26 files changed, 438 insertions, 451 deletions

diff --git a/include/asm-s390/airq.h b/include/asm-s390/airq.h
new file mode 100644
index 000000000000..41d028cb52a4
--- /dev/null
+++ b/include/asm-s390/airq.h
@@ -0,0 +1,19 @@
+/*
+ *  include/asm-s390/airq.h
+ *
+ *    Copyright IBM Corp. 2002,2007
+ *    Author(s): Ingo Adlung <adlung@de.ibm.com>
+ *		 Cornelia Huck <cornelia.huck@de.ibm.com>
+ *		 Arnd Bergmann <arndb@de.ibm.com>
+ *		 Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
+ */
+
+#ifndef _ASM_S390_AIRQ_H
+#define _ASM_S390_AIRQ_H
+
+typedef void (*adapter_int_handler_t)(void *, void *);
+
+void *s390_register_adapter_interrupt(adapter_int_handler_t, void *);
+void s390_unregister_adapter_interrupt(void *);
+
+#endif /* _ASM_S390_AIRQ_H */
diff --git a/include/asm-s390/bitops.h b/include/asm-s390/bitops.h
index 34d9a6357c38..882db054110c 100644
--- a/include/asm-s390/bitops.h
+++ b/include/asm-s390/bitops.h
@@ -440,242 +440,256 @@ __constant_test_bit(unsigned long nr, const volatile unsigned long *addr) {
  __test_bit((nr),(addr)) )
 
 /*
- * ffz = Find First Zero in word. Undefined if no zero exists,
- * so code should check against ~0UL first..
+ * Optimized find bit helper functions.
  */
-static inline unsigned long ffz(unsigned long word)
+
+/**
+ * __ffz_word_loop - find byte offset of first long != -1UL
+ * @addr: pointer to array of unsigned long
+ * @size: size of the array in bits
+ */
+static inline unsigned long __ffz_word_loop(const unsigned long *addr,
+					    unsigned long size)
 {
-        unsigned long bit = 0;
+	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
+	unsigned long bytes = 0;
 
+	asm volatile(
+#ifndef __s390x__
+		"	ahi	%1,31\n"
+		"	srl	%1,5\n"
+		"0:	c	%2,0(%0,%3)\n"
+		"	jne	1f\n"
+		"	la	%0,4(%0)\n"
+		"	brct	%1,0b\n"
+		"1:\n"
+#else
+		"	aghi	%1,63\n"
+		"	srlg	%1,%1,6\n"
+		"0:	cg	%2,0(%0,%3)\n"
+		"	jne	1f\n"
+		"	la	%0,8(%0)\n"
+		"	brct	%1,0b\n"
+		"1:\n"
+#endif
+		: "+a" (bytes), "+d" (size)
+		: "d" (-1UL), "a" (addr), "m" (*(addrtype *) addr)
+		: "cc" );
+	return bytes;
+}
+
+/**
+ * __ffs_word_loop - find byte offset of first long != 0UL
+ * @addr: pointer to array of unsigned long
+ * @size: size of the array in bits
+ */
+static inline unsigned long __ffs_word_loop(const unsigned long *addr,
+					    unsigned long size)
+{
+	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
+	unsigned long bytes = 0;
+
+	asm volatile(
+#ifndef __s390x__
+		"	ahi	%1,31\n"
+		"	srl	%1,5\n"
+		"0:	c	%2,0(%0,%3)\n"
+		"	jne	1f\n"
+		"	la	%0,4(%0)\n"
+		"	brct	%1,0b\n"
+		"1:\n"
+#else
+		"	aghi	%1,63\n"
+		"	srlg	%1,%1,6\n"
+		"0:	cg	%2,0(%0,%3)\n"
+		"	jne	1f\n"
+		"	la	%0,8(%0)\n"
+		"	brct	%1,0b\n"
+		"1:\n"
+#endif
+		: "+a" (bytes), "+a" (size)
+		: "d" (0UL), "a" (addr), "m" (*(addrtype *) addr)
+		: "cc" );
+	return bytes;
+}
+
+/**
+ * __ffz_word - add number of the first unset bit
+ * @nr: base value the bit number is added to
+ * @word: the word that is searched for unset bits
+ */
+static inline unsigned long __ffz_word(unsigned long nr, unsigned long word)
+{
 #ifdef __s390x__
 	if (likely((word & 0xffffffff) == 0xffffffff)) {
 		word >>= 32;
-		bit += 32;
+		nr += 32;
 	}
 #endif
 	if (likely((word & 0xffff) == 0xffff)) {
 		word >>= 16;
-		bit += 16;
+		nr += 16;
 	}
 	if (likely((word & 0xff) == 0xff)) {
 		word >>= 8;
-		bit += 8;
+		nr += 8;
 	}
-	return bit + _zb_findmap[word & 0xff];
+	return nr + _zb_findmap[(unsigned char) word];
 }
 
-/*
- * __ffs = find first bit in word. Undefined if no bit exists,
- * so code should check against 0UL first..
+/**
+ * __ffs_word - add number of the first set bit
+ * @nr: base value the bit number is added to
+ * @word: the word that is searched for set bits
  */
-static inline unsigned long __ffs (unsigned long word)
+static inline unsigned long __ffs_word(unsigned long nr, unsigned long word)
 {
-	unsigned long bit = 0;
-
 #ifdef __s390x__
 	if (likely((word & 0xffffffff) == 0)) {
 		word >>= 32;
-		bit += 32;
+		nr += 32;
 	}
 #endif
 	if (likely((word & 0xffff) == 0)) {
 		word >>= 16;
-		bit += 16;
+		nr += 16;
 	}
 	if (likely((word & 0xff) == 0)) {
 		word >>= 8;
-		bit += 8;
+		nr += 8;
 	}
-	return bit + _sb_findmap[word & 0xff];
+	return nr + _sb_findmap[(unsigned char) word];
 }
 
-/*
- * Find-bit routines..
- */
 
-#ifndef __s390x__
+/**
+ * __load_ulong_be - load big endian unsigned long
+ * @p: pointer to array of unsigned long
+ * @offset: byte offset of source value in the array
+ */
+static inline unsigned long __load_ulong_be(const unsigned long *p,
+					    unsigned long offset)
+{
+	p = (unsigned long *)((unsigned long) p + offset);
+	return *p;
+}
 
-static inline int
-find_first_zero_bit(const unsigned long * addr, unsigned long size)
+/**
+ * __load_ulong_le - load little endian unsigned long
+ * @p: pointer to array of unsigned long
+ * @offset: byte offset of source value in the array
+ */
+static inline unsigned long __load_ulong_le(const unsigned long *p,
+					    unsigned long offset)
 {
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-	unsigned long cmp, count;
-        unsigned int res;
+	unsigned long word;
 
