X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=include%2Fasm-mips%2Fsystem.h;h=5e1289c85ed958f414ec673a535e48b1c056e221;hb=refs%2Fheads%2Fvserver;hp=a421cdb1e0da06b7ce164de62d88d35dcb23480a;hpb=6a77f38946aaee1cd85eeec6cf4229b204c15071;p=linux-2.6.git diff --git a/include/asm-mips/system.h b/include/asm-mips/system.h index a421cdb1e..5e1289c85 100644 --- a/include/asm-mips/system.h +++ b/include/asm-mips/system.h @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 1994, 95, 96, 97, 98, 99, 2003 by Ralf Baechle + * Copyright (C) 1994, 95, 96, 97, 98, 99, 2003, 06 by Ralf Baechle * Copyright (C) 1996 by Paul M. Antoine * Copyright (C) 1999 Silicon Graphics * Kevin D. Kissell, kevink@mips.org and Carsten Langgaard, carstenl@mips.com @@ -12,139 +12,15 @@ #ifndef _ASM_SYSTEM_H #define _ASM_SYSTEM_H -#include #include +#include #include +#include #include -#include +#include #include -#include -/* - * read_barrier_depends - Flush all pending reads that subsequents reads - * depend on. - * - * No data-dependent reads from memory-like regions are ever reordered - * over this barrier. All reads preceding this primitive are guaranteed - * to access memory (but not necessarily other CPUs' caches) before any - * reads following this primitive that depend on the data return by - * any of the preceding reads. This primitive is much lighter weight than - * rmb() on most CPUs, and is never heavier weight than is - * rmb(). - * - * These ordering constraints are respected by both the local CPU - * and the compiler. - * - * Ordering is not guaranteed by anything other than these primitives, - * not even by data dependencies. See the documentation for - * memory_barrier() for examples and URLs to more information. - * - * For example, the following code would force ordering (the initial - * value of "a" is zero, "b" is one, and "p" is "&a"): - * - * - * CPU 0 CPU 1 - * - * b = 2; - * memory_barrier(); - * p = &b; q = p; - * read_barrier_depends(); - * d = *q; - * - * - * because the read of "*q" depends on the read of "p" and these - * two reads are separated by a read_barrier_depends(). However, - * the following code, with the same initial values for "a" and "b": - * - * - * CPU 0 CPU 1 - * - * a = 2; - * memory_barrier(); - * b = 3; y = b; - * read_barrier_depends(); - * x = a; - * - * - * does not enforce ordering, since there is no data dependency between - * the read of "a" and the read of "b". Therefore, on some CPUs, such - * as Alpha, "y" could be set to 3 and "x" to 0. Use rmb() - * in cases like thiswhere there are no data dependencies. - */ - -#define read_barrier_depends() do { } while(0) - -#ifdef CONFIG_CPU_HAS_SYNC -#define __sync() \ - __asm__ __volatile__( \ - ".set push\n\t" \ - ".set noreorder\n\t" \ - ".set mips2\n\t" \ - "sync\n\t" \ - ".set pop" \ - : /* no output */ \ - : /* no input */ \ - : "memory") -#else -#define __sync() do { } while(0) -#endif - -#define __fast_iob() \ - __asm__ __volatile__( \ - ".set push\n\t" \ - ".set noreorder\n\t" \ - "lw $0,%0\n\t" \ - "nop\n\t" \ - ".set pop" \ - : /* no output */ \ - : "m" (*(int *)CKSEG1) \ - : "memory") - -#define fast_wmb() __sync() -#define fast_rmb() __sync() -#define fast_mb() __sync() -#define fast_iob() \ - do { \ - __sync(); \ - __fast_iob(); \ - } while (0) - -#ifdef CONFIG_CPU_HAS_WB - -#include - -#define wmb() fast_wmb() -#define rmb() fast_rmb() -#define mb() wbflush() -#define iob() wbflush() - -#else /* !CONFIG_CPU_HAS_WB */ - -#define wmb() fast_wmb() -#define rmb() fast_rmb() -#define