Merge branch 'linus'

17 files changed, 193 insertions, 216 deletions

diff --git a/include/asm-ia64/atomic.h b/include/asm-ia64/atomic.h
index d3e0dfa99e1f..569ec7574baf 100644
--- a/include/asm-ia64/atomic.h
+++ b/include/asm-ia64/atomic.h
@@ -95,8 +95,14 @@ ia64_atomic64_sub (__s64 i, atomic64_t *v)
 ({								\
 	int c, old;						\
 	c = atomic_read(v);					\
-	while (c != (u) && (old = atomic_cmpxchg((v), c, c + (a))) != c) \
+	for (;;) {						\
+		if (unlikely(c == (u)))				\
+			break;					\
+		old = atomic_cmpxchg((v), c, c + (a));		\
+		if (likely(old == c))				\
+			break;					\
 		c = old;					\
+	}							\
 	c != (u);						\
 })
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
diff --git a/include/asm-ia64/cache.h b/include/asm-ia64/cache.h
index 40dd25195d65..f0a104db8f20 100644
--- a/include/asm-ia64/cache.h
+++ b/include/asm-ia64/cache.h
@@ -25,4 +25,6 @@
 # define SMP_CACHE_BYTES	(1 << 3)
 #endif
 
+#define __read_mostly __attribute__((__section__(".data.read_mostly")))
+
 #endif /* _ASM_IA64_CACHE_H */
diff --git a/include/asm-ia64/intel_intrin.h b/include/asm-ia64/intel_intrin.h
index a7122d850177..d069b6acddce 100644
--- a/include/asm-ia64/intel_intrin.h
+++ b/include/asm-ia64/intel_intrin.h
@@ -5,113 +5,10 @@
  *
  * Copyright (C) 2002,2003 Jun Nakajima <jun.nakajima@intel.com>
  * Copyright (C) 2002,2003 Suresh Siddha <suresh.b.siddha@intel.com>
+ * Copyright (C) 2005,2006 Hongjiu Lu <hongjiu.lu@intel.com>
  *
  */
-#include <asm/types.h>
-
-void  __lfetch(int lfhint, void *y);
-void  __lfetch_excl(int lfhint, void *y);
-void  __lfetch_fault(int lfhint, void *y);
-void  __lfetch_fault_excl(int lfhint, void *y);
-
-/* In the following, whichFloatReg should be an integer from 0-127 */
-void  __ldfs(const int whichFloatReg, void *src);
-void  __ldfd(const int whichFloatReg, void *src);
-void  __ldfe(const int whichFloatReg, void *src);
-void  __ldf8(const int whichFloatReg, void *src);
-void  __ldf_fill(const int whichFloatReg, void *src);
-void  __stfs(void *dst, const int whichFloatReg);
-void  __stfd(void *dst, const int whichFloatReg);
-void  __stfe(void *dst, const int whichFloatReg);
-void  __stf8(void *dst, const int whichFloatReg);
-void  __stf_spill(void *dst, const int whichFloatReg);
-
-void  __st1_rel(void *dst, const __s8  value);
-void  __st2_rel(void *dst, const __s16 value);
-void  __st4_rel(void *dst, const __s32 value);
-void  __st8_rel(void *dst, const __s64 value);
-__u8  __ld1_acq(void *src);
-__u16 __ld2_acq(void *src);
-__u32 __ld4_acq(void *src);
-__u64 __ld8_acq(void *src);
-
-__u64 __fetchadd4_acq(__u32 *addend, const int increment);
-__u64 __fetchadd4_rel(__u32 *addend, const int increment);
-__u64 __fetchadd8_acq(__u64 *addend, const int increment);
-__u64 __fetchadd8_rel(__u64 *addend, const int increment);
-
-__u64 __getf_exp(double d);
-
-/* OS Related Itanium(R) Intrinsics  */
-
-/* The names to use for whichReg and whichIndReg below come from
-   the include file asm/ia64regs.h */
-
-__u64 __getIndReg(const int whichIndReg, __s64 index);
-__u64 __getReg(const int whichReg);
-
-void  __setIndReg(const int whichIndReg, __s64 index, __u64 value);
-void  __setReg(const int whichReg, __u64 value);
-
-void  __mf(void);
-void  __mfa(void);
-void  __synci(void);
-void  __itcd(__s64 pa);
-void  __itci(__s64 pa);
-void  __itrd(__s64 whichTransReg, __s64 pa);
-void  __itri(__s64 whichTransReg, __s64 pa);
-void  __ptce(__s64 va);
-void  __ptcl(__s64 va, __s64 pagesz);
-void  __ptcg(__s64 va, __s64 pagesz);
-void  __ptcga(__s64 va, __s64 pagesz);
-void  __ptri(__s64 va, __s64 pagesz);
-void  __ptrd(__s64 va, __s64 pagesz);
-void  __invala (void);
-void  __invala_gr(const int whichGeneralReg /* 0-127 */ );
-void  __invala_fr(const int whichFloatReg /* 0-127 */ );
-void  __nop(const int);
-void  __fc(__u64 *addr);
-void  __sum(int mask);
-void  __rum(int mask);
-void  __ssm(int mask);
-void  __rsm(int mask);
-__u64 __thash(__s64);
-__u64 __ttag(__s64);
-__s64 __tpa(__s64);
-
-/* Intrinsics for implementing get/put_user macros */
-void __st_user(const char *tableName, __u64 addr, char size, char relocType, __u64 val);
-void __ld_user(const char *tableName, __u64 addr, char size, char relocType);
-
-/* This intrinsic does not generate code, it creates a barrier across which
- * the compiler will not schedule data access instructions.
- */
-void __memory_barrier(void);
-
-void __isrlz(void);
-void __dsrlz(void);
-
-__u64  _m64_mux1(__u64 a, const int n);
-__u64  __thash(__u64);
-
-/* Lock and Atomic Operation Related Intrinsics */
-__u64 _InterlockedExchange8(volatile __u8 *trgt, __u8 value);
-__u64 _InterlockedExchange16(volatile __u16 *trgt, __u16 value);
-__s64 _InterlockedExchange(volatile __u32 *trgt, __u32 value);
-__s64 _InterlockedExchange64(volatile __u64 *trgt, __u64 value);
-
-__u64 _InterlockedCompareExchange8_rel(volatile __u8 *dest, __u64 xchg, __u64 comp);
-__u64 _InterlockedCompareExchange8_acq(volatile __u8 *dest, __u64 xchg, __u64 comp);
-__u64 _InterlockedCompareExchange16_rel(volatile __u16 *dest, __u64 xchg, __u64 comp);
-__u64 _InterlockedCompareExchange16_acq(volatile __u16 *dest, __u64 xchg, __u64 comp);
-__u64 _InterlockedCompareExchange_rel(volatile __u32 *dest, __u64 xchg, __u64 comp);
-__u64 _InterlockedCompareExchange_acq(volatile __u32 *dest, __u64 xchg, __u64 comp);
-__u64 _InterlockedCompareExchange64_rel(volatile __u64 *dest, __u64 xchg, __u64 comp);
-__u64 _InterlockedCompareExchange64_acq(volatile __u64 *dest, __u64 xchg, __u64 comp);
-
-__s64 _m64_dep_mi(const int v, __s64 s, const int p, const int len);
-__s64 _m64_shrp(__s64 a, __s64 b, const int count);
-__s64 _m64_popcnt(__s64 a);
+#include <ia64intrin.h>
 
