobe_handler);
 
 /*
  * Function return probe trampoline:
@@ -355,8 +388,7 @@ static void __used kretprobe_trampoline_holder(void)
 /*
  * Called when the probe at kretprobe trampoline is hit
  */
-static int __kprobes trampoline_probe_handler(struct kprobe *p,
-					      struct pt_regs *regs)
+static int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 {
 	struct kretprobe_instance *ri;
 	struct hlist_head *head, empty_rp;
@@ -444,6 +476,7 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p,
 	 */
 	return 1;
 }
+NOKPROBE_SYMBOL(trampoline_probe_handler);
 
 /*
  * Called after single-stepping.  p->addr is the address of the
@@ -453,12 +486,30 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p,
  * single-stepped a copy of the instruction.  The address of this
  * copy is p->ainsn.insn.
  */
-static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs)
+static void resume_execution(struct kprobe *p, struct pt_regs *regs)
 {
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 	unsigned long ip = regs->psw.addr & PSW_ADDR_INSN;
 	int fixup = probe_get_fixup_type(p->ainsn.insn);
 
+	/* Check if the kprobes location is an enabled ftrace caller */
+	if (p->ainsn.is_ftrace_insn) {
+		struct ftrace_insn *insn = (struct ftrace_insn *) p->addr;
+		struct ftrace_insn call_insn;
+
+		ftrace_generate_call_insn(&call_insn, (unsigned long) p->addr);
+		/*
+		 * A kprobe on an enabled ftrace call site actually single
+		 * stepped an unconditional branch (ftrace nop equivalent).
+		 * Now we need to fixup things and pretend that a brasl r0,...
+		 * was executed instead.
+		 */
+		if (insn->disp == KPROBE_ON_FTRACE_CALL) {
+			ip += call_insn.disp * 2 - MCOUNT_INSN_SIZE;
+			regs->gprs[0] = (unsigned long)p->addr + sizeof(*insn);
+		}
+	}
+
 	if (fixup & FIXUP_PSW_NORMAL)
 		ip += (unsigned long) p->addr - (unsigned long) p->ainsn.insn;
 
@@ -476,8 +527,9 @@ static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs)
 
 	disable_singlestep(kcb, regs, ip);
 }
+NOKPROBE_SYMBOL(resume_execution);
 
-static int __kprobes post_kprobe_handler(struct pt_regs *regs)
+static int post_kprobe_handler(struct pt_regs *regs)
 {
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 	struct kprobe *p = kprobe_running();
@@ -504,8 +556,9 @@ static int __kprobes post_kprobe_handler(struct pt_regs *regs)
 
 	return 1;
 }
+NOKPROBE_SYMBOL(post_kprobe_handler);
 
-static int __kprobes kprobe_trap_handler(struct pt_regs *regs, int trapnr)
+static int kprobe_trap_handler(struct pt_regs *regs, int trapnr)
 {
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 	struct kprobe *p = kprobe_running();
@@ -567,8 +620,9 @@ static int __kprobes kprobe_trap_handler(struct pt_regs *regs, int trapnr)
 	}
 	return 0;
 }
+NOKPROBE_SYMBOL(kprobe_trap_handler);
 
-int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
+int kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 {
 	int ret;
 
@@ -579,12 +633,13 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
 		local_irq_restore(regs->psw.mask & ~PSW_MASK_PER);
 	return ret;
 }
+NOKPROBE_SYMBOL(kprobe_fault_handler);
 
 /*
  * Wrapper routine to for handling exceptions.
  */
-int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
-				       unsigned long val, void *data)
+int kprobe_exceptions_notify(struct notifier_block *self,
+			     unsigned long val, void *data)
 {
 	struct die_args *args = (struct die_args *) data;
 	struct pt_regs *regs = args->regs;
@@ -616,8 +671,9 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
 
 	return ret;
 }
+NOKPROBE_SYMBOL(kprobe_exceptions_notify);
 
