Merge branch 'percpu-for-linus' into percpu-for-next

Conflicts:
	arch/sparc/kernel/smp_64.c
	arch/x86/kernel/cpu/perf_counter.c
	arch/x86/kernel/setup_percpu.c
	drivers/cpufreq/cpufreq_ondemand.c
	mm/percpu.c

Conflicts in core and arch percpu codes are mostly from commit
ed78e1e078dd44249f88b1dd8c76dafb39567161 which substituted many
num_possible_cpus() with nr_cpu_ids.  As for-next branch has moved all
the first chunk allocators into mm/percpu.c, the changes are moved
from arch code to mm/percpu.c.

Signed-off-by: Tejun Heo <tj@kernel.org>

29 files changed, 233 insertions, 208 deletions

diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index 9038f39d9d73..06f8d5b5b0f9 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -16,6 +16,8 @@ config PARISC
 	select RTC_DRV_GENERIC
 	select INIT_ALL_POSSIBLE
 	select BUG
+	select HAVE_PERF_COUNTERS
+	select GENERIC_ATOMIC64 if !64BIT
 	help
 	  The PA-RISC microprocessor is designed by Hewlett-Packard and used
 	  in many of their workstations & servers (HP9000 700 and 800 series,
diff --git a/arch/parisc/include/asm/atomic.h b/arch/parisc/include/asm/atomic.h
index 7eeaff944360..8bc9e96699b2 100644
--- a/arch/parisc/include/asm/atomic.h
+++ b/arch/parisc/include/asm/atomic.h
@@ -222,13 +222,13 @@ static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
 
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
 
-#define atomic_add(i,v)	((void)(__atomic_add_return( ((int)(i)),(v))))
-#define atomic_sub(i,v)	((void)(__atomic_add_return(-((int)(i)),(v))))
+#define atomic_add(i,v)	((void)(__atomic_add_return( (i),(v))))
+#define atomic_sub(i,v)	((void)(__atomic_add_return(-(i),(v))))
 #define atomic_inc(v)	((void)(__atomic_add_return(   1,(v))))
 #define atomic_dec(v)	((void)(__atomic_add_return(  -1,(v))))
 
-#define atomic_add_return(i,v)	(__atomic_add_return( ((int)(i)),(v)))
-#define atomic_sub_return(i,v)	(__atomic_add_return(-((int)(i)),(v)))
+#define atomic_add_return(i,v)	(__atomic_add_return( (i),(v)))
+#define atomic_sub_return(i,v)	(__atomic_add_return(-(i),(v)))
 #define atomic_inc_return(v)	(__atomic_add_return(   1,(v)))
 #define atomic_dec_return(v)	(__atomic_add_return(  -1,(v)))
 
@@ -336,7 +336,11 @@ static __inline__ int atomic64_add_unless(atomic64_t *v, long a, long u)
 
 #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0)
 
-#endif /* CONFIG_64BIT */
+#else /* CONFIG_64BIT */
+
+#include <asm-generic/atomic64.h>
+
+#endif /* !CONFIG_64BIT */
 
 #include <asm-generic/atomic-long.h>
 
diff --git a/arch/parisc/include/asm/dma.h b/arch/parisc/include/asm/dma.h
index 31ad0f05af3d..f7a18f968703 100644
--- a/arch/parisc/include/asm/dma.h
+++ b/arch/parisc/include/asm/dma.h
@@ -1,5 +1,4 @@
-/* $Id: dma.h,v 1.2 1999/04/27 00:46:18 deller Exp $
- * linux/include/asm/dma.h: Defines for using and allocating dma channels.
+/* asm/dma.h: Defines for using and allocating dma channels.
  * Written by Hennus Bergman, 1992.
  * High DMA channel support & info by Hannu Savolainen
  * and John Boyd, Nov. 1992.
diff --git a/arch/parisc/include/asm/perf_counter.h b/arch/parisc/include/asm/perf_counter.h
new file mode 100644
index 000000000000..dc9e829f7013
--- /dev/null
+++ b/arch/parisc/include/asm/perf_counter.h
@@ -0,0 +1,7 @@
+#ifndef __ASM_PARISC_PERF_COUNTER_H
+#define __ASM_PARISC_PERF_COUNTER_H
+
+/* parisc only supports software counters through this interface. */
+static inline void set_perf_counter_pending(void) { }
+
+#endif /* __ASM_PARISC_PERF_COUNTER_H */
diff --git a/arch/parisc/include/asm/processor.h b/arch/parisc/include/asm/processor.h
index 9d64df8754ba..9ce66e9d1c2b 100644
--- a/arch/parisc/include/asm/processor.h
+++ b/arch/parisc/include/asm/processor.h
@@ -18,6 +18,7 @@
 #include <asm/types.h>
 #include <asm/system.h>
 #include <asm/percpu.h>
+
 #endif /* __ASSEMBLY__ */
 
 #define KERNEL_STACK_SIZE 	(4*PAGE_SIZE)
@@ -127,6 +128,8 @@ struct thread_struct {
 	unsigned long  flags;
 }; 
 
