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

Pull s390 updates from Martin Schwidefsky:
 "There are a couple of new things for s390 with this merge request:

   - a new scheduling domain "drawer" is added to reflect the unusual
     topology found on z13 machines.  Performance tests showed up to 8
     percent gain with the additional domain.

   - the new crc-32 checksum crypto module uses the vector-galois-field
     multiply and sum SIMD instruction to speed up crc-32 and crc-32c.

   - proper __ro_after_init support, this requires RO_AFTER_INIT_DATA in
     the generic vmlinux.lds linker script definitions.

   - kcov instrumentation support.  A prerequisite for that is the
     inline assembly basic block cleanup, which is the reason for the
     net/iucv/iucv.c change.

   - support for 2GB pages is added to the hugetlbfs backend.

  Then there are two removals:

   - the oprofile hardware sampling support is dead code and is removed.
     The oprofile user space uses the perf interface nowadays.

   - the ETR clock synchronization is removed, this has been superseeded
     be the STP clock synchronization.  And it always has been
     "interesting" code..

  And the usual bug fixes and cleanups"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (82 commits)
  s390/pci: Delete an unnecessary check before the function call "pci_dev_put"
  s390/smp: clean up a condition
  s390/cio/chp : Remove deprecated create_singlethread_workqueue
  s390/chsc: improve channel path descriptor determination
  s390/chsc: sanitize fmt check for chp_desc determination
  s390/cio: make fmt1 channel path descriptor optional
  s390/chsc: fix ioctl CHSC_INFO_CU command
  s390/cio/device_ops: fix kernel doc
  s390/cio: allow to reset channel measurement block
  s390/console: Make preferred console handling more consistent
  s390/mm: fix gmap tlb flush issues
  s390/mm: add support for 2GB hugepages
  s390: have unique symbol for __switch_to address
  s390/cpuinfo: show maximum thread id
  s390/ptrace: clarify bits in the per_struct
  s390: stack address vs thread_info
  s390: remove pointless load within __switch_to
  s390: enable kcov support
  s390/cpumf: use basic block for ecctr inline assembly
  s390/hypfs: use basic block for diag inline assembly
  ...

30 files changed, 578 insertions, 625 deletions

diff --git a/arch/s390/include/asm/cache.h b/arch/s390/include/asm/cache.h
index 22da3b34c655..05219a5e0b2f 100644
--- a/arch/s390/include/asm/cache.h
+++ b/arch/s390/include/asm/cache.h
@@ -13,9 +13,6 @@
 #define L1_CACHE_SHIFT     8
 #define NET_SKB_PAD	   32
 
-#define __read_mostly __attribute__((__section__(".data..read_mostly")))
-
-/* Read-only memory is marked before mark_rodata_ro() is called. */
-#define __ro_after_init __read_mostly
+#define __read_mostly __section(.data..read_mostly)
 
 #endif
diff --git a/arch/s390/include/asm/cio.h b/arch/s390/include/asm/cio.h
index d1e7b0a0feeb..f7ed88cc066e 100644
--- a/arch/s390/include/asm/cio.h
+++ b/arch/s390/include/asm/cio.h
@@ -320,7 +320,7 @@ struct cio_iplinfo {
 extern int cio_get_iplinfo(struct cio_iplinfo *iplinfo);
 
 /* Function from drivers/s390/cio/chsc.c */
-int chsc_sstpc(void *page, unsigned int op, u16 ctrl);
+int chsc_sstpc(void *page, unsigned int op, u16 ctrl, u64 *clock_delta);
 int chsc_sstpi(void *page, void *result, size_t size);
 
 #endif
diff --git a/arch/s390/include/asm/cpu_mf.h b/arch/s390/include/asm/cpu_mf.h
index 9dd04b9e9782..03516476127b 100644
--- a/arch/s390/include/asm/cpu_mf.h
+++ b/arch/s390/include/asm/cpu_mf.h
@@ -169,16 +169,27 @@ static inline int lcctl(u64 ctl)
 }
 
 /* Extract CPU counter */
-static inline int ecctr(u64 ctr, u64 *val)
+static inline int __ecctr(u64 ctr, u64 *content)
 {
-	register u64 content asm("4") = 0;
+	register u64 _content asm("4") = 0;
 	int cc;
 
