[PATCH] out-of-line locks / generic

This patch achieves out of line spinlocks by creating kernel/spinlock.c
and using the _raw_* inline locking functions.

Now, as much as this is supposed to be arch agnostic, there was still a
fair amount of rummaging about in archs, mostly for the cases where the
arch already has out of line locks and i wanted to avoid the extra call,
saving that extra call also makes lock profiling easier.  PPC32/64 was
an example of such an arch and i have added the necessary profile_pc()
function as an example. 

Size differences are with CONFIG_PREEMPT enabled since we wanted to
determine how much could be saved by moving that lot out of line too.

ppc64 = 259897 bytes:
   text    data     bss     dec     hex filename
5489808 1962724  709064 8161596  7c893c vmlinux-after
5749577 1962852  709064 8421493  808075 vmlinux-before

sparc64 = 193368 bytes:
  text    data     bss     dec     hex filename
3472037  633712  308920 4414669  435ccd vmlinux-after
3665285  633832  308920 4608037  465025 vmlinux-before

i386 = 416075 bytes
   text    data     bss     dec     hex filename
5808371  867442  326864 7002677  6ada35 vmlinux-after
6221254  870634  326864 7418752  713380 vmlinux-before

x86-64 = 282446 bytes
   text    data     bss     dec     hex filename
4598025 1450644  523632 6572301  64490d vmlinux-after
4881679 1449436  523632 6854747  68985b vmlinux-before

It has been compile tested (UP, SMP, PREEMPT) on i386, x86-64, sparc,
sparc64, ppc64, ppc32 and runtime tested on i386, x86-64 and sparc64.

Signed-off-by: Zwane Mwaikambo <zwane@fsmlabs.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

1 files changed, 160 insertions, 91 deletions

diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
index de381ec1ea13..947abd3ff9bd 100644
--- a/include/linux/spinlock.h
+++ b/include/linux/spinlock.h
@@ -19,18 +19,18 @@
 /*
  * Must define these before including other files, inline functions need them
  */
-#define LOCK_SECTION_NAME			\
-	".text.lock." __stringify(KBUILD_BASENAME)
+#define LOCK_SECTION_NAME                       \
+        ".text.lock." __stringify(KBUILD_BASENAME)
 
-#define LOCK_SECTION_START(extra)		\
-	".subsection 1\n\t"			\
-	extra					\
-	".ifndef " LOCK_SECTION_NAME "\n\t"	\
-	LOCK_SECTION_NAME ":\n\t"		\
-	".endif\n\t"
+#define LOCK_SECTION_START(extra)               \
+        ".subsection 1\n\t"                     \
+        extra                                   \
+        ".ifndef " LOCK_SECTION_NAME "\n\t"     \
+        LOCK_SECTION_NAME ":\n\t"               \
+        ".endif\n\t"
 
-#define LOCK_SECTION_END			\
-	".previous\n\t"
+#define LOCK_SECTION_END                        \
+        ".previous\n\t"
 
 /*
  * If CONFIG_SMP is set, pull in the _raw_* definitions
@@ -38,10 +38,41 @@
 #ifdef CONFIG_SMP
 #include <asm/spinlock.h>
 
+#define __lockfunc fastcall __attribute__((section(".spinlock.text")))
+
+int __lockfunc _spin_trylock(spinlock_t *lock);
+int __lockfunc _write_trylock(rwlock_t *lock);
+void __lockfunc _spin_lock(spinlock_t *lock);
+void __lockfunc _write_lock(rwlock_t *lock);
+void __lockfunc _spin_lock(spinlock_t *lock);
+void __lockfunc _read_lock(rwlock_t *lock);
+void __lockfunc _spin_unlock(spinlock_t *lock);
+void __lockfunc _write_unlock(rwlock_t *lock);
+void __lockfunc _read_unlock(rwlock_t *lock);
+unsigned long __lockfunc _spin_lock_irqsave(spinlock_t *lock);
+unsigned long __lockfunc _read_lock_irqsave(rwlock_t *lock);
+unsigned long __lockfunc _write_lock_irqsave(rwlock_t *lock);
+void __lockfunc _spin_lock_irq(spinlock_t *lock);
+void __lockfunc _spin_lock_bh(spinlock_t *lock);
+void __lockfunc _read_lock_irq(rwlock_t *lock);
+void __lockfunc _read_lock_bh(rwlock_t *lock);
+void __lockfunc _write_lock_irq(rwlock_t *lock);
+void __lockfunc _write_lock_bh(rwlock_t *lock);
+void __lockfunc _spin_unlock_irqrestore(spinlock_t *lock, unsigned long flags);
+void __lockfunc _spin_unlock_irq(spinlock_t *lock);
+void __lockfunc _spin_unlock_bh(spinlock_t *lock);
+void __lockfunc _read_unlock_irqrestore(rwlock_t *lock, unsigned long flags);
+void __lockfunc _read_unlock_irq(rwlock_t *lock);
+void __lockfunc _read_unlock_bh(rwlock_t *lock);
+void __lockfunc _write_unlock_irqrestore(rwlock_t *lock, unsigned long flags);
+void __lockfunc _write_unlock_irq(rwlock_t *lock);
+void __lockfunc _write_unlock_bh(rwlock_t *lock);
+int __lockfunc _spin_trylock_bh(spinlock_t *lock);
+
+extern unsigned long __lock_text_start;
+extern unsigned long __lock_text_end;
 #else
 