+#define task_pt_regs(tsk) ((struct pt_regs *)&((tsk)->thread.regs))
+
 /* Thread struct flags. */
 #define PARISC_UAC_NOPRINT	(1UL << 0)	/* see prctl and unaligned.c */
 #define PARISC_UAC_SIGBUS	(1UL << 1)
diff --git a/arch/parisc/include/asm/system.h b/arch/parisc/include/asm/system.h
index ee80c920b464..d91357bca5b4 100644
--- a/arch/parisc/include/asm/system.h
+++ b/arch/parisc/include/asm/system.h
@@ -168,8 +168,8 @@ static inline void set_eiem(unsigned long val)
 /* LDCW, the only atomic read-write operation PA-RISC has. *sigh*.  */
 #define __ldcw(a) ({						\
 	unsigned __ret;						\
-	__asm__ __volatile__(__LDCW " 0(%1),%0"			\
-		: "=r" (__ret) : "r" (a));			\
+	__asm__ __volatile__(__LDCW " 0(%2),%0"			\
+		: "=r" (__ret), "+m" (*(a)) : "r" (a));		\
 	__ret;							\
 })
 
diff --git a/arch/parisc/include/asm/thread_info.h b/arch/parisc/include/asm/thread_info.h
index 0407959da489..4ce0edfbe969 100644
--- a/arch/parisc/include/asm/thread_info.h
+++ b/arch/parisc/include/asm/thread_info.h
@@ -23,7 +23,7 @@ struct thread_info {
 	.flags		= 0,			\
 	.cpu		= 0,			\
 	.addr_limit	= KERNEL_DS,		\
-	.preempt_count	= 1,			\
+	.preempt_count	= INIT_PREEMPT_COUNT,	\
   	.restart_block	= {			\
 		.fn = do_no_restart_syscall	\
 	}					\
diff --git a/arch/parisc/include/asm/tlb.h b/arch/parisc/include/asm/tlb.h
index 383b1db310ee..07924903989e 100644
--- a/arch/parisc/include/asm/tlb.h
+++ b/arch/parisc/include/asm/tlb.h
@@ -21,7 +21,7 @@ do {	if (!(tlb)->fullmm)	\
 
 #include <asm-generic/tlb.h>
 
-#define __pmd_free_tlb(tlb, pmd)	pmd_free((tlb)->mm, pmd)
-#define __pte_free_tlb(tlb, pte)	pte_free((tlb)->mm, pte)
+#define __pmd_free_tlb(tlb, pmd, addr)	pmd_free((tlb)->mm, pmd)
+#define __pte_free_tlb(tlb, pte, addr)	pte_free((tlb)->mm, pte)
 
 #endif
diff --git a/arch/parisc/include/asm/tlbflush.h b/arch/parisc/include/asm/tlbflush.h
index 1f6fd4fc05b9..8f1a8100bf2d 100644
--- a/arch/parisc/include/asm/tlbflush.h
+++ b/arch/parisc/include/asm/tlbflush.h
@@ -12,14 +12,12 @@
  * N class systems, only one PxTLB inter processor broadcast can be
  * active at any one time on the Merced bus.  This tlb purge
  * synchronisation is fairly lightweight and harmless so we activate
- * it on all SMP systems not just the N class.  We also need to have
- * preemption disabled on uniprocessor machines, and spin_lock does that
- * nicely.
+ * it on all systems not just the N class.
  */
 extern spinlock_t pa_tlb_lock;
 
-#define purge_tlb_start(x) spin_lock(&pa_tlb_lock)
-#define purge_tlb_end(x) spin_unlock(&pa_tlb_lock)
+#define purge_tlb_start(flags)	spin_lock_irqsave(&pa_tlb_lock, flags)
+#define purge_tlb_end(flags)	spin_unlock_irqrestore(&pa_tlb_lock, flags)
 
 extern void flush_tlb_all(void);
 extern void flush_tlb_all_local(void *);
@@ -63,14 +61,16 @@ static inline void flush_tlb_mm(struct mm_struct *mm)
 static inline void flush_tlb_page(struct vm_area_struct *vma,
 	unsigned long addr)
 {
+	unsigned long flags;
+
 	/* For one page, it's not worth testing the split_tlb variable */
 
 	mb();
 	mtsp(vma->vm_mm->context,1);
-	purge_tlb_start();
+	purge_tlb_start(flags);
 	pdtlb(addr);
 	pitlb(addr);
-	purge_tlb_end();
+	purge_tlb_end(flags);
 }
 
 void __flush_tlb_range(unsigned long sid,
diff --git a/arch/parisc/include/asm/unistd.h b/arch/parisc/include/asm/unistd.h
index ef26b009dc5d..f3d3b8b012c4 100644
--- a/arch/parisc/include/asm/unistd.h
+++ b/arch/parisc/include/asm/unistd.h
@@ -807,8 +807,12 @@
 #define __NR_dup3		(__NR_Linux + 312)
 #define __NR_pipe2		(__NR_Linux + 313)
 #define __NR_inotify_init1	(__NR_Linux + 314)
+#define __NR_preadv		(__NR_Linux + 315)
+#define __NR_pwritev		(__NR_Linux + 316)
+#define __NR_rt_tgsigqueueinfo	(__NR_Linux + 317)
+#define __NR_perf_counter_open	(__NR_Linux + 318)
 
-#define __NR_Linux_syscalls	(__NR_inotify_init1 + 1)
+#define __NR_Linux_syscalls	(__NR_perf_counter_open + 1)
 
 
 #define __IGNORE_select		/* newselect */
diff --git a/arch/parisc/kernel/cache.c b/arch/parisc/kernel/cache.c
index 837530ea32e7..b6ed34de14e1 100644
--- a/arch/parisc/kernel/cache.c
+++ b/arch/parisc/kernel/cache.c
@@ -1,5 +1,4 @@
-/* $Id: cache.c,v 1.4 2000/01/25 00:11:38 prumpf Exp $
- *
+/*
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
@@ -398,12 +397,13 @@ EXPORT_SYMBOL(flush_kernel_icache_range_asm);
 
 void clear_user_page_asm(void *page, unsigned long vaddr)
 {
+	unsigned long flags;
 	/* This function is implemented in assembly in pacache.S */
 	extern void __clear_user_page_asm(void *page, unsigned long vaddr);
 
-	purge_tlb_start();
+	purge_tlb_start(flags);
 	__clear_user_page_asm(page, vaddr);
-	purge_tlb_end();
+	purge_tlb_end(flags);
 }
 