 	asm volatile (
 		"	.insn	rre,0xb2e40000,%0,%2\n"
 		"	ipm	%1\n"
 		"	srl	%1,28\n"
-		: "=d" (content), "=d" (cc) : "d" (ctr) : "cc");
+		: "=d" (_content), "=d" (cc) : "d" (ctr) : "cc");
+	*content = _content;
+	return cc;
+}
+
+/* Extract CPU counter */
+static inline int ecctr(u64 ctr, u64 *val)
+{
+	u64 content;
+	int cc;
+
+	cc = __ecctr(ctr, &content);
 	if (!cc)
 		*val = content;
 	return cc;
diff --git a/arch/s390/include/asm/diag.h b/arch/s390/include/asm/diag.h
index 5fac921c1c42..86cae09e076a 100644
--- a/arch/s390/include/asm/diag.h
+++ b/arch/s390/include/asm/diag.h
@@ -49,7 +49,7 @@ static inline void diag10_range(unsigned long start_pfn, unsigned long num_pfn)
 	diag_stat_inc(DIAG_STAT_X010);
 	asm volatile(
 		"0:	diag	%0,%1,0x10\n"
-		"1:\n"
+		"1:	nopr	%%r7\n"
 		EX_TABLE(0b, 1b)
 		EX_TABLE(1b, 1b)
 		: : "a" (start_addr), "a" (end_addr));
diff --git a/arch/s390/include/asm/etr.h b/arch/s390/include/asm/etr.h
deleted file mode 100644
index 105f90e63a0e..000000000000
--- a/arch/s390/include/asm/etr.h
+++ /dev/null
@@ -1,261 +0,0 @@
-/*
- *  Copyright IBM Corp. 2006
- *  Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
- */
-#ifndef __S390_ETR_H
-#define __S390_ETR_H
-
-/* ETR attachment control register */
-struct etr_eacr {
-	unsigned int e0		: 1;	/* port 0 stepping control */
-	unsigned int e1		: 1;	/* port 1 stepping control */
-	unsigned int _pad0	: 5;	/* must be 00100 */
-	unsigned int dp		: 1;	/* data port control */
-	unsigned int p0		: 1;	/* port 0 change recognition control */
-	unsigned int p1		: 1;	/* port 1 change recognition control */
-	unsigned int _pad1	: 3;	/* must be 000 */
-	unsigned int ea		: 1;	/* ETR alert control */
-	unsigned int es		: 1;	/* ETR sync check control */
-	unsigned int sl		: 1;	/* switch to local control */
-} __attribute__ ((packed));
-
-/* Port state returned by steai */
-enum etr_psc {
-	etr_psc_operational = 0,
-	etr_psc_semi_operational = 1,
-	etr_psc_protocol_error =  4,
-	etr_psc_no_symbols = 8,
-	etr_psc_no_signal = 12,
-	etr_psc_pps_mode = 13
-};
-
-/* Logical port state returned by stetr */
-enum etr_lpsc {
-	etr_lpsc_operational_step = 0,
-	etr_lpsc_operational_alt = 1,
-	etr_lpsc_semi_operational = 2,
-	etr_lpsc_protocol_error =  4,
-	etr_lpsc_no_symbol_sync = 8,
-	etr_lpsc_no_signal = 12,
-	etr_lpsc_pps_mode = 13
-};
-
-/* ETR status words */
-struct etr_esw {
-	struct etr_eacr eacr;		/* attachment control register */
-	unsigned int y		: 1;	/* stepping mode */
-	unsigned int _pad0	: 5;	/* must be 00000 */
-	unsigned int p		: 1;	/* stepping port number */
-	unsigned int q		: 1;	/* data port number */
-	unsigned int psc0	: 4;	/* port 0 state code */
-	unsigned int psc1	: 4;	/* port 1 state code */
-} __attribute__ ((packed));
-
-/* Second level data register status word */
-struct etr_slsw {
-	unsigned int vv1	: 1;	/* copy of validity bit data frame 1 */
-	unsigned int vv2	: 1;	/* copy of validity bit data frame 2 */
-	unsigned int vv3	: 1;	/* copy of validity bit data frame 3 */
-	unsigned int vv4	: 1;	/* copy of validity bit data frame 4 */
-	unsigned int _pad0	: 19;	/* must by all zeroes */
-	unsigned int n		: 1;	/* EAF port number */
-	unsigned int v1		: 1;	/* validity bit ETR data frame 1 */
-	unsigned int v2		: 1;	/* validity bit ETR data frame 2 */
-	unsigned int v3		: 1;	/* validity bit ETR data frame 3 */
-	unsigned int v4		: 1;	/* validity bit ETR data frame 4 */
-	unsigned int _pad1	: 4;	/* must be 0000 */
-} __attribute__ ((packed));
-
-/* ETR data frames */
-struct etr_edf1 {
-	unsigned int u		: 1;	/* untuned bit */
-	unsigned int _pad0	: 1;	/* must be 0 */
-	unsigned int r		: 1;	/* service request bit */
-	unsigned int _pad1	: 4;	/* must be 0000 */
-	unsigned int a		: 1;	/* time adjustment bit */
-	unsigned int net_id	: 8;	/* ETR network id */
-	unsigned int etr_id	: 8;	/* id of ETR which sends data frames */
-	unsigned int etr_pn	: 8;	/* port number of ETR output port */
-} __attribute__ ((packed));
-
-struct etr_edf2 {
-	unsigned int etv	: 32;	/* Upper 32 bits of TOD. */
-} __attribute__ ((packed));
-
-struct etr_edf3 {
-	unsigned int rc		: 8;	/* failure reason code */
-	unsigned int _pad0	: 3;	/* must be 000 */
-	unsigned int c		: 1;	/* ETR coupled bit */
-	unsigned int tc		: 4;	/* ETR type code */
-	unsigned int blto	: 8;	/* biased local time offset */
-					/* (blto - 128) * 15 = minutes */
-	unsigned int buo	: 8;	/* biased utc offset */
-					/* (buo - 128) = leap seconds */
-} __attribute__ ((packed));
-
-struct etr_edf4 {
-	unsigned int ed		: 8;	/* ETS device dependent data */
-	unsigned int _pad0	: 1;	/* must be 0 */
-	unsigned int buc	: 5;	/* biased ut1 correction */
-					/* (buc - 16) * 0.1 seconds */
-	unsigned int em		: 6;	/* ETS error magnitude */
-	unsigned int dc		: 6;	/* ETS drift code */
-	unsigned int sc		: 6;	/* ETS steering code */
-} __attribute__ ((packed));
-
-/*
- * ETR attachment information block, two formats
- * format 1 has 4 reserved words with a size of 64 bytes
- * format 2 has 16 reserved words with a size of 96 bytes
- */
-struct etr_aib {
-	struct etr_esw esw;
-	struct etr_slsw slsw;
-	unsigned long long tsp;
-	struct etr_edf1 edf1;
-	struct etr_edf2 edf2;
-	struct etr_edf3 edf3;
-	struct etr_edf4 edf4;
-	unsigned int reserved[16];
-} __attribute__ ((packed,aligned(8)));
-
-/* ETR interruption parameter */
-struct etr_irq_parm {
-	unsigned int _pad0	: 8;
-	unsigned int pc0	: 1;	/* port 0 state change */
-	unsigned int pc1	: 1;	/* port 1 state change */
-	unsigned int _pad1	: 3;
-	unsigned int eai	: 1;	/* ETR alert indication */
-	unsigned int _pad2	: 18;
-} __attribute__ ((packed));
-
-/* Query TOD offset result */
-struct etr_ptff_qto {
-	unsigned long long physical_clock;
-	unsigned long long tod_offset;
-	unsigned long long logical_tod_offset;
-	unsigned long long tod_epoch_difference;
-} __attribute__ ((packed));
-
-/* Inline assembly helper functions */
-static inline int etr_setr(struct etr_eacr *ctrl)
-{
-	int rc = -EOPNOTSUPP;
-
-	asm volatile(
-		"	.insn	s,0xb2160000,%1\n"
-		"0:	la	%0,0\n"
-		"1:\n"
-		EX_TABLE(0b,1b)
-		: "+d" (rc) : "Q" (*ctrl));
-	return rc;
-}
-
-/* Stores a format 1 aib with 64 bytes */
-static inline int etr_stetr(struct etr_aib *aib)
-{
-	int rc = -EOPNOTSUPP;
-
-	asm volatile(
-		"	.insn	s,0xb2170000,%1\n"
-		"0:	la	%0,0\n"
-		"1:\n"
-		EX_TABLE(0b,1b)
-		: "+d" (rc) : "Q" (*aib));
-	return rc;
-}
-
-/* Stores a format 2 aib with 96 bytes for specified port */
-static inline int etr_steai(struct etr_aib *aib, unsigned int func)
-{
-	register unsigned int reg0 asm("0") = func;
-	int rc = -EOPNOTSUPP;
-
-	asm volatile(
-		"	.insn	s,0xb2b30000,%1\n"
-		"0:	la	%0,0\n"
-		"1:\n"
-		EX_TABLE(0b,1b)
-		: "+d" (rc) : "Q" (*aib), "d" (reg0));
-	return rc;
-}
-
-/* Function codes for the steai instruction. */
-#define ETR_STEAI_STEPPING_PORT		0x10
-#define ETR_STEAI_ALTERNATE_PORT	0x11
-#define ETR_STEAI_PORT_0		0x12
-#define ETR_STEAI_PORT_1		0x13
-
-static inline int etr_ptff(void *ptff_block, unsigned int func)
-{
-	register unsigned int reg0 asm("0") = func;
-	register unsigned long reg1 asm("1") = (unsigned long) ptff_block;
-	int rc = -EOPNOTSUPP;
-
-	asm volatile(
-		"	.word	0x0104\n"
-		"	ipm	%0\n"
-		"	srl	%0,28\n"
-		: "=d" (rc), "=m" (ptff_block)
-		: "d" (reg0), "d" (reg1), "m" (ptff_block) : "cc");
-	return rc;
-}
-
-/* Function codes for the ptff instruction. */
-#define ETR_PTFF_QAF	0x00	/* query available functions */
-#define ETR_PTFF_QTO	0x01	/* query tod offset */
-#define ETR_PTFF_QSI	0x02	/* query steering information */
-#define ETR_PTFF_ATO	0x40	/* adjust tod offset */
-#define ETR_PTFF_STO	0x41	/* set tod offset */
-#define ETR_PTFF_SFS	0x42	/* set fine steering rate */
-#define ETR_PTFF_SGS	0x43	/* set gross steering rate */
-
-/* Functions needed by the machine check handler */
-int etr_switch_to_local(void);
-int etr_sync_check(void);
-void etr_queue_work(void);
-
-/* notifier for syncs */
-extern struct atomic_notifier_head s390_epoch_delta_notifier;
-
-/* STP interruption parameter */
-struct stp_irq_parm {
-	unsigned int _pad0	: 14;
-	unsigned int tsc	: 1;	/* Timing status change */
-	unsigned int lac	: 1;	/* Link availability change */
-	unsigned int tcpc	: 1;	/* Time control parameter change */
-	unsigned int _pad2	: 15;
-} __attribute__ ((packed));
-
-#define STP_OP_SYNC	1
-#define STP_OP_CTRL	3
-
-struct stp_sstpi {
-	unsigned int rsvd0;
-	unsigned int rsvd1 : 8;
-	unsigned int stratum : 8;
-	unsigned int vbits : 16;
-	unsigned int leaps : 16;
-	unsigned int tmd : 4;
-	unsigned int ctn : 4;
-	unsigned int rsvd2 : 3;
-	unsigned int c : 1;
-	unsigned int tst : 4;
-	unsigned int tzo : 16;
-	unsigned int dsto : 16;
-	unsigned int ctrl : 16;
-	unsigned int rsvd3 : 16;
-	unsigned int tto;
-	unsigned int rsvd4;
-	unsigned int ctnid[3];
-	unsigned int rsvd5;
-	unsigned int todoff[4];
-	unsigned int rsvd6[48];
-} __attribute__ ((packed));
-
-/* Functions needed by the machine check handler */
-int stp_sync_check(void);
-int stp_island_check(void);
-void stp_queue_work(void);
-
-#endif /* __S390_ETR_H */
diff --git a/arch/s390/include/asm/fcx.h b/arch/s390/include/asm/fcx.h
index 7ecb92b469b6..04cb4b4bcc5f 100644
--- a/arch/s390/include/asm/fcx.h
+++ b/arch/s390/include/asm/fcx.h
@@ -6,7 +6,7 @@
  */
 
 #ifndef _ASM_S390_FCX_H
-#define _ASM_S390_FCX_H _ASM_S390_FCX_H
+#define _ASM_S390_FCX_H
 
 #include <linux/types.h>
 
diff --git a/arch/s390/include/asm/fpu/api.h b/arch/s390/include/asm/fpu/api.h
index 8ae236b0f80b..6aba6fc406ad 100644
--- a/arch/s390/include/asm/fpu/api.h
+++ b/arch/s390/include/asm/fpu/api.h
@@ -1,6 +1,41 @@
 /*
  * In-kernel FPU support functions
  *
+ *
+ * Consider these guidelines before using in-kernel FPU functions:
+ *
+ *  1. Use kernel_fpu_begin() and kernel_fpu_end() to enclose all in-kernel
+ *     use of floating-point or vector registers and instructions.
+ *
+ *  2. For kernel_fpu_begin(), specify the vector register range you want to
+ *     use with the KERNEL_VXR_* constants. Consider these usage guidelines:
+ *
+ *     a) If your function typically runs in process-context, use the lower
+ *	  half of the vector registers, for example, specify KERNEL_VXR_LOW.
+ *     b) If your function typically runs in soft-irq or hard-irq context,
+ *	  prefer using the upper half of the vector registers, for example,
+ *	  specify KERNEL_VXR_HIGH.
+ *
+ *     If you adhere to these guidelines, an interrupted process context
+ *     does not require to save and restore vector registers because of
+ *     disjoint register ranges.
+ *
+ *     Also note that the __kernel_fpu_begin()/__kernel_fpu_end() functions
+ *     includes logic to save and restore up to 16 vector registers at once.
+ *
+ *  3. You can nest kernel_fpu_begin()/kernel_fpu_end() by using different
+ *     struct kernel_fpu states.  Vector registers that are in use by outer
+ *     levels are saved and restored.  You can minimize the save and restore
+ *     effort by choosing disjoint vector register ranges.
+ *
+ *  5. To use vector floating-point instructions, specify the KERNEL_FPC
+ *     flag to save and restore floating-point controls in addition to any
+ *     vector register range.
+ *
+ *  6. To use floating-point registers and instructions only, specify the
+ *     KERNEL_FPR flag.  This flag triggers a save and restore of vector
+ *     registers V0 to V15 and floating-point controls.
+ *
  * Copyright IBM Corp. 2015
  * Author(s): Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
  */
@@ -8,6 +43,8 @@
 #ifndef _ASM_S390_FPU_API_H
 #define _ASM_S390_FPU_API_H
 