 #define ia64_barrier()		__memory_barrier()
 
@@ -122,15 +19,16 @@ __s64 _m64_popcnt(__s64 a);
 #define ia64_getreg		__getReg
 #define ia64_setreg		__setReg
 
-#define ia64_hint(x)
+#define ia64_hint		__hint
+#define ia64_hint_pause		__hint_pause
 
-#define ia64_mux1_brcst	 0
-#define ia64_mux1_mix		 8
-#define ia64_mux1_shuf		 9
-#define ia64_mux1_alt		10
-#define ia64_mux1_rev		11
+#define ia64_mux1_brcst		_m64_mux1_brcst
+#define ia64_mux1_mix		_m64_mux1_mix
+#define ia64_mux1_shuf		_m64_mux1_shuf
+#define ia64_mux1_alt		_m64_mux1_alt
+#define ia64_mux1_rev		_m64_mux1_rev
 
-#define ia64_mux1		_m64_mux1
+#define ia64_mux1(x,v)		_m_to_int64(_m64_mux1(_m_from_int64(x), (v)))
 #define ia64_popcnt		_m64_popcnt
 #define ia64_getf_exp		__getf_exp
 #define ia64_shrp		_m64_shrp
@@ -158,7 +56,7 @@ __s64 _m64_popcnt(__s64 a);
 #define ia64_stf8		__stf8
 #define ia64_stf_spill		__stf_spill
 
-#define ia64_mf		__mf
+#define ia64_mf			__mf
 #define ia64_mfa		__mfa
 
 #define ia64_fetchadd4_acq	__fetchadd4_acq
@@ -234,10 +132,10 @@ __s64 _m64_popcnt(__s64 a);
 
 /* Values for lfhint in __lfetch and __lfetch_fault */
 
-#define ia64_lfhint_none   	0
-#define ia64_lfhint_nt1    	1
-#define ia64_lfhint_nt2    	2
-#define ia64_lfhint_nta    	3
+#define ia64_lfhint_none	__lfhint_none
+#define ia64_lfhint_nt1		__lfhint_nt1
+#define ia64_lfhint_nt2		__lfhint_nt2
+#define ia64_lfhint_nta		__lfhint_nta
 
