6 files changed, 105 insertions, 51 deletions

diff --git a/include/asm-ia64/acpi.h b/include/asm-ia64/acpi.h
index 4c06d455139c..3a544ffc5008 100644
--- a/include/asm-ia64/acpi.h
+++ b/include/asm-ia64/acpi.h
@@ -116,6 +116,11 @@ extern int __initdata nid_to_pxm_map[MAX_NUMNODES];
 
 extern u16 ia64_acpiid_to_sapicid[];
 
+/*
+ * Refer Intel ACPI _PDC support document for bit definitions
+ */
+#define ACPI_PDC_EST_CAPABILITY_SMP     0x8
+
 #endif /*__KERNEL__*/
 
 #endif /*_ASM_ACPI_H*/
diff --git a/include/asm-ia64/mmu.h b/include/asm-ia64/mmu.h
index ae1525352a25..611432ba579c 100644
--- a/include/asm-ia64/mmu.h
+++ b/include/asm-ia64/mmu.h
@@ -2,10 +2,12 @@
 #define __MMU_H
 
 /*
- * Type for a context number.  We declare it volatile to ensure proper ordering when it's
- * accessed outside of spinlock'd critical sections (e.g., as done in activate_mm() and
- * init_new_context()).
+ * Type for a context number.  We declare it volatile to ensure proper
+ * ordering when it's accessed outside of spinlock'd critical sections
+ * (e.g., as done in activate_mm() and init_new_context()).
  */
 typedef volatile unsigned long mm_context_t;
 
+typedef unsigned long nv_mm_context_t;
+
 #endif
diff --git a/include/asm-ia64/mmu_context.h b/include/asm-ia64/mmu_context.h
index e3e5fededb04..0680d163be97 100644
--- a/include/asm-ia64/mmu_context.h
+++ b/include/asm-ia64/mmu_context.h
@@ -55,34 +55,46 @@ static inline void
 delayed_tlb_flush (void)
 {
         extern void local_flush_tlb_all (void);
+        unsigned long flags;
 
         if (unlikely(__ia64_per_cpu_var(ia64_need_tlb_flush))) {
-                local_flush_tlb_all();
-                __ia64_per_cpu_var(ia64_need_tlb_flush) = 0;
+                spin_lock_irqsave(&ia64_ctx.lock, flags);
+                {
+                        if (__ia64_per_cpu_var(ia64_need_tlb_flush)) {
+                                local_flush_tlb_all();
+                                __ia64_per_cpu_var(ia64_need_tlb_flush) = 0;
+                        }
+                }
+                spin_unlock_irqrestore(&ia64_ctx.lock, flags);
         }
 }
 
-static inline mm_context_t
+static inline nv_mm_context_t
 get_mmu_context (struct mm_struct *mm)
 {
         unsigned long flags;
-        mm_context_t context = mm->context;
-
-        if (context)
-                return context;
-
-        spin_lock_irqsave(&ia64_ctx.lock, flags);
-        {
-                /* re-check, now that we've got the lock: */
-                context = mm->context;
-                if (context == 0) {
-                        cpus_clear(mm->cpu_vm_mask);
-                        if (ia64_ctx.next >= ia64_ctx.limit)
-                                wrap_mmu_context(mm);
-                        mm->context = context = ia64_ctx.next++;
+        nv_mm_context_t context = mm->context;
+
+        if (unlikely(!context)) {
+                spin_lock_irqsave(&ia64_ctx.lock, flags);
+                {
+                        /* re-check, now that we've got the lock: */
+                        context = mm->context;
+                        if (context == 0) {
+                                cpus_clear(mm->cpu_vm_mask);
+                                if (ia64_ctx.next >= ia64_ctx.limit)
+                                        wrap_mmu_context(mm);
+                                mm->context = context = ia64_ctx.next++;
+                        }
                 }
+                spin_unlock_irqrestore(&ia64_ctx.lock, flags);
         }
-        spin_unlock_irqrestore(&ia64_ctx.lock, flags);
+        /*
+         * Ensure we're not starting to use "context" before any old
+         * uses of it are gone from our TLB.
+         */
+        delayed_tlb_flush();
+
         return context;
 }
 