+#include <linux/preempt.h>
+
 void save_fpu_regs(void);
 
 static inline int test_fp_ctl(u32 fpc)
@@ -27,4 +64,42 @@ static inline int test_fp_ctl(u32 fpc)
 	return rc;
 }
 
+#define KERNEL_VXR_V0V7		1
+#define KERNEL_VXR_V8V15	2
+#define KERNEL_VXR_V16V23	4
+#define KERNEL_VXR_V24V31	8
+#define KERNEL_FPR		16
+#define KERNEL_FPC		256
+
+#define KERNEL_VXR_LOW		(KERNEL_VXR_V0V7|KERNEL_VXR_V8V15)
+#define KERNEL_VXR_MID		(KERNEL_VXR_V8V15|KERNEL_VXR_V16V23)
+#define KERNEL_VXR_HIGH		(KERNEL_VXR_V16V23|KERNEL_VXR_V24V31)
+
+#define KERNEL_FPU_MASK		(KERNEL_VXR_LOW|KERNEL_VXR_HIGH|KERNEL_FPR)
+
+struct kernel_fpu;
+
+/*
+ * Note the functions below must be called with preemption disabled.
+ * Do not enable preemption before calling __kernel_fpu_end() to prevent
+ * an corruption of an existing kernel FPU state.
+ *
+ * Prefer using the kernel_fpu_begin()/kernel_fpu_end() pair of functions.
+ */
+void __kernel_fpu_begin(struct kernel_fpu *state, u32 flags);
+void __kernel_fpu_end(struct kernel_fpu *state);
+
+
+static inline void kernel_fpu_begin(struct kernel_fpu *state, u32 flags)
+{
+	preempt_disable();
+	__kernel_fpu_begin(state, flags);
+}
+
+static inline void kernel_fpu_end(struct kernel_fpu *state)
+{
+	__kernel_fpu_end(state);
+	preempt_enable();
+}
+
 #endif /* _ASM_S390_FPU_API_H */
diff --git a/arch/s390/include/asm/fpu/types.h b/arch/s390/include/asm/fpu/types.h
index fe937c9b6471..bce255ead72b 100644
--- a/arch/s390/include/asm/fpu/types.h
+++ b/arch/s390/include/asm/fpu/types.h
@@ -24,4 +24,14 @@ struct fpu {
 /* VX array structure for address operand constraints in inline assemblies */
 struct vx_array { __vector128 _[__NUM_VXRS]; };
 
+/* In-kernel FPU state structure */
+struct kernel_fpu {
+	u32	    mask;
+	u32	    fpc;
+	union {
+		freg_t fprs[__NUM_FPRS];
+		__vector128 vxrs[__NUM_VXRS];
+	};
+};
+
 #endif /* _ASM_S390_FPU_TYPES_H */
diff --git a/arch/s390/include/asm/hugetlb.h b/arch/s390/include/asm/hugetlb.h
index d9be7c0c1291..4c7fac75090e 100644
--- a/arch/s390/include/asm/hugetlb.h
+++ b/arch/s390/include/asm/hugetlb.h
@@ -41,7 +41,10 @@ static inline int prepare_hugepage_range(struct file *file,
 static inline void huge_pte_clear(struct mm_struct *mm, unsigned long addr,
 				  pte_t *ptep)
 {
-	pte_val(*ptep) = _SEGMENT_ENTRY_EMPTY;
+	if ((pte_val(*ptep) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R3)
+		pte_val(*ptep) = _REGION3_ENTRY_EMPTY;
+	else
+		pte_val(*ptep) = _SEGMENT_ENTRY_EMPTY;
 }
 
 static inline void huge_ptep_clear_flush(struct vm_area_struct *vma,
diff --git a/arch/s390/include/asm/ipl.h b/arch/s390/include/asm/ipl.h
index 6fc44dca193e..4da22b2f0521 100644
--- a/arch/s390/include/asm/ipl.h
+++ b/arch/s390/include/asm/ipl.h
@@ -141,11 +141,11 @@ extern void setup_ipl(void);
  * DIAG 308 support
  */
 enum diag308_subcode  {
-	DIAG308_REL_HSA	= 2,
-	DIAG308_IPL	= 3,
-	DIAG308_DUMP	= 4,
-	DIAG308_SET	= 5,
-	DIAG308_STORE	= 6,
+	DIAG308_REL_HSA = 2,
+	DIAG308_LOAD_CLEAR = 3,
+	DIAG308_LOAD_NORMAL_DUMP = 4,
+	DIAG308_SET = 5,
+	DIAG308_STORE = 6,
 };
 
 enum diag308_ipl_type {
diff --git a/arch/s390/include/asm/irq.h b/arch/s390/include/asm/irq.h
index f97b055de76a..70c9bce766f5 100644
--- a/arch/s390/include/asm/irq.h
+++ b/arch/s390/include/asm/irq.h
@@ -7,11 +7,8 @@
 
 #define NR_IRQS_BASE	3
 
-#ifdef CONFIG_PCI_NR_MSI
-# define NR_IRQS	(NR_IRQS_BASE + CONFIG_PCI_NR_MSI)
-#else
-# define NR_IRQS	NR_IRQS_BASE
-#endif
+#define NR_IRQS	NR_IRQS_BASE
+#define NR_IRQS_LEGACY NR_IRQS_BASE
 
 /* External interruption codes */
 #define EXT_IRQ_INTERRUPT_KEY	0x0040
diff --git a/arch/s390/include/asm/jump_label.h b/arch/s390/include/asm/jump_label.h
index 7f9fd5e3f1bf..9be198f5ee79 100644
--- a/arch/s390/include/asm/jump_label.h
+++ b/arch/s390/include/asm/jump_label.h
@@ -4,6 +4,7 @@
 #ifndef __ASSEMBLY__
 
 #include <linux/types.h>
+#include <linux/stringify.h>
 
 #define JUMP_LABEL_NOP_SIZE 6
 #define JUMP_LABEL_NOP_OFFSET 2
diff --git a/arch/s390/include/asm/kprobes.h b/arch/s390/include/asm/kprobes.h
index b47ad3b642cc..591e5a5279b0 100644
--- a/arch/s390/include/asm/kprobes.h
+++ b/arch/s390/include/asm/kprobes.h
@@ -43,9 +43,9 @@ typedef u16 kprobe_opcode_t;
 #define MAX_INSN_SIZE		0x0003
 #define MAX_STACK_SIZE		64
 #define MIN_STACK_SIZE(ADDR) (((MAX_STACK_SIZE) < \
-	(((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR))) \
+	(((unsigned long)task_stack_page(current)) + THREAD_SIZE - (ADDR))) \
 	? (MAX_STACK_SIZE) \
-	: (((unsigned long)current_thread_info()) + THREAD_SIZE - (ADDR)))
+	: (((unsigned long)task_stack_page(current)) + THREAD_SIZE - (ADDR)))
 
 #define kretprobe_blacklist_size 0
 
diff --git a/arch/s390/include/asm/mathemu.h b/arch/s390/include/asm/mathemu.h
deleted file mode 100644
index 614dfaf47f71..000000000000
--- a/arch/s390/include/asm/mathemu.h
+++ /dev/null
@@ -1,28 +0,0 @@
-/*
- *    IEEE floating point emulation.
- *
- *  S390 version
- *    Copyright IBM Corp. 1999
- *    Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
- */
-
-#ifndef __MATHEMU__
-#define __MATHEMU__
-
-extern int math_emu_b3(__u8 *, struct pt_regs *);
-extern int math_emu_ed(__u8 *, struct pt_regs *);
-extern int math_emu_ldr(__u8 *);
-extern int math_emu_ler(__u8 *);
-extern int math_emu_std(__u8 *, struct pt_regs *);
-extern int math_emu_ld(__u8 *, struct pt_regs *);
-extern int math_emu_ste(__u8 *, struct pt_regs *);
-extern int math_emu_le(__u8 *, struct pt_regs *);
-extern int math_emu_lfpc(__u8 *, struct pt_regs *);
-extern int math_emu_stfpc(__u8 *, struct pt_regs *);
-extern int math_emu_srnm(__u8 *, struct pt_regs *);
-
-#endif                                 /* __MATHEMU__                      */
-
-
-
-
diff --git a/arch/s390/include/asm/mmu.h b/arch/s390/include/asm/mmu.h
index 081b2ad99d73..18226437a832 100644
--- a/arch/s390/include/asm/mmu.h
+++ b/arch/s390/include/asm/mmu.h
@@ -6,7 +6,7 @@
 
 typedef struct {
 	cpumask_t cpu_attach_mask;
-	atomic_t attach_count;
+	atomic_t flush_count;
 	unsigned int flush_mm;
 	spinlock_t list_lock;
 	struct list_head pgtable_list;
diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h
index c837b79b455d..f77c638bf397 100644
--- a/arch/s390/include/asm/mmu_context.h
+++ b/arch/s390/include/asm/mmu_context.h
@@ -19,7 +19,7 @@ static inline int init_new_context(struct task_struct *tsk,
 	INIT_LIST_HEAD(&mm->context.pgtable_list);
 	INIT_LIST_HEAD(&mm->context.gmap_list);
 	cpumask_clear(&mm->context.cpu_attach_mask);
-	atomic_set(&mm->context.attach_count, 0);
+	atomic_set(&mm->context.flush_count, 0);
 	mm->context.flush_mm = 0;
 #ifdef CONFIG_PGSTE
 	mm->context.alloc_pgste = page_table_allocate_pgste;
@@ -90,15 +90,12 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 	S390_lowcore.user_asce = next->context.asce;
 	if (prev == next)
 		return;
-	if (MACHINE_HAS_TLB_LC)
-		cpumask_set_cpu(cpu, &next->context.cpu_attach_mask);
+	cpumask_set_cpu(cpu, &next->context.cpu_attach_mask);
+	cpumask_set_cpu(cpu, mm_cpumask(next));
 	/* Clear old ASCE by loading the kernel ASCE. */
 	__ctl_load(S390_lowcore.kernel_asce, 1, 1);
 	__ctl_load(S390_lowcore.kernel_asce, 7, 7);
-	atomic_inc(&next->context.attach_count);
-	atomic_dec(&prev->context.attach_count);
-	if (MACHINE_HAS_TLB_LC)
-		cpumask_clear_cpu(cpu, &prev->context.cpu_attach_mask);
+	cpumask_clear_cpu(cpu, &prev->context.cpu_attach_mask);
 }
 
