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

Pull s390 updates from Martin Schwidefsky:
 "The most notable change for this pull request is the ftrace rework
  from Heiko.  It brings a small performance improvement and the ground
  work to support a new gcc option to replace the mcount blocks with a
  single nop.

  Two new s390 specific system calls are added to emulate user space
  mmio for PCI, an artifact of the how PCI memory is accessed.

  Two patches for the memory management with changes to common code.
  For KVM mm_forbids_zeropage is added which disables the empty zero
  page for an mm that is used by a KVM process.  And an optimization,
  pmdp_get_and_clear_full is added analog to ptep_get_and_clear_full.

  Some micro optimization for the cmpxchg and the spinlock code.

  And as usual bug fixes and cleanups"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (46 commits)
  s390/cputime: fix 31-bit compile
  s390/scm_block: make the number of reqs per HW req configurable
  s390/scm_block: handle multiple requests in one HW request
  s390/scm_block: allocate aidaw pages only when necessary
  s390/scm_block: use mempool to manage aidaw requests
  s390/eadm: change timeout value
  s390/mm: fix memory leak of ptlock in pmd_free_tlb
  s390: use local symbol names in entry[64].S
  s390/ptrace: always include vector registers in core files
  s390/simd: clear vector register pointer on fork/clone
  s390: translate cputime magic constants to macros
  s390/idle: convert open coded idle time seqcount
  s390/idle: add missing irq off lockdep annotation
  s390/debug: avoid function call for debug_sprintf_*
  s390/kprobes: fix instruction copy for out of line execution
  s390: remove diag 44 calls from cpu_relax()
  s390/dasd: retry partition detection
  s390/dasd: fix list corruption for sleep_on requests
  s390/dasd: fix infinite term I/O loop
  s390/dasd: remove unused code
  ...

1 files changed, 83 insertions, 32 deletions

diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c
index 99a567b70d1..eabfb459451 100644
--- a/arch/s390/kernel/ptrace.c
+++ b/arch/s390/kernel/ptrace.c
@@ -248,14 +248,27 @@ static unsigned long __peek_user(struct task_struct *child, addr_t addr)
 		 */
 		tmp = 0;
 
+	} else if (addr == (addr_t) &dummy->regs.fp_regs.fpc) {
+		/*
+		 * floating point control reg. is in the thread structure
+		 */
+		tmp = child->thread.fp_regs.fpc;
+		tmp <<= BITS_PER_LONG - 32;
+
 	} else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) {
-		/* 
-		 * floating point regs. are stored in the thread structure
+		/*
+		 * floating point regs. are either in child->thread.fp_regs
+		 * or the child->thread.vxrs array
 		 */
-		offset = addr - (addr_t) &dummy->regs.fp_regs;
-		tmp = *(addr_t *)((addr_t) &child->thread.fp_regs + offset);
-		if (addr == (addr_t) &dummy->regs.fp_regs.fpc)
-			tmp <<= BITS_PER_LONG - 32;
+		offset = addr - (addr_t) &dummy->regs.fp_regs.fprs;
+#ifdef CONFIG_64BIT
+		if (child->thread.vxrs)
+			tmp = *(addr_t *)
+			       ((addr_t) child->thread.vxrs + 2*offset);
+		else
+#endif
+			tmp = *(addr_t *)
+			       ((addr_t) &child->thread.fp_regs.fprs + offset);
 
 	} else if (addr < (addr_t) (&dummy->regs.per_info + 1)) {
 		/*
@@ -383,16 +396,29 @@ static int __poke_user(struct task_struct *child, addr_t addr, addr_t data)
 		 */
 		return 0;
 
+	} else if (addr == (addr_t) &dummy->regs.fp_regs.fpc) {
+		/*
+		 * floating point control reg. is in the thread structure
+		 */
+		if ((unsigned int) data != 0 ||
+		    test_fp_ctl(data >> (BITS_PER_LONG - 32)))
+			return -EINVAL;
+		child->thread.fp_regs.fpc = data >> (BITS_PER_LONG - 32);
+
 	} else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) {
 		/*
-		 * floating point regs. are stored in the thread structure
+		 * floating point regs. are either in child->thread.fp_regs
+		 * or the child->thread.vxrs array
 		 */
-		if (addr == (addr_t) &dummy->regs.fp_regs.fpc)
-			if ((unsigned int) data != 0 ||
-			    test_fp_ctl(data >> (BITS_PER_LONG - 32)))
-				return -EINVAL;
-		offset = addr - (addr_t) &dummy->regs.fp_regs;
-		*(addr_t *)((addr_t) &child->thread.fp_regs + offset) = data;
+		offset = addr - (addr_t) &dummy->regs.fp_regs.fprs;
+#ifdef CONFIG_64BIT
+		if (child->thread.vxrs)
+			*(addr_t *)((addr_t)
+				child->thread.vxrs + 2*offset) = data;
+		else
+#endif
+			*(addr_t *)((addr_t)
+				&child->thread.fp_regs.fprs + offset) = data;
 
 	} else if (addr < (addr_t) (&dummy->regs.per_info + 1)) {
 		/*
@@ -611,12 +637,26 @@ static u32 __peek_user_compat(struct task_struct *child, addr_t addr)
 		 */
 		tmp = 0;
 