@@ -104,7 +116,7 @@ destroy_context (struct mm_struct *mm)
 }
 
 static inline void
-reload_context (mm_context_t context)
+reload_context (nv_mm_context_t context)
 {
         unsigned long rid;
         unsigned long rid_incr = 0;
@@ -138,7 +150,7 @@ reload_context (mm_context_t context)
 static inline void
 activate_context (struct mm_struct *mm)
 {
-        mm_context_t context;
+        nv_mm_context_t context;
 
         do {
                 context = get_mmu_context(mm);
@@ -157,8 +169,6 @@ activate_context (struct mm_struct *mm)
 static inline void
 activate_mm (struct mm_struct *prev, struct mm_struct *next)
 {
-        delayed_tlb_flush();
-
         /*
          * We may get interrupts here, but that's OK because interrupt handlers cannot
          * touch user-space.
diff --git a/include/asm-ia64/pal.h b/include/asm-ia64/pal.h
index 2303a10ee595..e828377ad295 100644
--- a/include/asm-ia64/pal.h
+++ b/include/asm-ia64/pal.h
@@ -75,6 +75,8 @@
 #define PAL_CACHE_READ          259     /* read tag & data of cacheline for diagnostic testing */
 #define PAL_CACHE_WRITE         260     /* write tag & data of cacheline for diagnostic testing */
 #define PAL_VM_TR_READ          261     /* read contents of translation register */
+#define PAL_GET_PSTATE          262     /* get the current P-state */
+#define PAL_SET_PSTATE          263     /* set the P-state */
 
 #ifndef __ASSEMBLY__
 
@@ -1111,6 +1113,25 @@ ia64_pal_halt_info (pal_power_mgmt_info_u_t *power_buf)
         return iprv.status;
 }
 
+/* Get the current P-state information */
+static inline s64
+ia64_pal_get_pstate (u64 *pstate_index)
+{
+        struct ia64_pal_retval iprv;
+        PAL_CALL_STK(iprv, PAL_GET_PSTATE, 0, 0, 0);
+        *pstate_index = iprv.v0;
+        return iprv.status;
+}
+
+/* Set the P-state */
+static inline s64
+ia64_pal_set_pstate (u64 pstate_index)
+{
+        struct ia64_pal_retval iprv;
+        PAL_CALL_STK(iprv, PAL_SET_PSTATE, pstate_index, 0, 0);
+        return iprv.status;
+}
+
 /* Cause the processor to enter LIGHT HALT state, where prefetching and execution are
  * suspended, but cache and TLB coherency is maintained.
  */
diff --git a/include/asm-ia64/rwsem.h b/include/asm-ia64/rwsem.h
index 6ece5061dc19..e18b5ab0cb75 100644
--- a/include/asm-ia64/rwsem.h
+++ b/include/asm-ia64/rwsem.h
@@ -3,6 +3,7 @@
  *
  * Copyright (C) 2003 Ken Chen <kenneth.w.chen@intel.com>
  * Copyright (C) 2003 Asit Mallick <asit.k.mallick@intel.com>
+ * Copyright (C) 2005 Christoph Lameter <clameter@sgi.com>
  *
  * Based on asm-i386/rwsem.h and other architecture implementation.
  *
@@ -11,9 +12,9 @@
  *
  * The lock count is initialized to 0 (no active and no waiting lockers).
  *
- * When a writer subtracts WRITE_BIAS, it'll get 0xffff0001 for the case
- * of an uncontended lock. Readers increment by 1 and see a positive value
- * when uncontended, negative if there are writers (and maybe) readers
+ * When a writer subtracts WRITE_BIAS, it'll get 0xffffffff00000001 for
+ * the case of an uncontended lock. Readers increment by 1 and see a positive
+ * value when uncontended, negative if there are writers (and maybe) readers
  * waiting (in which case it goes to sleep).
  */
 
@@ -29,7 +30,7 @@
  * the semaphore definition
  */
 struct rw_semaphore {
-        signed int              count;
+        signed long             count;
         spinlock_t              wait_lock;
         struct list_head        wait_list;
 #if RWSEM_DEBUG
@@ -37,10 +38,10 @@ struct rw_semaphore {
 #endif
 };
 
-#define RWSEM_UNLOCKED_VALUE            0x00000000
-#define RWSEM_ACTIVE_BIAS               0x00000001
-#define RWSEM_ACTIVE_MASK               0x0000ffff
-#define RWSEM_WAITING_BIAS              (-0x00010000)
+#define RWSEM_UNLOCKED_VALUE            __IA64_UL_CONST(0x0000000000000000)
+#define RWSEM_ACTIVE_BIAS               __IA64_UL_CONST(0x0000000000000001)
+#define RWSEM_ACTIVE_MASK               __IA64_UL_CONST(0x00000000ffffffff)
+#define RWSEM_WAITING_BIAS              -__IA64_UL_CONST(0x0000000100000000)
 #define RWSEM_ACTIVE_READ_BIAS          RWSEM_ACTIVE_BIAS
 #define RWSEM_ACTIVE_WRITE_BIAS         (RWSEM_WAITING_BIAS + RWSEM_ACTIVE_BIAS)
 
@@ -83,7 +84,7 @@ init_rwsem (struct rw_semaphore *sem)
 static inline void
 __down_read (struct rw_semaphore *sem)
 {
-        int result = ia64_fetchadd4_acq((unsigned int *)&sem->count, 1);
+        long result = ia64_fetchadd8_acq((unsigned long *)&sem->count, 1);
 
         if (result < 0)
                 rwsem_down_read_failed(sem);
@@ -95,7 +96,7 @@ __down_read (struct rw_semaphore *sem)
 static inline void
 __down_write (struct rw_semaphore *sem)
 {
-        int old, new;
+        long old, new;
 
         do {
                 old = sem->count;
@@ -112,7 +113,7 @@ __down_write (struct rw_semaphore *sem)
 static inline void
 __up_read (struct rw_semaphore *sem)
 {
-        int result = ia64_fetchadd4_rel((unsigned int *)&sem->count, -1);
+        long result = ia64_fetchadd8_rel((unsigned long *)&sem->count, -1);
 
         if (result < 0 && (--result & RWSEM_ACTIVE_MASK) == 0)
                 rwsem_wake(sem);
@@ -124,7 +125,7 @@ __up_read (struct rw_semaphore *sem)
 static inline void
 __up_write (struct rw_semaphore *sem)
 {
-        int old, new;
+        long old, new;
 
         do {
                 old = sem->count;
@@ -141,7 +142,7 @@ __up_write (struct rw_semaphore *sem)
 static inline int
 __down_read_trylock (struct rw_semaphore *sem)
 {
-        int tmp;
+        long tmp;
         while ((tmp = sem->count) >= 0) {
                 if (tmp == cmpxchg_acq(&sem->count, tmp, tmp+1)) {
                         return 1;
@@ -156,7 +157,7 @@ __down_read_trylock (struct rw_semaphore *sem)
 static inline int
 __down_write_trylock (struct rw_semaphore *sem)
 {
-        int tmp = cmpxchg_acq(&sem->count, RWSEM_UNLOCKED_VALUE,
+        long tmp = cmpxchg_acq(&sem->count, RWSEM_UNLOCKED_VALUE,
                               RWSEM_ACTIVE_WRITE_BIAS);
         return tmp == RWSEM_UNLOCKED_VALUE;
 }
@@ -167,7 +168,7 @@ __down_write_trylock (struct rw_semaphore *sem)
 static inline void
 __downgrade_write (struct rw_semaphore *sem)
 {
-        int old, new;
+        long old, new;
 
         do {
                 old = sem->count;
@@ -182,7 +183,7 @@ __downgrade_write (struct rw_semaphore *sem)
  * Implement atomic add functionality.  These used to be "inline" functions, but GCC v3.1
  * doesn't quite optimize this stuff right and ends up with bad calls to fetchandadd.
  */
-#define rwsem_atomic_add(delta, sem)    atomic_add(delta, (atomic_t *)(&(sem)->count))
-#define rwsem_atomic_update(delta, sem) atomic_add_return(delta, (atomic_t *)(&(sem)->count))
+#define rwsem_atomic_add(delta, sem)    atomic64_add(delta, (atomic64_t *)(&(sem)->count))
+#define rwsem_atomic_update(delta, sem) atomic64_add_return(delta, (atomic64_t *)(&(sem)->count))
 
 #endif /* _ASM_IA64_RWSEM_H */
diff --git a/include/asm-ia64/spinlock.h b/include/asm-ia64/spinlock.h
index 909936f25512..d2430aa0d49d 100644
--- a/include/asm-ia64/spinlock.h
+++ b/include/asm-ia64/spinlock.h
@@ -93,7 +93,15 @@ _raw_spin_lock_flags (spinlock_t *lock, unsigned long flags)
 # endif /* CONFIG_MCKINLEY */
 #endif
 }
+
 #define _raw_spin_lock(lock) _raw_spin_lock_flags(lock, 0)
+
+/* Unlock by doing an ordered store and releasing the cacheline with nta */
+static inline void _raw_spin_unlock(spinlock_t *x) {
+        barrier();
+        asm volatile ("st4.rel.nta [%0] = r0\n\t" :: "r"(x));
+}
+
 #else /* !ASM_SUPPORTED */
 #define _raw_spin_lock_flags(lock, flags) _raw_spin_lock(lock)
 # define _raw_spin_lock(x)                                                              \
@@ -109,16 +117,16 @@ do {											\
                 } while (ia64_spinlock_val);                                            \
         }                                                                               \
 } while (0)
+#define _raw_spin_unlock(x)     do { barrier(); ((spinlock_t *) x)->lock = 0; } while (0)
 #endif /* !ASM_SUPPORTED */
 
 #define spin_is_locked(x)       ((x)->lock != 0)
-#define _raw_spin_unlock(x)     do { barrier(); ((spinlock_t *) x)->lock = 0; } while (0)
 #define _raw_spin_trylock(x)    (cmpxchg_acq(&(x)->lock, 0, 1) == 0)
 #define spin_unlock_wait(x)     do { barrier(); } while ((x)->lock)
 
 typedef struct {
-        volatile unsigned int read_counter      : 31;
-        volatile unsigned int write_lock        :  1;
+        volatile unsigned int read_counter      : 24;
+        volatile unsigned int write_lock        :  8;
 #ifdef CONFIG_PREEMPT
         unsigned int break_lock;
 #endif
@@ -174,6 +182,13 @@ do {										\
         (result == 0);                                                          \
 })
 
+static inline void _raw_write_unlock(rwlock_t *x)
+{
+        u8 *y = (u8 *)x;
+        barrier();
+        asm volatile ("st1.rel.nta [%0] = r0\n\t" :: "r"(y+3) : "memory" );
+}
+
 #else /* !ASM_SUPPORTED */
 
 #define _raw_write_lock(l)                                                              \
@@ -195,14 +210,14 @@ do {										\
         (ia64_val == 0);                                                \
 })
 
+static inline void _raw_write_unlock(rwlock_t *x)
+{
+        barrier();
+        x->write_lock = 0;
+}
+
 #endif /* !ASM_SUPPORTED */
 
 #define _raw_read_trylock(lock) generic_raw_read_trylock(lock)
 
-#define _raw_write_unlock(x)                                                            \
-({                                                                                      \
-        smp_mb__before_clear_bit();     /* need barrier before releasing lock... */     \
-        clear_bit(31, (x));                                                             \
-})
-
 #endif /*  _ASM_IA64_SPINLOCK_H */