 #define finish_arch_post_lock_switch finish_arch_post_lock_switch
@@ -110,10 +107,9 @@ static inline void finish_arch_post_lock_switch(void)
 	load_kernel_asce();
 	if (mm) {
 		preempt_disable();
-		while (atomic_read(&mm->context.attach_count) >> 16)
+		while (atomic_read(&mm->context.flush_count))
 			cpu_relax();
 
-		cpumask_set_cpu(smp_processor_id(), mm_cpumask(mm));
 		if (mm->context.flush_mm)
 			__tlb_flush_mm(mm);
 		preempt_enable();
@@ -128,7 +124,6 @@ static inline void activate_mm(struct mm_struct *prev,
                                struct mm_struct *next)
 {
 	switch_mm(prev, next, current);
-	cpumask_set_cpu(smp_processor_id(), mm_cpumask(next));
 	set_user_asce(next);
 }
 
diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h
index 53eacbd4f09b..b2146c4119b2 100644
--- a/arch/s390/include/asm/page.h
+++ b/arch/s390/include/asm/page.h
@@ -21,6 +21,7 @@
 #define HPAGE_SIZE	(1UL << HPAGE_SHIFT)
 #define HPAGE_MASK	(~(HPAGE_SIZE - 1))
 #define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT - PAGE_SHIFT)
+#define HUGE_MAX_HSTATE		2
 
 #define ARCH_HAS_SETCLEAR_HUGE_PTE
 #define ARCH_HAS_HUGE_PTE_TYPE
@@ -30,11 +31,12 @@
 #include <asm/setup.h>
 #ifndef __ASSEMBLY__
 
+void __storage_key_init_range(unsigned long start, unsigned long end);
+
 static inline void storage_key_init_range(unsigned long start, unsigned long end)
 {
-#if PAGE_DEFAULT_KEY
-	__storage_key_init_range(start, end);
-#endif
+	if (PAGE_DEFAULT_KEY)
+		__storage_key_init_range(start, end);
 }
 
 #define clear_page(page)	memset((page), 0, PAGE_SIZE)
diff --git a/arch/s390/include/asm/perf_event.h b/arch/s390/include/asm/perf_event.h
index 1f7ff85c5e4c..c64c0befd3f3 100644
--- a/arch/s390/include/asm/perf_event.h
+++ b/arch/s390/include/asm/perf_event.h
@@ -86,16 +86,4 @@ struct sf_raw_sample {
 	u8		    padding[];	  /* Padding to next multiple of 8 */
 } __packed;
 
-/* Perf hardware reserve and release functions */
-#ifdef CONFIG_PERF_EVENTS
-int perf_reserve_sampling(void);
-void perf_release_sampling(void);
-#else /* CONFIG_PERF_EVENTS */
-static inline int perf_reserve_sampling(void)
-{
-	return 0;
-}
-static inline void perf_release_sampling(void) {}
-#endif /* CONFIG_PERF_EVENTS */
-
 #endif /* _ASM_S390_PERF_EVENT_H */
diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
index 18d2beb89340..ea1533e07271 100644
--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h
@@ -28,12 +28,33 @@
 #include <linux/mm_types.h>
 #include <linux/page-flags.h>
 #include <linux/radix-tree.h>
+#include <linux/atomic.h>
 #include <asm/bug.h>
 #include <asm/page.h>
 
-extern pgd_t swapper_pg_dir[] __attribute__ ((aligned (4096)));
+extern pgd_t swapper_pg_dir[];
 extern void paging_init(void);
 extern void vmem_map_init(void);
+pmd_t *vmem_pmd_alloc(void);
+pte_t *vmem_pte_alloc(void);
+
+enum {
+	PG_DIRECT_MAP_4K = 0,
+	PG_DIRECT_MAP_1M,
+	PG_DIRECT_MAP_2G,
+	PG_DIRECT_MAP_MAX
+};
+
+extern atomic_long_t direct_pages_count[PG_DIRECT_MAP_MAX];
+
+static inline void update_page_count(int level, long count)
+{
+	if (IS_ENABLED(CONFIG_PROC_FS))
+		atomic_long_add(count, &direct_pages_count[level]);
+}
+
+struct seq_file;
+void arch_report_meminfo(struct seq_file *m);
 
 /*
  * The S390 doesn't have any external MMU info: the kernel page
@@ -270,8 +291,23 @@ static inline int is_module_addr(void *addr)
 #define _REGION3_ENTRY		(_REGION_ENTRY_TYPE_R3 | _REGION_ENTRY_LENGTH)
 #define _REGION3_ENTRY_EMPTY	(_REGION_ENTRY_TYPE_R3 | _REGION_ENTRY_INVALID)
 
-#define _REGION3_ENTRY_LARGE	0x400	/* RTTE-format control, large page  */
-#define _REGION3_ENTRY_RO	0x200	/* page protection bit		    */
+#define _REGION3_ENTRY_ORIGIN_LARGE ~0x7fffffffUL /* large page address	     */
+#define _REGION3_ENTRY_ORIGIN  ~0x7ffUL/* region third table origin	     */
+
+#define _REGION3_ENTRY_DIRTY	0x2000	/* SW region dirty bit */
+#define _REGION3_ENTRY_YOUNG	0x1000	/* SW region young bit */
+#define _REGION3_ENTRY_LARGE	0x0400	/* RTTE-format control, large page  */
+#define _REGION3_ENTRY_READ	0x0002	/* SW region read bit */
+#define _REGION3_ENTRY_WRITE	0x0001	/* SW region write bit */
+
+#ifdef CONFIG_MEM_SOFT_DIRTY
+#define _REGION3_ENTRY_SOFT_DIRTY 0x4000 /* SW region soft dirty bit */
+#else
+#define _REGION3_ENTRY_SOFT_DIRTY 0x0000 /* SW region soft dirty bit */
+#endif
+
+#define _REGION_ENTRY_BITS	 0xfffffffffffff227UL
+#define _REGION_ENTRY_BITS_LARGE 0xffffffff8000fe27UL
 
 /* Bits in the segment table entry */
 #define _SEGMENT_ENTRY_BITS	0xfffffffffffffe33UL
@@ -297,7 +333,8 @@ static inline int is_module_addr(void *addr)
 #endif
 
 /*
- * Segment table entry encoding (R = read-only, I = invalid, y = young bit):
+ * Segment table and region3 table entry encoding
+ * (R = read-only, I = invalid, y = young bit):
  *				dy..R...I...rw
  * prot-none, clean, old	00..1...1...00
  * prot-none, clean, young	01..1...1...00
@@ -391,6 +428,33 @@ static inline int is_module_addr(void *addr)
 				 _SEGMENT_ENTRY_READ)
 #define SEGMENT_WRITE	__pgprot(_SEGMENT_ENTRY_READ | \
 				 _SEGMENT_ENTRY_WRITE)
+#define SEGMENT_KERNEL	__pgprot(_SEGMENT_ENTRY |	\
+				 _SEGMENT_ENTRY_LARGE |	\
+				 _SEGMENT_ENTRY_READ |	\
+				 _SEGMENT_ENTRY_WRITE | \
+				 _SEGMENT_ENTRY_YOUNG | \
+				 _SEGMENT_ENTRY_DIRTY)
+#define SEGMENT_KERNEL_RO __pgprot(_SEGMENT_ENTRY |	\
+				 _SEGMENT_ENTRY_LARGE |	\
+				 _SEGMENT_ENTRY_READ |	\
+				 _SEGMENT_ENTRY_YOUNG |	\
+				 _SEGMENT_ENTRY_PROTECT)
+
+/*
+ * Region3 entry (large page) protection definitions.
+ */
+
+#define REGION3_KERNEL	__pgprot(_REGION_ENTRY_TYPE_R3 | \
+				 _REGION3_ENTRY_LARGE |	 \
+				 _REGION3_ENTRY_READ |	 \
+				 _REGION3_ENTRY_WRITE |	 \
+				 _REGION3_ENTRY_YOUNG |	 \
+				 _REGION3_ENTRY_DIRTY)
+#define REGION3_KERNEL_RO __pgprot(_REGION_ENTRY_TYPE_R3 | \
+				   _REGION3_ENTRY_LARGE |  \
+				   _REGION3_ENTRY_READ |   \
+				   _REGION3_ENTRY_YOUNG |  \
+				   _REGION_ENTRY_PROTECT)
 
 static inline int mm_has_pgste(struct mm_struct *mm)
 {
@@ -424,6 +488,53 @@ static inline int mm_use_skey(struct mm_struct *mm)
 	return 0;
 }
 