mb() fast_mb() -#define iob() fast_iob() - -#endif /* !CONFIG_CPU_HAS_WB */ - -#ifdef CONFIG_SMP -#define smp_mb() mb() -#define smp_rmb() rmb() -#define smp_wmb() wmb() -#define smp_read_barrier_depends() read_barrier_depends() -#else -#define smp_mb() barrier() -#define smp_rmb() barrier() -#define smp_wmb() barrier() -#define smp_read_barrier_depends() do { } while(0) -#endif - -#define set_mb(var, value) \ -do { var = value; mb(); } while (0) - -#define set_wmb(var, value) \ -do { var = value; wmb(); } while (0) /* * switch_to(n) should switch tasks to task nr n, first @@ -154,14 +30,56 @@ extern asmlinkage void *resume(void *last, void *next, void *next_ti); struct task_struct; -#define switch_to(prev,next,last) \ -do { \ - (last) = resume(prev, next, next->thread_info); \ +#ifdef CONFIG_MIPS_MT_FPAFF + +/* + * Handle the scheduler resume end of FPU affinity management. We do this + * inline to try to keep the overhead down. If we have been forced to run on + * a "CPU" with an FPU because of a previous high level of FP computation, + * but did not actually use the FPU during the most recent time-slice (CU1 + * isn't set), we undo the restriction on cpus_allowed. + * + * We're not calling set_cpus_allowed() here, because we have no need to + * force prompt migration - we're already switching the current CPU to a + * different thread. + */ + +#define switch_to(prev,next,last) \ +do { \ + if (cpu_has_fpu && \ + (prev->thread.mflags & MF_FPUBOUND) && \ + (!(KSTK_STATUS(prev) & ST0_CU1))) { \ + prev->thread.mflags &= ~MF_FPUBOUND; \ + prev->cpus_allowed = prev->thread.user_cpus_allowed; \ + } \ + if (cpu_has_dsp) \ + __save_dsp(prev); \ + next->thread.emulated_fp = 0; \ + (last) = resume(prev, next, next->thread_info); \ + if (cpu_has_dsp) \ + __restore_dsp(current); \ +} while(0) + +#else +#define switch_to(prev,next,last) \ +do { \ + if (cpu_has_dsp) \ + __save_dsp(prev); \ + (last) = resume(prev, next, task_thread_info(next)); \ + if (cpu_has_dsp) \ + __restore_dsp(current); \ } while(0) +#endif -#define ROT_IN_PIECES \ - " .set noreorder \n" \ - " .set reorder \n" +/* + * On SMP systems, when the scheduler does migration-cost autodetection, + * it needs a way to flush as much of the CPU's caches as possible. + * + * TODO: fill this in! + */ +static inline void sched_cacheflush(void) +{ +} static inline unsigned long __xchg_u32(volatile int * m, unsigned int val) { @@ -171,14 +89,14 @@ static inline unsigned long __xchg_u32(volatile int * m, unsigned int val) unsigned long dummy; __asm__ __volatile__( + " .set mips3 \n" "1: ll %0, %3 # xchg_u32 \n" + " .set mips0 \n" " move %2, %z4 \n" + " .set mips3 \n" " sc %2, %1 \n" " beqzl %2, 1b \n" - ROT_IN_PIECES -#ifdef CONFIG_SMP - " sync \n" -#endif + " .set mips0 \n" : "=&r" (retval), "=m" (*m), "=&r" (dummy) : "R" (*m), "Jr" (val) : "memory"); @@ -186,13 +104,14 @@ static inline unsigned long __xchg_u32(volatile int * m, unsigned int val) unsigned long dummy; __asm__ __volatile__( + " .set mips3 \n" "1: ll %0, %3 # xchg_u32 \n" + " .set mips0 \n" " move %2, %z4 \n" + " .set mips3 \n" " sc %2, %1 \n" " beqz %2, 1b \n" -#ifdef CONFIG_SMP - " sync \n" -#endif + " .set mips0 \n" : "=&r" (retval), "=m" (*m), "=&r" (dummy) : "R" (*m), "Jr" (val) : "memory"); @@ -205,10 +124,12 @@ static inline unsigned long __xchg_u32(volatile int * m, unsigned int val) local_irq_restore(flags); /* implies memory barrier */ } + smp_mb(); + return retval; } -#ifdef CONFIG_MIPS64 +#ifdef CONFIG_64BIT static inline __u64 __xchg_u64(volatile __u64 * m, __u64 