 #define FLUSH_THRESHOLD 0x80000 /* 0.5MB */
@@ -444,20 +444,24 @@ extern void clear_user_page_asm(void *page, unsigned long vaddr);
 
 void clear_user_page(void *page, unsigned long vaddr, struct page *pg)
 {
+	unsigned long flags;
+
 	purge_kernel_dcache_page((unsigned long)page);
-	purge_tlb_start();
+	purge_tlb_start(flags);
 	pdtlb_kernel(page);
-	purge_tlb_end();
+	purge_tlb_end(flags);
 	clear_user_page_asm(page, vaddr);
 }
 EXPORT_SYMBOL(clear_user_page);
 
 void flush_kernel_dcache_page_addr(void *addr)
 {
+	unsigned long flags;
+
 	flush_kernel_dcache_page_asm(addr);
-	purge_tlb_start();
+	purge_tlb_start(flags);
 	pdtlb_kernel(addr);
-	purge_tlb_end();
+	purge_tlb_end(flags);
 }
 EXPORT_SYMBOL(flush_kernel_dcache_page_addr);
 
@@ -490,8 +494,10 @@ void __flush_tlb_range(unsigned long sid, unsigned long start,
 	if (npages >= 512)  /* 2MB of space: arbitrary, should be tuned */
 		flush_tlb_all();
 	else {
+		unsigned long flags;
+
 		mtsp(sid, 1);
-		purge_tlb_start();
+		purge_tlb_start(flags);
 		if (split_tlb) {
 			while (npages--) {
 				pdtlb(start);
@@ -504,7 +510,7 @@ void __flush_tlb_range(unsigned long sid, unsigned long start,
 				start += PAGE_SIZE;
 			}
 		}
-		purge_tlb_end();
+		purge_tlb_end(flags);
 	}
 }
 
diff --git a/arch/parisc/kernel/entry.S b/arch/parisc/kernel/entry.S
index ae3e70cd1e14..e552e547cb93 100644
--- a/arch/parisc/kernel/entry.S
+++ b/arch/parisc/kernel/entry.S
@@ -553,7 +553,7 @@
 	 * on most of those machines only handles cache transactions.
 	 */
 	extrd,u,*=	\pte,_PAGE_NO_CACHE_BIT+32,1,%r0
-	depi		1,12,1,\prot
+	depdi		1,12,1,\prot
 
 	/* Drop prot bits and convert to page addr for iitlbt and idtlbt */
 	convert_for_tlb_insert20 \pte
diff --git a/arch/parisc/kernel/inventory.c b/arch/parisc/kernel/inventory.c
index bd1f7f1ff74e..d228d8237879 100644
--- a/arch/parisc/kernel/inventory.c
+++ b/arch/parisc/kernel/inventory.c
@@ -170,23 +170,27 @@ static void __init pagezero_memconfig(void)
 static int __init 
 pat_query_module(ulong pcell_loc, ulong mod_index)
 {
-	pdc_pat_cell_mod_maddr_block_t pa_pdc_cell;
+	pdc_pat_cell_mod_maddr_block_t *pa_pdc_cell;
 	unsigned long bytecnt;
 	unsigned long temp;	/* 64-bit scratch value */
 	long status;		/* PDC return value status */
 	struct parisc_device *dev;
 
+	pa_pdc_cell = kmalloc(sizeof (*pa_pdc_cell), GFP_KERNEL);
+	if (!pa_pdc_cell)
+		panic("couldn't allocate memory for PDC_PAT_CELL!");
+
 	/* return cell module (PA or Processor view) */
 	status = pdc_pat_cell_module(&bytecnt, pcell_loc, mod_index,
-				     PA_VIEW, &pa_pdc_cell);
+				     PA_VIEW, pa_pdc_cell);
 
 	if (status != PDC_OK) {
 		/* no more cell modules or error */
 		return status;
 	}
 
-	temp = pa_pdc_cell.cba;
-	dev = alloc_pa_dev(PAT_GET_CBA(temp), &pa_pdc_cell.mod_path);
+	temp = pa_pdc_cell->cba;
+	dev = alloc_pa_dev(PAT_GET_CBA(temp), &(pa_pdc_cell->mod_path));
 	if (!dev) {
 		return PDC_OK;
 	}
@@ -203,8 +207,8 @@ pat_query_module(ulong pcell_loc, ulong mod_index)
 
 	/* save generic info returned from the call */
 	/* REVISIT: who is the consumer of this? not sure yet... */
-	dev->mod_info = pa_pdc_cell.mod_info;	/* pass to PAT_GET_ENTITY() */
-	dev->pmod_loc = pa_pdc_cell.mod_location;
+	dev->mod_info = pa_pdc_cell->mod_info;	/* pass to PAT_GET_ENTITY() */
+	dev->pmod_loc = pa_pdc_cell->mod_location;
 
 	register_parisc_device(dev);	/* advertise device */
 
@@ -216,14 +220,14 @@ pat_query_module(ulong pcell_loc, ulong mod_index)
 
 	case PAT_ENTITY_PROC:
 		printk(KERN_DEBUG "PAT_ENTITY_PROC: id_eid 0x%lx\n",
-			pa_pdc_cell.mod[0]);
+			pa_pdc_cell->mod[0]);
 		break;
 