+static inline void csp(unsigned int *ptr, unsigned int old, unsigned int new)
+{
+	register unsigned long reg2 asm("2") = old;
+	register unsigned long reg3 asm("3") = new;
+	unsigned long address = (unsigned long)ptr | 1;
+
+	asm volatile(
+		"	csp	%0,%3"
+		: "+d" (reg2), "+m" (*ptr)
+		: "d" (reg3), "d" (address)
+		: "cc");
+}
+
+static inline void cspg(unsigned long *ptr, unsigned long old, unsigned long new)
+{
+	register unsigned long reg2 asm("2") = old;
+	register unsigned long reg3 asm("3") = new;
+	unsigned long address = (unsigned long)ptr | 1;
+
+	asm volatile(
+		"	.insn	rre,0xb98a0000,%0,%3"
+		: "+d" (reg2), "+m" (*ptr)
+		: "d" (reg3), "d" (address)
+		: "cc");
+}
+
+#define CRDTE_DTT_PAGE		0x00UL
+#define CRDTE_DTT_SEGMENT	0x10UL
+#define CRDTE_DTT_REGION3	0x14UL
+#define CRDTE_DTT_REGION2	0x18UL
+#define CRDTE_DTT_REGION1	0x1cUL
+
+static inline void crdte(unsigned long old, unsigned long new,
+			 unsigned long table, unsigned long dtt,
+			 unsigned long address, unsigned long asce)
+{
+	register unsigned long reg2 asm("2") = old;
+	register unsigned long reg3 asm("3") = new;
+	register unsigned long reg4 asm("4") = table | dtt;
+	register unsigned long reg5 asm("5") = address;
+
+	asm volatile(".insn rrf,0xb98f0000,%0,%2,%4,0"
+		     : "+d" (reg2)
+		     : "d" (reg3), "d" (reg4), "d" (reg5), "a" (asce)
+		     : "memory", "cc");
+}
+
 /*
  * pgd/pmd/pte query functions
  */
@@ -465,7 +576,7 @@ static inline int pud_none(pud_t pud)
 {
 	if ((pud_val(pud) & _REGION_ENTRY_TYPE_MASK) < _REGION_ENTRY_TYPE_R3)
 		return 0;
-	return (pud_val(pud) & _REGION_ENTRY_INVALID) != 0UL;
+	return pud_val(pud) == _REGION3_ENTRY_EMPTY;
 }
 
 static inline int pud_large(pud_t pud)
@@ -475,17 +586,35 @@ static inline int pud_large(pud_t pud)
 	return !!(pud_val(pud) & _REGION3_ENTRY_LARGE);
 }
 
+static inline unsigned long pud_pfn(pud_t pud)
+{
+	unsigned long origin_mask;
+
+	origin_mask = _REGION3_ENTRY_ORIGIN;
+	if (pud_large(pud))
+		origin_mask = _REGION3_ENTRY_ORIGIN_LARGE;
+	return (pud_val(pud) & origin_mask) >> PAGE_SHIFT;
+}
+
+static inline int pmd_large(pmd_t pmd)
+{
+	return (pmd_val(pmd) & _SEGMENT_ENTRY_LARGE) != 0;
+}
+
+static inline int pmd_bad(pmd_t pmd)
+{
+	if (pmd_large(pmd))
+		return (pmd_val(pmd) & ~_SEGMENT_ENTRY_BITS_LARGE) != 0;
+	return (pmd_val(pmd) & ~_SEGMENT_ENTRY_BITS) != 0;
+}
+
 static inline int pud_bad(pud_t pud)
 {
-	/*
-	 * With dynamic page table levels the pud can be a region table
-	 * entry or a segment table entry. Check for the bit that are
-	 * invalid for either table entry.
-	 */
-	unsigned long mask =
-		~_SEGMENT_ENTRY_ORIGIN & ~_REGION_ENTRY_INVALID &
-		~_REGION_ENTRY_TYPE_MASK & ~_REGION_ENTRY_LENGTH;
-	return (pud_val(pud) & mask) != 0;
+	if ((pud_val(pud) & _REGION_ENTRY_TYPE_MASK) < _REGION_ENTRY_TYPE_R3)
+		return pmd_bad(__pmd(pud_val(pud)));
+	if (pud_large(pud))
+		return (pud_val(pud) & ~_REGION_ENTRY_BITS_LARGE) != 0;
+	return (pud_val(pud) & ~_REGION_ENTRY_BITS) != 0;
 }
 
 static inline int pmd_present(pmd_t pmd)
@@ -498,11 +627,6 @@ static inline int pmd_none(pmd_t pmd)
 	return pmd_val(pmd) == _SEGMENT_ENTRY_INVALID;
 }
 
-static inline int pmd_large(pmd_t pmd)
-{
-	return (pmd_val(pmd) & _SEGMENT_ENTRY_LARGE) != 0;
-}
-
 static inline unsigned long pmd_pfn(pmd_t pmd)
 {
 	unsigned long origin_mask;
@@ -513,13 +637,6 @@ static inline unsigned long pmd_pfn(pmd_t pmd)
 	return (pmd_val(pmd) & origin_mask) >> PAGE_SHIFT;
 }
 
-static inline int pmd_bad(pmd_t pmd)
-{
-	if (pmd_large(pmd))
-		return (pmd_val(pmd) & ~_SEGMENT_ENTRY_BITS_LARGE) != 0;
-	return (pmd_val(pmd) & ~_SEGMENT_ENTRY_BITS) != 0;
-}
-
 #define __HAVE_ARCH_PMD_WRITE
 static inline int pmd_write(pmd_t pmd)
 {
@@ -963,6 +1080,7 @@ static inline pmd_t *pmd_offset(pud_t *pud, unsigned long address)
 #define pte_page(x) pfn_to_page(pte_pfn(x))
 
 #define pmd_page(pmd) pfn_to_page(pmd_pfn(pmd))
+#define pud_page(pud) pfn_to_page(pud_pfn(pud))
 
 /* Find an entry in the lowest level page table.. */
 #define pte_offset(pmd, addr) ((pte_t *) pmd_deref(*(pmd)) + pte_index(addr))
@@ -970,20 +1088,6 @@ static inline pmd_t *pmd_offset(pud_t *pud, unsigned long address)
 #define pte_offset_map(pmd, address) pte_offset_kernel(pmd, address)
 #define pte_unmap(pte) do { } while (0)
 
-#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLB_PAGE)
-static inline unsigned long massage_pgprot_pmd(pgprot_t pgprot)
-{
-	/*
-	 * pgprot is PAGE_NONE, PAGE_READ, or PAGE_WRITE (see __Pxxx / __Sxxx)
-	 * Convert to segment table entry format.
-	 */
-	if (pgprot_val(pgprot) == pgprot_val(PAGE_NONE))
-		return pgprot_val(SEGMENT_NONE);
-	if (pgprot_val(pgprot) == pgprot_val(PAGE_READ))
-		return pgprot_val(SEGMENT_READ);
-	return pgprot_val(SEGMENT_WRITE);
-}
-
 static inline pmd_t pmd_wrprotect(pmd_t pmd)
 {
 	pmd_val(pmd) &= ~_SEGMENT_ENTRY_WRITE;
@@ -1020,6 +1124,56 @@ static inline pmd_t pmd_mkdirty(pmd_t pmd)
 	return pmd;
 }
 
+static inline pud_t pud_wrprotect(pud_t pud)
+{
+	pud_val(pud) &= ~_REGION3_ENTRY_WRITE;
+	pud_val(pud) |= _REGION_ENTRY_PROTECT;
+	return pud;
+}
+
+static inline pud_t pud_mkwrite(pud_t pud)
+{
+	pud_val(pud) |= _REGION3_ENTRY_WRITE;
+	if (pud_large(pud) && !(pud_val(pud) & _REGION3_ENTRY_DIRTY))
+		return pud;
+	pud_val(pud) &= ~_REGION_ENTRY_PROTECT;
+	return pud;
+}
+
+static inline pud_t pud_mkclean(pud_t pud)
+{
+	if (pud_large(pud)) {
+		pud_val(pud) &= ~_REGION3_ENTRY_DIRTY;
+		pud_val(pud) |= _REGION_ENTRY_PROTECT;
+	}
+	return pud;
+}
+
+static inline pud_t pud_mkdirty(pud_t pud)
+{
+	if (pud_large(pud)) {
+		pud_val(pud) |= _REGION3_ENTRY_DIRTY |
+				_REGION3_ENTRY_SOFT_DIRTY;
+		if (pud_val(pud) & _REGION3_ENTRY_WRITE)
+			pud_val(pud) &= ~_REGION_ENTRY_PROTECT;
+	}
+	return pud;
+}
+
+#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLB_PAGE)
+static inline unsigned long massage_pgprot_pmd(pgprot_t pgprot)
+{
+	/*
+	 * pgprot is PAGE_NONE, PAGE_READ, or PAGE_WRITE (see __Pxxx / __Sxxx)
+	 * Convert to segment table entry format.
+	 */
+	if (pgprot_val(pgprot) == pgprot_val(PAGE_NONE))
+		return pgprot_val(SEGMENT_NONE);
+	if (pgprot_val(pgprot) == pgprot_val(PAGE_READ))
+		return pgprot_val(SEGMENT_READ);
+	return pgprot_val(SEGMENT_WRITE);
+}
+
 static inline pmd_t pmd_mkyoung(pmd_t pmd)
 {
 	if (pmd_large(pmd)) {
@@ -1068,15 +1222,8 @@ static inline pmd_t mk_pmd_phys(unsigned long physpage, pgprot_t pgprot)
 
 static inline void __pmdp_csp(pmd_t *pmdp)
 {
-	register unsigned long reg2 asm("2") = pmd_val(*pmdp);
-	register unsigned long reg3 asm("3") = pmd_val(*pmdp) |
-					       _SEGMENT_ENTRY_INVALID;
-	register unsigned long reg4 asm("4") = ((unsigned long) pmdp) + 5;
-
-	asm volatile(
-		"	csp %1,%3"
-		: "=m" (*pmdp)
-		: "d" (reg2), "d" (reg3), "d" (reg4), "m" (*pmdp) : "cc");
+	csp((unsigned int *)pmdp + 1, pmd_val(*pmdp),
+	    pmd_val(*pmdp) | _SEGMENT_ENTRY_INVALID);
 }
 
 static inline void __pmdp_idte(unsigned long address, pmd_t *pmdp)
@@ -1091,6 +1238,19 @@ static inline void __pmdp_idte(unsigned long address, pmd_t *pmdp)
 		: "cc" );
 }
 
