From 1075cf7a959f72833e54dd2d4f885617e58e3e0a Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 30 Jan 2008 13:30:34 +0100 Subject: [PATCH] x86: merge spinlock.h variants Merge them finally together Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- include/asm-x86/spinlock.h | 211 +++++++++++++++++++++++++++++++++- include/asm-x86/spinlock_32.h | 208 --------------------------------- include/asm-x86/spinlock_64.h | 186 ------------------------------ 3 files changed, 209 insertions(+), 396 deletions(-) delete mode 100644 include/asm-x86/spinlock_32.h delete mode 100644 include/asm-x86/spinlock_64.h diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h index e1d555a3dfe5..afd4b80ff0ad 100644 --- a/include/asm-x86/spinlock.h +++ b/include/asm-x86/spinlock.h @@ -1,6 +1,22 @@ #ifndef _X86_SPINLOCK_H_ #define _X86_SPINLOCK_H_ +#include +#include +#include +#include + +/* + * Your basic SMP spinlocks, allowing only a single CPU anywhere + * + * Simple spin lock operations. There are two variants, one clears IRQ's + * on the local processor, one does not. + * + * We make no fairness assumptions. They have a cost. + * + * (the type definitions are in asm/spinlock_types.h) + */ + #ifdef CONFIG_PARAVIRT #include #else @@ -11,9 +27,200 @@ #endif /* CONFIG_PARAVIRT */ #ifdef CONFIG_X86_32 -# include "spinlock_32.h" +typedef char _slock_t; +# define LOCK_INS_DEC "decb" +# define LOCK_INS_XCH "xchgb" +# define LOCK_INS_MOV "movb" +# define LOCK_INS_CMP "cmpb" +# define LOCK_PTR_REG "a" #else -# include "spinlock_64.h" +typedef int _slock_t; +# define LOCK_INS_DEC "decl" +# define LOCK_INS_XCH "xchgl" +# define LOCK_INS_MOV "movl" +# define LOCK_INS_CMP "cmpl" +# define LOCK_PTR_REG "D" #endif +static inline int __raw_spin_is_locked(raw_spinlock_t *lock) +{ + return *(volatile _slock_t *)(&(lock)->slock) <= 0; +} + +static inline void __raw_spin_lock(raw_spinlock_t *lock) +{ + asm volatile( + "\n1:\t" + LOCK_PREFIX " ; " LOCK_INS_DEC " %0\n\t" + "jns 3f\n" + "2:\t" + "rep;nop\n\t" + LOCK_INS_CMP " $0,%0\n\t" + "jle 2b\n\t" + "jmp 1b\n" + "3:\n\t" + : "+m" (lock->slock) : : "memory"); +} + +/* + * It is easier for the lock validator if interrupts are not re-enabled + * in the middle of a lock-acquire. This is a performance feature anyway + * so we turn it off: + * + * NOTE: there's an irqs-on section here, which normally would have to be + * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use this variant. + */ +#ifndef CONFIG_PROVE_LOCKING +static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, + unsigned long flags) +{ + asm volatile( + "\n1:\t" + LOCK_PREFIX " ; " LOCK_INS_DEC " %[slock]\n\t" + "jns 5f\n" + "testl $0x200, %[flags]\n\t" + "jz 4f\n\t" + STI_STRING "\n" + "3:\t" + "rep;nop\n\t" + LOCK_INS_CMP " $0, %[slock]\n\t" + "jle 3b\n\t" + CLI_STRING "\n\t" + "jmp 1b\n" + "4:\t" + "rep;nop\n\t" + LOCK_INS_CMP " $0, %[slock]\n\t" + "jg 1b\n\t" + "jmp 4b\n" + "5:\n\t" + : [slock] "+m" (lock->slock) + : [flags] "r" ((u32)flags) + CLI_STI_INPUT_ARGS + : "memory" CLI_STI_CLOBBERS); +} +#endif + +static inline int __raw_spin_trylock(raw_spinlock_t *lock) +{ + _slock_t oldval; + + asm volatile( + LOCK_INS_XCH " %0,%1" + :"=q" (oldval), "+m" (lock->slock) + :"0" (0) : "memory"); + + return oldval > 0; +} + +/* + * __raw_spin_unlock based on writing $1 to the low byte. + * This method works. Despite all the confusion. + * (except on PPro SMP or if we are using OOSTORE, so we use xchgb there) + * (PPro errata 66, 92) + */ +#if defined(X86_64) || \ + (!defined(CONFIG_X86_OOSTORE) && !defined(CONFIG_X86_PPRO_FENCE)) + +static inline void __raw_spin_unlock(raw_spinlock_t *lock) +{ + asm volatile(LOCK_INS_MOV " $1,%0" : "=m" (lock->slock) :: "memory"); +} + +#else + +static inline void __raw_spin_unlock(raw_spinlock_t *lock) +{ + unsigned char oldval = 1; + + asm volatile("xchgb %b0, %1" + : "=q" (oldval), "+m" (lock->slock) + : "0" (oldval) : "memory"); +} + +#endif + +static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock) +{ + while (__raw_spin_is_locked(lock)) + cpu_relax(); +} + +/* + * Read-write spinlocks, allowing multiple readers + * but only one writer. + * + * NOTE! it is quite common to have readers in interrupts + * but no interrupt writers. For those circumstances we + * can "mix" irq-safe locks - any writer needs to get a + * irq-safe write-lock, but readers can get non-irqsafe + * read-locks. + * + * On x86, we implement read-write locks as a 32-bit counter + * with the high bit (sign) being the "contended" bit. + */ + +static inline int __raw_read_can_lock(raw_rwlock_t *lock) +{ + return (int)(lock)->lock > 0; +} + +static inline int __raw_write_can_lock(raw_rwlock_t *lock) +{ + return (lock)->lock == RW_LOCK_BIAS; +} + +static inline void __raw_read_lock(raw_rwlock_t *rw) +{ + asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t" + "jns 1f\n" + "call __read_lock_failed\n\t" + "1:\n" + ::LOCK_PTR_REG (rw) : "memory"); +} + +static inline void __raw_write_lock(raw_rwlock_t *rw) +{ + asm volatile(LOCK_PREFIX " subl %1,(%0)\n\t" + "jz 1f\n" + "call __write_lock_failed\n\t" + "1:\n" + ::LOCK_PTR_REG (rw), "i" (RW_LOCK_BIAS) : "memory"); +} + +static inline int __raw_read_trylock(raw_rwlock_t *lock) +{ + atomic_t *count = (atomic_t *)lock; + + atomic_dec(count); + if (atomic_read(count) >= 0) + return 1; + atomic_inc(count); + return 0; +} + +static inline int __raw_write_trylock(raw_rwlock_t *lock) +{ + atomic_t *count = (atomic_t *)lock; + + if (atomic_sub_and_test(RW_LOCK_BIAS, count)) + return 1; + atomic_add(RW_LOCK_BIAS, count); + return 0; +} + +static inline void __raw_read_unlock(raw_rwlock_t *rw) +{ + asm volatile(LOCK_PREFIX "incl %0" :"+m" (rw->lock) : : "memory"); +} + +static inline void __raw_write_unlock(raw_rwlock_t *rw) +{ + asm volatile(LOCK_PREFIX "addl %1, %0" + : "+m" (rw->lock) : "i" (RW_LOCK_BIAS) : "memory"); +} + +#define _raw_spin_relax(lock) cpu_relax() +#define _raw_read_relax(lock) cpu_relax() +#define _raw_write_relax(lock) cpu_relax() + #endif diff --git a/include/asm-x86/spinlock_32.h b/include/asm-x86/spinlock_32.h deleted file mode 100644 index 2de9b8b89903..000000000000 --- a/include/asm-x86/spinlock_32.h +++ /dev/null @@ -1,208 +0,0 @@ -#ifndef __ASM_SPINLOCK_H -#define __ASM_SPINLOCK_H - -#include -#include -#include -#include - -/* - * Your basic SMP spinlocks, allowing only a single CPU anywhere - * - * Simple spin lock operations. There are two variants, one clears IRQ's - * on the local processor, one does not. - * - * We make no fairness assumptions. They have a cost. - * - * (the type definitions are in asm/spinlock_types.h) - */ - -typedef char _slock_t; -#define LOCK_INS_DEC "decb" -#define LOCK_INS_XCH "xchgb" -#define LOCK_INS_MOV "movb" -#define LOCK_INS_CMP "cmpb" -#define LOCK_PTR_REG "a" - -static inline int __raw_spin_is_locked(raw_spinlock_t *lock) -{ - return *(volatile _slock_t *)(&(lock)->slock) <= 0; -} - -static inline void __raw_spin_lock(raw_spinlock_t *lock) -{ - asm volatile( - "\n1:\t" - LOCK_PREFIX " ; " LOCK_INS_DEC " %0\n\t" - "jns 3f\n" - "2:\t" - "rep;nop\n\t" - LOCK_INS_CMP " $0,%0\n\t" - "jle 2b\n\t" - "jmp 1b\n" - "3:\n\t" - : "+m" (lock->slock) : : "memory"); -} - -/* - * It is easier for the lock validator if interrupts are not re-enabled - * in the middle of a lock-acquire. This is a performance feature anyway - * so we turn it off: - * - * NOTE: there's an irqs-on section here, which normally would have to be - * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use this variant. - */ -#ifndef CONFIG_PROVE_LOCKING -static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, - unsigned long flags) -{ - asm volatile( - "\n1:\t" - LOCK_PREFIX " ; " LOCK_INS_DEC " %[slock]\n\t" - "jns 5f\n" - "testl $0x200, %[flags]\n\t" - "jz 4f\n\t" - STI_STRING "\n" - "3:\t" - "rep;nop\n\t" - LOCK_INS_CMP " $0, %[slock]\n\t" - "jle 3b\n\t" - CLI_STRING "\n\t" - "jmp 1b\n" - "4:\t" - "rep;nop\n\t" - LOCK_INS_CMP " $0, %[slock]\n\t" - "jg 1b\n\t" - "jmp 4b\n" - "5:\n\t" - : [slock] "+m" (lock->slock) - : [flags] "r" ((u32)flags) - CLI_STI_INPUT_ARGS - : "memory" CLI_STI_CLOBBERS); -} -#endif - -static inline int __raw_spin_trylock(raw_spinlock_t *lock) -{ - _slock_t oldval; - - asm volatile( - LOCK_INS_XCH " %0,%1" - :"=q" (oldval), "+m" (lock->slock) - :"0" (0) : "memory"); - - return oldval > 0; -} - -/* - * __raw_spin_unlock based on writing $1 to the low byte. - * This method works. Despite all the confusion. - * (except on PPro SMP or if we are using OOSTORE, so we use xchgb there) - * (PPro errata 66, 92) - */ - -#if !defined(CONFIG_X86_OOSTORE) && !defined(CONFIG_X86_PPRO_FENCE) - -static inline void __raw_spin_unlock(raw_spinlock_t *lock) -{ - asm volatile(LOCK_INS_MOV " $1,%0" : "=m" (lock->slock) :: "memory"); -} - -#else - -static inline void __raw_spin_unlock(raw_spinlock_t *lock) -{ - unsigned char oldval = 1; - - asm volatile("xchgb %b0, %1" - : "=q" (oldval), "+m" (lock->slock) - : "0" (oldval) : "memory"); -} - -#endif - -static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock) -{ - while (__raw_spin_is_locked(lock)) - cpu_relax(); -} - -/* - * Read-write spinlocks, allowing multiple readers - * but only one writer. - * - * NOTE! it is quite common to have readers in interrupts - * but no interrupt writers. For those circumstances we - * can "mix" irq-safe locks - any writer needs to get a - * irq-safe write-lock, but readers can get non-irqsafe - * read-locks. - * - * On x86, we implement read-write locks as a 32-bit counter - * with the high bit (sign) being the "contended" bit. - */ - -static inline int __raw_read_can_lock(raw_rwlock_t *lock) -{ - return (int)(lock)->lock > 0; -} - -static inline int __raw_write_can_lock(raw_rwlock_t *lock) -{ - return (lock)->lock == RW_LOCK_BIAS; -} - -static inline void __raw_read_lock(raw_rwlock_t *rw) -{ - asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t" - "jns 1f\n" - "call __read_lock_failed\n\t" - "1:\n" - ::LOCK_PTR_REG (rw) : "memory"); -} - -static inline void __raw_write_lock(raw_rwlock_t *rw) -{ - asm volatile(LOCK_PREFIX " subl %1,(%0)\n\t" - "jz 1f\n" - "call __write_lock_failed\n\t" - "1:\n" - ::LOCK_PTR_REG (rw), "i" (RW_LOCK_BIAS) : "memory"); -} - -static inline int __raw_read_trylock(raw_rwlock_t *lock) -{ - atomic_t *count = (atomic_t *)lock; - - atomic_dec(count); - if (atomic_read(count) >= 0) - return 1; - atomic_inc(count); - return 0; -} - -static inline int __raw_write_trylock(raw_rwlock_t *lock) -{ - atomic_t *count = (atomic_t *)lock; - - if (atomic_sub_and_test(RW_LOCK_BIAS, count)) - return 1; - atomic_add(RW_LOCK_BIAS, count); - return 0; -} - -static inline void __raw_read_unlock(raw_rwlock_t *rw) -{ - asm volatile(LOCK_PREFIX "incl %0" :"+m" (rw->lock) : : "memory"); -} - -static inline void __raw_write_unlock(raw_rwlock_t *rw) -{ - asm volatile(LOCK_PREFIX "addl %1, %0" - : "+m" (rw->lock) : "i" (RW_LOCK_BIAS) : "memory"); -} - -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() - -#endif /* __ASM_SPINLOCK_H */ diff --git a/include/asm-x86/spinlock_64.h b/include/asm-x86/spinlock_64.h deleted file mode 100644 index f5ba90b7335c..000000000000 --- a/include/asm-x86/spinlock_64.h +++ /dev/null @@ -1,186 +0,0 @@ -#ifndef __ASM_SPINLOCK_H -#define __ASM_SPINLOCK_H - -#include -#include -#include -#include - -/* - * Your basic SMP spinlocks, allowing only a single CPU anywhere - * - * Simple spin lock operations. There are two variants, one clears IRQ's - * on the local processor, one does not. - * - * We make no fairness assumptions. They have a cost. - * - * (the type definitions are in asm/spinlock_types.h) - */ - -typedef int _slock_t; -#define LOCK_INS_DEC "decl" -#define LOCK_INS_XCH "xchgl" -#define LOCK_INS_MOV "movl" -#define LOCK_INS_CMP "cmpl" -#define LOCK_PTR_REG "D" - -static inline int __raw_spin_is_locked(raw_spinlock_t *lock) -{ - return *(volatile _slock_t *)(&(lock)->slock) <= 0; -} - -static inline void __raw_spin_lock(raw_spinlock_t *lock) -{ - asm volatile( - "\n1:\t" - LOCK_PREFIX " ; " LOCK_INS_DEC " %0\n\t" - "jns 3f\n" - "2:\t" - "rep;nop\n\t" - LOCK_INS_CMP " $0,%0\n\t" - "jle 2b\n\t" - "jmp 1b\n" - "3:\n\t" - : "+m" (lock->slock) : : "memory"); -} - -/* - * It is easier for the lock validator if interrupts are not re-enabled - * in the middle of a lock-acquire. This is a performance feature anyway - * so we turn it off: - * - * NOTE: there's an irqs-on section here, which normally would have to be - * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use this variant. - */ -#ifndef CONFIG_PROVE_LOCKING -static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, - unsigned long flags) -{ - asm volatile( - "\n1:\t" - LOCK_PREFIX " ; " LOCK_INS_DEC " %[slock]\n\t" - "jns 5f\n" - "testl $0x200, %[flags]\n\t" - "jz 4f\n\t" - STI_STRING "\n" - "3:\t" - "rep;nop\n\t" - LOCK_INS_CMP " $0, %[slock]\n\t" - "jle 3b\n\t" - CLI_STRING "\n\t" - "jmp 1b\n" - "4:\t" - "rep;nop\n\t" - LOCK_INS_CMP " $0, %[slock]\n\t" - "jg 1b\n\t" - "jmp 4b\n" - "5:\n\t" - : [slock] "+m" (lock->slock) - : [flags] "r" ((u32)flags) - CLI_STI_INPUT_ARGS - : "memory" CLI_STI_CLOBBERS); -} -#endif - -static inline int __raw_spin_trylock(raw_spinlock_t *lock) -{ - _slock_t oldval; - - asm volatile( - LOCK_INS_XCH " %0,%1" - :"=q" (oldval), "+m" (lock->slock) - :"0" (0) : "memory"); - - return oldval > 0; -} - -static inline void __raw_spin_unlock(raw_spinlock_t *lock) -{ - asm volatile(LOCK_INS_MOV " $1,%0" : "=m" (lock->slock) :: "memory"); -} - -static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock) -{ - while (__raw_spin_is_locked(lock)) - cpu_relax(); -} - -/* - * Read-write spinlocks, allowing multiple readers - * but only one writer. - * - * NOTE! it is quite common to have readers in interrupts - * but no interrupt writers. For those circumstances we - * can "mix" irq-safe locks - any writer needs to get a - * irq-safe write-lock, but readers can get non-irqsafe - * read-locks. - * - * On x86, we implement read-write locks as a 32-bit counter - * with the high bit (sign) being the "contended" bit. - */ - -static inline int __raw_read_can_lock(raw_rwlock_t *lock) -{ - return (int)(lock)->lock > 0; -} - -static inline int __raw_write_can_lock(raw_rwlock_t *lock) -{ - return (lock)->lock == RW_LOCK_BIAS; -} - -static inline void __raw_read_lock(raw_rwlock_t *rw) -{ - asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t" - "jns 1f\n" - "call __read_lock_failed\n\t" - "1:\n" - ::LOCK_PTR_REG (rw) : "memory"); -} - -static inline void __raw_write_lock(raw_rwlock_t *rw) -{ - asm volatile(LOCK_PREFIX " subl %1,(%0)\n\t" - "jz 1f\n" - "call __write_lock_failed\n\t" - "1:\n" - ::LOCK_PTR_REG (rw), "i" (RW_LOCK_BIAS) : "memory"); -} - -static inline int __raw_read_trylock(raw_rwlock_t *lock) -{ - atomic_t *count = (atomic_t *)lock; - - atomic_dec(count); - if (atomic_read(count) >= 0) - return 1; - atomic_inc(count); - return 0; -} - -static inline int __raw_write_trylock(raw_rwlock_t *lock) -{ - atomic_t *count = (atomic_t *)lock; - - if (atomic_sub_and_test(RW_LOCK_BIAS, count)) - return 1; - atomic_add(RW_LOCK_BIAS, count); - return 0; -} - -static inline void __raw_read_unlock(raw_rwlock_t *rw) -{ - asm volatile(LOCK_PREFIX "incl %0" :"+m" (rw->lock) : : "memory"); -} - -static inline void __raw_write_unlock(raw_rwlock_t *rw) -{ - asm volatile(LOCK_PREFIX "addl %1, %0" - : "+m" (rw->lock) : "i" (RW_LOCK_BIAS) : "memory"); -} - -#define _raw_spin_relax(lock) cpu_relax() -#define _raw_read_relax(lock) cpu_relax() -#define _raw_write_relax(lock) cpu_relax() - -#endif /* __ASM_SPINLOCK_H */