+	} else if (addr == (addr_t) &dummy32->regs.fp_regs.fpc) {
+		/*
+		 * floating point control reg. is in the thread structure
+		 */
+		tmp = child->thread.fp_regs.fpc;
+
 	} else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) {
 		/*
-		 * floating point regs. are stored in the thread structure 
+		 * floating point regs. are either in child->thread.fp_regs
+		 * or the child->thread.vxrs array
 		 */
-	        offset = addr - (addr_t) &dummy32->regs.fp_regs;
-		tmp = *(__u32 *)((addr_t) &child->thread.fp_regs + offset);
+		offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs;
+#ifdef CONFIG_64BIT
+		if (child->thread.vxrs)
+			tmp = *(__u32 *)
+			       ((addr_t) child->thread.vxrs + 2*offset);
+		else
+#endif
+			tmp = *(__u32 *)
+			       ((addr_t) &child->thread.fp_regs.fprs + offset);
 
 	} else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) {
 		/*
@@ -722,15 +762,28 @@ static int __poke_user_compat(struct task_struct *child,
 		 */
 		return 0;
 
-	} else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) {
+	} else if (addr == (addr_t) &dummy32->regs.fp_regs.fpc) {
 		/*
-		 * floating point regs. are stored in the thread structure 
+		 * floating point control reg. is in the thread structure
 		 */
-		if (addr == (addr_t) &dummy32->regs.fp_regs.fpc &&
-		    test_fp_ctl(tmp))
+		if (test_fp_ctl(tmp))
 			return -EINVAL;
-	        offset = addr - (addr_t) &dummy32->regs.fp_regs;
-		*(__u32 *)((addr_t) &child->thread.fp_regs + offset) = tmp;
+		child->thread.fp_regs.fpc = data;
+
+	} else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) {
+		/*
+		 * floating point regs. are either in child->thread.fp_regs
+		 * or the child->thread.vxrs array
+		 */
+		offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs;
+#ifdef CONFIG_64BIT
+		if (child->thread.vxrs)
+			*(__u32 *)((addr_t)
+				child->thread.vxrs + 2*offset) = tmp;
+		else
+#endif
+			*(__u32 *)((addr_t)
+				&child->thread.fp_regs.fprs + offset) = tmp;
 
 	} else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) {
 		/*
@@ -1038,12 +1091,6 @@ static int s390_tdb_set(struct task_struct *target,
 	return 0;
 }
 
-static int s390_vxrs_active(struct task_struct *target,
-			      const struct user_regset *regset)
-{
-	return !!target->thread.vxrs;
-}
-
 static int s390_vxrs_low_get(struct task_struct *target,
 			     const struct user_regset *regset,
 			     unsigned int pos, unsigned int count,
@@ -1052,6 +1099,8 @@ static int s390_vxrs_low_get(struct task_struct *target,
 	__u64 vxrs[__NUM_VXRS_LOW];
 	int i;
 
+	if (!MACHINE_HAS_VX)
+		return -ENODEV;
 	if (target->thread.vxrs) {
 		if (target == current)
 			save_vx_regs(target->thread.vxrs);
@@ -1070,6 +1119,8 @@ static int s390_vxrs_low_set(struct task_struct *target,
 	__u64 vxrs[__NUM_VXRS_LOW];
 	int i, rc;
 
+	if (!MACHINE_HAS_VX)
+		return -ENODEV;
 	if (!target->thread.vxrs) {
 		rc = alloc_vector_registers(target);
 		if (rc)
@@ -1095,6 +1146,8 @@ static int s390_vxrs_high_get(struct task_struct *target,
 {
 	__vector128 vxrs[__NUM_VXRS_HIGH];
 
+	if (!MACHINE_HAS_VX)
+		return -ENODEV;
 	if (target->thread.vxrs) {
 		if (target == current)
 			save_vx_regs(target->thread.vxrs);
@@ -1112,6 +1165,8 @@ static int s390_vxrs_high_set(struct task_struct *target,
 {
 	int rc;
 
+	if (!MACHINE_HAS_VX)
+		return -ENODEV;
 	if (!target->thread.vxrs) {
 		rc = alloc_vector_registers(target);
 		if (rc)
@@ -1196,7 +1251,6 @@ static const struct user_regset s390_regsets[] = {
 		.n = __NUM_VXRS_LOW,
 		.size = sizeof(__u64),
 		.align = sizeof(__u64),
-		.active = s390_vxrs_active,
 		.get = s390_vxrs_low_get,
 		.set = s390_vxrs_low_set,
 	},
@@ -1205,7 +1259,6 @@ static const struct user_regset s390_regsets[] = {
 		.n = __NUM_VXRS_HIGH,
 		.size = sizeof(__vector128),
 		.align = sizeof(__vector128),
-		.active = s390_vxrs_active,
 		.get = s390_vxrs_high_get,
 		.set = s390_vxrs_high_set,
 	},
@@ -1419,7 +1472,6 @@ static const struct user_regset s390_compat_regsets[] = {
 		.n = __NUM_VXRS_LOW,
 		.size = sizeof(__u64),
 		.align = sizeof(__u64),
-		.active = s390_vxrs_active,
 		.get = s390_vxrs_low_get,
 		.set = s390_vxrs_low_set,
 	},
@@ -1428,7 +1480,6 @@ static const struct user_regset s390_compat_regsets[] = {
 		.n = __NUM_VXRS_HIGH,
 		.size = sizeof(__vector128),
 		.align = sizeof(__vector128),
-		.active = s390_vxrs_active,
 		.get = s390_vxrs_high_get,
 		.set = s390_vxrs_high_set,
 	},