+static inline void __pudp_idte(unsigned long address, pud_t *pudp)
+{
+	unsigned long r3o;
+
+	r3o = (unsigned long) pudp - pud_index(address) * sizeof(pud_t);
+	r3o |= _ASCE_TYPE_REGION3;
+	asm volatile(
+		"	.insn	rrf,0xb98e0000,%2,%3,0,0"
+		: "=m" (*pudp)
+		: "m" (*pudp), "a" (r3o), "a" ((address & PUD_MASK))
+		: "cc");
+}
+
 static inline void __pmdp_idte_local(unsigned long address, pmd_t *pmdp)
 {
 	unsigned long sto;
@@ -1103,8 +1263,22 @@ static inline void __pmdp_idte_local(unsigned long address, pmd_t *pmdp)
 		: "cc" );
 }
 
+static inline void __pudp_idte_local(unsigned long address, pud_t *pudp)
+{
+	unsigned long r3o;
+
+	r3o = (unsigned long) pudp - pud_index(address) * sizeof(pud_t);
+	r3o |= _ASCE_TYPE_REGION3;
+	asm volatile(
+		"	.insn	rrf,0xb98e0000,%2,%3,0,1"
+		: "=m" (*pudp)
+		: "m" (*pudp), "a" (r3o), "a" ((address & PUD_MASK))
+		: "cc");
+}
+
 pmd_t pmdp_xchg_direct(struct mm_struct *, unsigned long, pmd_t *, pmd_t);
 pmd_t pmdp_xchg_lazy(struct mm_struct *, unsigned long, pmd_t *, pmd_t);
+pud_t pudp_xchg_direct(struct mm_struct *, unsigned long, pud_t *, pud_t);
 
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 
diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h
index 9d4d311d7e52..09529202ea77 100644
--- a/arch/s390/include/asm/processor.h
+++ b/arch/s390/include/asm/processor.h
@@ -77,7 +77,10 @@ static inline void get_cpu_id(struct cpuid *ptr)
 	asm volatile("stidp %0" : "=Q" (*ptr));
 }
 
-extern void s390_adjust_jiffies(void);
+void s390_adjust_jiffies(void);
+void s390_update_cpu_mhz(void);
+void cpu_detect_mhz_feature(void);
+
 extern const struct seq_operations cpuinfo_op;
 extern int sysctl_ieee_emulation_warnings;
 extern void execve_tail(void);
@@ -233,6 +236,18 @@ void cpu_relax(void);
 
 #define cpu_relax_lowlatency()  barrier()
 
+#define ECAG_CACHE_ATTRIBUTE	0
+#define ECAG_CPU_ATTRIBUTE	1
+
+static inline unsigned long __ecag(unsigned int asi, unsigned char parm)
+{
+	unsigned long val;
+
+	asm volatile(".insn	rsy,0xeb000000004c,%0,0,0(%1)" /* ecag */
+		     : "=d" (val) : "a" (asi << 8 | parm));
+	return val;
+}
+
 static inline void psw_set_key(unsigned int key)
 {
 	asm volatile("spka 0(%0)" : : "d" (key));
diff --git a/arch/s390/include/asm/sections.h b/arch/s390/include/asm/sections.h
index fbd9116eb17b..5ce29fe100ba 100644
--- a/arch/s390/include/asm/sections.h
+++ b/arch/s390/include/asm/sections.h
@@ -4,5 +4,6 @@
 #include <asm-generic/sections.h>
 
 extern char _eshared[], _ehead[];
+extern char __start_ro_after_init[], __end_ro_after_init[];
 
 #endif
diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h
index c0f0efbb6ab5..5e8d57e1cc5e 100644
--- a/arch/s390/include/asm/setup.h
+++ b/arch/s390/include/asm/setup.h
@@ -86,9 +86,13 @@ extern char vmpoff_cmd[];
 #define CONSOLE_IS_SCLP		(console_mode == 1)
 #define CONSOLE_IS_3215		(console_mode == 2)
 #define CONSOLE_IS_3270		(console_mode == 3)
+#define CONSOLE_IS_VT220	(console_mode == 4)
+#define CONSOLE_IS_HVC		(console_mode == 5)
 #define SET_CONSOLE_SCLP	do { console_mode = 1; } while (0)
 #define SET_CONSOLE_3215	do { console_mode = 2; } while (0)
 #define SET_CONSOLE_3270	do { console_mode = 3; } while (0)
+#define SET_CONSOLE_VT220	do { console_mode = 4; } while (0)
+#define SET_CONSOLE_HVC		do { console_mode = 5; } while (0)
 
