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
  ...

17 files changed, 176 insertions, 283 deletions

diff --git a/arch/s390/include/asm/cmpxchg.h b/arch/s390/include/asm/cmpxchg.h
index 4236408070e5..6259895fcd97 100644
--- a/arch/s390/include/asm/cmpxchg.h
+++ b/arch/s390/include/asm/cmpxchg.h
@@ -11,200 +11,28 @@
 #include <linux/types.h>
 #include <linux/bug.h>
 
-extern void __xchg_called_with_bad_pointer(void);
-
-static inline unsigned long __xchg(unsigned long x, void *ptr, int size)
-{
-	unsigned long addr, old;
-	int shift;
-
-	switch (size) {
-	case 1:
-		addr = (unsigned long) ptr;
-		shift = (3 ^ (addr & 3)) << 3;
-		addr ^= addr & 3;
-		asm volatile(
-			"	l	%0,%4\n"
-			"0:	lr	0,%0\n"
-			"	nr	0,%3\n"
-			"	or	0,%2\n"
-			"	cs	%0,0,%4\n"
-			"	jl	0b\n"
-			: "=&d" (old), "=Q" (*(int *) addr)
-			: "d" ((x & 0xff) << shift), "d" (~(0xff << shift)),
-			  "Q" (*(int *) addr) : "memory", "cc", "0");
-		return old >> shift;
-	case 2:
-		addr = (unsigned long) ptr;
-		shift = (2 ^ (addr & 2)) << 3;
-		addr ^= addr & 2;
-		asm volatile(
-			"	l	%0,%4\n"
-			"0:	lr	0,%0\n"
-			"	nr	0,%3\n"
-			"	or	0,%2\n"
-			"	cs	%0,0,%4\n"
-			"	jl	0b\n"
-			: "=&d" (old), "=Q" (*(int *) addr)
-			: "d" ((x & 0xffff) << shift), "d" (~(0xffff << shift)),
-			  "Q" (*(int *) addr) : "memory", "cc", "0");
-		return old >> shift;
-	case 4:
-		asm volatile(
-			"	l	%0,%3\n"
-			"0:	cs	%0,%2,%3\n"
-			"	jl	0b\n"
-			: "=&d" (old), "=Q" (*(int *) ptr)
-			: "d" (x), "Q" (*(int *) ptr)
-			: "memory", "cc");
-		return old;
-#ifdef CONFIG_64BIT
-	case 8:
-		asm volatile(
-			"	lg	%0,%3\n"
-			"0:	csg	%0,%2,%3\n"
-			"	jl	0b\n"
-			: "=&d" (old), "=m" (*(long *) ptr)
-			: "d" (x), "Q" (*(long *) ptr)
-			: "memory", "cc");
-		return old;
-#endif /* CONFIG_64BIT */
-	}
-	__xchg_called_with_bad_pointer();
-	return x;
-}
-
-#define xchg(ptr, x)							  \
-({									  \
-	__typeof__(*(ptr)) __ret;					  \
-	__ret = (__typeof__(*(ptr)))					  \
-		__xchg((unsigned long)(x), (void *)(ptr), sizeof(*(ptr)));\
-	__ret;								  \
+#define cmpxchg(ptr, o, n)						\
+({									\
+	__typeof__(*(ptr)) __o = (o);					\
+	__typeof__(*(ptr)) __n = (n);					\
+	(__typeof__(*(ptr))) __sync_val_compare_and_swap((ptr),__o,__n);\
 })
 