-#define _raw_spin_lock_flags(lock, flags) _raw_spin_lock(lock)
-
 #if !defined(CONFIG_PREEMPT) && !defined(CONFIG_DEBUG_SPINLOCK)
 # define atomic_dec_and_lock(atomic,lock) atomic_dec_and_test(atomic)
 # define ATOMIC_DEC_AND_LOCK
@@ -162,9 +193,9 @@ typedef struct {
 #define spin_lock_init(lock)	do { (void)(lock); } while(0)
 #define _raw_spin_lock(lock)	do { (void)(lock); } while(0)
 #define spin_is_locked(lock)	((void)(lock), 0)
-#define _raw_spin_trylock(lock)	((void)(lock), 1)
-#define spin_unlock_wait(lock)	do { (void)(lock); } while(0)
-#define _raw_spin_unlock(lock)	do { (void)(lock); } while(0)
+#define _raw_spin_trylock(lock)	(((void)(lock), 1))
+#define spin_unlock_wait(lock)	(void)(lock);
+#define _raw_spin_unlock(lock) do { (void)(lock); } while(0)
 #endif /* CONFIG_DEBUG_SPINLOCK */
 
 /* RW spinlocks: No debug version */
@@ -184,141 +215,116 @@ typedef struct {
 #define _raw_write_unlock(lock)	do { (void)(lock); } while(0)
 #define _raw_write_trylock(lock) ({ (void)(lock); (1); })
 
-#endif /* !SMP */
-
-/*
- * Define the various spin_lock and rw_lock methods.  Note we define these
- * regardless of whether CONFIG_SMP or CONFIG_PREEMPT are set. The various
- * methods are defined as nops in the case they are not required.
- */
-#define spin_trylock(lock)	({preempt_disable(); _raw_spin_trylock(lock) ? \
+#define _spin_trylock(lock)	({preempt_disable(); _raw_spin_trylock(lock) ? \
 				1 : ({preempt_enable(); 0;});})
 
-#define write_trylock(lock)	({preempt_disable();_raw_write_trylock(lock) ? \
+#define _write_trylock(lock)	({preempt_disable(); _raw_write_trylock(lock) ? \
 				1 : ({preempt_enable(); 0;});})
 
-/* Where's read_trylock? */
-
-#if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT)
-void __preempt_spin_lock(spinlock_t *lock);
-void __preempt_write_lock(rwlock_t *lock);
+#define _spin_trylock_bh(lock)	({preempt_disable(); local_bh_disable(); \
+				_raw_spin_trylock(lock) ? \
+				1 : ({preempt_enable(); local_bh_enable(); 0;});})
 
-#define spin_lock(lock) \
-do { \
-	preempt_disable(); \
-	if (unlikely(!_raw_spin_trylock(lock))) \
-		__preempt_spin_lock(lock); \
-} while (0)
-
-#define write_lock(lock) \
-do { \
-	preempt_disable(); \
-	if (unlikely(!_raw_write_trylock(lock))) \
-		__preempt_write_lock(lock); \
-} while (0)
-
-#else
-#define spin_lock(lock)	\
+#define _spin_lock(lock)	\
 do { \
 	preempt_disable(); \
 	_raw_spin_lock(lock); \
 } while(0)
 
-#define write_lock(lock) \
+#define _write_lock(lock) \
 do { \
 	preempt_disable(); \
 	_raw_write_lock(lock); \
 } while(0)
-#endif
-
-#define read_lock(lock)	\
+ 
+#define _read_lock(lock)	\
 do { \
 	preempt_disable(); \
 	_raw_read_lock(lock); \
 } while(0)
 
-#define spin_unlock(lock) \
+#define _spin_unlock(lock) \
 do { \
 	_raw_spin_unlock(lock); \
 	preempt_enable(); \
 } while (0)
 