 #define NSS_NAME_SIZE	8
 extern char kernel_nss_name[];
diff --git a/arch/s390/include/asm/sfp-machine.h b/arch/s390/include/asm/sfp-machine.h
deleted file mode 100644
index 4e16aede4b06..000000000000
--- a/arch/s390/include/asm/sfp-machine.h
+++ /dev/null
@@ -1,142 +0,0 @@
-/* Machine-dependent software floating-point definitions.
-   S/390 kernel version.
-   Copyright (C) 1997,1998,1999 Free Software Foundation, Inc.
-   This file is part of the GNU C Library.
-   Contributed by Richard Henderson (rth@cygnus.com),
-		  Jakub Jelinek (jj@ultra.linux.cz),
-		  David S. Miller (davem@redhat.com) and
-		  Peter Maydell (pmaydell@chiark.greenend.org.uk).
-
-   The GNU C Library is free software; you can redistribute it and/or
-   modify it under the terms of the GNU Library General Public License as
-   published by the Free Software Foundation; either version 2 of the
-   License, or (at your option) any later version.
-
-   The GNU C Library is distributed in the hope that it will be useful,
-   but WITHOUT ANY WARRANTY; without even the implied warranty of
-   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
-   Library General Public License for more details.
-
-   You should have received a copy of the GNU Library General Public
-   License along with the GNU C Library; see the file COPYING.LIB.  If
-   not, write to the Free Software Foundation, Inc.,
-   59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.  */
-
-#ifndef _SFP_MACHINE_H
-#define _SFP_MACHINE_H
-   
-
-#define _FP_W_TYPE_SIZE		32
-#define _FP_W_TYPE		unsigned int
-#define _FP_WS_TYPE		signed int
-#define _FP_I_TYPE		int
-
-#define _FP_MUL_MEAT_S(R,X,Y)					\
-  _FP_MUL_MEAT_1_wide(_FP_WFRACBITS_S,R,X,Y,umul_ppmm)
-#define _FP_MUL_MEAT_D(R,X,Y)					\
-  _FP_MUL_MEAT_2_wide(_FP_WFRACBITS_D,R,X,Y,umul_ppmm)
-#define _FP_MUL_MEAT_Q(R,X,Y)					\
-  _FP_MUL_MEAT_4_wide(_FP_WFRACBITS_Q,R,X,Y,umul_ppmm)
-
-#define _FP_DIV_MEAT_S(R,X,Y)	_FP_DIV_MEAT_1_udiv(S,R,X,Y)
-#define _FP_DIV_MEAT_D(R,X,Y)	_FP_DIV_MEAT_2_udiv(D,R,X,Y)
-#define _FP_DIV_MEAT_Q(R,X,Y)	_FP_DIV_MEAT_4_udiv(Q,R,X,Y)
-
-#define _FP_NANFRAC_S		((_FP_QNANBIT_S << 1) - 1)
-#define _FP_NANFRAC_D		((_FP_QNANBIT_D << 1) - 1), -1
-#define _FP_NANFRAC_Q		((_FP_QNANBIT_Q << 1) - 1), -1, -1, -1
-#define _FP_NANSIGN_S		0
-#define _FP_NANSIGN_D		0
-#define _FP_NANSIGN_Q		0
-
-#define _FP_KEEPNANFRACP 1
-
-/*
- * If one NaN is signaling and the other is not,
- * we choose that one, otherwise we choose X.
- */
-#define _FP_CHOOSENAN(fs, wc, R, X, Y, OP)                      \
-  do {                                                          \
-    if ((_FP_FRAC_HIGH_RAW_##fs(X) & _FP_QNANBIT_##fs)          \
-        && !(_FP_FRAC_HIGH_RAW_##fs(Y) & _FP_QNANBIT_##fs))     \
-      {                                                         \
-        R##_s = Y##_s;                                          \
-        _FP_FRAC_COPY_##wc(R,Y);                                \
-      }                                                         \
-    else                                                        \
-      {                                                         \
-        R##_s = X##_s;                                          \
-        _FP_FRAC_COPY_##wc(R,X);                                \
-      }                                                         \
-    R##_c = FP_CLS_NAN;                                         \
-  } while (0)
-
-/* Some assembly to speed things up. */
-#define __FP_FRAC_ADD_3(r2,r1,r0,x2,x1,x0,y2,y1,y0) ({		\
-	unsigned int __r2 = (x2) + (y2);			\
-	unsigned int __r1 = (x1);				\
-	unsigned int __r0 = (x0);				\
-	asm volatile(						\
-		"	alr	%2,%3\n"			\
-		"	brc	12,0f\n"			\
-		"	lhi	0,1\n"				\
-		"	alr	%1,0\n"				\
-		"	brc	12,0f\n"			\
-		"	alr	%0,0\n"				\
-		"0:"						\
-		: "+&d" (__r2), "+&d" (__r1), "+&d" (__r0)	\
-		: "d" (y0), "i" (1) : "cc", "0" );		\
-	asm volatile(						\
-		"	alr	%1,%2\n"			\
-		"	brc	12,0f\n"			\
-		"	ahi	%0,1\n"				\
-		"0:"						\
-		: "+&d" (__r2), "+&d" (__r1)			\
-		: "d" (y1) : "cc");				\
-	(r2) = __r2;						\
-	(r1) = __r1;						\
-	(r0) = __r0;						\
-})
-
-#define __FP_FRAC_SUB_3(r2,r1,r0,x2,x1,x0,y2,y1,y0) ({		\
-	unsigned int __r2 = (x2) - (y2);			\
-	unsigned int __r1 = (x1);				\
-	unsigned int __r0 = (x0);				\
-	asm volatile(						\
-		"	slr   %2,%3\n"				\
-		"	brc	3,0f\n"				\
-		"	lhi	0,1\n"				\
-		"	slr	%1,0\n"				\
-		"	brc	3,0f\n"				\
-		"	slr	%0,0\n"				\
-		"0:"						\
-		: "+&d" (__r2), "+&d" (__r1), "+&d" (__r0)	\
-		: "d" (y0) : "cc", "0");			\
-	asm volatile(						\
-		"	slr	%1,%2\n"			\
-		"	brc	3,0f\n"				\
-		"	ahi	%0,-1\n"			\
-		"0:"						\
-		: "+&d" (__r2), "+&d" (__r1)			\
-		: "d" (y1) : "cc");				\
-	(r2) = __r2;						\
-	(r1) = __r1;						\
-	(r0) = __r0;						\
-})
-
-#define __FP_FRAC_DEC_3(x2,x1,x0,y2,y1,y0) __FP_FRAC_SUB_3(x2,x1,x0,x2,x1,x0,y2,y1,y0)
-
-/* Obtain the current rounding mode. */
-#define FP_ROUNDMODE	mode
-
-/* Exception flags. */
-#define FP_EX_INVALID		0x800000
-#define FP_EX_DIVZERO		0x400000
-#define FP_EX_OVERFLOW		0x200000
-#define FP_EX_UNDERFLOW		0x100000
-#define FP_EX_INEXACT		0x080000
-
-/* We write the results always */
-#define FP_INHIBIT_RESULTS 0
-
-#endif
diff --git a/arch/s390/include/asm/sfp-util.h b/arch/s390/include/asm/sfp-util.h
deleted file mode 100644
index c8b7cf9d6279..000000000000
--- a/arch/s390/include/asm/sfp-util.h
+++ /dev/null
@@ -1,67 +0,0 @@
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/types.h>
-#include <asm/byteorder.h>
-
-#define add_ssaaaa(sh, sl, ah, al, bh, bl) ({		\
-	unsigned int __sh = (ah);			\
-	unsigned int __sl = (al);			\
-	asm volatile(					\
-		"	alr	%1,%3\n"		\
-		"	brc	12,0f\n"		\
-		"	ahi	%0,1\n"			\
-		"0:	alr  %0,%2"			\
-		: "+&d" (__sh), "+d" (__sl)		\
-		: "d" (bh), "d" (bl) : "cc");		\
-	(sh) = __sh;					\
-	(sl) = __sl;					\
-})
-
-#define sub_ddmmss(sh, sl, ah, al, bh, bl) ({		\
-	unsigned int __sh = (ah);			\
-	unsigned int __sl = (al);			\
-	asm volatile(					\
-		"	slr	%1,%3\n"		\
-		"	brc	3,0f\n"			\
-		"	ahi	%0,-1\n"		\
-		"0:	slr	%0,%2"			\
-		: "+&d" (__sh), "+d" (__sl)		\
-		: "d" (bh), "d" (bl) : "cc");		\
-	(sh) = __sh;					\
-	(sl) = __sl;					\
-})
-
-/* a umul b = a mul b + (a>=2<<31) ? b<<32:0 + (b>=2<<31) ? a<<32:0 */
-#define umul_ppmm(wh, wl, u, v) ({			\
-	unsigned int __wh = u;				\
-	unsigned int __wl = v;				\
-	asm volatile(					\
-		"	ltr	1,%0\n"			\
-		"	mr	0,%1\n"			\
-		"	jnm	0f\n"				\
-		"	alr	0,%1\n"			\
-		"0:	ltr	%1,%1\n"			\
-		"	jnm	1f\n"				\
-		"	alr	0,%0\n"			\
-		"1:	lr	%0,0\n"			\
-		"	lr	%1,1\n"			\
-		: "+d" (__wh), "+d" (__wl)		\
-		: : "0", "1", "cc");			\
-	wh = __wh;					\
-	wl = __wl;					\
-})
-
-#define udiv_qrnnd(q, r, n1, n0, d)			\
-  do { unsigned long __n;				\
-       unsigned int __r, __d;				\
-    __n = ((unsigned long)(n1) << 32) + n0;		\
-    __d = (d);						\
-    (q) = __n / __d;					\
-    (r) = __n % __d;					\
-  } while (0)
-
-#define UDIV_NEEDS_NORMALIZATION 0
-
-#define abort() BUG()
-
-#define __BYTE_ORDER __BIG_ENDIAN
diff --git a/arch/s390/include/asm/sigp.h b/arch/s390/include/asm/sigp.h
index 1c8f33fca356..72df5f2de6b0 100644
--- a/arch/s390/include/asm/sigp.h
+++ b/arch/s390/include/asm/sigp.h
@@ -37,8 +37,8 @@
 
 #ifndef __ASSEMBLY__
 
-static inline int __pcpu_sigp(u16 addr, u8 order, unsigned long parm,
-			      u32 *status)
+static inline int ____pcpu_sigp(u16 addr, u8 order, unsigned long parm,
+				u32 *status)
 {
 	register unsigned long reg1 asm ("1") = parm;
 	int cc;
@@ -48,8 +48,19 @@ static inline int __pcpu_sigp(u16 addr, u8 order, unsigned long parm,
 		"	ipm	%0\n"
 		"	srl	%0,28\n"
 		: "=d" (cc), "+d" (reg1) : "d" (addr), "a" (order) : "cc");
+	*status = reg1;
+	return cc;
+}
+
+static inline int __pcpu_sigp(u16 addr, u8 order, unsigned long parm,
+			      u32 *status)
+{
+	u32 _status;
+	int cc;
+
+	cc = ____pcpu_sigp(addr, order, parm, &_status);
 	if (status && cc == 1)
-		*status = reg1;
+		*status = _status;
 	return cc;
 }
 
diff --git a/arch/s390/include/asm/stp.h b/arch/s390/include/asm/stp.h
new file mode 100644
index 000000000000..7689727585b2
--- /dev/null
+++ b/arch/s390/include/asm/stp.h
@@ -0,0 +1,51 @@
+/*
+ *  Copyright IBM Corp. 2006
+ *  Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
+ */
+#ifndef __S390_STP_H
+#define __S390_STP_H
+
+/* notifier for syncs */
+extern struct atomic_notifier_head s390_epoch_delta_notifier;
+
+/* STP interruption parameter */
+struct stp_irq_parm {
+	unsigned int _pad0	: 14;
+	unsigned int tsc	: 1;	/* Timing status change */
+	unsigned int lac	: 1;	/* Link availability change */
+	unsigned int tcpc	: 1;	/* Time control parameter change */
+	unsigned int _pad2	: 15;
+} __attribute__ ((packed));
+
+#define STP_OP_SYNC	1
+#define STP_OP_CTRL	3
+
+struct stp_sstpi {
+	unsigned int rsvd0;
+	unsigned int rsvd1 : 8;
+	unsigned int stratum : 8;
+	unsigned int vbits : 16;
+	unsigned int leaps : 16;
+	unsigned int tmd : 4;
+	unsigned int ctn : 4;
+	unsigned int rsvd2 : 3;
+	unsigned int c : 1;
+	unsigned int tst : 4;
+	unsigned int tzo : 16;
+	unsigned int dsto : 16;
+	unsigned int ctrl : 16;
+	unsigned int rsvd3 : 16;
+	unsigned int tto;
+	unsigned int rsvd4;
+	unsigned int ctnid[3];
+	unsigned int rsvd5;
+	unsigned int todoff[4];
+	unsigned int rsvd6[48];
+} __attribute__ ((packed));
+
+/* Functions needed by the machine check handler */
+int stp_sync_check(void);
+int stp_island_check(void);
+void stp_queue_work(void);
+
+#endif /* __S390_STP_H */
diff --git a/arch/s390/include/asm/timex.h b/arch/s390/include/asm/timex.h
index dcb6312a0b91..0bb08f341c09 100644
--- a/arch/s390/include/asm/timex.h
+++ b/arch/s390/include/asm/timex.h
@@ -52,6 +52,70 @@ static inline void store_clock_comparator(__u64 *time)
 
 void clock_comparator_work(void);
 