 	case PAT_ENTITY_MEM:
 		printk(KERN_DEBUG 
 			"PAT_ENTITY_MEM: amount 0x%lx min_gni_base 0x%lx min_gni_len 0x%lx\n",
-			pa_pdc_cell.mod[0], pa_pdc_cell.mod[1], 
-			pa_pdc_cell.mod[2]);
+			pa_pdc_cell->mod[0], pa_pdc_cell->mod[1],
+			pa_pdc_cell->mod[2]);
 		break;
 	case PAT_ENTITY_CA:
 		printk(KERN_DEBUG "PAT_ENTITY_CA: %ld\n", pcell_loc);
@@ -243,23 +247,26 @@ pat_query_module(ulong pcell_loc, ulong mod_index)
  print_ranges:
 		pdc_pat_cell_module(&bytecnt, pcell_loc, mod_index,
 				    IO_VIEW, &io_pdc_cell);
-		printk(KERN_DEBUG "ranges %ld\n", pa_pdc_cell.mod[1]);
-		for (i = 0; i < pa_pdc_cell.mod[1]; i++) {
+		printk(KERN_DEBUG "ranges %ld\n", pa_pdc_cell->mod[1]);
+		for (i = 0; i < pa_pdc_cell->mod[1]; i++) {
 			printk(KERN_DEBUG 
 				"  PA_VIEW %ld: 0x%016lx 0x%016lx 0x%016lx\n", 
-				i, pa_pdc_cell.mod[2 + i * 3],	/* type */
-				pa_pdc_cell.mod[3 + i * 3],	/* start */
-				pa_pdc_cell.mod[4 + i * 3]);	/* finish (ie end) */
+				i, pa_pdc_cell->mod[2 + i * 3],	/* type */
+				pa_pdc_cell->mod[3 + i * 3],	/* start */
+				pa_pdc_cell->mod[4 + i * 3]);	/* finish (ie end) */
 			printk(KERN_DEBUG 
 				"  IO_VIEW %ld: 0x%016lx 0x%016lx 0x%016lx\n", 
-				i, io_pdc_cell.mod[2 + i * 3],	/* type */
-				io_pdc_cell.mod[3 + i * 3],	/* start */
-				io_pdc_cell.mod[4 + i * 3]);	/* finish (ie end) */
+				i, io_pdc_cell->mod[2 + i * 3],	/* type */
+				io_pdc_cell->mod[3 + i * 3],	/* start */
+				io_pdc_cell->mod[4 + i * 3]);	/* finish (ie end) */
 		}
 		printk(KERN_DEBUG "\n");
 		break;
 	}
 #endif /* DEBUG_PAT */
+
+	kfree(pa_pdc_cell);
+
 	return PDC_OK;
 }
 
diff --git a/arch/parisc/kernel/irq.c b/arch/parisc/kernel/irq.c
index 8007f1e65729..330f536a9324 100644
--- a/arch/parisc/kernel/irq.c
+++ b/arch/parisc/kernel/irq.c
@@ -120,7 +120,7 @@ int cpu_check_affinity(unsigned int irq, const struct cpumask *dest)
 	if (CHECK_IRQ_PER_CPU(irq)) {
 		/* Bad linux design decision.  The mask has already
 		 * been set; we must reset it */
-		cpumask_setall(&irq_desc[irq].affinity);
+		cpumask_setall(irq_desc[irq].affinity);
 		return -EINVAL;
 	}
 
@@ -138,13 +138,13 @@ static int cpu_set_affinity_irq(unsigned int irq, const struct cpumask *dest)
 	if (cpu_dest < 0)
 		return -1;
 
-	cpumask_copy(&irq_desc[irq].affinity, dest);
+	cpumask_copy(irq_desc[irq].affinity, dest);
 
 	return 0;
 }
 #endif
 
-static struct hw_interrupt_type cpu_interrupt_type = {
+static struct irq_chip cpu_interrupt_type = {
 	.typename	= "CPU",
 	.startup	= cpu_startup_irq,
 	.shutdown	= cpu_disable_irq,
@@ -299,7 +299,7 @@ int txn_alloc_irq(unsigned int bits_wide)
 unsigned long txn_affinity_addr(unsigned int irq, int cpu)
 {
 #ifdef CONFIG_SMP
-	cpumask_copy(&irq_desc[irq].affinity, cpumask_of(cpu));
+	cpumask_copy(irq_desc[irq].affinity, cpumask_of(cpu));
 #endif
 
 	return per_cpu(cpu_data, cpu).txn_addr;
@@ -356,7 +356,7 @@ void do_cpu_irq_mask(struct pt_regs *regs)
 	irq = eirr_to_irq(eirr_val);
 
 #ifdef CONFIG_SMP
-	cpumask_copy(&dest, &irq_desc[irq].affinity);
+	cpumask_copy(&dest, irq_desc[irq].affinity);
 	if (CHECK_IRQ_PER_CPU(irq_desc[irq].status) &&
 	    !cpu_isset(smp_processor_id(), dest)) {
 		int cpu = first_cpu(dest);
diff --git a/arch/parisc/kernel/module.c b/arch/parisc/kernel/module.c
index ef5caf2e6ed0..61ee0eec4e69 100644
--- a/arch/parisc/kernel/module.c
+++ b/arch/parisc/kernel/module.c
@@ -86,8 +86,12 @@
  * the bottom of the table, which has a maximum signed displacement of
  * 0x3fff; however, since we're only going forward, this becomes
  * 0x1fff, and thus, since each GOT entry is 8 bytes long we can have
- * at most 1023 entries */
-#define MAX_GOTS	1023
+ * at most 1023 entries.
+ * To overcome this 14bit displacement with some kernel modules, we'll
+ * use instead the unusal 16bit displacement method (see reassemble_16a)
+ * which gives us a maximum positive displacement of 0x7fff, and as such
+ * allows us to allocate up to 4095 GOT entries. */
+#define MAX_GOTS	4095
 
 /* three functions to determine where in the module core
  * or init pieces the location is */
@@ -145,12 +149,40 @@ struct stub_entry {
 /* The reassemble_* functions prepare an immediate value for
    insertion into an opcode. pa-risc uses all sorts of weird bitfields
    in the instruction to hold the value.  */
+static inline int sign_unext(int x, int len)
+{
+	int len_ones;
+
+	len_ones = (1 << len) - 1;
+	return x & len_ones;
+}
+
+static inline int low_sign_unext(int x, int len)
+{
+	int sign, temp;
+
+	sign = (x >> (len-1)) & 1;
+	temp = sign_unext(x, len-1);
+	return (temp << 1) | sign;
+}
+
 static inline int reassemble_14(int as14)
 {
 	return (((as14 & 0x1fff) << 1) |
 		((as14 & 0x2000) >> 13));
 }
 