-/*
- * Atomic compare and exchange.	 Compare OLD with MEM, if identical,
- * store NEW in MEM.  Return the initial value in MEM.	Success is
- * indicated by comparing RETURN with OLD.
- */
-
-#define __HAVE_ARCH_CMPXCHG
-
-extern void __cmpxchg_called_with_bad_pointer(void);
-
-static inline unsigned long __cmpxchg(void *ptr, unsigned long old,
-				      unsigned long new, int size)
-{
-	unsigned long addr, prev, tmp;
-	int shift;
-
-	switch (size) {
-	case 1:
-		addr = (unsigned long) ptr;
-		shift = (3 ^ (addr & 3)) << 3;
-		addr ^= addr & 3;
-		asm volatile(
-			"	l	%0,%2\n"
-			"0:	nr	%0,%5\n"
-			"	lr	%1,%0\n"
-			"	or	%0,%3\n"
-			"	or	%1,%4\n"
-			"	cs	%0,%1,%2\n"
-			"	jnl	1f\n"
-			"	xr	%1,%0\n"
-			"	nr	%1,%5\n"
-			"	jnz	0b\n"
-			"1:"
-			: "=&d" (prev), "=&d" (tmp), "+Q" (*(int *) addr)
-			: "d" ((old & 0xff) << shift),
-			  "d" ((new & 0xff) << shift),
-			  "d" (~(0xff << shift))
-			: "memory", "cc");
-		return prev >> shift;
-	case 2:
-		addr = (unsigned long) ptr;
-		shift = (2 ^ (addr & 2)) << 3;
-		addr ^= addr & 2;
-		asm volatile(
-			"	l	%0,%2\n"
-			"0:	nr	%0,%5\n"
-			"	lr	%1,%0\n"
-			"	or	%0,%3\n"
-			"	or	%1,%4\n"
-			"	cs	%0,%1,%2\n"
-			"	jnl	1f\n"
-			"	xr	%1,%0\n"
-			"	nr	%1,%5\n"
-			"	jnz	0b\n"
-			"1:"
-			: "=&d" (prev), "=&d" (tmp), "+Q" (*(int *) addr)
-			: "d" ((old & 0xffff) << shift),
-			  "d" ((new & 0xffff) << shift),
-			  "d" (~(0xffff << shift))
-			: "memory", "cc");
-		return prev >> shift;
-	case 4:
-		asm volatile(
-			"	cs	%0,%3,%1\n"
-			: "=&d" (prev), "=Q" (*(int *) ptr)
-			: "0" (old), "d" (new), "Q" (*(int *) ptr)
-			: "memory", "cc");
-		return prev;
-#ifdef CONFIG_64BIT
-	case 8:
-		asm volatile(
-			"	csg	%0,%3,%1\n"
-			: "=&d" (prev), "=Q" (*(long *) ptr)
-			: "0" (old), "d" (new), "Q" (*(long *) ptr)
-			: "memory", "cc");
-		return prev;
-#endif /* CONFIG_64BIT */
-	}
-	__cmpxchg_called_with_bad_pointer();
-	return old;
-}
-
-#define cmpxchg(ptr, o, n)						 \
-({									 \
-	__typeof__(*(ptr)) __ret;					 \
-	__ret = (__typeof__(*(ptr)))					 \
-		__cmpxchg((ptr), (unsigned long)(o), (unsigned long)(n), \
-			  sizeof(*(ptr)));				 \
-	__ret;								 \
-})
+#define cmpxchg64	cmpxchg
+#define cmpxchg_local	cmpxchg
+#define cmpxchg64_local	cmpxchg
 
-#ifdef CONFIG_64BIT
-#define cmpxchg64(ptr, o, n)						\
+#define xchg(ptr, x)							\
 ({									\
-	cmpxchg((ptr), (o), (n));					\
+	__typeof__(ptr) __ptr = (ptr);					\
+	__typeof__(*(ptr)) __old;					\
+	do {								\
+		__old = *__ptr;						\
+	} while (!__sync_bool_compare_and_swap(__ptr, __old, x));	\
+	__old;								\
 })
-#else /* CONFIG_64BIT */
-static inline unsigned long long __cmpxchg64(void *ptr,
-					     unsigned long long old,
-					     unsigned long long new)
-{
-	register_pair rp_old = {.pair = old};
-	register_pair rp_new = {.pair = new};
-	unsigned long long *ullptr = ptr;
 
-	asm volatile(
-		"	cds	%0,%2,%1"
-		: "+d" (rp_old), "+Q" (*ullptr)
-		: "d" (rp_new)
-		: "memory", "cc");
-	return rp_old.pair;
-}
-
-#define cmpxchg64(ptr, o, n)				\
-({							\
-	__typeof__(*(ptr)) __ret;			\
-	__ret = (__typeof__(*(ptr)))			\
-		__cmpxchg64((ptr),			\
-			    (unsigned long long)(o),	\
-			    (unsigned long long)(n));	\
-	__ret;						\
-})
-#endif /* CONFIG_64BIT */
+#define __HAVE_ARCH_CMPXCHG
 