-        if (!size)
-                return 0;
+	p = (unsigned long *)((unsigned long) p + offset);
+#ifndef __s390x__
 	asm volatile(
-		"	lhi	%1,-1\n"
-		"	lr	%2,%3\n"
-		"	slr	%0,%0\n"
-		"	ahi	%2,31\n"
-		"	srl	%2,5\n"
-		"0:	c	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	la	%0,4(%0)\n"
-		"	brct	%2,0b\n"
-		"	lr	%0,%3\n"
-		"	j	4f\n"
-		"1:	l	%2,0(%0,%4)\n"
-		"	sll	%0,3\n"
-		"	lhi	%1,0xff\n"
-		"	tml	%2,0xffff\n"
-		"	jno	2f\n"
-		"	ahi	%0,16\n"
-		"	srl	%2,16\n"
-		"2:	tml	%2,0x00ff\n"
-		"	jno	3f\n"
-		"	ahi	%0,8\n"
-		"	srl	%2,8\n"
-		"3:	nr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	alr	%0,%2\n"
-		"4:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (addr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) addr) : "cc");
-        return (res < size) ? res : size;
+		"	ic	%0,0(%1)\n"
+		"	icm	%0,2,1(%1)\n"
+		"	icm	%0,4,2(%1)\n"
+		"	icm	%0,8,3(%1)"
+		: "=&d" (word) : "a" (p), "m" (*p) : "cc");
+#else
+	asm volatile(
+		"	lrvg	%0,%1"
+		: "=d" (word) : "m" (*p) );
+#endif
+	return word;
 }
 
-static inline int
-find_first_bit(const unsigned long * addr, unsigned long size)
+/*
+ * The various find bit functions.
+ */
+
+/*
+ * ffz - find first zero in word.
+ * @word: The word to search
+ *
+ * Undefined if no zero exists, so code should check against ~0UL first.
+ */
+static inline unsigned long ffz(unsigned long word)
 {
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-	unsigned long cmp, count;
-        unsigned int res;
+	return __ffz_word(0, word);
+}
 
-        if (!size)
-                return 0;
-	asm volatile(
-		"	slr	%1,%1\n"
-		"	lr	%2,%3\n"
-		"	slr	%0,%0\n"
-		"	ahi	%2,31\n"
-		"	srl	%2,5\n"
-		"0:	c	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	la	%0,4(%0)\n"
-		"	brct	%2,0b\n"
-		"	lr	%0,%3\n"
-		"	j	4f\n"
-		"1:	l	%2,0(%0,%4)\n"
-		"	sll	%0,3\n"
-		"	lhi	%1,0xff\n"
-		"	tml	%2,0xffff\n"
-		"	jnz	2f\n"
-		"	ahi	%0,16\n"
-		"	srl	%2,16\n"
-		"2:	tml	%2,0x00ff\n"
-		"	jnz	3f\n"
-		"	ahi	%0,8\n"
-		"	srl	%2,8\n"
-		"3:	nr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	alr	%0,%2\n"
-		"4:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (addr), "a" (&_sb_findmap),
-		  "m" (*(addrtype *) addr) : "cc");
-        return (res < size) ? res : size;
+/**
+ * __ffs - find first bit in word.
+ * @word: The word to search
+ *
+ * Undefined if no bit exists, so code should check against 0 first.
+ */
+static inline unsigned long __ffs (unsigned long word)
+{
+	return __ffs_word(0, word);
 }
 
-#else /* __s390x__ */
+/**
+ * ffs - find first bit set
+ * @x: the word to search
+ *
+ * This is defined the same way as
+ * the libc and compiler builtin ffs routines, therefore
+ * differs in spirit from the above ffz (man ffs).
+ */
+static inline int ffs(int x)
+{
+	if (!x)
+		return 0;
+	return __ffs_word(1, x);
+}
 
-static inline unsigned long
-find_first_zero_bit(const unsigned long * addr, unsigned long size)
+/**
+ * find_first_zero_bit - find the first zero bit in a memory region
+ * @addr: The address to start the search at
+ * @size: The maximum size to search
+ *
+ * Returns the bit-number of the first zero bit, not the number of the byte
+ * containing a bit.
+ */
+static inline unsigned long find_first_zero_bit(const unsigned long *addr,
+						unsigned long size)
 {
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-        unsigned long res, cmp, count;
+	unsigned long bytes, bits;
 
         if (!size)
                 return 0;
-	asm volatile(
-		"	lghi	%1,-1\n"
-		"	lgr	%2,%3\n"
-		"	slgr	%0,%0\n"
-		"	aghi	%2,63\n"
-		"	srlg	%2,%2,6\n"
-		"0:	cg	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	la	%0,8(%0)\n"
-		"	brct	%2,0b\n"
-		"	lgr	%0,%3\n"
-		"	j	5f\n"
-		"1:	lg	%2,0(%0,%4)\n"
-		"	sllg	%0,%0,3\n"
-		"	clr	%2,%1\n"
-		"	jne	2f\n"
-		"	aghi	%0,32\n"
-		"	srlg	%2,%2,32\n"
-		"2:	lghi	%1,0xff\n"
-		"	tmll	%2,0xffff\n"
-		"	jno	3f\n"
-		"	aghi	%0,16\n"
-		"	srl	%2,16\n"
-		"3:	tmll	%2,0x00ff\n"
-		"	jno	4f\n"
-		"	aghi	%0,8\n"
-		"	srl	%2,8\n"
-		"4:	ngr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	algr	%0,%2\n"
-		"5:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (addr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) addr) : "cc");
-        return (res < size) ? res : size;
-}
-
-static inline unsigned long
-find_first_bit(const unsigned long * addr, unsigned long size)
+	bytes = __ffz_word_loop(addr, size);
+	bits = __ffz_word(bytes*8, __load_ulong_be(addr, bytes));
+	return (bits < size) ? bits : size;
+}
+
+/**
+ * find_first_bit - find the first set bit in a memory region
+ * @addr: The address to start the search at
+ * @size: The maximum size to search
+ *
+ * Returns the bit-number of the first set bit, not the number of the byte
+ * containing a bit.
+ */
+static inline unsigned long find_first_bit(const unsigned long * addr,
+					   unsigned long size)
 {
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-        unsigned long res, cmp, count;
+	unsigned long bytes, bits;
 
         if (!size)
                 return 0;
-	asm volatile(
-		"	slgr	%1,%1\n"
-		"	lgr	%2,%3\n"
-		"	slgr	%0,%0\n"
-		"	aghi	%2,63\n"
-		"	srlg	%2,%2,6\n"
-		"0:	cg	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	aghi	%0,8\n"
-		"	brct	%2,0b\n"
-		"	lgr	%0,%3\n"
-		"	j	5f\n"
-		"1:	lg	%2,0(%0,%4)\n"
-		"	sllg	%0,%0,3\n"
-		"	clr	%2,%1\n"
-		"	jne	2f\n"
-		"	aghi	%0,32\n"
-		"	srlg	%2,%2,32\n"
-		"2:	lghi	%1,0xff\n"
-		"	tmll	%2,0xffff\n"
-		"	jnz	3f\n"
-		"	aghi	%0,16\n"
-		"	srl	%2,16\n"
-		"3:	tmll	%2,0x00ff\n"
-		"	jnz	4f\n"
-		"	aghi	%0,8\n"
-		"	srl	%2,8\n"
-		"4:	ngr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	algr	%0,%2\n"
-		"5:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (addr), "a" (&_sb_findmap),
-		  "m" (*(addrtype *) addr) : "cc");
-        return (res < size) ? res : size;
+	bytes = __ffs_word_loop(addr, size);
+	bits = __ffs_word(bytes*8, __load_ulong_be(addr, bytes));
+	return (bits < size) ? bits : size;
 }
 