-int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+int setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 {
 	struct jprobe *jp = container_of(p, struct jprobe, kp);
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
@@ -635,13 +691,15 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	memcpy(kcb->jprobes_stack, (void *) stack, MIN_STACK_SIZE(stack));
 	return 1;
 }
+NOKPROBE_SYMBOL(setjmp_pre_handler);
 
-void __kprobes jprobe_return(void)
+void jprobe_return(void)
 {
 	asm volatile(".word 0x0002");
 }
+NOKPROBE_SYMBOL(jprobe_return);
 
-int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
+int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 {
 	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
 	unsigned long stack;
@@ -655,6 +713,7 @@ int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	preempt_enable_no_resched();
 	return 1;
 }
+NOKPROBE_SYMBOL(longjmp_break_handler);
 
 static struct kprobe trampoline = {
 	.addr = (kprobe_opcode_t *) &kretprobe_trampoline,
@@ -666,7 +725,8 @@ int __init arch_init_kprobes(void)
 	return register_kprobe(&trampoline);
 }
 
-int __kprobes arch_trampoline_kprobe(struct kprobe *p)
+int arch_trampoline_kprobe(struct kprobe *p)
 {
 	return p->addr == (kprobe_opcode_t *) &kretprobe_trampoline;
 }
+NOKPROBE_SYMBOL(arch_trampoline_kprobe);
diff --git a/arch/s390/kernel/mcount.S b/arch/s390/kernel/mcount.S
index 4300ea374826..b6dfc5bfcb89 100644
--- a/arch/s390/kernel/mcount.S
+++ b/arch/s390/kernel/mcount.S
@@ -27,6 +27,7 @@ ENTRY(ftrace_caller)
 	.globl	ftrace_regs_caller
 	.set	ftrace_regs_caller,ftrace_caller
 	lgr	%r1,%r15
+	aghi	%r0,MCOUNT_RETURN_FIXUP
 	aghi	%r15,-STACK_FRAME_SIZE
 	stg	%r1,__SF_BACKCHAIN(%r15)
 	stg	%r1,(STACK_PTREGS_GPRS+15*8)(%r15)
diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c
index b878f12a9597..c3f8d157cb0d 100644
--- a/arch/s390/kernel/perf_cpum_sf.c
+++ b/arch/s390/kernel/perf_cpum_sf.c
@@ -1383,7 +1383,6 @@ static int cpumsf_pmu_add(struct perf_event *event, int flags)
 		cpuhw->lsctl.ed = 1;
 
 	/* Set in_use flag and store event */
-	event->hw.idx = 0;	  /* only one sampling event per CPU supported */
 	cpuhw->event = event;
 	cpuhw->flags |= PMU_F_IN_USE;
 
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index ed84cc224899..aa7a83948c7b 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -61,7 +61,7 @@ unsigned long thread_saved_pc(struct task_struct *tsk)
 	return sf->gprs[8];
 }
 
-extern void __kprobes kernel_thread_starter(void);
+extern void kernel_thread_starter(void);
 
 /*
  * Free current thread data structures etc..
@@ -153,6 +153,7 @@ int copy_thread(unsigned long clone_flags, unsigned long new_stackp,
 	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];
diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index 99a567b70d16..eabfb4594517 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -248,14 +248,27 @@ static unsigned long __peek_user(struct task_struct *child, addr_t addr)
 		 */
 		tmp = 0;
 