 #define ia64_lfetch		__lfetch
 #define ia64_lfetch_excl	__lfetch_excl
@@ -254,4 +152,6 @@ do {							\
 	}						\
 } while (0)
 
+#define __builtin_trap()	__break(0);
+
 #endif /* _ASM_IA64_INTEL_INTRIN_H */
diff --git a/include/asm-ia64/machvec.h b/include/asm-ia64/machvec.h
index ca5ea994d688..c3e4ed8a3e17 100644
--- a/include/asm-ia64/machvec.h
+++ b/include/asm-ia64/machvec.h
@@ -20,6 +20,7 @@ struct scatterlist;
 struct page;
 struct mm_struct;
 struct pci_bus;
+struct task_struct;
 
 typedef void ia64_mv_setup_t (char **);
 typedef void ia64_mv_cpu_init_t (void);
@@ -34,6 +35,7 @@ typedef int ia64_mv_pci_legacy_read_t (struct pci_bus *, u16 port, u32 *val,
 				       u8 size);
 typedef int ia64_mv_pci_legacy_write_t (struct pci_bus *, u16 port, u32 val,
 					u8 size);
+typedef void ia64_mv_migrate_t(struct task_struct * task);
 
 /* DMA-mapping interface: */
 typedef void ia64_mv_dma_init (void);
@@ -85,6 +87,11 @@ machvec_noop_mm (struct mm_struct *mm)
 {
 }
 
+static inline void
+machvec_noop_task (struct task_struct *task)
+{
+}
+
 extern void machvec_setup (char **);
 extern void machvec_timer_interrupt (int, void *, struct pt_regs *);
 extern void machvec_dma_sync_single (struct device *, dma_addr_t, size_t, int);
@@ -146,6 +153,7 @@ extern void machvec_tlb_migrate_finish (struct mm_struct *);
 #  define platform_readw_relaxed        ia64_mv.readw_relaxed
 #  define platform_readl_relaxed        ia64_mv.readl_relaxed
 #  define platform_readq_relaxed        ia64_mv.readq_relaxed
+#  define platform_migrate		ia64_mv.migrate
 # endif
 
 /* __attribute__((__aligned__(16))) is required to make size of the
@@ -194,6 +202,7 @@ struct ia64_machine_vector {
 	ia64_mv_readw_relaxed_t *readw_relaxed;
 	ia64_mv_readl_relaxed_t *readl_relaxed;
 	ia64_mv_readq_relaxed_t *readq_relaxed;
+	ia64_mv_migrate_t *migrate;
 } __attribute__((__aligned__(16))); /* align attrib? see above comment */
 
 #define MACHVEC_INIT(name)			\
@@ -238,6 +247,7 @@ struct ia64_machine_vector {
 	platform_readw_relaxed,			\
 	platform_readl_relaxed,			\
 	platform_readq_relaxed,			\
+	platform_migrate,			\
 }
 
 extern struct ia64_machine_vector ia64_mv;
@@ -386,5 +396,8 @@ extern ia64_mv_dma_supported		swiotlb_dma_supported;
 #ifndef platform_readq_relaxed
 # define platform_readq_relaxed	__ia64_readq_relaxed
 #endif
+#ifndef platform_migrate
+# define platform_migrate machvec_noop_task
+#endif
 
 #endif /* _ASM_IA64_MACHVEC_H */
diff --git a/include/asm-ia64/machvec_sn2.h b/include/asm-ia64/machvec_sn2.h
index 03d00faf03b5..da1d43755afe 100644
--- a/include/asm-ia64/machvec_sn2.h
+++ b/include/asm-ia64/machvec_sn2.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2002-2003, 2006 Silicon Graphics, Inc.  All Rights Reserved.
+ * Copyright (c) 2002-2003,2006 Silicon Graphics, Inc.  All Rights Reserved.
  * 
  * This program is free software; you can redistribute it and/or modify it 
  * under the terms of version 2 of the GNU General Public License 
@@ -66,6 +66,7 @@ extern ia64_mv_dma_sync_single_for_device sn_dma_sync_single_for_device;
 extern ia64_mv_dma_sync_sg_for_device	sn_dma_sync_sg_for_device;
 extern ia64_mv_dma_mapping_error	sn_dma_mapping_error;
 extern ia64_mv_dma_supported		sn_dma_supported;
+extern ia64_mv_migrate_t		sn_migrate;
 
 /*
  * This stuff has dual use!
@@ -115,6 +116,7 @@ extern ia64_mv_dma_supported		sn_dma_supported;
 #define platform_dma_sync_sg_for_device	sn_dma_sync_sg_for_device
 #define platform_dma_mapping_error		sn_dma_mapping_error
 #define platform_dma_supported		sn_dma_supported
+#define platform_migrate		sn_migrate
 