-#endif /* __s390x__ */
-
-static inline int
-find_next_zero_bit (const unsigned long * addr, unsigned long size,
-		    unsigned long offset)
+/**
+ * find_next_zero_bit - find the first zero bit in a memory region
+ * @addr: The address to base the search on
+ * @offset: The bitnumber to start searching at
+ * @size: The maximum size to search
+ */
+static inline int find_next_zero_bit (const unsigned long * addr,
+				      unsigned long size,
+				      unsigned long offset)
 {
         const unsigned long *p;
 	unsigned long bit, set;
@@ -688,10 +702,10 @@ find_next_zero_bit (const unsigned long * addr, unsigned long size,
 	p = addr + offset / __BITOPS_WORDSIZE;
 	if (bit) {
 		/*
-		 * s390 version of ffz returns __BITOPS_WORDSIZE
+		 * __ffz_word returns __BITOPS_WORDSIZE
 		 * if no zero bit is present in the word.
 		 */
-		set = ffz(*p >> bit) + bit;
+		set = __ffz_word(0, *p >> bit) + bit;
 		if (set >= size)
 			return size + offset;
 		if (set < __BITOPS_WORDSIZE)
@@ -703,9 +717,15 @@ find_next_zero_bit (const unsigned long * addr, unsigned long size,
 	return offset + find_first_zero_bit(p, size);
 }
 
-static inline int
-find_next_bit (const unsigned long * addr, unsigned long size,
-	       unsigned long offset)
+/**
+ * find_next_bit - find the first set bit in a memory region
+ * @addr: The address to base the search on
+ * @offset: The bitnumber to start searching at
+ * @size: The maximum size to search
+ */
+static inline int find_next_bit (const unsigned long * addr,
+				 unsigned long size,
+				 unsigned long offset)
 {
         const unsigned long *p;
 	unsigned long bit, set;
@@ -718,10 +738,10 @@ find_next_bit (const unsigned long * addr, unsigned long size,
 	p = addr + offset / __BITOPS_WORDSIZE;
 	if (bit) {
 		/*
-		 * s390 version of __ffs returns __BITOPS_WORDSIZE
+		 * __ffs_word returns __BITOPS_WORDSIZE
 		 * if no one bit is present in the word.
 		 */
-		set = __ffs(*p & (~0UL << bit));
+		set = __ffs_word(0, *p & (~0UL << bit));
 		if (set >= size)
 			return size + offset;
 		if (set < __BITOPS_WORDSIZE)
@@ -744,8 +764,6 @@ static inline int sched_find_first_bit(unsigned long *b)
 	return find_first_bit(b, 140);
 }
 
-#include <asm-generic/bitops/ffs.h>
-
 #include <asm-generic/bitops/fls.h>
 #include <asm-generic/bitops/fls64.h>
 
@@ -773,105 +791,22 @@ static inline int sched_find_first_bit(unsigned long *b)
 #define ext2_test_bit(nr, addr)      \
 	test_bit((nr)^(__BITOPS_WORDSIZE - 8), (unsigned long *)addr)
 
-#ifndef __s390x__
-
-static inline int 
-ext2_find_first_zero_bit(void *vaddr, unsigned int size)
+static inline int ext2_find_first_zero_bit(void *vaddr, unsigned int size)
 {
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-	unsigned long cmp, count;
-        unsigned int res;
+	unsigned long bytes, bits;
 
         if (!size)
                 return 0;
-	asm volatile(
-		"	lhi	%1,-1\n"
-		"	lr	%2,%3\n"
-		"	ahi	%2,31\n"
-		"	srl	%2,5\n"
-		"	slr	%0,%0\n"
-		"0:	cl	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	ahi	%0,4\n"
-		"	brct	%2,0b\n"
-		"	lr	%0,%3\n"
-		"	j	4f\n"
-		"1:	l	%2,0(%0,%4)\n"
-		"	sll	%0,3\n"
-		"	ahi	%0,24\n"
-		"	lhi	%1,0xff\n"
-		"	tmh	%2,0xffff\n"
-		"	jo	2f\n"
-		"	ahi	%0,-16\n"
-		"	srl	%2,16\n"
-		"2:	tml	%2,0xff00\n"
-		"	jo	3f\n"
-		"	ahi	%0,-8\n"
-		"	srl	%2,8\n"
-		"3:	nr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	alr	%0,%2\n"
-		"4:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (vaddr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) vaddr) : "cc");
-        return (res < size) ? res : size;
+	bytes = __ffz_word_loop(vaddr, size);
+	bits = __ffz_word(bytes*8, __load_ulong_le(vaddr, bytes));
+	return (bits < size) ? bits : size;
 }
 
-#else /* __s390x__ */
-
-static inline unsigned long
-ext2_find_first_zero_bit(void *vaddr, unsigned long size)
-{
-	typedef struct { long _[__BITOPS_WORDS(size)]; } addrtype;
-        unsigned long res, cmp, count;
-
-        if (!size)
-                return 0;
-	asm volatile(
-		"	lghi	%1,-1\n"
-		"	lgr	%2,%3\n"
-		"	aghi	%2,63\n"
-		"	srlg	%2,%2,6\n"
-		"	slgr	%0,%0\n"
-		"0:	clg	%1,0(%0,%4)\n"
-		"	jne	1f\n"
-		"	aghi	%0,8\n"
-		"	brct	%2,0b\n"
-		"	lgr	%0,%3\n"
-		"	j	5f\n"
-		"1:	cl	%1,0(%0,%4)\n"
-		"	jne	2f\n"
-		"	aghi	%0,4\n"
-		"2:	l	%2,0(%0,%4)\n"
-		"	sllg	%0,%0,3\n"
-		"	aghi	%0,24\n"
-		"	lghi	%1,0xff\n"
-		"	tmlh	%2,0xffff\n"
-		"	jo	3f\n"
-		"	aghi	%0,-16\n"
-		"	srl	%2,16\n"
-		"3:	tmll	%2,0xff00\n"
-		"	jo	4f\n"
-		"	aghi	%0,-8\n"
-		"	srl	%2,8\n"
-		"4:	ngr	%2,%1\n"
-		"	ic	%2,0(%2,%5)\n"
-		"	algr	%0,%2\n"
-		"5:"
-		: "=&a" (res), "=&d" (cmp), "=&a" (count)
-		: "a" (size), "a" (vaddr), "a" (&_zb_findmap),
-		  "m" (*(addrtype *) vaddr) : "cc");
-        return (res < size) ? res : size;
-}
-
-#endif /* __s390x__ */
-
-static inline int
-ext2_find_next_zero_bit(void *vaddr, unsigned long size, unsigned long offset)
+static inline int ext2_find_next_zero_bit(void *vaddr, unsigned long size,
+					  unsigned long offset)
 {
         unsigned long *addr = vaddr, *p;
-	unsigned long word, bit, set;
+	unsigned long bit, set;
 
         if (offset >= size)
                 return size;
@@ -880,23 +815,11 @@ ext2_find_next_zero_bit(void *vaddr, unsigned long size, unsigned long offset)
 	size -= offset;
 	p = addr + offset / __BITOPS_WORDSIZE;
         if (bit) {
-#ifndef __s390x__
-		asm volatile(
-			"	ic	%0,0(%1)\n"
-			"	icm	%0,2,1(%1)\n"
-			"	icm	%0,4,2(%1)\n"
-			"	icm	%0,8,3(%1)"
-			: "=&a" (word) : "a" (p), "m" (*p) : "cc");
-#else
-		asm volatile(
-			"	lrvg	%0,%1"
-			: "=a" (word) : "m" (*p) );
-#endif
 		/*
 		 * s390 version of ffz returns __BITOPS_WORDSIZE
 		 * if no zero bit is present in the word.
 		 */
-		set = ffz(word >> bit) + bit;
+		set = ffz(__load_ulong_le(p, 0) >> bit) + bit;
 		if (set >= size)
 			return size + offset;
 		if (set < __BITOPS_WORDSIZE)
@@ -908,6 +831,47 @@ ext2_find_next_zero_bit(void *vaddr, unsigned long size, unsigned long offset)
 	return offset + ext2_find_first_zero_bit(p, size);
 }
 