+	} else if (addr == (addr_t) &dummy->regs.fp_regs.fpc) {
+		/*
+		 * floating point control reg. is in the thread structure
+		 */
+		tmp = child->thread.fp_regs.fpc;
+		tmp <<= BITS_PER_LONG - 32;
+
 	} else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) {
-		/* 
-		 * floating point regs. are stored in the thread structure
+		/*
+		 * floating point regs. are either in child->thread.fp_regs
+		 * or the child->thread.vxrs array
 		 */
-		offset = addr - (addr_t) &dummy->regs.fp_regs;
-		tmp = *(addr_t *)((addr_t) &child->thread.fp_regs + offset);
-		if (addr == (addr_t) &dummy->regs.fp_regs.fpc)
-			tmp <<= BITS_PER_LONG - 32;
+		offset = addr - (addr_t) &dummy->regs.fp_regs.fprs;
+#ifdef CONFIG_64BIT
+		if (child->thread.vxrs)
+			tmp = *(addr_t *)
+			       ((addr_t) child->thread.vxrs + 2*offset);
+		else
+#endif
+			tmp = *(addr_t *)
+			       ((addr_t) &child->thread.fp_regs.fprs + offset);
 
 	} else if (addr < (addr_t) (&dummy->regs.per_info + 1)) {
 		/*
@@ -383,16 +396,29 @@ static int __poke_user(struct task_struct *child, addr_t addr, addr_t data)
 		 */
 		return 0;
 
+	} else if (addr == (addr_t) &dummy->regs.fp_regs.fpc) {
+		/*
+		 * floating point control reg. is in the thread structure
+		 */
+		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);
+
 	} else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) {
 		/*
-		 * floating point regs. are stored in the thread structure
+		 * floating point regs. are either in child->thread.fp_regs
+		 * or the child->thread.vxrs array
 		 */
-		if (addr == (addr_t) &dummy->regs.fp_regs.fpc)
-			if ((unsigned int) data != 0 ||
-			    test_fp_ctl(data >> (BITS_PER_LONG - 32)))
-				return -EINVAL;
-		offset = addr - (addr_t) &dummy->regs.fp_regs;
-		*(addr_t *)((addr_t) &child->thread.fp_regs + offset) = data;
+		offset = addr - (addr_t) &dummy->regs.fp_regs.fprs;
+#ifdef CONFIG_64BIT
+		if (child->thread.vxrs)
+			*(addr_t *)((addr_t)
+				child->thread.vxrs + 2*offset) = data;
+		else
+#endif
+			*(addr_t *)((addr_t)
+				&child->thread.fp_regs.fprs + offset) = data;
 
 	} else if (addr < (addr_t) (&dummy->regs.per_info + 1)) {
 		/*
@@ -611,12 +637,26 @@ static u32 __peek_user_compat(struct task_struct *child, addr_t addr)
 		 */
 		tmp = 0;
 
+	} else if (addr == (addr_t) &dummy32->regs.fp_regs.fpc) {
+		/*
+		 * floating point control reg. is in the thread structure
+		 */
+		tmp = child->thread.fp_regs.fpc;
+
 	} else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) {
 		/*
-		 * floating point regs. are stored in the thread structure 
+		 * floating point regs. are either in child->thread.fp_regs
+		 * or the child->thread.vxrs array
 		 */
-	        offset = addr - (addr_t) &dummy32->regs.fp_regs;
-		tmp = *(__u32 *)((addr_t) &child->thread.fp_regs + offset);
+		offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs;
+#ifdef CONFIG_64BIT
+		if (child->thread.vxrs)
+			tmp = *(__u32 *)
+			       ((addr_t) child->thread.vxrs + 2*offset);
+		else
+#endif
+			tmp = *(__u32 *)
+			       ((addr_t) &child->thread.fp_regs.fprs + offset);
 
 	} else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) {
 		/*
@@ -722,15 +762,28 @@ static int __poke_user_compat(struct task_struct *child,
 		 */
 		return 0;
 
-	} else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) {
+	} else if (addr == (addr_t) &dummy32->regs.fp_regs.fpc) {
 		/*
-		 * floating point regs. are stored in the thread structure 
+		 * floating point control reg. is in the thread structure
 		 */
-		if (addr == (addr_t) &dummy32->regs.fp_regs.fpc &&
-		    test_fp_ctl(tmp))
+		if (test_fp_ctl(tmp))
 			return -EINVAL;
-	        offset = addr - (addr_t) &dummy32->regs.fp_regs;
-		*(__u32 *)((addr_t) &child->thread.fp_regs + offset) = tmp;
+		child->thread.fp_regs.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
+		 */
+		offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs;
+#ifdef CONFIG_64BIT
+		if (child->thread.vxrs)
+			*(__u32 *)((addr_t)
+				child->thread.vxrs + 2*offset) = tmp;
+		else
+#endif
+			*(__u32 *)((addr_t)
+				&child->thread.fp_regs.fprs + offset) = tmp;
 