 #define __cmpxchg_double_op(p1, p2, o1, o2, n1, n2, insn)		\
 ({									\
@@ -265,40 +93,4 @@ extern void __cmpxchg_double_called_with_bad_pointer(void);
 
 #define system_has_cmpxchg_double()	1
 
-#include <asm-generic/cmpxchg-local.h>
-
-static inline unsigned long __cmpxchg_local(void *ptr,
-					    unsigned long old,
-					    unsigned long new, int size)
-{
-	switch (size) {
-	case 1:
-	case 2:
-	case 4:
-#ifdef CONFIG_64BIT
-	case 8:
-#endif
-		return __cmpxchg(ptr, old, new, size);
-	default:
-		return __cmpxchg_local_generic(ptr, old, new, size);
-	}
-
-	return old;
-}
-
-/*
- * cmpxchg_local and cmpxchg64_local are atomic wrt current CPU. Always make
- * them available.
- */
-#define cmpxchg_local(ptr, o, n)					\
-({									\
-	__typeof__(*(ptr)) __ret;					\
-	__ret = (__typeof__(*(ptr)))					\
-		__cmpxchg_local((ptr), (unsigned long)(o),		\
-				(unsigned long)(n), sizeof(*(ptr)));	\
-	__ret;								\
-})
-
-#define cmpxchg64_local(ptr, o, n)	cmpxchg64((ptr), (o), (n))
-
 #endif /* __ASM_CMPXCHG_H */
diff --git a/arch/s390/include/asm/cputime.h b/arch/s390/include/asm/cputime.h
index f8c196984853..b91e960e4045 100644
--- a/arch/s390/include/asm/cputime.h
+++ b/arch/s390/include/asm/cputime.h
@@ -10,6 +10,8 @@
 #include <linux/types.h>
 #include <asm/div64.h>
 
+#define CPUTIME_PER_USEC 4096ULL
+#define CPUTIME_PER_SEC (CPUTIME_PER_USEC * USEC_PER_SEC)
 
 /* We want to use full resolution of the CPU timer: 2**-12 micro-seconds. */
 
@@ -38,24 +40,24 @@ static inline unsigned long __div(unsigned long long n, unsigned long base)
  */
 static inline unsigned long cputime_to_jiffies(const cputime_t cputime)
 {
-	return __div((__force unsigned long long) cputime, 4096000000ULL / HZ);
+	return __div((__force unsigned long long) cputime, CPUTIME_PER_SEC / HZ);
 }
 
 static inline cputime_t jiffies_to_cputime(const unsigned int jif)
 {
-	return (__force cputime_t)(jif * (4096000000ULL / HZ));
+	return (__force cputime_t)(jif * (CPUTIME_PER_SEC / HZ));
 }
 
 static inline u64 cputime64_to_jiffies64(cputime64_t cputime)
 {
 	unsigned long long jif = (__force unsigned long long) cputime;
-	do_div(jif, 4096000000ULL / HZ);
+	do_div(jif, CPUTIME_PER_SEC / HZ);
 	return jif;
 }
 
 static inline cputime64_t jiffies64_to_cputime64(const u64 jif)
 {
-	return (__force cputime64_t)(jif * (4096000000ULL / HZ));
+	return (__force cputime64_t)(jif * (CPUTIME_PER_SEC / HZ));
 }
 
 /*
@@ -68,7 +70,7 @@ static inline unsigned int cputime_to_usecs(const cputime_t cputime)
 
 static inline cputime_t usecs_to_cputime(const unsigned int m)
 {
-	return (__force cputime_t)(m * 4096ULL);
+	return (__force cputime_t)(m * CPUTIME_PER_USEC);
 }
 
 #define usecs_to_cputime64(m)		usecs_to_cputime(m)
@@ -78,12 +80,12 @@ static inline cputime_t usecs_to_cputime(const unsigned int m)
  */
 static inline unsigned int cputime_to_secs(const cputime_t cputime)
 {
-	return __div((__force unsigned long long) cputime, 2048000000) >> 1;
+	return __div((__force unsigned long long) cputime, CPUTIME_PER_SEC / 2) >> 1;
 }
 
 static inline cputime_t secs_to_cputime(const unsigned int s)
 {
-	return (__force cputime_t)(s * 4096000000ULL);
+	return (__force cputime_t)(s * CPUTIME_PER_SEC);
 }
 