+static inline int reassemble_16a(int as16)
+{
+	int s, t;
+
+	/* Unusual 16-bit encoding, for wide mode only.  */
+	t = (as16 << 1) & 0xffff;
+	s = (as16 & 0x8000);
+	return (t ^ s ^ (s >> 1)) | (s >> 15);
+}
+
+
 static inline int reassemble_17(int as17)
 {
 	return (((as17 & 0x10000) >> 16) |
@@ -407,6 +439,7 @@ static Elf_Addr get_stub(struct module *me, unsigned long value, long addend,
 	enum elf_stub_type stub_type, Elf_Addr loc0, unsigned int targetsec)
 {
 	struct stub_entry *stub;
+	int __maybe_unused d;
 
 	/* initialize stub_offset to point in front of the section */
 	if (!me->arch.section[targetsec].stub_offset) {
@@ -460,12 +493,19 @@ static Elf_Addr get_stub(struct module *me, unsigned long value, long addend,
  */
 	switch (stub_type) {
 	case ELF_STUB_GOT:
-		stub->insns[0] = 0x537b0000;	/* ldd 0(%dp),%dp	*/
+		d = get_got(me, value, addend);
+		if (d <= 15) {
+			/* Format 5 */
+			stub->insns[0] = 0x0f6010db; /* ldd 0(%dp),%dp	*/
+			stub->insns[0] |= low_sign_unext(d, 5) << 16;
+		} else {
+			/* Format 3 */
+			stub->insns[0] = 0x537b0000; /* ldd 0(%dp),%dp	*/
+			stub->insns[0] |= reassemble_16a(d);
+		}
 		stub->insns[1] = 0x53610020;	/* ldd 10(%dp),%r1	*/
 		stub->insns[2] = 0xe820d000;	/* bve (%r1)		*/
 		stub->insns[3] = 0x537b0030;	/* ldd 18(%dp),%dp	*/
-
-		stub->insns[0] |= reassemble_14(get_got(me, value, addend) & 0x3fff);
 		break;
 	case ELF_STUB_MILLI:
 		stub->insns[0] = 0x20200000;	/* ldil 0,%r1		*/
diff --git a/arch/parisc/kernel/pci-dma.c b/arch/parisc/kernel/pci-dma.c
index 7d927eac932b..c07f618ff7da 100644
--- a/arch/parisc/kernel/pci-dma.c
+++ b/arch/parisc/kernel/pci-dma.c
@@ -90,12 +90,14 @@ static inline int map_pte_uncached(pte_t * pte,
 	if (end > PMD_SIZE)
 		end = PMD_SIZE;
 	do {
+		unsigned long flags;
+
 		if (!pte_none(*pte))
 			printk(KERN_ERR "map_pte_uncached: page already exists\n");
 		set_pte(pte, __mk_pte(*paddr_ptr, PAGE_KERNEL_UNC));
-		purge_tlb_start();
+		purge_tlb_start(flags);
 		pdtlb_kernel(orig_vaddr);
-		purge_tlb_end();
+		purge_tlb_end(flags);
 		vaddr += PAGE_SIZE;
 		orig_vaddr += PAGE_SIZE;
 		(*paddr_ptr) += PAGE_SIZE;
@@ -168,11 +170,13 @@ static inline void unmap_uncached_pte(pmd_t * pmd, unsigned long vaddr,
 	if (end > PMD_SIZE)
 		end = PMD_SIZE;
 	do {
+		unsigned long flags;
 		pte_t page = *pte;
+
 		pte_clear(&init_mm, vaddr, pte);
-		purge_tlb_start();
+		purge_tlb_start(flags);
 		pdtlb_kernel(orig_vaddr);
-		purge_tlb_end();
+		purge_tlb_end(flags);
 		vaddr += PAGE_SIZE;
 		orig_vaddr += PAGE_SIZE;
 		pte++;
diff --git a/arch/parisc/kernel/pci.c b/arch/parisc/kernel/pci.c
index 6936386c9861..f7064abc3bb6 100644
--- a/arch/parisc/kernel/pci.c
+++ b/arch/parisc/kernel/pci.c
@@ -1,5 +1,4 @@
-/* $Id: pci.c,v 1.6 2000/01/29 00:12:05 grundler Exp $
- *
+/*
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
diff --git a/arch/parisc/kernel/process.c b/arch/parisc/kernel/process.c
index 61c07078c072..1f3aa8db0203 100644
--- a/arch/parisc/kernel/process.c
+++ b/arch/parisc/kernel/process.c
@@ -156,7 +156,7 @@ void machine_power_off(void)
 	 * software. The user has to press the button himself. */
 
 	printk(KERN_EMERG "System shut down completed.\n"
-	       KERN_EMERG "Please power this system off now.");
+	       "Please power this system off now.");
 }
 
 void (*pm_power_off)(void) = machine_power_off;
diff --git a/arch/parisc/kernel/processor.c b/arch/parisc/kernel/processor.c
index e09d0f7fb6b0..c8fb61ed32f4 100644
--- a/arch/parisc/kernel/processor.c
+++ b/arch/parisc/kernel/processor.c
@@ -1,5 +1,4 @@
-/*    $Id: processor.c,v 1.1 2002/07/20 16:27:06 rhirst Exp $
- *
+/*
  *    Initial setup-routines for HP 9000 based hardware.
  *
  *    Copyright (C) 1991, 1992, 1995  Linus Torvalds
@@ -121,22 +120,28 @@ static int __cpuinit processor_probe(struct parisc_device *dev)
 	if (is_pdc_pat()) {
 		ulong status;
 		unsigned long bytecnt;
-	        pdc_pat_cell_mod_maddr_block_t pa_pdc_cell;
+	        pdc_pat_cell_mod_maddr_block_t *pa_pdc_cell;
 #undef USE_PAT_CPUID
 #ifdef USE_PAT_CPUID
 		struct pdc_pat_cpu_num cpu_info;
 #endif
 