 	} else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) {
 		/*
@@ -1038,12 +1091,6 @@ static int s390_tdb_set(struct task_struct *target,
 	return 0;
 }
 
-static int s390_vxrs_active(struct task_struct *target,
-			      const struct user_regset *regset)
-{
-	return !!target->thread.vxrs;
-}
-
 static int s390_vxrs_low_get(struct task_struct *target,
 			     const struct user_regset *regset,
 			     unsigned int pos, unsigned int count,
@@ -1052,6 +1099,8 @@ static int s390_vxrs_low_get(struct task_struct *target,
 	__u64 vxrs[__NUM_VXRS_LOW];
 	int i;
 
+	if (!MACHINE_HAS_VX)
+		return -ENODEV;
 	if (target->thread.vxrs) {
 		if (target == current)
 			save_vx_regs(target->thread.vxrs);
@@ -1070,6 +1119,8 @@ static int s390_vxrs_low_set(struct task_struct *target,
 	__u64 vxrs[__NUM_VXRS_LOW];
 	int i, rc;
 
+	if (!MACHINE_HAS_VX)
+		return -ENODEV;
 	if (!target->thread.vxrs) {
 		rc = alloc_vector_registers(target);
 		if (rc)
@@ -1095,6 +1146,8 @@ static int s390_vxrs_high_get(struct task_struct *target,
 {
 	__vector128 vxrs[__NUM_VXRS_HIGH];
 
+	if (!MACHINE_HAS_VX)
+		return -ENODEV;
 	if (target->thread.vxrs) {
 		if (target == current)
 			save_vx_regs(target->thread.vxrs);
@@ -1112,6 +1165,8 @@ static int s390_vxrs_high_set(struct task_struct *target,
 {
 	int rc;
 
+	if (!MACHINE_HAS_VX)
+		return -ENODEV;
 	if (!target->thread.vxrs) {
 		rc = alloc_vector_registers(target);
 		if (rc)
@@ -1196,7 +1251,6 @@ static const struct user_regset s390_regsets[] = {
 		.n = __NUM_VXRS_LOW,
 		.size = sizeof(__u64),
 		.align = sizeof(__u64),
-		.active = s390_vxrs_active,
 		.get = s390_vxrs_low_get,
 		.set = s390_vxrs_low_set,
 	},
@@ -1205,7 +1259,6 @@ static const struct user_regset s390_regsets[] = {
 		.n = __NUM_VXRS_HIGH,
 		.size = sizeof(__vector128),
 		.align = sizeof(__vector128),
-		.active = s390_vxrs_active,
 		.get = s390_vxrs_high_get,
 		.set = s390_vxrs_high_set,
 	},
@@ -1419,7 +1472,6 @@ static const struct user_regset s390_compat_regsets[] = {
 		.n = __NUM_VXRS_LOW,
 		.size = sizeof(__u64),
 		.align = sizeof(__u64),
-		.active = s390_vxrs_active,
 		.get = s390_vxrs_low_get,
 		.set = s390_vxrs_low_set,
 	},
@@ -1428,7 +1480,6 @@ static const struct user_regset s390_compat_regsets[] = {
 		.n = __NUM_VXRS_HIGH,
 		.size = sizeof(__vector128),
 		.align = sizeof(__vector128),
-		.active = s390_vxrs_active,
 		.get = s390_vxrs_high_get,
 		.set = s390_vxrs_high_set,
 	},
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index e80d9ff9a56d..4e532c67832f 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -41,7 +41,6 @@
 #include <linux/ctype.h>
 #include <linux/reboot.h>
 #include <linux/topology.h>
-#include <linux/ftrace.h>
 #include <linux/kexec.h>
 #include <linux/crash_dump.h>
 #include <linux/memory.h>
@@ -356,7 +355,6 @@ static void __init setup_lowcore(void)
 	lc->steal_timer = S390_lowcore.steal_timer;
 	lc->last_update_timer = S390_lowcore.last_update_timer;
 	lc->last_update_clock = S390_lowcore.last_update_clock;
-	lc->ftrace_func = S390_lowcore.ftrace_func;
 