 #include <asm/sn/io.h>
 
diff --git a/include/asm-ia64/mca.h b/include/asm-ia64/mca.h
index c7d9c9ed38ba..bfbbb8da79c7 100644
--- a/include/asm-ia64/mca.h
+++ b/include/asm-ia64/mca.h
@@ -131,6 +131,8 @@ struct ia64_mca_cpu {
 /* Array of physical addresses of each CPU's MCA area.  */
 extern unsigned long __per_cpu_mca[NR_CPUS];
 
+extern int cpe_vector;
+extern int ia64_cpe_irq;
 extern void ia64_mca_init(void);
 extern void ia64_mca_cpu_init(void *);
 extern void ia64_os_mca_dispatch(void);
diff --git a/include/asm-ia64/mutex.h b/include/asm-ia64/mutex.h
index 458c1f7fbc18..5a3224f6af38 100644
--- a/include/asm-ia64/mutex.h
+++ b/include/asm-ia64/mutex.h
@@ -1,9 +1,92 @@
 /*
- * Pull in the generic implementation for the mutex fastpath.
+ * ia64 implementation of the mutex fastpath.
  *
- * TODO: implement optimized primitives instead, or leave the generic
- * implementation in place, or pick the atomic_xchg() based generic
- * implementation. (see asm-generic/mutex-xchg.h for details)
+ * Copyright (C) 2006 Ken Chen <kenneth.w.chen@intel.com>
+ *
+ */
+
+#ifndef _ASM_MUTEX_H
+#define _ASM_MUTEX_H
+
+/**
+ *  __mutex_fastpath_lock - try to take the lock by moving the count
+ *                          from 1 to a 0 value
+ *  @count: pointer of type atomic_t
+ *  @fail_fn: function to call if the original value was not 1
+ *
+ * Change the count from 1 to a value lower than 1, and call <fail_fn> if
+ * it wasn't 1 originally. This function MUST leave the value lower than
+ * 1 even when the "1" assertion wasn't true.
+ */
+static inline void
+__mutex_fastpath_lock(atomic_t *count, void (*fail_fn)(atomic_t *))
+{
+	if (unlikely(ia64_fetchadd4_acq(count, -1) != 1))
+		fail_fn(count);
+}
+
+/**
+ *  __mutex_fastpath_lock_retval - try to take the lock by moving the count
+ *                                 from 1 to a 0 value
+ *  @count: pointer of type atomic_t
+ *  @fail_fn: function to call if the original value was not 1
+ *
+ * Change the count from 1 to a value lower than 1, and call <fail_fn> if
+ * it wasn't 1 originally. This function returns 0 if the fastpath succeeds,
+ * or anything the slow path function returns.
+ */
+static inline int
+__mutex_fastpath_lock_retval(atomic_t *count, int (*fail_fn)(atomic_t *))
+{
+	if (unlikely(ia64_fetchadd4_acq(count, -1) != 1))
+		return fail_fn(count);
+	return 0;
+}
+
+/**
+ *  __mutex_fastpath_unlock - try to promote the count from 0 to 1
+ *  @count: pointer of type atomic_t
+ *  @fail_fn: function to call if the original value was not 0
+ *
+ * Try to promote the count from 0 to 1. If it wasn't 0, call <fail_fn>.
+ * In the failure case, this function is allowed to either set the value to
+ * 1, or to set it to a value lower than 1.
+ *
+ * If the implementation sets it to a value of lower than 1, then the
+ * __mutex_slowpath_needs_to_unlock() macro needs to return 1, it needs
+ * to return 0 otherwise.
+ */
+static inline void
+__mutex_fastpath_unlock(atomic_t *count, void (*fail_fn)(atomic_t *))
+{
+	int ret = ia64_fetchadd4_rel(count, 1);
+	if (unlikely(ret < 0))
+		fail_fn(count);
+}
+
+#define __mutex_slowpath_needs_to_unlock()		1
+
+/**
+ * __mutex_fastpath_trylock - try to acquire the mutex, without waiting
+ *
+ *  @count: pointer of type atomic_t
+ *  @fail_fn: fallback function
+ *
+ * Change the count from 1 to a value lower than 1, and return 0 (failure)
+ * if it wasn't 1 originally, or return 1 (success) otherwise. This function
+ * MUST leave the value lower than 1 even when the "1" assertion wasn't true.
+ * Additionally, if the value was < 0 originally, this function must not leave
+ * it to 0 on failure.
+ *
+ * If the architecture has no effective trylock variant, it should call the
+ * <fail_fn> spinlock-based trylock variant unconditionally.
  */
+static inline int
+__mutex_fastpath_trylock(atomic_t *count, int (*fail_fn)(atomic_t *))
+{
+	if (likely(cmpxchg_acq(count, 1, 0)) == 1)
+		return 1;
+	return 0;
+}
 