+		pa_pdc_cell = kmalloc(sizeof (*pa_pdc_cell), GFP_KERNEL);
+		if (!pa_pdc_cell)
+			panic("couldn't allocate memory for PDC_PAT_CELL!");
+
 		status = pdc_pat_cell_module(&bytecnt, dev->pcell_loc,
-			dev->mod_index, PA_VIEW, &pa_pdc_cell);
+			dev->mod_index, PA_VIEW, pa_pdc_cell);
 
 		BUG_ON(PDC_OK != status);
 
 		/* verify it's the same as what do_pat_inventory() found */
-		BUG_ON(dev->mod_info != pa_pdc_cell.mod_info);
-		BUG_ON(dev->pmod_loc != pa_pdc_cell.mod_location);
+		BUG_ON(dev->mod_info != pa_pdc_cell->mod_info);
+		BUG_ON(dev->pmod_loc != pa_pdc_cell->mod_location);
+
+		txn_addr = pa_pdc_cell->mod[0];   /* id_eid for IO sapic */
 
-		txn_addr = pa_pdc_cell.mod[0];   /* id_eid for IO sapic */
+		kfree(pa_pdc_cell);
 
 #ifdef USE_PAT_CPUID
 /* We need contiguous numbers for cpuid. Firmware's notion
diff --git a/arch/parisc/kernel/setup.c b/arch/parisc/kernel/setup.c
index 82131ca8e05c..cb71f3dac995 100644
--- a/arch/parisc/kernel/setup.c
+++ b/arch/parisc/kernel/setup.c
@@ -1,5 +1,4 @@
-/*    $Id: setup.c,v 1.8 2000/02/02 04:42:38 prumpf Exp $
- *
+/*
  *    Initial setup-routines for HP 9000 based hardware.
  *
  *    Copyright (C) 1991, 1992, 1995  Linus Torvalds
diff --git a/arch/parisc/kernel/sys_parisc32.c b/arch/parisc/kernel/sys_parisc32.c
index 1adb40c81669..92a0acaa0d12 100644
--- a/arch/parisc/kernel/sys_parisc32.c
+++ b/arch/parisc/kernel/sys_parisc32.c
@@ -174,68 +174,6 @@ asmlinkage long sys32_sched_rr_get_interval(pid_t pid,
 	return ret;
 }
 
-/*** copied from mips64 ***/
-/*
- * Ooo, nasty.  We need here to frob 32-bit unsigned longs to
- * 64-bit unsigned longs.
- */
-
-static inline int
-get_fd_set32(unsigned long n, u32 *ufdset, unsigned long *fdset)
-{
-	n = (n + 8*sizeof(u32) - 1) / (8*sizeof(u32));
-	if (ufdset) {
-		unsigned long odd;
-
-		if (!access_ok(VERIFY_WRITE, ufdset, n*sizeof(u32)))
-			return -EFAULT;
-
-		odd = n & 1UL;
-		n &= ~1UL;
-		while (n) {
-			unsigned long h, l;
-			__get_user(l, ufdset);
-			__get_user(h, ufdset+1);
-			ufdset += 2;
-			*fdset++ = h << 32 | l;
-			n -= 2;
-		}
-		if (odd)
-			__get_user(*fdset, ufdset);
-	} else {
-		/* Tricky, must clear full unsigned long in the
-		 * kernel fdset at the end, this makes sure that
-		 * actually happens.
-		 */
-		memset(fdset, 0, ((n + 1) & ~1)*sizeof(u32));
-	}
-	return 0;
-}
-
-static inline void
-set_fd_set32(unsigned long n, u32 *ufdset, unsigned long *fdset)
-{
-	unsigned long odd;
-	n = (n + 8*sizeof(u32) - 1) / (8*sizeof(u32));
-
-	if (!ufdset)
-		return;
-
-	odd = n & 1UL;
-	n &= ~1UL;
-	while (n) {
-		unsigned long h, l;
-		l = *fdset++;
-		h = l >> 32;
-		__put_user(l, ufdset);
-		__put_user(h, ufdset+1);
-		ufdset += 2;
-		n -= 2;
-	}
-	if (odd)
-		__put_user(*fdset, ufdset);
-}
-
 struct msgbuf32 {
     int mtype;
     char mtext[1];
diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S
index 03b9a01bc16c..cf145eb026b3 100644
--- a/arch/parisc/kernel/syscall_table.S
+++ b/arch/parisc/kernel/syscall_table.S
@@ -413,6 +413,10 @@
 	ENTRY_SAME(dup3)
 	ENTRY_SAME(pipe2)
 	ENTRY_SAME(inotify_init1)
+	ENTRY_COMP(preadv)		/* 315 */
+	ENTRY_COMP(pwritev)
+	ENTRY_COMP(rt_tgsigqueueinfo)
+	ENTRY_SAME(perf_counter_open)
 
 	/* Nothing yet */
 
diff --git a/arch/parisc/kernel/time.c b/arch/parisc/kernel/time.c
index d4dd05674c62..a79c6f9e7e2c 100644
--- a/arch/parisc/kernel/time.c
+++ b/arch/parisc/kernel/time.c
@@ -56,9 +56,9 @@ static unsigned long clocktick __read_mostly;	/* timer cycles per tick */
  */
 irqreturn_t __irq_entry timer_interrupt(int irq, void *dev_id)
 {
-	unsigned long now;
+	unsigned long now, now2;
 	unsigned long next_tick;
-	unsigned long cycles_elapsed, ticks_elapsed;
+	unsigned long cycles_elapsed, ticks_elapsed = 1;
 	unsigned long cycles_remainder;
 	unsigned int cpu = smp_processor_id();
 	struct cpuinfo_parisc *cpuinfo = &per_cpu(cpu_data, cpu);
@@ -71,44 +71,24 @@ irqreturn_t __irq_entry timer_interrupt(int irq, void *dev_id)
 	/* Initialize next_tick to the expected tick time. */
 	next_tick = cpuinfo->it_value;
 