-#define write_unlock(lock) \
+#define _write_unlock(lock) \
 do { \
 	_raw_write_unlock(lock); \
 	preempt_enable(); \
 } while(0)
 
-#define read_unlock(lock) \
+#define _read_unlock(lock) \
 do { \
 	_raw_read_unlock(lock); \
 	preempt_enable(); \
 } while(0)
 
-#define spin_lock_irqsave(lock, flags) \
-do { \
+#define _spin_lock_irqsave(lock, flags) \
+do {	\
 	local_irq_save(flags); \
 	preempt_disable(); \
-	_raw_spin_lock_flags(lock, flags); \
+	_raw_spin_lock(lock); \
 } while (0)
 
-#define spin_lock_irq(lock) \
+#define _spin_lock_irq(lock) \
 do { \
 	local_irq_disable(); \
 	preempt_disable(); \
 	_raw_spin_lock(lock); \
 } while (0)
 
-#define spin_lock_bh(lock) \
+#define _spin_lock_bh(lock) \
 do { \
 	local_bh_disable(); \
 	preempt_disable(); \
 	_raw_spin_lock(lock); \
 } while (0)
 
-#define read_lock_irqsave(lock, flags) \
-do { \
+#define _read_lock_irqsave(lock, flags) \
+do {	\
 	local_irq_save(flags); \
 	preempt_disable(); \
 	_raw_read_lock(lock); \
 } while (0)
 
-#define read_lock_irq(lock) \
+#define _read_lock_irq(lock) \
 do { \
 	local_irq_disable(); \
 	preempt_disable(); \
 	_raw_read_lock(lock); \
 } while (0)
 
-#define read_lock_bh(lock) \
+#define _read_lock_bh(lock) \
 do { \
 	local_bh_disable(); \
 	preempt_disable(); \
 	_raw_read_lock(lock); \
 } while (0)
 
-#define write_lock_irqsave(lock, flags) \
-do { \
+#define _write_lock_irqsave(lock, flags) \
+do {	\
 	local_irq_save(flags); \
 	preempt_disable(); \
 	_raw_write_lock(lock); \
 } while (0)
 
-#define write_lock_irq(lock) \
+#define _write_lock_irq(lock) \
 do { \
 	local_irq_disable(); \
 	preempt_disable(); \
 	_raw_write_lock(lock); \
 } while (0)
 
-#define write_lock_bh(lock) \
+#define _write_lock_bh(lock) \
 do { \
 	local_bh_disable(); \
 	preempt_disable(); \
 	_raw_write_lock(lock); \
 } while (0)
 
-#define spin_unlock_irqrestore(lock, flags) \
+#define _spin_unlock_irqrestore(lock, flags) \
 do { \
 	_raw_spin_unlock(lock); \
 	local_irq_restore(flags); \
@@ -331,65 +337,128 @@ do { \
 	local_irq_restore(flags); \
 } while (0)
 
-#define spin_unlock_irq(lock) \
+#define _spin_unlock_irq(lock) \
 do { \
 	_raw_spin_unlock(lock); \
 	local_irq_enable(); \
 	preempt_enable(); \
 } while (0)
 
-#define spin_unlock_bh(lock) \
+#define _spin_unlock_bh(lock) \
 do { \
 	_raw_spin_unlock(lock); \
 	preempt_enable(); \
 	local_bh_enable(); \
 } while (0)
 
-#define read_unlock_irqrestore(lock, flags) \
+#define _write_unlock_bh(lock) \
 do { \
-	_raw_read_unlock(lock); \
-	local_irq_restore(flags); \
+	_raw_write_unlock(lock); \
 	preempt_enable(); \
+	local_bh_enable(); \
 } while (0)
 
-#define read_unlock_irq(lock) \
+#define _read_unlock_irqrestore(lock, flags) \
 do { \
 	_raw_read_unlock(lock); \
-	local_irq_enable(); \
+	local_irq_restore(flags); \
 	preempt_enable(); \
 } while (0)
 
-#define read_unlock_bh(lock) \
+#define _write_unlock_irqrestore(lock, flags) \
 do { \
-	_raw_read_unlock(lock); \
+	_raw_write_unlock(lock); \
+	local_irq_restore(flags); \
 	preempt_enable(); \
-	local_bh_enable(); \
 } while (0)
 
-#define write_unlock_irqrestore(lock, flags) \
+#define _read_unlock_irq(lock)	\
 do { \
-	_raw_write_unlock(lock); \
-	local_irq_restore(flags); \
-	preempt_enable(); \
+	_raw_read_unlock(lock);	\
+	local_irq_enable();	\
+	preempt_enable();	\
 } while (0)
 