-#include <asm-generic/mutex-dec.h>
+#endif
diff --git a/include/asm-ia64/page.h b/include/asm-ia64/page.h
index 5e6362a786b7..3ab27333dae4 100644
--- a/include/asm-ia64/page.h
+++ b/include/asm-ia64/page.h
@@ -57,6 +57,8 @@
 
 # define HAVE_ARCH_HUGETLB_UNMAPPED_AREA
 # define ARCH_HAS_HUGEPAGE_ONLY_RANGE
+# define ARCH_HAS_PREPARE_HUGEPAGE_RANGE
+# define ARCH_HAS_HUGETLB_FREE_PGD_RANGE
 #endif /* CONFIG_HUGETLB_PAGE */
 
 #ifdef __ASSEMBLY__
diff --git a/include/asm-ia64/pgtable.h b/include/asm-ia64/pgtable.h
index e2560c58384b..c0f8144f2349 100644
--- a/include/asm-ia64/pgtable.h
+++ b/include/asm-ia64/pgtable.h
@@ -314,7 +314,7 @@ ia64_phys_addr_valid (unsigned long addr)
 #define pte_mkyoung(pte)	(__pte(pte_val(pte) | _PAGE_A))
 #define pte_mkclean(pte)	(__pte(pte_val(pte) & ~_PAGE_D))
 #define pte_mkdirty(pte)	(__pte(pte_val(pte) | _PAGE_D))
-#define pte_mkhuge(pte)		(__pte(pte_val(pte) | _PAGE_P))
+#define pte_mkhuge(pte)		(__pte(pte_val(pte)))
 
 /*
  * Macro to a page protection value as "uncacheable".  Note that "protection" is really a
@@ -505,9 +505,6 @@ extern struct page *zero_page_memmap_ptr;
 #define HUGETLB_PGDIR_SHIFT	(HPAGE_SHIFT + 2*(PAGE_SHIFT-3))
 #define HUGETLB_PGDIR_SIZE	(__IA64_UL(1) << HUGETLB_PGDIR_SHIFT)
 #define HUGETLB_PGDIR_MASK	(~(HUGETLB_PGDIR_SIZE-1))
-struct mmu_gather;
-void hugetlb_free_pgd_range(struct mmu_gather **tlb, unsigned long addr,
-		unsigned long end, unsigned long floor, unsigned long ceiling);
 #endif
 
 /*
diff --git a/include/asm-ia64/processor.h b/include/asm-ia64/processor.h
index 23c8e1be1911..128fefd8056f 100644
--- a/include/asm-ia64/processor.h
+++ b/include/asm-ia64/processor.h
@@ -50,7 +50,8 @@
 #define IA64_THREAD_PM_VALID	(__IA64_UL(1) << 2)	/* performance registers valid? */
 #define IA64_THREAD_UAC_NOPRINT	(__IA64_UL(1) << 3)	/* don't log unaligned accesses */
 #define IA64_THREAD_UAC_SIGBUS	(__IA64_UL(1) << 4)	/* generate SIGBUS on unaligned acc. */
-							/* bit 5 is currently unused */
+#define IA64_THREAD_MIGRATION	(__IA64_UL(1) << 5)	/* require migration
+							   sync at ctx sw */
 #define IA64_THREAD_FPEMU_NOPRINT (__IA64_UL(1) << 6)	/* don't log any fpswa faults */
 #define IA64_THREAD_FPEMU_SIGFPE  (__IA64_UL(1) << 7)	/* send a SIGFPE for fpswa faults */
 
diff --git a/include/asm-ia64/signal.h b/include/asm-ia64/signal.h
index 608168d713d3..5e328ed5d01d 100644
--- a/include/asm-ia64/signal.h
+++ b/include/asm-ia64/signal.h
@@ -158,8 +158,6 @@ struct k_sigaction {
 
 #define ptrace_signal_deliver(regs, cookie) do { } while (0)
 
-void set_sigdelayed(pid_t pid, int signo, int code, void __user *addr);
-
 #endif /* __KERNEL__ */
 
 # endif /* !__ASSEMBLY__ */
diff --git a/include/asm-ia64/sn/addrs.h b/include/asm-ia64/sn/addrs.h
index 2c32e4b77b54..1d9efe541662 100644
--- a/include/asm-ia64/sn/addrs.h
+++ b/include/asm-ia64/sn/addrs.h
@@ -283,5 +283,13 @@
 #define REMOTE_HUB_L(n, a)		HUB_L(REMOTE_HUB_ADDR((n), (a)))
 #define REMOTE_HUB_S(n, a, d)		HUB_S(REMOTE_HUB_ADDR((n), (a)), (d))
 
+/*
+ * Coretalk address breakdown
+ */
+#define CTALK_NASID_SHFT		40
+#define CTALK_NASID_MASK		(0x3FFFULL << CTALK_NASID_SHFT)
+#define CTALK_CID_SHFT			38
+#define CTALK_CID_MASK			(0x3ULL << CTALK_CID_SHFT)
+#define CTALK_NODE_OFFSET		0x3FFFFFFFFF
 