-	/* Get current interval timer.
-	 * CR16 reads as 64 bits in CPU wide mode.
-	 * CR16 reads as 32 bits in CPU narrow mode.
-	 */
+	/* Get current cycle counter (Control Register 16). */
 	now = mfctl(16);
 
 	cycles_elapsed = now - next_tick;
 
-	if ((cycles_elapsed >> 5) < cpt) {
+	if ((cycles_elapsed >> 6) < cpt) {
 		/* use "cheap" math (add/subtract) instead
 		 * of the more expensive div/mul method
 		 */
 		cycles_remainder = cycles_elapsed;
-		ticks_elapsed = 1;
 		while (cycles_remainder > cpt) {
 			cycles_remainder -= cpt;
 			ticks_elapsed++;
 		}
 	} else {
+		/* TODO: Reduce this to one fdiv op */
 		cycles_remainder = cycles_elapsed % cpt;
-		ticks_elapsed = 1 + cycles_elapsed / cpt;
-	}
-
-	/* Can we differentiate between "early CR16" (aka Scenario 1) and
-	 * "long delay" (aka Scenario 3)? I don't think so.
-	 *
-	 * We expected timer_interrupt to be delivered at least a few hundred
-	 * cycles after the IT fires. But it's arbitrary how much time passes
-	 * before we call it "late". I've picked one second.
-	 */
-	if (unlikely(ticks_elapsed > HZ)) {
-		/* Scenario 3: very long delay?  bad in any case */
-		printk (KERN_CRIT "timer_interrupt(CPU %d): delayed!"
-			" cycles %lX rem %lX "
-			" next/now %lX/%lX\n",
-			cpu,
-			cycles_elapsed, cycles_remainder,
-			next_tick, now );
+		ticks_elapsed += cycles_elapsed / cpt;
 	}
 
 	/* convert from "division remainder" to "remainder of clock tick" */
@@ -122,18 +102,56 @@ irqreturn_t __irq_entry timer_interrupt(int irq, void *dev_id)
 
 	cpuinfo->it_value = next_tick;
 
-	/* Skip one clocktick on purpose if we are likely to miss next_tick.
-	 * We want to avoid the new next_tick being less than CR16.
-	 * If that happened, itimer wouldn't fire until CR16 wrapped.
-	 * We'll catch the tick we missed on the tick after that.
+	/* Program the IT when to deliver the next interrupt.
+	 * Only bottom 32-bits of next_tick are writable in CR16!
 	 */
-	if (!(cycles_remainder >> 13))
-		next_tick += cpt;
-
-	/* Program the IT when to deliver the next interrupt. */
-	/* Only bottom 32-bits of next_tick are written to cr16.  */
 	mtctl(next_tick, 16);
 
+	/* Skip one clocktick on purpose if we missed next_tick.
+	 * The new CR16 must be "later" than current CR16 otherwise
+	 * itimer would not fire until CR16 wrapped - e.g 4 seconds
+	 * later on a 1Ghz processor. We'll account for the missed
+	 * tick on the next timer interrupt.
+	 *
+	 * "next_tick - now" will always give the difference regardless
+	 * if one or the other wrapped. If "now" is "bigger" we'll end up
+	 * with a very large unsigned number.
+	 */
+	now2 = mfctl(16);
+	if (next_tick - now2 > cpt)
+		mtctl(next_tick+cpt, 16);
+
+#if 1
+/*
+ * GGG: DEBUG code for how many cycles programming CR16 used.
+ */
+	if (unlikely(now2 - now > 0x3000)) 	/* 12K cycles */
+		printk (KERN_CRIT "timer_interrupt(CPU %d): SLOW! 0x%lx cycles!"
+			" cyc %lX rem %lX "
+			" next/now %lX/%lX\n",
+			cpu, now2 - now, cycles_elapsed, cycles_remainder,
+			next_tick, now );
+#endif
+
+	/* Can we differentiate between "early CR16" (aka Scenario 1) and
+	 * "long delay" (aka Scenario 3)? I don't think so.
+	 *
+	 * Timer_interrupt will be delivered at least a few hundred cycles
+	 * after the IT fires. But it's arbitrary how much time passes
+	 * before we call it "late". I've picked one second.
+	 *
+	 * It's important NO printk's are between reading CR16 and
+	 * setting up the next value. May introduce huge variance.
+	 */
+	if (unlikely(ticks_elapsed > HZ)) {
+		/* Scenario 3: very long delay?  bad in any case */
+		printk (KERN_CRIT "timer_interrupt(CPU %d): delayed!"
+			" cycles %lX rem %lX "
+			" next/now %lX/%lX\n",
+			cpu,
+			cycles_elapsed, cycles_remainder,
+			next_tick, now );
+	}
 
 	/* Done mucking with unreliable delivery of interrupts.
 	 * Go do system house keeping.
@@ -173,7 +191,7 @@ EXPORT_SYMBOL(profile_pc);
 
 /* clock source code */
 
-static cycle_t read_cr16(void)
+static cycle_t read_cr16(struct clocksource *cs)
 {
 	return get_cycles();
 }
diff --git a/arch/parisc/kernel/traps.c b/arch/parisc/kernel/traps.c
index c32f5d6d778e..528f0ff9b273 100644
--- a/arch/parisc/kernel/traps.c
+++ b/arch/parisc/kernel/traps.c
@@ -250,15 +250,14 @@ void die_if_kernel(char *str, struct pt_regs *regs, long err)
 	oops_enter();
 