-#define write_unlock_irq(lock) \
+#define _read_unlock_bh(lock)	\
 do { \
-	_raw_write_unlock(lock); \
-	local_irq_enable(); \
-	preempt_enable(); \
+	_raw_read_unlock(lock);	\
+	local_bh_enable();	\
+	preempt_enable();	\
 } while (0)
 
-#define write_unlock_bh(lock) \
+#define _write_unlock_irq(lock)	\
 do { \
-	_raw_write_unlock(lock); \
-	preempt_enable(); \
-	local_bh_enable(); \
+	_raw_write_unlock(lock);	\
+	local_irq_enable();	\
+	preempt_enable();	\
 } while (0)
 
-#define spin_trylock_bh(lock)	({ local_bh_disable(); preempt_disable(); \
-				_raw_spin_trylock(lock) ? 1 : \
-				({preempt_enable(); local_bh_enable(); 0;});})
+#endif /* !SMP */
+
+/*
+ * Define the various spin_lock and rw_lock methods.  Note we define these
+ * regardless of whether CONFIG_SMP or CONFIG_PREEMPT are set. The various
+ * methods are defined as nops in the case they are not required.
+ */
+#define spin_trylock(lock)	_spin_trylock(lock)
+#define write_trylock(lock)	_write_trylock(lock)
+
+/* Where's read_trylock? */
+
+#if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT)
+void __preempt_spin_lock(spinlock_t *lock);
+void __preempt_write_lock(rwlock_t *lock);
+#endif
+
+#define spin_lock(lock)		_spin_lock(lock)
+#define write_lock(lock)	_write_lock(lock)
+#define read_lock(lock)		_read_lock(lock)
+#define spin_unlock(lock)	_spin_unlock(lock)
+#define write_unlock(lock)	_write_unlock(lock)
+#define read_unlock(lock)	_read_unlock(lock)
+
+#ifdef CONFIG_SMP
+#define spin_lock_irqsave(lock, flags)	flags = _spin_lock_irqsave(lock)
+#define read_lock_irqsave(lock, flags)	flags = _read_lock_irqsave(lock)
+#define write_lock_irqsave(lock, flags)	flags = _write_lock_irqsave(lock)
+#else
+#define spin_lock_irqsave(lock, flags)	_spin_lock_irqsave(lock, flags)
+#define read_lock_irqsave(lock, flags)	_read_lock_irqsave(lock, flags)
+#define write_lock_irqsave(lock, flags)	_write_lock_irqsave(lock, flags)
+#endif
+
+#define spin_lock_irq(lock)		_spin_lock_irq(lock)
+#define spin_lock_bh(lock)		_spin_lock_bh(lock)
+
+#define read_lock_irq(lock)		_read_lock_irq(lock)
+#define read_lock_bh(lock)		_read_lock_bh(lock)
+
+#define write_lock_irq(lock)		_write_lock_irq(lock)
+#define write_lock_bh(lock)		_write_lock_bh(lock)
+#define spin_unlock_irqrestore(lock, flags)	_spin_unlock_irqrestore(lock, flags)
+#define spin_unlock_irq(lock)		_spin_unlock_irq(lock)
+#define spin_unlock_bh(lock)		_spin_unlock_bh(lock)
+
+#define read_unlock_irqrestore(lock, flags)	_read_unlock_irqrestore(lock, flags)
+#define read_unlock_irq(lock)			_read_unlock_irq(lock)
+#define read_unlock_bh(lock)			_read_unlock_bh(lock)
+
+#define write_unlock_irqrestore(lock, flags)	_write_unlock_irqrestore(lock, flags)
+#define write_unlock_irq(lock)			_write_unlock_irq(lock)
+#define write_unlock_bh(lock)			_write_unlock_bh(lock)
+
+#define spin_trylock_bh(lock)			_spin_trylock_bh(lock)
+
+#ifdef CONFIG_LOCKMETER
+extern void _metered_spin_lock   (spinlock_t *lock);
+extern void _metered_spin_unlock (spinlock_t *lock);
+extern int  _metered_spin_trylock(spinlock_t *lock);
+extern void _metered_read_lock    (rwlock_t *lock);
+extern void _metered_read_unlock  (rwlock_t *lock);
+extern void _metered_write_lock   (rwlock_t *lock);
+extern void _metered_write_unlock (rwlock_t *lock);
+extern int  _metered_write_trylock(rwlock_t *lock);
+#endif
 
 /* "lock on reference count zero" */
 #ifndef ATOMIC_DEC_AND_LOCK