+static inline unsigned long ext2_find_first_bit(void *vaddr,
+						unsigned long size)
+{
+	unsigned long bytes, bits;
+
+	if (!size)
+		return 0;
+	bytes = __ffs_word_loop(vaddr, size);
+	bits = __ffs_word(bytes*8, __load_ulong_le(vaddr, bytes));
+	return (bits < size) ? bits : size;
+}
+
+static inline int ext2_find_next_bit(void *vaddr, unsigned long size,
+				     unsigned long offset)
+{
+	unsigned long *addr = vaddr, *p;
+	unsigned long bit, set;
+
+	if (offset >= size)
+		return size;
+	bit = offset & (__BITOPS_WORDSIZE - 1);
+	offset -= bit;
+	size -= offset;
+	p = addr + offset / __BITOPS_WORDSIZE;
+	if (bit) {
+		/*
+		 * s390 version of ffz returns __BITOPS_WORDSIZE
+		 * if no zero bit is present in the word.
+		 */
+		set = ffs(__load_ulong_le(p, 0) >> bit) + bit;
+		if (set >= size)
+			return size + offset;
+		if (set < __BITOPS_WORDSIZE)
+			return set + offset;
+		offset += __BITOPS_WORDSIZE;
+		size -= __BITOPS_WORDSIZE;
+		p++;
+	}
+	return offset + ext2_find_first_bit(p, size);
+}
+
 #include <asm-generic/bitops/minix.h>
 
 #endif /* __KERNEL__ */
diff --git a/include/asm-s390/cacheflush.h b/include/asm-s390/cacheflush.h
index f7cade8083f3..49d5af916d01 100644
--- a/include/asm-s390/cacheflush.h
+++ b/include/asm-s390/cacheflush.h
@@ -24,4 +24,8 @@
 #define copy_from_user_page(vma, page, vaddr, dst, src, len) \
 	memcpy(dst, src, len)
 
+#ifdef CONFIG_DEBUG_PAGEALLOC
+void kernel_map_pages(struct page *page, int numpages, int enable);
+#endif
+
 #endif /* _S390_CACHEFLUSH_H */
diff --git a/include/asm-s390/ccwgroup.h b/include/asm-s390/ccwgroup.h
index 7109c7cab87e..289053ef5e60 100644
--- a/include/asm-s390/ccwgroup.h
+++ b/include/asm-s390/ccwgroup.h
@@ -37,6 +37,7 @@ struct ccwgroup_device {
  * @remove: function called on remove
  * @set_online: function called when device is set online
  * @set_offline: function called when device is set offline
+ * @shutdown: function called when device is shut down
  * @driver: embedded driver structure
  */
 struct ccwgroup_driver {
@@ -49,6 +50,7 @@ struct ccwgroup_driver {
 	void (*remove) (struct ccwgroup_device *);
 	int (*set_online) (struct ccwgroup_device *);
 	int (*set_offline) (struct ccwgroup_device *);
+	void (*shutdown)(struct ccwgroup_device *);
 
 	struct device_driver driver;
 };
diff --git a/include/asm-s390/cio.h b/include/asm-s390/cio.h
index 2f08c16e44ad..123b557c3ff4 100644
--- a/include/asm-s390/cio.h
+++ b/include/asm-s390/cio.h
@@ -24,8 +24,8 @@
  * @fmt: format
  * @pfch: prefetch
  * @isic: initial-status interruption control
- * @alcc: adress-limit checking control
- * @ssi: supress-suspended interruption
+ * @alcc: address-limit checking control
+ * @ssi: suppress-suspended interruption
  * @zcc: zero condition code
  * @ectl: extended control
  * @pno: path not operational
diff --git a/include/asm-s390/compat.h b/include/asm-s390/compat.h
index 7f4ad623f7d5..de065b32381a 100644
--- a/include/asm-s390/compat.h
+++ b/include/asm-s390/compat.h
@@ -149,7 +149,7 @@ typedef u32		compat_sigset_word;
  * A pointer passed in from user mode. This should not
  * be used for syscall parameters, just declare them
  * as pointers because the syscall entry code will have
- * appropriately comverted them already.
+ * appropriately converted them already.
  */
 typedef	u32		compat_uptr_t;
 
diff --git a/include/asm-s390/cputime.h b/include/asm-s390/cputime.h
index 4b3ef7cad115..133ce054fc89 100644
--- a/include/asm-s390/cputime.h
+++ b/include/asm-s390/cputime.h
@@ -54,6 +54,7 @@ __div(unsigned long long n, unsigned int base)
 #define cputime_lt(__a, __b)		((__a) <  (__b))
 #define cputime_le(__a, __b)		((__a) <= (__b))
 #define cputime_to_jiffies(__ct)	(__div((__ct), 1000000 / HZ))
+#define cputime_to_scaled(__ct)		(__ct)
 #define jiffies_to_cputime(__hz)	((cputime_t)(__hz) * (1000000 / HZ))
 
 #define cputime64_zero			(0ULL)
diff --git a/include/asm-s390/dasd.h b/include/asm-s390/dasd.h
index 604f68fa6f56..3f002e13d024 100644
--- a/include/asm-s390/dasd.h
+++ b/include/asm-s390/dasd.h
@@ -105,7 +105,7 @@ typedef struct dasd_information_t {
 } dasd_information_t;
 
 /*
- * Read Subsystem Data - Perfomance Statistics
+ * Read Subsystem Data - Performance Statistics
  */ 
 typedef struct dasd_rssd_perf_stats_t {
 	unsigned char  invalid:1;
diff --git a/include/asm-s390/ipl.h b/include/asm-s390/ipl.h
index 2c40fd3a137f..c1b2e50392bb 100644
--- a/include/asm-s390/ipl.h
+++ b/include/asm-s390/ipl.h
@@ -83,6 +83,8 @@ extern u32 dump_prefix_page;
 extern unsigned int zfcpdump_prefix_array[];
 
 extern void do_reipl(void);
+extern void do_halt(void);
+extern void do_poff(void);
 extern void ipl_save_parameters(void);
 
 enum {
@@ -118,7 +120,7 @@ struct ipl_info
 };
 
 extern struct ipl_info ipl_info;
-extern void setup_ipl_info(void);
+extern void setup_ipl(void);
 
 /*
  * DIAG 308 support
@@ -141,6 +143,10 @@ enum diag308_opt {
 	DIAG308_IPL_OPT_DUMP	= 0x20,
 };
 
+enum diag308_flags {
+	DIAG308_FLAGS_LP_VALID	= 0x80,
+};
+
 enum diag308_rc {
 	DIAG308_RC_OK	= 1,
 };
diff --git a/include/asm-s390/mmu_context.h b/include/asm-s390/mmu_context.h
index 05b842126b99..a77d4ba3c8eb 100644
--- a/include/asm-s390/mmu_context.h
+++ b/include/asm-s390/mmu_context.h
@@ -12,10 +12,15 @@
 #include <asm/pgalloc.h>
 #include <asm-generic/mm_hooks.h>
 