+void __init ptff_init(void);
+
+extern unsigned char ptff_function_mask[16];
+extern unsigned long lpar_offset;
+extern unsigned long initial_leap_seconds;
+
+/* Function codes for the ptff instruction. */
+#define PTFF_QAF	0x00	/* query available functions */
+#define PTFF_QTO	0x01	/* query tod offset */
+#define PTFF_QSI	0x02	/* query steering information */
+#define PTFF_QUI	0x04	/* query UTC information */
+#define PTFF_ATO	0x40	/* adjust tod offset */
+#define PTFF_STO	0x41	/* set tod offset */
+#define PTFF_SFS	0x42	/* set fine steering rate */
+#define PTFF_SGS	0x43	/* set gross steering rate */
+
+/* Query TOD offset result */
+struct ptff_qto {
+	unsigned long long physical_clock;
+	unsigned long long tod_offset;
+	unsigned long long logical_tod_offset;
+	unsigned long long tod_epoch_difference;
+} __packed;
+
+static inline int ptff_query(unsigned int nr)
+{
+	unsigned char *ptr;
+
+	ptr = ptff_function_mask + (nr >> 3);
+	return (*ptr & (0x80 >> (nr & 7))) != 0;
+}
+
+/* Query UTC information result */
+struct ptff_qui {
+	unsigned int tm : 2;
+	unsigned int ts : 2;
+	unsigned int : 28;
+	unsigned int pad_0x04;
+	unsigned long leap_event;
+	short old_leap;
+	short new_leap;
+	unsigned int pad_0x14;
+	unsigned long prt[5];
+	unsigned long cst[3];
+	unsigned int skew;
+	unsigned int pad_0x5c[41];
+} __packed;
+
+static inline int ptff(void *ptff_block, size_t len, unsigned int func)
+{
+	typedef struct { char _[len]; } addrtype;
+	register unsigned int reg0 asm("0") = func;
+	register unsigned long reg1 asm("1") = (unsigned long) ptff_block;
+	int rc;
+
+	asm volatile(
+		"	.word	0x0104\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
+		: "=d" (rc), "+m" (*(addrtype *) ptff_block)
+		: "d" (reg0), "d" (reg1) : "cc");
+	return rc;
+}
+
 static inline unsigned long long local_tick_disable(void)
 {
 	unsigned long long old;
@@ -105,7 +169,7 @@ static inline cycles_t get_cycles(void)
 	return (cycles_t) get_tod_clock() >> 2;
 }
 
-int get_sync_clock(unsigned long long *clock);
+int get_phys_clock(unsigned long long *clock);
 void init_cpu_timer(void);
 unsigned long long monotonic_clock(void);
 
diff --git a/arch/s390/include/asm/tlbflush.h b/arch/s390/include/asm/tlbflush.h
index a2e6ef32e054..1a691ef740cf 100644
--- a/arch/s390/include/asm/tlbflush.h
+++ b/arch/s390/include/asm/tlbflush.h
@@ -5,6 +5,7 @@
 #include <linux/sched.h>
 #include <asm/processor.h>
 #include <asm/pgalloc.h>
+#include <asm/pgtable.h>
 
 /*
  * Flush all TLB entries on the local CPU.
@@ -44,17 +45,9 @@ void smp_ptlb_all(void);
  */
 static inline void __tlb_flush_global(void)
 {
-	register unsigned long reg2 asm("2");
-	register unsigned long reg3 asm("3");
-	register unsigned long reg4 asm("4");
-	long dummy;
-
-	dummy = 0;
-	reg2 = reg3 = 0;
-	reg4 = ((unsigned long) &dummy) + 1;
-	asm volatile(
-		"	csp	%0,%2"
-		: : "d" (reg2), "d" (reg3), "d" (reg4), "m" (dummy) : "cc" );
+	unsigned int dummy = 0;
+
+	csp(&dummy, 0, 0);
 }
 
 /*
@@ -64,7 +57,7 @@ static inline void __tlb_flush_global(void)
 static inline void __tlb_flush_full(struct mm_struct *mm)
 {
 	preempt_disable();
-	atomic_add(0x10000, &mm->context.attach_count);
+	atomic_inc(&mm->context.flush_count);
 	if (cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id()))) {
 		/* Local TLB flush */
 		__tlb_flush_local();
@@ -76,21 +69,19 @@ static inline void __tlb_flush_full(struct mm_struct *mm)
 			cpumask_copy(mm_cpumask(mm),
 				     &mm->context.cpu_attach_mask);
 	}
-	atomic_sub(0x10000, &mm->context.attach_count);
+	atomic_dec(&mm->context.flush_count);
 	preempt_enable();
 }
 
 /*
- * Flush TLB entries for a specific ASCE on all CPUs.
+ * Flush TLB entries for a specific ASCE on all CPUs. Should never be used
+ * when more than one asce (e.g. gmap) ran on this mm.
  */
 static inline void __tlb_flush_asce(struct mm_struct *mm, unsigned long asce)
 {
-	int active, count;
-
 	preempt_disable();
-	active = (mm == current->active_mm) ? 1 : 0;
-	count = atomic_add_return(0x10000, &mm->context.attach_count);
-	if (MACHINE_HAS_TLB_LC && (count & 0xffff) <= active &&
+	atomic_inc(&mm->context.flush_count);
+	if (MACHINE_HAS_TLB_LC &&
 	    cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id()))) {
 		__tlb_flush_idte_local(asce);
 	} else {
@@ -103,7 +94,7 @@ static inline void __tlb_flush_asce(struct mm_struct *mm, unsigned long asce)
 			cpumask_copy(mm_cpumask(mm),
 				     &mm->context.cpu_attach_mask);
 	}
-	atomic_sub(0x10000, &mm->context.attach_count);
+	atomic_dec(&mm->context.flush_count);
 	preempt_enable();
 }
 
diff --git a/arch/s390/include/asm/topology.h b/arch/s390/include/asm/topology.h
index 6b53962e807e..f15f5571ca2b 100644
--- a/arch/s390/include/asm/topology.h
+++ b/arch/s390/include/asm/topology.h
@@ -14,10 +14,12 @@ struct cpu_topology_s390 {
 	unsigned short core_id;
 	unsigned short socket_id;
 	unsigned short book_id;
+	unsigned short drawer_id;
 	unsigned short node_id;
 	cpumask_t thread_mask;
 	cpumask_t core_mask;
 	cpumask_t book_mask;
+	cpumask_t drawer_mask;
 };
 
 DECLARE_PER_CPU(struct cpu_topology_s390, cpu_topology);
@@ -30,6 +32,8 @@ DECLARE_PER_CPU(struct cpu_topology_s390, cpu_topology);
 #define topology_core_cpumask(cpu)	  (&per_cpu(cpu_topology, cpu).core_mask)
 #define topology_book_id(cpu)		  (per_cpu(cpu_topology, cpu).book_id)
 #define topology_book_cpumask(cpu)	  (&per_cpu(cpu_topology, cpu).book_mask)
+#define topology_drawer_id(cpu)		  (per_cpu(cpu_topology, cpu).drawer_id)
+#define topology_drawer_cpumask(cpu)	  (&per_cpu(cpu_topology, cpu).drawer_mask)
 
 #define mc_capable() 1
 
diff --git a/arch/s390/include/asm/uaccess.h b/arch/s390/include/asm/uaccess.h
index e0900ddf91dd..9b49cf1daa8f 100644
--- a/arch/s390/include/asm/uaccess.h
+++ b/arch/s390/include/asm/uaccess.h
@@ -151,8 +151,65 @@ unsigned long __must_check __copy_to_user(void __user *to, const void *from,
 	__rc;							\
 })
 
-#define __put_user_fn(x, ptr, size) __put_get_user_asm(ptr, x, size, 0x810000UL)
-#define __get_user_fn(x, ptr, size) __put_get_user_asm(x, ptr, size, 0x81UL)
+static inline int __put_user_fn(void *x, void __user *ptr, unsigned long size)
+{
+	unsigned long spec = 0x810000UL;
+	int rc;
+
+	switch (size) {
+	case 1:
+		rc = __put_get_user_asm((unsigned char __user *)ptr,
+					(unsigned char *)x,
+					size, spec);
+		break;
+	case 2:
+		rc = __put_get_user_asm((unsigned short __user *)ptr,
+					(unsigned short *)x,
+					size, spec);
+		break;
+	case 4:
+		rc = __put_get_user_asm((unsigned int __user *)ptr,
+					(unsigned int *)x,
+					size, spec);
+		break;
+	case 8:
+		rc = __put_get_user_asm((unsigned long __user *)ptr,
+					(unsigned long *)x,
+					size, spec);
+		break;
+	};
+	return rc;
+}
+
+static inline int __get_user_fn(void *x, const void __user *ptr, unsigned long size)
+{
+	unsigned long spec = 0x81UL;
+	int rc;
+
+	switch (size) {
+	case 1:
+		rc = __put_get_user_asm((unsigned char *)x,
+					(unsigned char __user *)ptr,
+					size, spec);
+		break;
+	case 2:
+		rc = __put_get_user_asm((unsigned short *)x,
+					(unsigned short __user *)ptr,
+					size, spec);
+		break;
+	case 4:
+		rc = __put_get_user_asm((unsigned int *)x,
+					(unsigned int __user *)ptr,
+					size, spec);
+		break;
+	case 8:
+		rc = __put_get_user_asm((unsigned long *)x,
+					(unsigned long __user *)ptr,
+					size, spec);
+		break;
+	};
+	return rc;
+}
 
 #else /* CONFIG_HAVE_MARCH_Z10_FEATURES */
 
@@ -191,7 +248,7 @@ static inline int __get_user_fn(void *x, const void __user *ptr, unsigned long s
 		__put_user_bad();				\
 		break;						\
 	 }							\
-	__pu_err;						\
+	__builtin_expect(__pu_err, 0);				\
 })
 
 #define put_user(x, ptr)					\
@@ -240,7 +297,7 @@ int __put_user_bad(void) __attribute__((noreturn));
 		__get_user_bad();				\
 		break;						\
 	}							\
-	__gu_err;						\
+	__builtin_expect(__gu_err, 0);				\
 })
 
 #define get_user(x, ptr)					\