 #endif /* _ASM_IA64_SN_ADDRS_H */
diff --git a/include/asm-ia64/sn/rw_mmr.h b/include/asm-ia64/sn/rw_mmr.h
index f40fd1a5510d..2d78f4c5a45e 100644
--- a/include/asm-ia64/sn/rw_mmr.h
+++ b/include/asm-ia64/sn/rw_mmr.h
@@ -3,15 +3,14 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 2002-2004 Silicon Graphics, Inc.  All Rights Reserved.
+ * Copyright (C) 2002-2006 Silicon Graphics, Inc.  All Rights Reserved.
  */
 #ifndef _ASM_IA64_SN_RW_MMR_H
 #define _ASM_IA64_SN_RW_MMR_H
 
 
 /*
- * This file contains macros used to access MMR registers via
- * uncached physical addresses.
+ * This file that access MMRs via uncached physical addresses.
  * 	pio_phys_read_mmr  - read an MMR
  * 	pio_phys_write_mmr - write an MMR
  * 	pio_atomic_phys_write_mmrs - atomically write 1 or 2 MMRs with psr.ic=0
@@ -22,53 +21,8 @@
  */
 
 
-extern inline long
-pio_phys_read_mmr(volatile long *mmr) 
-{
-	long val;
-        asm volatile
-            ("mov r2=psr;;"
-             "rsm psr.i | psr.dt;;"
-             "srlz.i;;"
-             "ld8.acq %0=[%1];;"
-             "mov psr.l=r2;;"
-             "srlz.i;;"
-             : "=r"(val)
-             : "r"(mmr)
-	     : "r2");
-        return val;
-}
-
-
-
-extern inline void
-pio_phys_write_mmr(volatile long *mmr, long val) 
-{
-        asm volatile
-            ("mov r2=psr;;"
-             "rsm psr.i | psr.dt;;"
-             "srlz.i;;"
-             "st8.rel [%0]=%1;;"
-             "mov psr.l=r2;;"
-             "srlz.i;;"
-	     :: "r"(mmr), "r"(val)
-             : "r2", "memory");
-}            
-
-extern inline void
-pio_atomic_phys_write_mmrs(volatile long *mmr1, long val1, volatile long *mmr2, long val2) 
-{
-        asm volatile
-            ("mov r2=psr;;"
-             "rsm psr.i | psr.dt | psr.ic;;"
-	     "cmp.ne p9,p0=%2,r0;"
-             "srlz.i;;"
-             "st8.rel [%0]=%1;"
-             "(p9) st8.rel [%2]=%3;;"
-             "mov psr.l=r2;;"
-             "srlz.i;;"
-	     :: "r"(mmr1), "r"(val1), "r"(mmr2), "r"(val2)
-             : "p9", "r2", "memory");
-}            
+extern long pio_phys_read_mmr(volatile long *mmr); 
+extern void pio_phys_write_mmr(volatile long *mmr, long val);
+extern void pio_atomic_phys_write_mmrs(volatile long *mmr1, long val1, volatile long *mmr2, long val2); 
 
 #endif /* _ASM_IA64_SN_RW_MMR_H */
diff --git a/include/asm-ia64/sn/tioce.h b/include/asm-ia64/sn/tioce.h
index d4c990712eac..893468e1b41b 100644
--- a/include/asm-ia64/sn/tioce.h
+++ b/include/asm-ia64/sn/tioce.h
@@ -11,7 +11,7 @@
 
 /* CE ASIC part & mfgr information  */
 #define TIOCE_PART_NUM			0xCE00
-#define TIOCE_MFGR_NUM			0x36
+#define TIOCE_SRC_ID			0x01
 #define TIOCE_REV_A			0x1
 
 /* CE Virtual PPB Vendor/Device IDs */
@@ -20,7 +20,7 @@
 
 /* CE Host Bridge Vendor/Device IDs */
 #define CE_HOST_BRIDGE_VENDOR_ID	0x10a9
-#define CE_HOST_BRIDGE_DEVICE_ID	0x4003
+#define CE_HOST_BRIDGE_DEVICE_ID	0x4001
 
 
 #define TIOCE_NUM_M40_ATES		4096
@@ -463,6 +463,25 @@ typedef volatile struct tioce {
 	u64	ce_end_of_struct;			/* 0x044400 */
 } tioce_t;
 