 /*
@@ -91,8 +93,8 @@ static inline cputime_t secs_to_cputime(const unsigned int s)
  */
 static inline cputime_t timespec_to_cputime(const struct timespec *value)
 {
-	unsigned long long ret = value->tv_sec * 4096000000ULL;
-	return (__force cputime_t)(ret + value->tv_nsec * 4096 / 1000);
+	unsigned long long ret = value->tv_sec * CPUTIME_PER_SEC;
+	return (__force cputime_t)(ret + __div(value->tv_nsec * CPUTIME_PER_USEC, NSEC_PER_USEC));
 }
 
 static inline void cputime_to_timespec(const cputime_t cputime,
@@ -103,12 +105,12 @@ static inline void cputime_to_timespec(const cputime_t cputime,
 	register_pair rp;
 
 	rp.pair = __cputime >> 1;
-	asm ("dr %0,%1" : "+d" (rp) : "d" (2048000000UL));
-	value->tv_nsec = rp.subreg.even * 1000 / 4096;
+	asm ("dr %0,%1" : "+d" (rp) : "d" (CPUTIME_PER_SEC / 2));
+	value->tv_nsec = rp.subreg.even * NSEC_PER_USEC / CPUTIME_PER_USEC;
 	value->tv_sec = rp.subreg.odd;
 #else
-	value->tv_nsec = (__cputime % 4096000000ULL) * 1000 / 4096;
-	value->tv_sec = __cputime / 4096000000ULL;
+	value->tv_nsec = (__cputime % CPUTIME_PER_SEC) * NSEC_PER_USEC / CPUTIME_PER_USEC;
+	value->tv_sec = __cputime / CPUTIME_PER_SEC;
 #endif
 }
 
@@ -119,8 +121,8 @@ static inline void cputime_to_timespec(const cputime_t cputime,
  */
 static inline cputime_t timeval_to_cputime(const struct timeval *value)
 {
-	unsigned long long ret = value->tv_sec * 4096000000ULL;
-	return (__force cputime_t)(ret + value->tv_usec * 4096ULL);
+	unsigned long long ret = value->tv_sec * CPUTIME_PER_SEC;
+	return (__force cputime_t)(ret + value->tv_usec * CPUTIME_PER_USEC);
 }
 
 static inline void cputime_to_timeval(const cputime_t cputime,
@@ -131,12 +133,12 @@ static inline void cputime_to_timeval(const cputime_t cputime,
 	register_pair rp;
 
 	rp.pair = __cputime >> 1;
-	asm ("dr %0,%1" : "+d" (rp) : "d" (2048000000UL));
-	value->tv_usec = rp.subreg.even / 4096;
+	asm ("dr %0,%1" : "+d" (rp) : "d" (CPUTIME_PER_USEC / 2));
+	value->tv_usec = rp.subreg.even / CPUTIME_PER_USEC;
 	value->tv_sec = rp.subreg.odd;
 #else
-	value->tv_usec = (__cputime % 4096000000ULL) / 4096;
-	value->tv_sec = __cputime / 4096000000ULL;
+	value->tv_usec = (__cputime % CPUTIME_PER_SEC) / CPUTIME_PER_USEC;
+	value->tv_sec = __cputime / CPUTIME_PER_SEC;
 #endif
 }
 
@@ -146,13 +148,13 @@ static inline void cputime_to_timeval(const cputime_t cputime,
 static inline clock_t cputime_to_clock_t(cputime_t cputime)
 {
 	unsigned long long clock = (__force unsigned long long) cputime;
-	do_div(clock, 4096000000ULL / USER_HZ);
+	do_div(clock, CPUTIME_PER_SEC / USER_HZ);
 	return clock;
 }
 
 static inline cputime_t clock_t_to_cputime(unsigned long x)
 {
-	return (__force cputime_t)(x * (4096000000ULL / USER_HZ));
+	return (__force cputime_t)(x * (CPUTIME_PER_SEC / USER_HZ));
 }
 