-/*
- * get a new mmu context.. S390 don't know about contexts.
- */
-#define init_new_context(tsk,mm)        0
+static inline int init_new_context(struct task_struct *tsk,
+				   struct mm_struct *mm)
+{
+	mm->context = _ASCE_TABLE_LENGTH | _ASCE_USER_BITS;
+#ifdef CONFIG_64BIT
+	mm->context |= _ASCE_TYPE_REGION3;
+#endif
+	return 0;
+}
 
 #define destroy_context(mm)             do { } while (0)
 
@@ -27,19 +32,11 @@
 
 static inline void update_mm(struct mm_struct *mm, struct task_struct *tsk)
 {
-	pgd_t *pgd = mm->pgd;
-	unsigned long asce_bits;
-
-	/* Calculate asce bits from the first pgd table entry. */
-	asce_bits = _ASCE_TABLE_LENGTH | _ASCE_USER_BITS;
-#ifdef CONFIG_64BIT
-	asce_bits |= _ASCE_TYPE_REGION3;
-#endif
-	S390_lowcore.user_asce = asce_bits | __pa(pgd);
+	S390_lowcore.user_asce = mm->context | __pa(mm->pgd);
 	if (switch_amode) {
 		/* Load primary space page table origin. */
-		pgd_t *shadow_pgd = get_shadow_table(pgd) ? : pgd;
-		S390_lowcore.user_exec_asce = asce_bits | __pa(shadow_pgd);
+		pgd_t *shadow_pgd = get_shadow_table(mm->pgd) ? : mm->pgd;
+		S390_lowcore.user_exec_asce = mm->context | __pa(shadow_pgd);
 		asm volatile(LCTL_OPCODE" 1,1,%0\n"
 			     : : "m" (S390_lowcore.user_exec_asce) );
 	} else
diff --git a/include/asm-s390/percpu.h b/include/asm-s390/percpu.h
index 545857e64443..408d60b4f75b 100644
--- a/include/asm-s390/percpu.h
+++ b/include/asm-s390/percpu.h
@@ -4,8 +4,6 @@
 #include <linux/compiler.h>
 #include <asm/lowcore.h>
 
-#define __GENERIC_PER_CPU
-
 /*
  * s390 uses its own implementation for per cpu data, the offset of
  * the cpu local data area is cached in the cpu's lowcore memory.
@@ -15,67 +13,25 @@
  */
 #if defined(__s390x__) && defined(MODULE)
 
-#define __reloc_hide(var,offset) (*({			\
+#define SHIFT_PERCPU_PTR(ptr,offset) (({			\
 	extern int simple_identifier_##var(void);	\
 	unsigned long *__ptr;				\
-	asm ( "larl %0,per_cpu__"#var"@GOTENT"		\
-	    : "=a" (__ptr) : "X" (per_cpu__##var) );	\
-	(typeof(&per_cpu__##var))((*__ptr) + (offset));	}))
+	asm ( "larl %0, %1@GOTENT"		\
+	    : "=a" (__ptr) : "X" (ptr) );		\
+	(typeof(ptr))((*__ptr) + (offset));	}))
 
 #else
 
-#define __reloc_hide(var, offset) (*({				\
+#define SHIFT_PERCPU_PTR(ptr, offset) (({				\
 	extern int simple_identifier_##var(void);		\
 	unsigned long __ptr;					\
-	asm ( "" : "=a" (__ptr) : "0" (&per_cpu__##var) );	\
-	(typeof(&per_cpu__##var)) (__ptr + (offset)); }))
+	asm ( "" : "=a" (__ptr) : "0" (ptr) );			\
+	(typeof(ptr)) (__ptr + (offset)); }))
 
 #endif
 
-#ifdef CONFIG_SMP
-
-extern unsigned long __per_cpu_offset[NR_CPUS];
-
-/* Separate out the type, so (int[3], foo) works. */
-#define DEFINE_PER_CPU(type, name) \
-    __attribute__((__section__(".data.percpu"))) \
-    __typeof__(type) per_cpu__##name
-
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)		\
-    __attribute__((__section__(".data.percpu.shared_aligned"))) \
-    __typeof__(type) per_cpu__##name				\
-    ____cacheline_aligned_in_smp
-
-#define __get_cpu_var(var) __reloc_hide(var,S390_lowcore.percpu_offset)
-#define __raw_get_cpu_var(var) __reloc_hide(var,S390_lowcore.percpu_offset)
-#define per_cpu(var,cpu) __reloc_hide(var,__per_cpu_offset[cpu])
-#define per_cpu_offset(x) (__per_cpu_offset[x])
-
-/* A macro to avoid #include hell... */
-#define percpu_modcopy(pcpudst, src, size)			\
-do {								\
-	unsigned int __i;					\
-	for_each_possible_cpu(__i)				\
-		memcpy((pcpudst)+__per_cpu_offset[__i],		\
-		       (src), (size));				\
-} while (0)
-
-#else /* ! SMP */
-
-#define DEFINE_PER_CPU(type, name) \
-    __typeof__(type) per_cpu__##name
-#define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)	\
-    DEFINE_PER_CPU(type, name)
-
-#define __get_cpu_var(var) __reloc_hide(var,0)
-#define __raw_get_cpu_var(var) __reloc_hide(var,0)
-#define per_cpu(var,cpu) __reloc_hide(var,0)
-
-#endif /* SMP */
-
-#define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
+#define __my_cpu_offset S390_lowcore.percpu_offset
 
-#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var)
-#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var)
+#include <asm-generic/percpu.h>
 
 #endif /* __ARCH_S390_PERCPU__ */
diff --git a/include/asm-s390/pgalloc.h b/include/asm-s390/pgalloc.h
index 709dd1740956..6f6619ba8980 100644
--- a/include/asm-s390/pgalloc.h
+++ b/include/asm-s390/pgalloc.h
@@ -57,10 +57,10 @@ static inline unsigned long pgd_entry_type(struct mm_struct *mm)
 }
 
 #define pud_alloc_one(mm,address)		({ BUG(); ((pud_t *)2); })
-#define pud_free(x)				do { } while (0)
+#define pud_free(mm, x)				do { } while (0)
 
 #define pmd_alloc_one(mm,address)		({ BUG(); ((pmd_t *)2); })
-#define pmd_free(x)				do { } while (0)
+#define pmd_free(mm, x)				do { } while (0)
 
 #define pgd_populate(mm, pgd, pud)		BUG()
 #define pgd_populate_kernel(mm, pgd, pud)	BUG()
@@ -76,7 +76,7 @@ static inline unsigned long pgd_entry_type(struct mm_struct *mm)
 }
 
 #define pud_alloc_one(mm,address)		({ BUG(); ((pud_t *)2); })
-#define pud_free(x)				do { } while (0)
+#define pud_free(mm, x)				do { } while (0)
 
 static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long vmaddr)
 {
@@ -85,7 +85,7 @@ static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long vmaddr)
 		crst_table_init(crst, _SEGMENT_ENTRY_EMPTY);
 	return (pmd_t *) crst;
 }
-#define pmd_free(pmd) crst_table_free((unsigned long *) pmd)
+#define pmd_free(mm, pmd) crst_table_free((unsigned long *)pmd)
 
 #define pgd_populate(mm, pgd, pud)		BUG()
 #define pgd_populate_kernel(mm, pgd, pud)	BUG()
@@ -115,7 +115,7 @@ static inline pgd_t *pgd_alloc(struct mm_struct *mm)
 		crst_table_init(crst, pgd_entry_type(mm));
 	return (pgd_t *) crst;
 }
-#define pgd_free(pgd) crst_table_free((unsigned long *) pgd)
+#define pgd_free(mm, pgd) crst_table_free((unsigned long *) pgd)
 
 static inline void 
 pmd_populate_kernel(struct mm_struct *mm, pmd_t *pmd, pte_t *pte)
@@ -151,9 +151,9 @@ pmd_populate(struct mm_struct *mm, pmd_t *pmd, struct page *page)
 #define pte_alloc_one(mm, vmaddr) \
 	virt_to_page(page_table_alloc(s390_noexec))
 
-#define pte_free_kernel(pte) \
+#define pte_free_kernel(mm, pte) \
 	page_table_free((unsigned long *) pte)
-#define pte_free(pte) \
+#define pte_free(mm, pte) \
 	page_table_free((unsigned long *) page_to_phys((struct page *) pte))
 
 #endif /* _S390_PGALLOC_H */
diff --git a/include/asm-s390/pgtable.h b/include/asm-s390/pgtable.h
index f2cc25b74adf..3f520754e71c 100644
--- a/include/asm-s390/pgtable.h
+++ b/include/asm-s390/pgtable.h
@@ -104,42 +104,34 @@ extern char empty_zero_page[PAGE_SIZE];
 
 #ifndef __ASSEMBLY__
 /*
- * Just any arbitrary offset to the start of the vmalloc VM area: the
- * current 8MB value just means that there will be a 8MB "hole" after the
- * physical memory until the kernel virtual memory starts.  That means that
- * any out-of-bounds memory accesses will hopefully be caught.
- * The vmalloc() routines leaves a hole of 4kB between each vmalloced
- * area for the same reason. ;)
- * vmalloc area starts at 4GB to prevent syscall table entry exchanging
- * from modules.
- */
-extern unsigned long vmalloc_end;
-
-#ifdef CONFIG_64BIT
-#define VMALLOC_ADDR	(max(0x100000000UL, (unsigned long) high_memory))
-#else
-#define VMALLOC_ADDR	((unsigned long) high_memory)
-#endif
-#define VMALLOC_OFFSET	(8*1024*1024)
-#define VMALLOC_START	((VMALLOC_ADDR + VMALLOC_OFFSET) & ~(VMALLOC_OFFSET-1))
-#define VMALLOC_END	vmalloc_end
-
-/*
- * We need some free virtual space to be able to do vmalloc.
- * VMALLOC_MIN_SIZE defines the minimum size of the vmalloc
- * area. On a machine with 2GB memory we make sure that we
- * have at least 128MB free space for vmalloc. On a machine
- * with 4TB we make sure we have at least 128GB.
+ * The vmalloc area will always be on the topmost area of the kernel
+ * mapping. We reserve 96MB (31bit) / 1GB (64bit) for vmalloc,
+ * which should be enough for any sane case.
+ * By putting vmalloc at the top, we maximise the gap between physical
+ * memory and vmalloc to catch misplaced memory accesses. As a side
+ * effect, this also makes sure that 64 bit module code cannot be used
+ * as system call address.
  */
 #ifndef __s390x__
-#define VMALLOC_MIN_SIZE	0x8000000UL
-#define VMALLOC_END_INIT	0x80000000UL
+#define VMALLOC_START	0x78000000UL
+#define VMALLOC_END	0x7e000000UL
+#define VMEM_MAP_END	0x80000000UL
 #else /* __s390x__ */
-#define VMALLOC_MIN_SIZE	0x2000000000UL
-#define VMALLOC_END_INIT	0x40000000000UL
+#define VMALLOC_START	0x3e000000000UL
+#define VMALLOC_END	0x3e040000000UL
+#define VMEM_MAP_END	0x40000000000UL
 #endif /* __s390x__ */
 
 /*
+ * VMEM_MAX_PHYS is the highest physical address that can be added to the 1:1
+ * mapping. This needs to be calculated at compile time since the size of the
+ * VMEM_MAP is static but the size of struct page can change.
+ */
+#define VMEM_MAX_PHYS	min(VMALLOC_START, ((VMEM_MAP_END - VMALLOC_END) / \
+			  sizeof(struct page) * PAGE_SIZE) & ~((16 << 20) - 1))
+#define VMEM_MAP	((struct page *) VMALLOC_END)
+
+/*
  * A 31 bit pagetable entry of S390 has following format:
  *  |   PFRA          |    |  OS  |
  * 0                   0IP0
@@ -453,12 +445,12 @@ static inline int pgd_bad(pgd_t pgd)	 { return 0; }
 
 static inline int pud_present(pud_t pud)
 {
-	return pud_val(pud) & _REGION_ENTRY_ORIGIN;
+	return (pud_val(pud) & _REGION_ENTRY_ORIGIN) != 0UL;
 }
 
 static inline int pud_none(pud_t pud)
 {
-	return pud_val(pud) & _REGION_ENTRY_INV;
+	return (pud_val(pud) & _REGION_ENTRY_INV) != 0UL;
 }
 
 static inline int pud_bad(pud_t pud)
@@ -471,12 +463,12 @@ static inline int pud_bad(pud_t pud)
 
 static inline int pmd_present(pmd_t pmd)
 {
-	return pmd_val(pmd) & _SEGMENT_ENTRY_ORIGIN;
+	return (pmd_val(pmd) & _SEGMENT_ENTRY_ORIGIN) != 0UL;
 }
 
 static inline int pmd_none(pmd_t pmd)
 {
-	return pmd_val(pmd) & _SEGMENT_ENTRY_INV;
+	return (pmd_val(pmd) & _SEGMENT_ENTRY_INV) != 0UL;
 }
 
 static inline int pmd_bad(pmd_t pmd)
diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h
index 21d40a19355e..4f744609cd11 100644
--- a/include/asm-s390/processor.h
+++ b/include/asm-s390/processor.h
@@ -59,9 +59,6 @@ extern void s390_adjust_jiffies(void);
 extern void print_cpu_info(struct cpuinfo_S390 *);
 extern int get_cpu_capability(unsigned int *);
 
-/* Lazy FPU handling on uni-processor */
-extern struct task_struct *last_task_used_math;
-
 /*
  * User space process size: 2GB for 31 bit, 4TB for 64 bit.
  */
@@ -73,8 +70,9 @@ extern struct task_struct *last_task_used_math;
 
 #else /* __s390x__ */
 
-# define TASK_SIZE		(test_thread_flag(TIF_31BIT) ? \
+# define TASK_SIZE_OF(tsk)	(test_tsk_thread_flag(tsk, TIF_31BIT) ? \
 					(0x80000000UL) : (0x40000000000UL))
+# define TASK_SIZE		TASK_SIZE_OF(current)
 # define TASK_UNMAPPED_BASE	(TASK_SIZE / 2)
 # define DEFAULT_TASK_SIZE	(0x40000000000UL)
 
@@ -95,7 +93,6 @@ struct thread_struct {
         unsigned long ksp;              /* kernel stack pointer             */
 	mm_segment_t mm_segment;
         unsigned long prot_addr;        /* address of protection-excep.     */
-        unsigned int error_code;        /* error-code of last prog-excep.   */
         unsigned int trap_no;
         per_struct per_info;
 	/* Used to give failing instruction back to user for ieee exceptions */
diff --git a/include/asm-s390/ptrace.h b/include/asm-s390/ptrace.h
index 332ee73688fc..61f6952f2e35 100644
--- a/include/asm-s390/ptrace.h
+++ b/include/asm-s390/ptrace.h
@@ -465,6 +465,14 @@ struct user_regs_struct
 #ifdef __KERNEL__
 #define __ARCH_SYS_PTRACE	1
 
+/*
+ * These are defined as per linux/ptrace.h, which see.
+ */
+#define arch_has_single_step()	(1)
+struct task_struct;
+extern void user_enable_single_step(struct task_struct *);
+extern void user_disable_single_step(struct task_struct *);
+
 #define user_mode(regs) (((regs)->psw.mask & PSW_MASK_PSTATE) != 0)
 #define instruction_pointer(regs) ((regs)->psw.addr & PSW_ADDR_INSN)
 #define regs_return_value(regs)((regs)->gprs[2])
diff --git a/include/asm-s390/qdio.h b/include/asm-s390/qdio.h
index 74db1dc10a7d..4b8ff55f680e 100644
--- a/include/asm-s390/qdio.h
+++ b/include/asm-s390/qdio.h
@@ -184,7 +184,7 @@ struct qdr {
 #endif /* QDIO_32_BIT */
 	unsigned long qiba;             /* queue-information-block address */
 	unsigned int  res8;             /* reserved */
-	unsigned int  qkey    :  4;     /* queue-informatio-block key */
+	unsigned int  qkey    :  4;	/* queue-information-block key */
 	unsigned int  res9    : 28;     /* reserved */
 /*	union _qd {*/ /* why this? */
 		struct qdesfmt0 qdf0[126];
diff --git a/include/asm-s390/rwsem.h b/include/asm-s390/rwsem.h
index 90f4eccaa290..9d2a17971805 100644
--- a/include/asm-s390/rwsem.h
+++ b/include/asm-s390/rwsem.h
@@ -91,8 +91,8 @@ struct rw_semaphore {
 #endif
 
 #define __RWSEM_INITIALIZER(name) \
-{ RWSEM_UNLOCKED_VALUE, SPIN_LOCK_UNLOCKED, LIST_HEAD_INIT((name).wait_list) \
-  __RWSEM_DEP_MAP_INIT(name) }
+ { RWSEM_UNLOCKED_VALUE, __SPIN_LOCK_UNLOCKED((name).wait.lock), \
+   LIST_HEAD_INIT((name).wait_list) __RWSEM_DEP_MAP_INIT(name) }
 
 #define DECLARE_RWSEM(name) \
 	struct rw_semaphore name = __RWSEM_INITIALIZER(name)
diff --git a/include/asm-s390/sclp.h b/include/asm-s390/sclp.h
index cb9faf1ea5cf..b5f2843013a3 100644
--- a/include/asm-s390/sclp.h
+++ b/include/asm-s390/sclp.h
@@ -27,7 +27,25 @@ struct sclp_ipl_info {
 	char loadparm[LOADPARM_LEN];
 };
 
-void sclp_readinfo_early(void);
+struct sclp_cpu_entry {
+	u8 address;
+	u8 reserved0[13];
+	u8 type;
+	u8 reserved1;
+} __attribute__((packed));
+
+struct sclp_cpu_info {
+	unsigned int configured;
+	unsigned int standby;
+	unsigned int combined;
+	int has_cpu_type;
+	struct sclp_cpu_entry cpu[255];
+};
+
+int sclp_get_cpu_info(struct sclp_cpu_info *info);
+int sclp_cpu_configure(u8 cpu);
+int sclp_cpu_deconfigure(u8 cpu);
+void sclp_read_info_early(void);
 void sclp_facilities_detect(void);
 unsigned long long sclp_memory_detect(void);
 int sclp_sdias_blk_count(void);
diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h
index 07708c07701e..c7b74326a527 100644
--- a/include/asm-s390/smp.h
+++ b/include/asm-s390/smp.h
@@ -35,8 +35,6 @@ extern void machine_restart_smp(char *);
 extern void machine_halt_smp(void);
 extern void machine_power_off_smp(void);
 
-extern void smp_setup_cpu_possible_map(void);
-
 #define NO_PROC_ID		0xFF		/* No processor magic marker */
 
 /*
@@ -92,6 +90,8 @@ extern void __cpu_die (unsigned int cpu);
 extern void cpu_die (void) __attribute__ ((noreturn));
 extern int __cpu_up (unsigned int cpu);
 
+extern int smp_call_function_mask(cpumask_t mask, void (*func)(void *),
+	void *info, int wait);
 #endif
 
 #ifndef CONFIG_SMP
@@ -103,7 +103,6 @@ static inline void smp_send_stop(void)
 
 #define hard_smp_processor_id()		0
 #define smp_cpu_not_running(cpu)	1
-#define smp_setup_cpu_possible_map()	do { } while (0)
 #endif
 
 extern union save_area *zfcpdump_save_areas[NR_CPUS + 1];
diff --git a/include/asm-s390/socket.h b/include/asm-s390/socket.h
index 1161ebe3dec9..c786ab623b2d 100644
--- a/include/asm-s390/socket.h
+++ b/include/asm-s390/socket.h
@@ -60,4 +60,6 @@
 #define SO_TIMESTAMPNS		35
 #define SCM_TIMESTAMPNS		SO_TIMESTAMPNS
 
+#define SO_MARK			36
+
 #endif /* _ASM_SOCKET_H */
diff --git a/include/asm-s390/spinlock.h b/include/asm-s390/spinlock.h
index 3fd43826fd0b..df84ae96915f 100644
--- a/include/asm-s390/spinlock.h
+++ b/include/asm-s390/spinlock.h
@@ -53,44 +53,48 @@ _raw_compare_and_swap(volatile unsigned int *lock,
  */
 
 #define __raw_spin_is_locked(x) ((x)->owner_cpu != 0)
-#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
 #define __raw_spin_unlock_wait(lock) \
 	do { while (__raw_spin_is_locked(lock)) \
 		 _raw_spin_relax(lock); } while (0)
 
-extern void _raw_spin_lock_wait(raw_spinlock_t *, unsigned int pc);
-extern int _raw_spin_trylock_retry(raw_spinlock_t *, unsigned int pc);
+extern void _raw_spin_lock_wait(raw_spinlock_t *);
+extern void _raw_spin_lock_wait_flags(raw_spinlock_t *, unsigned long flags);
+extern int _raw_spin_trylock_retry(raw_spinlock_t *);
 extern void _raw_spin_relax(raw_spinlock_t *lock);
 
 static inline void __raw_spin_lock(raw_spinlock_t *lp)
 {
-	unsigned long pc = 1 | (unsigned long) __builtin_return_address(0);
 	int old;
 
 	old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id());
-	if (likely(old == 0)) {
-		lp->owner_pc = pc;
+	if (likely(old == 0))
 		return;
-	}
-	_raw_spin_lock_wait(lp, pc);
+	_raw_spin_lock_wait(lp);
+}
+
+static inline void __raw_spin_lock_flags(raw_spinlock_t *lp,
+					 unsigned long flags)
+{
+	int old;
+
+	old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id());
+	if (likely(old == 0))
+		return;
+	_raw_spin_lock_wait_flags(lp, flags);
 }
 
 static inline int __raw_spin_trylock(raw_spinlock_t *lp)
 {
-	unsigned long pc = 1 | (unsigned long) __builtin_return_address(0);
 	int old;
 
 	old = _raw_compare_and_swap(&lp->owner_cpu, 0, ~smp_processor_id());
-	if (likely(old == 0)) {
-		lp->owner_pc = pc;
+	if (likely(old == 0))
 		return 1;
-	}
-	return _raw_spin_trylock_retry(lp, pc);
+	return _raw_spin_trylock_retry(lp);
 }
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lp)
 {
-	lp->owner_pc = 0;
 	_raw_compare_and_swap(&lp->owner_cpu, lp->owner_cpu, 0);
 }
 		
diff --git a/include/asm-s390/spinlock_types.h b/include/asm-s390/spinlock_types.h
index b7ac13f7aa37..654abc40de04 100644
--- a/include/asm-s390/spinlock_types.h
+++ b/include/asm-s390/spinlock_types.h
@@ -7,7 +7,6 @@
 
 typedef struct {
 	volatile unsigned int owner_cpu;
-	volatile unsigned int owner_pc;
 } __attribute__ ((aligned (4))) raw_spinlock_t;
 
 #define __RAW_SPIN_LOCK_UNLOCKED	{ 0 }
diff --git a/include/asm-s390/system.h b/include/asm-s390/system.h
index d866d3385556..44bda786eef7 100644
--- a/include/asm-s390/system.h
+++ b/include/asm-s390/system.h
@@ -388,6 +388,11 @@ extern void (*_machine_power_off)(void);
 
 #define arch_align_stack(x) (x)
 
+#ifdef CONFIG_TRACE_IRQFLAGS
+extern psw_t sysc_restore_trace_psw;
+extern psw_t io_restore_trace_psw;
+#endif
+
 #endif /* __KERNEL__ */
 
 #endif
diff --git a/include/asm-s390/tlb.h b/include/asm-s390/tlb.h
index 618693cfc10f..985de2b88279 100644
--- a/include/asm-s390/tlb.h
+++ b/include/asm-s390/tlb.h
@@ -65,9 +65,9 @@ static inline void tlb_flush_mmu(struct mmu_gather *tlb,
 	if (!tlb->fullmm && (tlb->nr_ptes > 0 || tlb->nr_pmds < TLB_NR_PTRS))
 		__tlb_flush_mm(tlb->mm);
 	while (tlb->nr_ptes > 0)
-		pte_free(tlb->array[--tlb->nr_ptes]);
+		pte_free(tlb->mm, tlb->array[--tlb->nr_ptes]);
 	while (tlb->nr_pmds < TLB_NR_PTRS)
-		pmd_free((pmd_t *) tlb->array[tlb->nr_pmds++]);
+		pmd_free(tlb->mm, (pmd_t *) tlb->array[tlb->nr_pmds++]);
 }
 
 static inline void tlb_finish_mmu(struct mmu_gather *tlb,
@@ -102,7 +102,7 @@ static inline void pte_free_tlb(struct mmu_gather *tlb, struct page *page)
 		if (tlb->nr_ptes >= tlb->nr_pmds)
 			tlb_flush_mmu(tlb, 0, 0);
 	} else
-		pte_free(page);
+		pte_free(tlb->mm, page);
 }
 
 /*
@@ -117,7 +117,7 @@ static inline void pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd)
 		if (tlb->nr_ptes >= tlb->nr_pmds)
 			tlb_flush_mmu(tlb, 0, 0);
 	} else
-		pmd_free(pmd);
+		pmd_free(tlb->mm, pmd);
 #endif
 }
 
diff --git a/include/asm-s390/tlbflush.h b/include/asm-s390/tlbflush.h
index a69bd2490d52..70fa5ae58180 100644
--- a/include/asm-s390/tlbflush.h
+++ b/include/asm-s390/tlbflush.h
@@ -42,11 +42,11 @@ static inline void __tlb_flush_global(void)
 /*
  * Flush all tlb entries of a page table on all cpus.
  */
-static inline void __tlb_flush_idte(pgd_t *pgd)
+static inline void __tlb_flush_idte(unsigned long asce)
 {
 	asm volatile(
 		"	.insn	rrf,0xb98e0000,0,%0,%1,0"
-		: : "a" (2048), "a" (__pa(pgd) & PAGE_MASK) : "cc" );
+		: : "a" (2048), "a" (asce) : "cc" );
 }
 
 static inline void __tlb_flush_mm(struct mm_struct * mm)
@@ -61,11 +61,11 @@ static inline void __tlb_flush_mm(struct mm_struct * mm)
 	 * only ran on the local cpu.
 	 */
 	if (MACHINE_HAS_IDTE) {
-		pgd_t *shadow_pgd = get_shadow_table(mm->pgd);
+		pgd_t *shadow = get_shadow_table(mm->pgd);
 
-		if (shadow_pgd)
-			__tlb_flush_idte(shadow_pgd);
-		__tlb_flush_idte(mm->pgd);
+		if (shadow)
+			__tlb_flush_idte((unsigned long) shadow | mm->context);
+		__tlb_flush_idte((unsigned long) mm->pgd | mm->context);
 		return;
 	}
 	preempt_disable();
@@ -106,9 +106,23 @@ static inline void __tlb_flush_mm_cond(struct mm_struct * mm)
  */
 #define flush_tlb()				do { } while (0)
 #define flush_tlb_all()				do { } while (0)
-#define flush_tlb_mm(mm)			__tlb_flush_mm_cond(mm)
 #define flush_tlb_page(vma, addr)		do { } while (0)
-#define flush_tlb_range(vma, start, end)	__tlb_flush_mm_cond(mm)
-#define flush_tlb_kernel_range(start, end)	__tlb_flush_mm(&init_mm)
+
+static inline void flush_tlb_mm(struct mm_struct *mm)
+{
+	__tlb_flush_mm_cond(mm);
+}
+
+static inline void flush_tlb_range(struct vm_area_struct *vma,
+				   unsigned long start, unsigned long end)
+{
+	__tlb_flush_mm_cond(vma->vm_mm);
+}
+
+static inline void flush_tlb_kernel_range(unsigned long start,
+					  unsigned long end)
+{
+	__tlb_flush_mm(&init_mm);
+}
 
 #endif /* _S390_TLBFLUSH_H */
diff --git a/include/asm-s390/zcrypt.h b/include/asm-s390/zcrypt.h
index a5dada617751..f228f1b86877 100644
--- a/include/asm-s390/zcrypt.h
+++ b/include/asm-s390/zcrypt.h
@@ -117,7 +117,7 @@ struct CPRBX {
 	unsigned char	padx004[16 - sizeof (char *)];
 	unsigned char *	req_extb;	/* request extension block 'addr'*/
 	unsigned char	padx005[16 - sizeof (char *)];
-	unsigned char *	rpl_extb;	/* reply extension block 'addres'*/
+	unsigned char *	rpl_extb;	/* reply extension block 'address'*/
 	unsigned short	ccp_rtcode;	/* server return code		 */
 	unsigned short	ccp_rscode;	/* server reason code		 */
 	unsigned int	mac_data_len;	/* Mac Data Length		 */