+/* ce_lsiX_gb_cfg1 register bit masks & shifts */
+#define CE_LSI_GB_CFG1_RXL0S_THS_SHFT	0
+#define CE_LSI_GB_CFG1_RXL0S_THS_MASK	(0xffULL << 0)
+#define CE_LSI_GB_CFG1_RXL0S_SMP_SHFT	8
+#define CE_LSI_GB_CFG1_RXL0S_SMP_MASK	(0xfULL << 8);
+#define CE_LSI_GB_CFG1_RXL0S_ADJ_SHFT	12
+#define CE_LSI_GB_CFG1_RXL0S_ADJ_MASK	(0x7ULL << 12)
+#define CE_LSI_GB_CFG1_RXL0S_FLT_SHFT	15
+#define CE_LSI_GB_CFG1_RXL0S_FLT_MASK	(0x1ULL << 15)
+#define CE_LSI_GB_CFG1_LPBK_SEL_SHFT	16
+#define CE_LSI_GB_CFG1_LPBK_SEL_MASK	(0x3ULL << 16)
+#define CE_LSI_GB_CFG1_LPBK_EN_SHFT	18
+#define CE_LSI_GB_CFG1_LPBK_EN_MASK	(0x1ULL << 18)
+#define CE_LSI_GB_CFG1_RVRS_LB_SHFT	19
+#define CE_LSI_GB_CFG1_RVRS_LB_MASK	(0x1ULL << 19)
+#define CE_LSI_GB_CFG1_RVRS_CLK_SHFT	20
+#define CE_LSI_GB_CFG1_RVRS_CLK_MASK	(0x3ULL << 20)
+#define CE_LSI_GB_CFG1_SLF_TS_SHFT	24
+#define CE_LSI_GB_CFG1_SLF_TS_MASK	(0xfULL << 24)
 
 /* ce_adm_int_mask/ce_adm_int_status register bit defines */
 #define CE_ADM_INT_CE_ERROR_SHFT		0
@@ -592,6 +611,11 @@ typedef volatile struct tioce {
 #define CE_URE_RD_MRG_ENABLE		(0x1ULL << 0)
 #define CE_URE_WRT_MRG_ENABLE1		(0x1ULL << 4)
 #define CE_URE_WRT_MRG_ENABLE2		(0x1ULL << 5)
+#define CE_URE_WRT_MRG_TIMER_SHFT	12
+#define CE_URE_WRT_MRG_TIMER_MASK	(0x7FFULL << CE_URE_WRT_MRG_TIMER_SHFT)
+#define CE_URE_WRT_MRG_TIMER(x)		(((u64)(x) << \
+					  CE_URE_WRT_MRG_TIMER_SHFT) & \
+					 CE_URE_WRT_MRG_TIMER_MASK)
 #define CE_URE_RSPQ_BYPASS_DISABLE	(0x1ULL << 24)
 #define CE_URE_UPS_DAT1_PAR_DISABLE	(0x1ULL << 32)
 #define CE_URE_UPS_HDR1_PAR_DISABLE	(0x1ULL << 33)
@@ -653,8 +677,12 @@ typedef volatile struct tioce {
 #define CE_URE_SI			(0x1ULL << 0)
 #define CE_URE_ELAL_SHFT		4
 #define CE_URE_ELAL_MASK		(0x7ULL << CE_URE_ELAL_SHFT)
+#define CE_URE_ELAL_SET(n)		(((u64)(n) << CE_URE_ELAL_SHFT) & \
+					 CE_URE_ELAL_MASK)
 #define CE_URE_ELAL1_SHFT		8
 #define CE_URE_ELAL1_MASK		(0x7ULL << CE_URE_ELAL1_SHFT)
+#define CE_URE_ELAL1_SET(n)		(((u64)(n) << CE_URE_ELAL1_SHFT) & \
+					 CE_URE_ELAL1_MASK)
 #define CE_URE_SCC			(0x1ULL << 12)
 #define CE_URE_PN1_SHFT			16
 #define CE_URE_PN1_MASK			(0xFFULL << CE_URE_PN1_SHFT)
@@ -675,8 +703,12 @@ typedef volatile struct tioce {
 #define CE_URE_HPC			(0x1ULL << 6)
 #define CE_URE_SPLV_SHFT		7
 #define CE_URE_SPLV_MASK		(0xFFULL << CE_URE_SPLV_SHFT)
+#define CE_URE_SPLV_SET(n)		(((u64)(n) << CE_URE_SPLV_SHFT) & \
+					 CE_URE_SPLV_MASK)
 #define CE_URE_SPLS_SHFT		15
 #define CE_URE_SPLS_MASK		(0x3ULL << CE_URE_SPLS_SHFT)
+#define CE_URE_SPLS_SET(n)		(((u64)(n) << CE_URE_SPLS_SHFT) & \
+					 CE_URE_SPLS_MASK)
 #define CE_URE_PSN1_SHFT		19
 #define CE_URE_PSN1_MASK		(0x1FFFULL << CE_URE_PSN1_SHFT)
 #define CE_URE_PSN2_SHFT		32
diff --git a/include/asm-ia64/sn/xpc.h b/include/asm-ia64/sn/xpc.h
index df7f5f4f3cde..aa3b8ace9030 100644
--- a/include/asm-ia64/sn/xpc.h
+++ b/include/asm-ia64/sn/xpc.h
@@ -1227,28 +1227,6 @@ xpc_map_bte_errors(bte_result_t error)
 