 	/* Amuse the user in a SPARC fashion */
-	if (err) printk(
-KERN_CRIT "      _______________________________ \n"
-KERN_CRIT "     < Your System ate a SPARC! Gah! >\n"
-KERN_CRIT "      ------------------------------- \n"
-KERN_CRIT "             \\   ^__^\n"
-KERN_CRIT "              \\  (xx)\\_______\n"
-KERN_CRIT "                 (__)\\       )\\/\\\n"
-KERN_CRIT "                  U  ||----w |\n"
-KERN_CRIT "                     ||     ||\n");
+	if (err) printk(KERN_CRIT
+			"      _______________________________ \n"
+			"     < Your System ate a SPARC! Gah! >\n"
+			"      ------------------------------- \n"
+			"             \\   ^__^\n"
+			"                 (__)\\       )\\/\\\n"
+			"                  U  ||----w |\n"
+			"                     ||     ||\n");
 	
 	/* unlock the pdc lock if necessary */
 	pdc_emergency_unlock();
@@ -797,7 +796,8 @@ void notrace handle_interruption(int code, struct pt_regs *regs)
 		else
 			printk(KERN_DEBUG "User Fault (long pointer) (fault %d) ",
 			       code);
-		printk("pid=%d command='%s'\n", task_pid_nr(current), current->comm);
+		printk(KERN_CONT "pid=%d command='%s'\n",
+		       task_pid_nr(current), current->comm);
 		show_regs(regs);
 #endif
 		si.si_signo = SIGSEGV;
diff --git a/arch/parisc/lib/checksum.c b/arch/parisc/lib/checksum.c
index 462696d30d3b..ae66d31f9ecf 100644
--- a/arch/parisc/lib/checksum.c
+++ b/arch/parisc/lib/checksum.c
@@ -13,8 +13,6 @@
  *		modify it under the terms of the GNU General Public License
  *		as published by the Free Software Foundation; either version
  *		2 of the License, or (at your option) any later version.
- *
- * $Id: checksum.c,v 1.3 1997/12/01 17:57:34 ralf Exp $
  */
 #include <linux/module.h>
 #include <linux/types.h>
diff --git a/arch/parisc/lib/memcpy.c b/arch/parisc/lib/memcpy.c
index bbda909c866e..abf41f4632a9 100644
--- a/arch/parisc/lib/memcpy.c
+++ b/arch/parisc/lib/memcpy.c
@@ -405,7 +405,7 @@ byte_copy:
 
 unaligned_copy:
 	/* possibly we are aligned on a word, but not on a double... */
-	if (likely(t1 & (sizeof(unsigned int)-1)) == 0) {
+	if (likely((t1 & (sizeof(unsigned int)-1)) == 0)) {
 		t2 = src & (sizeof(unsigned int) - 1);
 
 		if (unlikely(t2 != 0)) {
diff --git a/arch/parisc/math-emu/decode_exc.c b/arch/parisc/math-emu/decode_exc.c
index 66c8a9f6a27e..3ca1c6149218 100644
--- a/arch/parisc/math-emu/decode_exc.c
+++ b/arch/parisc/math-emu/decode_exc.c
@@ -40,7 +40,7 @@
  * END_DESC
 */
 
-
+#include <linux/kernel.h>
 #include "float.h"
 #include "sgl_float.h"
 #include "dbl_float.h"
diff --git a/arch/parisc/mm/fault.c b/arch/parisc/mm/fault.c
index bfb6dd6ab380..c6afbfc95770 100644
--- a/arch/parisc/mm/fault.c
+++ b/arch/parisc/mm/fault.c
@@ -1,5 +1,4 @@
-/* $Id: fault.c,v 1.5 2000/01/26 16:20:29 jsm Exp $
- *
+/*
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
index 4356ceb1e366..b0831d9e35cb 100644
--- a/arch/parisc/mm/init.c
+++ b/arch/parisc/mm/init.c
@@ -370,34 +370,22 @@ static void __init setup_bootmem(void)
 
 void free_initmem(void)
 {
-	unsigned long addr, init_begin, init_end;
-
-	printk(KERN_INFO "Freeing unused kernel memory: ");
+	unsigned long addr;
+	unsigned long init_begin = (unsigned long)__init_begin;
+	unsigned long init_end = (unsigned long)__init_end;
 
 #ifdef CONFIG_DEBUG_KERNEL
 	/* Attempt to catch anyone trying to execute code here
 	 * by filling the page with BRK insns.
-	 * 
-	 * If we disable interrupts for all CPUs, then IPI stops working.
-	 * Kinda breaks the global cache flushing.
 	 */
-	local_irq_disable();
-
-	memset(__init_begin, 0x00,
-		(unsigned long)__init_end - (unsigned long)__init_begin);
-
-	flush_data_cache();
-	asm volatile("sync" : : );
-	flush_icache_range((unsigned long)__init_begin, (unsigned long)__init_end);
-	asm volatile("sync" : : );
-
-	local_irq_enable();
+	memset((void *)init_begin, 0x00, init_end - init_begin);
+	flush_icache_range(init_begin, init_end);
 #endif
 	
 	/* align __init_begin and __init_end to page size,
 	   ignoring linker script where we might have tried to save RAM */
-	init_begin = PAGE_ALIGN((unsigned long)(__init_begin));
-	init_end   = PAGE_ALIGN((unsigned long)(__init_end));
+	init_begin = PAGE_ALIGN(init_begin);
+	init_end = PAGE_ALIGN(init_end);
 	for (addr = init_begin; addr < init_end; addr += PAGE_SIZE) {
 		ClearPageReserved(virt_to_page(addr));
 		init_page_count(virt_to_page(addr));
@@ -409,7 +397,8 @@ void free_initmem(void)
 	/* set up a new led state on systems shipped LED State panel */
 	pdc_chassis_send_status(PDC_CHASSIS_DIRECT_BCOMPLETE);
 	
-	printk("%luk freed\n", (init_end - init_begin) >> 10);
+	printk(KERN_INFO "Freeing unused kernel memory: %luk freed\n",
+		(init_end - init_begin) >> 10);
 }