 /*
@@ -161,7 +163,7 @@ static inline cputime_t clock_t_to_cputime(unsigned long x)
 static inline clock_t cputime64_to_clock_t(cputime64_t cputime)
 {
 	unsigned long long clock = (__force unsigned long long) cputime;
-	do_div(clock, 4096000000ULL / USER_HZ);
+	do_div(clock, CPUTIME_PER_SEC / USER_HZ);
 	return clock;
 }
 
diff --git a/arch/s390/include/asm/debug.h b/arch/s390/include/asm/debug.h
index 530c15eb01e9..0206c8052328 100644
--- a/arch/s390/include/asm/debug.h
+++ b/arch/s390/include/asm/debug.h
@@ -151,9 +151,21 @@ debug_text_event(debug_info_t* id, int level, const char* txt)
  * stored in the s390dbf. See Documentation/s390/s390dbf.txt for more details!
  */
 extern 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, ...)
 	__attribute__ ((format(printf, 3, 4)));
 
+#define debug_sprintf_event(_id, _level, _fmt, ...)			\
+({									\
+	debug_entry_t *__ret;						\
+	debug_info_t *__id = _id;					\
+	int __level = _level;						\
+	if ((!__id) || (__level > __id->level))				\
+		__ret = NULL;						\
+	else								\
+		__ret = __debug_sprintf_event(__id, __level,		\
+					      _fmt, ## __VA_ARGS__);	\
+	__ret;								\
+})
 
 static inline debug_entry_t*
 debug_exception(debug_info_t* id, int level, void* data, int length)
@@ -194,9 +206,22 @@ debug_text_exception(debug_info_t* id, int level, const char* txt)
  * stored in the s390dbf. See Documentation/s390/s390dbf.txt for more details!
  */
 extern 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, ...)
 	__attribute__ ((format(printf, 3, 4)));
 
+#define debug_sprintf_exception(_id, _level, _fmt, ...)			\
+({									\
+	debug_entry_t *__ret;						\
+	debug_info_t *__id = _id;					\
+	int __level = _level;						\
+	if ((!__id) || (__level > __id->level))				\
+		__ret = NULL;						\
+	else								\
+		__ret = __debug_sprintf_exception(__id, __level,	\
+						  _fmt, ## __VA_ARGS__);\
+	__ret;								\
+})
+
 int debug_register_view(debug_info_t* id, struct debug_view* view);
 int debug_unregister_view(debug_info_t* id, struct debug_view* view);
 
diff --git a/arch/s390/include/asm/ftrace.h b/arch/s390/include/asm/ftrace.h
index 3aef8afec336..abb618f1ead2 100644
--- a/arch/s390/include/asm/ftrace.h
+++ b/arch/s390/include/asm/ftrace.h
@@ -1,25 +1,69 @@
 #ifndef _ASM_S390_FTRACE_H
 #define _ASM_S390_FTRACE_H
 
+#define ARCH_SUPPORTS_FTRACE_OPS 1
+
+#define MCOUNT_INSN_SIZE	24
+#define MCOUNT_RETURN_FIXUP	18
+
 #ifndef __ASSEMBLY__
 
-extern void _mcount(void);
+#define ftrace_return_address(n) __builtin_return_address(n)
+
+void _mcount(void);
+void ftrace_caller(void);
+
 extern char ftrace_graph_caller_end;
+extern unsigned long ftrace_plt;
 
 struct dyn_arch_ftrace { };
 
-#define MCOUNT_ADDR ((long)_mcount)
+#define MCOUNT_ADDR ((unsigned long)_mcount)
+#define FTRACE_ADDR ((unsigned long)ftrace_caller)
 
+#define KPROBE_ON_FTRACE_NOP	0
+#define KPROBE_ON_FTRACE_CALL	1
 
 static inline unsigned long ftrace_call_adjust(unsigned long addr)
 {
 	return addr;
 }
 
-#endif /* __ASSEMBLY__ */
+struct ftrace_insn {
+	u16 opc;
+	s32 disp;
+} __packed;
+
+static inline void ftrace_generate_nop_insn(struct ftrace_insn *insn)
+{
+#ifdef CONFIG_FUNCTION_TRACER
+	/* jg .+24 */
+	insn->opc = 0xc0f4;
+	insn->disp = MCOUNT_INSN_SIZE / 2;
+#endif
+}
 
-#define MCOUNT_INSN_SIZE  18
+static inline int is_ftrace_nop(struct ftrace_insn *insn)
+{
+#ifdef CONFIG_FUNCTION_TRACER
+	if (insn->disp == MCOUNT_INSN_SIZE / 2)
+		return 1;
+#endif
+	return 0;
+}
 