 
 
-static inline void *
-xpc_kmalloc_cacheline_aligned(size_t size, gfp_t flags, void **base)
-{
-	/* see if kmalloc will give us cachline aligned memory by default */
-	*base = kmalloc(size, flags);
-	if (*base == NULL) {
-		return NULL;
-	}
-	if ((u64) *base == L1_CACHE_ALIGN((u64) *base)) {
-		return *base;
-	}
-	kfree(*base);
-
-	/* nope, we'll have to do it ourselves */
-	*base = kmalloc(size + L1_CACHE_BYTES, flags);
-	if (*base == NULL) {
-		return NULL;
-	}
-	return (void *) L1_CACHE_ALIGN((u64) *base);
-}
-
-
 /*
  * Check to see if there is any channel activity to/from the specified
  * partition.
diff --git a/include/asm-ia64/system.h b/include/asm-ia64/system.h
index 062538715623..cd4233d66f15 100644
--- a/include/asm-ia64/system.h
+++ b/include/asm-ia64/system.h
@@ -244,6 +244,13 @@ extern void ia64_load_extra (struct task_struct *task);
 		__ia64_save_fpu((prev)->thread.fph);				\
 	}									\
 	__switch_to(prev, next, last);						\
+	/* "next" in old context is "current" in new context */			\
+	if (unlikely((current->thread.flags & IA64_THREAD_MIGRATION) &&	       \
+		     (task_cpu(current) !=				       \
+		      		      task_thread_info(current)->last_cpu))) { \
+		platform_migrate(current);				       \
+		task_thread_info(current)->last_cpu = task_cpu(current);       \
+	}								       \
 } while (0)
 #else
 # define switch_to(prev,next,last)	__switch_to(prev, next, last)
diff --git a/include/asm-ia64/thread_info.h b/include/asm-ia64/thread_info.h
index 1d6518fe1f02..56394a2c7055 100644
--- a/include/asm-ia64/thread_info.h
+++ b/include/asm-ia64/thread_info.h
@@ -26,16 +26,10 @@ struct thread_info {
 	struct exec_domain *exec_domain;/* execution domain */
 	__u32 flags;			/* thread_info flags (see TIF_*) */
 	__u32 cpu;			/* current CPU */
+	__u32 last_cpu;			/* Last CPU thread ran on */
 	mm_segment_t addr_limit;	/* user-level address space limit */
 	int preempt_count;		/* 0=premptable, <0=BUG; will also serve as bh-counter */
 	struct restart_block restart_block;
-	struct {
-		int signo;
-		int code;
-		void __user *addr;
-		unsigned long start_time;
-		pid_t pid;
-	} sigdelayed;			/* Saved information for TIF_SIGDELAYED */
 };
 
 #define THREAD_SIZE			KERNEL_STACK_SIZE
@@ -89,7 +83,6 @@ struct thread_info {
 #define TIF_NEED_RESCHED	2	/* rescheduling necessary */
 #define TIF_SYSCALL_TRACE	3	/* syscall trace active */
 #define TIF_SYSCALL_AUDIT	4	/* syscall auditing active */
-#define TIF_SIGDELAYED		5	/* signal delayed from MCA/INIT/NMI/PMI context */
 #define TIF_POLLING_NRFLAG	16	/* true if poll_idle() is polling TIF_NEED_RESCHED */
 #define TIF_MEMDIE		17
 #define TIF_MCA_INIT		18	/* this task is processing MCA or INIT */
@@ -101,13 +94,12 @@ struct thread_info {
 #define _TIF_NOTIFY_RESUME	(1 << TIF_NOTIFY_RESUME)
 #define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
 #define _TIF_NEED_RESCHED	(1 << TIF_NEED_RESCHED)
-#define _TIF_SIGDELAYED		(1 << TIF_SIGDELAYED)
 #define _TIF_POLLING_NRFLAG	(1 << TIF_POLLING_NRFLAG)
 #define _TIF_MCA_INIT		(1 << TIF_MCA_INIT)
 #define _TIF_DB_DISABLED	(1 << TIF_DB_DISABLED)
 
 /* "work to do on user-return" bits */
-#define TIF_ALLWORK_MASK	(_TIF_NOTIFY_RESUME|_TIF_SIGPENDING|_TIF_NEED_RESCHED|_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SIGDELAYED)
+#define TIF_ALLWORK_MASK	(_TIF_NOTIFY_RESUME|_TIF_SIGPENDING|_TIF_NEED_RESCHED|_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT)
 /* like TIF_ALLWORK_BITS but sans TIF_SYSCALL_TRACE or TIF_SYSCALL_AUDIT */
 #define TIF_WORK_MASK		(TIF_ALLWORK_MASK&~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT))