[IA64] Remove rwsem limitation of 32k waiters

We ran into the limit with the maximum number of waiters at one of our sites.

This patch increases the number of possible waiters from 2^15 to 2^31 by using
a long for the counter in struct rw_semaphore. S390 and alpha already do this.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Acked-by: Kenneth Chen <kenneth.w.chen@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

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
+ * When a writer subtracts WRITE_BIAS, it'll get 0xffffffff00000001 for
- * of an uncontended lock. Readers increment by 1 and see a positive value
+ * the case of an uncontended lock. Readers increment by 1 and see a positive
- * when uncontended, negative if there are writers (and maybe) readers
+ * 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_UNLOCKED_VALUE		__IA64_UL_CONST(0x0000000000000000)
-#define RWSEM_ACTIVE_BIAS		0x00000001
+#define RWSEM_ACTIVE_BIAS		__IA64_UL_CONST(0x0000000000000001)
-#define RWSEM_ACTIVE_MASK		0x0000ffff
+#define RWSEM_ACTIVE_MASK		__IA64_UL_CONST(0x00000000ffffffff)
-#define RWSEM_WAITING_BIAS		(-0x00010000)
+#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_add(delta, sem)	atomic64_add(delta, (atomic64_t *)(&(sem)->count))
-#define rwsem_atomic_update(delta, sem)	atomic_add_return(delta, (atomic_t *)(&(sem)->count))
+#define rwsem_atomic_update(delta, sem)	atomic64_add_return(delta, (atomic64_t *)(&(sem)->count))
 
 #endif /* _ASM_IA64_RWSEM_H */