-#define ARCH_SUPPORTS_FTRACE_OPS 1
+static inline void ftrace_generate_call_insn(struct ftrace_insn *insn,
+					     unsigned long ip)
+{
+#ifdef CONFIG_FUNCTION_TRACER
+	unsigned long target;
+
+	/* brasl r0,ftrace_caller */
+	target = is_module_addr((void *) ip) ? ftrace_plt : FTRACE_ADDR;
+	insn->opc = 0xc005;
+	insn->disp = (target - ip) / 2;
+#endif
+}
 
+#endif /* __ASSEMBLY__ */
 #endif /* _ASM_S390_FTRACE_H */
diff --git a/arch/s390/include/asm/idle.h b/arch/s390/include/asm/idle.h
index 6af037f574b8..113cd963dbbe 100644
--- a/arch/s390/include/asm/idle.h
+++ b/arch/s390/include/asm/idle.h
@@ -9,9 +9,10 @@
 
 #include <linux/types.h>
 #include <linux/device.h>
+#include <linux/seqlock.h>
 
 struct s390_idle_data {
-	unsigned int sequence;
+	seqcount_t seqcount;
 	unsigned long long idle_count;
 	unsigned long long idle_time;
 	unsigned long long clock_idle_enter;
diff --git a/arch/s390/include/asm/io.h b/arch/s390/include/asm/io.h
index 6ad9013c67e7..30fd5c84680e 100644
--- a/arch/s390/include/asm/io.h
+++ b/arch/s390/include/asm/io.h
@@ -39,6 +39,15 @@ static inline void iounmap(volatile void __iomem *addr)
 {
 }
 
+static inline void __iomem *ioport_map(unsigned long port, unsigned int nr)
+{
+	return NULL;
+}
+
+static inline void ioport_unmap(void __iomem *p)
+{
+}
+
 /*
  * s390 needs a private implementation of pci_iomap since ioremap with its
  * offset parameter isn't sufficient. That's because BAR spaces are not
diff --git a/arch/s390/include/asm/irq.h b/arch/s390/include/asm/irq.h
index b0d5f0a97a01..343ea7c987aa 100644
--- a/arch/s390/include/asm/irq.h
+++ b/arch/s390/include/asm/irq.h
@@ -1,11 +1,11 @@
 #ifndef _ASM_IRQ_H
 #define _ASM_IRQ_H
 
-#define EXT_INTERRUPT	1
-#define IO_INTERRUPT	2
-#define THIN_INTERRUPT	3
+#define EXT_INTERRUPT	0
+#define IO_INTERRUPT	1
+#define THIN_INTERRUPT	2
 
-#define NR_IRQS_BASE	4
+#define NR_IRQS_BASE	3
 
 #ifdef CONFIG_PCI_NR_MSI
 # define NR_IRQS	(NR_IRQS_BASE + CONFIG_PCI_NR_MSI)
@@ -13,9 +13,6 @@
 # define NR_IRQS	NR_IRQS_BASE
 #endif
 
-/* This number is used when no interrupt has been assigned */
-#define NO_IRQ		0
-
 /* External interruption codes */
 #define EXT_IRQ_INTERRUPT_KEY	0x0040
 #define EXT_IRQ_CLK_COMP	0x1004
diff --git a/arch/s390/include/asm/kprobes.h b/arch/s390/include/asm/kprobes.h
index 98629173ce3b..b47ad3b642cc 100644
--- a/arch/s390/include/asm/kprobes.h
+++ b/arch/s390/include/asm/kprobes.h
@@ -60,6 +60,7 @@ typedef u16 kprobe_opcode_t;
 struct arch_specific_insn {
 	/* copy of original instruction */
 	kprobe_opcode_t *insn;
+	unsigned int is_ftrace_insn : 1;
 };
 
 struct prev_kprobe {
diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h
index 6cc51fe84410..34fbcac61133 100644
--- a/arch/s390/include/asm/lowcore.h
+++ b/arch/s390/include/asm/lowcore.h
@@ -147,7 +147,7 @@ struct _lowcore {
 	__u32	softirq_pending;		/* 0x02ec */
 	__u32	percpu_offset;			/* 0x02f0 */
 	__u32	machine_flags;			/* 0x02f4 */
-	__u32	ftrace_func;			/* 0x02f8 */
+	__u8	pad_0x02f8[0x02fc-0x02f8];	/* 0x02f8 */
 	__u32	spinlock_lockval;		/* 0x02fc */
 
 	__u8	pad_0x0300[0x0e00-0x0300];	/* 0x0300 */
@@ -297,7 +297,7 @@ struct _lowcore {
 	__u64	percpu_offset;			/* 0x0378 */
 	__u64	vdso_per_cpu_data;		/* 0x0380 */
 	__u64	machine_flags;			/* 0x0388 */
-	__u64	ftrace_func;			/* 0x0390 */
+	__u8	pad_0x0390[0x0398-0x0390];	/* 0x0390 */
 	__u64	gmap;				/* 0x0398 */
 	__u32	spinlock_lockval;		/* 0x03a0 */
 	__u8	pad_0x03a0[0x0400-0x03a4];	/* 0x03a4 */
diff --git a/arch/s390/include/asm/pci.h b/arch/s390/include/asm/pci.h
index c030900320e0..ef803c202d42 100644
--- a/arch/s390/include/asm/pci.h
+++ b/arch/s390/include/asm/pci.h
@@ -50,10 +50,6 @@ struct zpci_fmb {
 	atomic64_t unmapped_pages;
 } __packed __aligned(16);
 
-#define ZPCI_MSI_VEC_BITS	11
-#define ZPCI_MSI_VEC_MAX	(1 << ZPCI_MSI_VEC_BITS)
-#define ZPCI_MSI_VEC_MASK	(ZPCI_MSI_VEC_MAX - 1)
-
 enum zpci_state {
 	ZPCI_FN_STATE_RESERVED,
 	ZPCI_FN_STATE_STANDBY,
@@ -90,6 +86,7 @@ struct zpci_dev {
 
 	/* IRQ stuff */
 	u64		msi_addr;	/* MSI address */
+	unsigned int	max_msi;	/* maximum number of MSI's */
 	struct airq_iv *aibv;		/* adapter interrupt bit vector */
 	unsigned int	aisb;		/* number of the summary bit */
 
diff --git a/arch/s390/include/asm/pci_io.h b/arch/s390/include/asm/pci_io.h
index d194d544d694..f664e96f48c7 100644
--- a/arch/s390/include/asm/pci_io.h
+++ b/arch/s390/include/asm/pci_io.h
@@ -139,7 +139,8 @@ static inline int zpci_memcpy_fromio(void *dst,
 	int size, rc = 0;
 
 	while (n > 0) {
-		size = zpci_get_max_write_size((u64) src, (u64) dst, n, 8);
+		size = zpci_get_max_write_size((u64 __force) src,
+					       (u64) dst, n, 8);
 		req = ZPCI_CREATE_REQ(entry->fh, entry->bar, size);
 		rc = zpci_read_single(req, dst, offset, size);
 		if (rc)
@@ -162,7 +163,8 @@ static inline int zpci_memcpy_toio(volatile void __iomem *dst,
 		return -EINVAL;
 
 	while (n > 0) {
-		size = zpci_get_max_write_size((u64) dst, (u64) src, n, 128);
+		size = zpci_get_max_write_size((u64 __force) dst,
+					       (u64) src, n, 128);
 		req = ZPCI_CREATE_REQ(entry->fh, entry->bar, size);
 
 		if (size > 8) /* main path */
diff --git a/arch/s390/include/asm/pgalloc.h b/arch/s390/include/asm/pgalloc.h
index d39a31c3cdf2..e510b9460efa 100644
--- a/arch/s390/include/asm/pgalloc.h
+++ b/arch/s390/include/asm/pgalloc.h
@@ -22,8 +22,6 @@ unsigned long *page_table_alloc(struct mm_struct *);
 void page_table_free(struct mm_struct *, unsigned long *);
 void page_table_free_rcu(struct mmu_gather *, unsigned long *, unsigned long);
 
-void page_table_reset_pgste(struct mm_struct *, unsigned long, unsigned long,
-			    bool init_skey);
 int set_guest_storage_key(struct mm_struct *mm, unsigned long addr,
 			  unsigned long key, bool nq);
 
diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
index 57c882761dea..5e102422c9ab 100644
--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h
@@ -133,6 +133,18 @@ extern unsigned long MODULES_END;
 #define MODULES_LEN	(1UL << 31)
 #endif
 
+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;
+}
+
 /*
  * A 31 bit pagetable entry of S390 has following format:
  *  |   PFRA          |    |  OS  |
@@ -479,6 +491,11 @@ static inline int mm_has_pgste(struct mm_struct *mm)
 	return 0;
 }
 
+/*
+ * In the case that a guest uses storage keys
+ * faults should no longer be backed by zero pages
+ */
+#define mm_forbids_zeropage mm_use_skey
 static inline int mm_use_skey(struct mm_struct *mm)
 {
 #ifdef CONFIG_PGSTE
@@ -1634,6 +1651,19 @@ static inline pmd_t pmdp_get_and_clear(struct mm_struct *mm,
 	return pmd;
 }
 
+#define __HAVE_ARCH_PMDP_GET_AND_CLEAR_FULL
+static inline pmd_t pmdp_get_and_clear_full(struct mm_struct *mm,
+					    unsigned long address,
+					    pmd_t *pmdp, int full)
+{
+	pmd_t pmd = *pmdp;
+
+	if (!full)
+		pmdp_flush_lazy(mm, address, pmdp);
+	pmd_clear(pmdp);
+	return pmd;
+}
+
 #define __HAVE_ARCH_PMDP_CLEAR_FLUSH
 static inline pmd_t pmdp_clear_flush(struct vm_area_struct *vma,
 				     unsigned long address, pmd_t *pmdp)
@@ -1746,7 +1776,8 @@ static inline pte_t mk_swap_pte(unsigned long type, unsigned long offset)
 extern int vmem_add_mapping(unsigned long start, unsigned long size);
 extern int vmem_remove_mapping(unsigned long start, unsigned long size);
 extern int s390_enable_sie(void);
-extern void s390_enable_skey(void);
+extern int s390_enable_skey(void);
+extern void s390_reset_cmma(struct mm_struct *mm);
 
 /*
  * No page table caches to initialise
diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h
index d559bdb03d18..bed05ea7ec27 100644
--- a/arch/s390/include/asm/processor.h
+++ b/arch/s390/include/asm/processor.h
@@ -217,8 +217,6 @@ static inline unsigned short stap(void)
  */
 static inline void cpu_relax(void)
 {
-	if (MACHINE_HAS_DIAG44)
-		asm volatile("diag 0,0,68");
 	barrier();
 }
 
diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h
index d6bdf906caa5..0e37cd041241 100644
--- a/arch/s390/include/asm/spinlock.h
+++ b/arch/s390/include/asm/spinlock.h
@@ -18,14 +18,7 @@ extern int spin_retry;
 static inline int
 _raw_compare_and_swap(unsigned int *lock, unsigned int old, unsigned int new)
 {
-	unsigned int old_expected = old;
-
-	asm volatile(
-		"	cs	%0,%3,%1"
-		: "=d" (old), "=Q" (*lock)
-		: "0" (old), "d" (new), "Q" (*lock)
-		: "cc", "memory" );
-	return old == old_expected;
+	return __sync_bool_compare_and_swap(lock, old, new);
 }
 
 /*
diff --git a/arch/s390/include/asm/tlb.h b/arch/s390/include/asm/tlb.h
index 572c59949004..06d8741ad6f4 100644
--- a/arch/s390/include/asm/tlb.h
+++ b/arch/s390/include/asm/tlb.h
@@ -121,6 +121,7 @@ static inline void pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd,
 #ifdef CONFIG_64BIT
 	if (tlb->mm->context.asce_limit <= (1UL << 31))
 		return;
+	pgtable_pmd_page_dtor(virt_to_page(pmd));
 	tlb_remove_table(tlb, pmd);
 #endif
 }
diff --git a/arch/s390/include/uapi/asm/unistd.h b/arch/s390/include/uapi/asm/unistd.h
index 4197c89c52d4..2b446cf0cc65 100644
--- a/arch/s390/include/uapi/asm/unistd.h
+++ b/arch/s390/include/uapi/asm/unistd.h
@@ -287,7 +287,9 @@
 #define __NR_getrandom		349
 #define __NR_memfd_create	350
 #define __NR_bpf		351
-#define NR_syscalls 352
+#define __NR_s390_pci_mmio_write	352
+#define __NR_s390_pci_mmio_read		353
+#define NR_syscalls 354
 
 /* 
  * There are some system calls that are not present on 64 bit, some