 	restart_stack = __alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0);
 	restart_stack += ASYNC_SIZE;
diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c
index 0c1a0ff0a558..6a2ac257d98f 100644
--- a/arch/s390/kernel/signal.c
+++ b/arch/s390/kernel/signal.c
@@ -371,7 +371,7 @@ static int setup_frame(int sig, struct k_sigaction *ka,
 		restorer = (unsigned long) ka->sa.sa_restorer | PSW_ADDR_AMODE;
 	} else {
 		/* Signal frame without vector registers are short ! */
-		__u16 __user *svc = (void *) frame + frame_size - 2;
+		__u16 __user *svc = (void __user *) frame + frame_size - 2;
 		if (__put_user(S390_SYSCALL_OPCODE | __NR_sigreturn, svc))
 			return -EFAULT;
 		restorer = (unsigned long) svc | PSW_ADDR_AMODE;
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 6fd9e60101f1..0b499f5cbe19 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -236,7 +236,6 @@ static void pcpu_prepare_secondary(struct pcpu *pcpu, int cpu)
 	lc->percpu_offset = __per_cpu_offset[cpu];
 	lc->kernel_asce = S390_lowcore.kernel_asce;
 	lc->machine_flags = S390_lowcore.machine_flags;
-	lc->ftrace_func = S390_lowcore.ftrace_func;
 	lc->user_timer = lc->system_timer = lc->steal_timer = 0;
 	__ctl_store(lc->cregs_save_area, 0, 15);
 	save_access_regs((unsigned int *) lc->access_regs_save_area);
diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
index 9f7087fd58de..a2987243bc76 100644
--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -360,3 +360,5 @@ SYSCALL(sys_seccomp,sys_seccomp,compat_sys_seccomp)
 SYSCALL(sys_getrandom,sys_getrandom,compat_sys_getrandom)
 SYSCALL(sys_memfd_create,sys_memfd_create,compat_sys_memfd_create) /* 350 */
 SYSCALL(sys_bpf,sys_bpf,compat_sys_bpf)
+SYSCALL(sys_ni_syscall,sys_s390_pci_mmio_write,compat_sys_s390_pci_mmio_write)
+SYSCALL(sys_ni_syscall,sys_s390_pci_mmio_read,compat_sys_s390_pci_mmio_read)
diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 005d665fe4a5..20660dddb2d6 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -61,10 +61,11 @@ static DEFINE_PER_CPU(struct clock_event_device, comparators);
 /*
  * Scheduler clock - returns current time in nanosec units.
  */
-unsigned long long notrace __kprobes sched_clock(void)
+unsigned long long notrace sched_clock(void)
 {
 	return tod_to_ns(get_tod_clock_monotonic());
 }
+NOKPROBE_SYMBOL(sched_clock);
 
 /*
  * Monotonic_clock - returns # of nanoseconds passed since time_init()
diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index 9ff5ecba26ab..f081cf1157c3 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -49,7 +49,8 @@ static inline void report_user_fault(struct pt_regs *regs, int signr)
 		return;
 	if (!printk_ratelimit())
 		return;
-	printk("User process fault: interruption code 0x%X ", regs->int_code);
+	printk("User process fault: interruption code %04x ilc:%d ",
+	       regs->int_code & 0xffff, regs->int_code >> 17);
 	print_vma_addr("in ", regs->psw.addr & PSW_ADDR_INSN);
 	printk("\n");
 	show_regs(regs);
@@ -87,16 +88,16 @@ void do_report_trap(struct pt_regs *regs, int si_signo, int si_code, char *str)
         }
 }
 
-static void __kprobes do_trap(struct pt_regs *regs, int si_signo, int si_code,
-			      char *str)
+static void do_trap(struct pt_regs *regs, int si_signo, int si_code, char *str)
 {
 	if (notify_die(DIE_TRAP, str, regs, 0,
 		       regs->int_code, si_signo) == NOTIFY_STOP)
 		return;
 	do_report_trap(regs, si_signo, si_code, str);
 }
+NOKPROBE_SYMBOL(do_trap);
 
-void __kprobes do_per_trap(struct pt_regs *regs)
+void do_per_trap(struct pt_regs *regs)
 {
 	siginfo_t info;
 
@@ -111,6 +112,7 @@ void __kprobes do_per_trap(struct pt_regs *regs)
 		(void __force __user *) current->thread.per_event.address;
 	force_sig_info(SIGTRAP, &info, current);
 }
+NOKPROBE_SYMBOL(do_per_trap);
 
 void default_trap_handler(struct pt_regs *regs)
 {
@@ -151,8 +153,6 @@ DO_ERROR_INFO(privileged_op, SIGILL, ILL_PRVOPC,
 	      "privileged operation")
 DO_ERROR_INFO(special_op_exception, SIGILL, ILL_ILLOPN,
 	      "special operation exception")
-DO_ERROR_INFO(translation_exception, SIGILL, ILL_ILLOPN,
-	      "translation exception")
 
 #ifdef CONFIG_64BIT
 DO_ERROR_INFO(transaction_exception, SIGILL, ILL_ILLOPN,
@@ -179,7 +179,13 @@ static inline void do_fp_trap(struct pt_regs *regs, int fpc)
 	do_trap(regs, SIGFPE, si_code, "floating point exception");
 }
 
-void __kprobes illegal_op(struct pt_regs *regs)
+void translation_exception(struct pt_regs *regs)
+{
+	/* May never happen. */
+	die(regs, "Translation exception");
+}
+
+void illegal_op(struct pt_regs *regs)
 {
 	siginfo_t info;
         __u8 opcode[6];
@@ -252,7 +258,7 @@ void __kprobes illegal_op(struct pt_regs *regs)
 	if (signal)
 		do_trap(regs, signal, ILL_ILLOPC, "illegal operation");
 }
-
+NOKPROBE_SYMBOL(illegal_op);
 
 #ifdef CONFIG_MATHEMU
 void specification_exception(struct pt_regs *regs)
@@ -469,7 +475,7 @@ void space_switch_exception(struct pt_regs *regs)
 	do_trap(regs, SIGILL, ILL_PRVOPC, "space switch event");
 }
 
-void __kprobes kernel_stack_overflow(struct pt_regs * regs)
+void kernel_stack_overflow(struct pt_regs *regs)
 {
 	bust_spinlocks(1);
 	printk("Kernel stack overflow.\n");
@@ -477,6 +483,7 @@ void __kprobes kernel_stack_overflow(struct pt_regs * regs)
 	bust_spinlocks(0);
 	panic("Corrupt kernel stack, can't continue.");
 }
+NOKPROBE_SYMBOL(kernel_stack_overflow);
 
 void __init trap_init(void)
 {
author	Linus Torvalds	2014-12-11 17:30:55 -0800
committer	Linus Torvalds	2014-12-11 17:30:55 -0800
commit	27afc5dbda52ee3dbcd0bda7375c917c6936b470 (patch)
tree	47591400f85590d48fa71bbfa50e0707e20e4bd0 /arch/s390/kernel
parent	70e71ca0af244f48a5dcf56dc435243792e3a495 (diff)
parent	351997810131565fe62aec2c366deccbf6bda3f4 (diff)