val) { __u64 retval; @@ -217,14 +138,12 @@ static inline __u64 __xchg_u64(volatile __u64 * m, __u64 val) unsigned long dummy; __asm__ __volatile__( + " .set mips3 \n" "1: lld %0, %3 # xchg_u64 \n" " move %2, %z4 \n" " scd %2, %1 \n" " beqzl %2, 1b \n" - ROT_IN_PIECES -#ifdef CONFIG_SMP - " sync \n" -#endif + " .set mips0 \n" : "=&r" (retval), "=m" (*m), "=&r" (dummy) : "R" (*m), "Jr" (val) : "memory"); @@ -232,13 +151,12 @@ static inline __u64 __xchg_u64(volatile __u64 * m, __u64 val) unsigned long dummy; __asm__ __volatile__( + " .set mips3 \n" "1: lld %0, %3 # xchg_u64 \n" " move %2, %z4 \n" " scd %2, %1 \n" " beqz %2, 1b \n" -#ifdef CONFIG_SMP - " sync \n" -#endif + " .set mips0 \n" : "=&r" (retval), "=m" (*m), "=&r" (dummy) : "R" (*m), "Jr" (val) : "memory"); @@ -251,6 +169,8 @@ static inline __u64 __xchg_u64(volatile __u64 * m, __u64 val) local_irq_restore(flags); /* implies memory barrier */ } + smp_mb(); + return retval; } #else @@ -265,10 +185,10 @@ extern void __xchg_called_with_bad_pointer(void); static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int size) { switch (size) { - case 4: - return __xchg_u32(ptr, x); - case 8: - return __xchg_u64(ptr, x); + case 4: + return __xchg_u32(ptr, x); + case 8: + return __xchg_u64(ptr, x); } __xchg_called_with_bad_pointer(); return x; @@ -286,35 +206,36 @@ static inline unsigned long __cmpxchg_u32(volatile int * m, unsigned long old, if (cpu_has_llsc && R10000_LLSC_WAR) { __asm__ __volatile__( + " .set push \n" " .set noat \n" + " .set mips3 \n" "1: ll %0, %2 # __cmpxchg_u32 \n" " bne %0, %z3, 2f \n" + " .set mips0 \n" " move $1, %z4 \n" + " .set mips3 \n" " sc $1, %1 \n" " beqzl $1, 1b \n" - ROT_IN_PIECES -#ifdef CONFIG_SMP - " sync \n" -#endif "2: \n" - " .set at \n" - : "=&r" (retval), "=m" (*m) + " .set pop \n" + : "=&r" (retval), "=R" (*m) : "R" (*m), "Jr" (old), "Jr" (new) : "memory"); } else if (cpu_has_llsc) { __asm__ __volatile__( + " .set push \n" " .set noat \n" + " .set mips3 \n" "1: ll %0, %2 # __cmpxchg_u32 \n" " bne %0, %z3, 2f \n" + " .set mips0 \n" " move $1, %z4 \n" + " .set mips3 \n" " sc $1, %1 \n" " beqz $1, 1b \n" -#ifdef CONFIG_SMP - " sync \n" -#endif "2: \n" - " .set at \n" - : "=&r" (retval), "=m" (*m) + " .set pop \n" + : "=&r" (retval), "=R" (*m) : "R" (*m), "Jr" (old), "Jr" (new) : "memory"); } else { @@ -327,46 +248,45 @@ static inline unsigned long __cmpxchg_u32(volatile int * m, unsigned long old, local_irq_restore(flags); /* implies memory barrier */ } + smp_mb(); + return retval; } -#ifdef CONFIG_MIPS64 +#ifdef CONFIG_64BIT static inline unsigned long __cmpxchg_u64(volatile int * m, unsigned long old, unsigned long new) { __u64 retval; - if (cpu_has_llsc) { + if (cpu_has_llsc && R10000_LLSC_WAR) { __asm__ __volatile__( + " .set push \n" " .set noat \n" + " .set mips3 \n" "1: lld %0, %2 # __cmpxchg_u64 \n" " bne %0, %z3, 2f \n" " move $1, %z4 \n" " scd $1, %1 \n" " beqzl $1, 1b \n" - ROT_IN_PIECES -#ifdef CONFIG_SMP - " sync \n" -#endif "2: \n" - " .set at \n" - : "=&r" (retval), "=m" (*m) + " .set pop \n" + : "=&r" (retval), "=R" (*m) : "R" (*m), "Jr" (old), "Jr" (new) : "memory"); } else if (cpu_has_llsc) { __asm__ __volatile__( + " .set push \n" " .set noat \n" + " .set mips3 \n" "1: lld %0, %2 # __cmpxchg_u64 \n" " bne %0, %z3, 2f \n" " move $1, %z4 \n" " scd $1, %1 \n" " beqz $1, 1b \n" -#ifdef CONFIG_SMP - " sync \n" -#endif "2: \n" - " .set at \n" - : "=&r" (retval), "=m" (*m) + " .set pop \n" + : "=&r" (retval), "=R" (*m) : "R" (*m), "Jr" (old), "Jr" (new) : "memory"); } else { @@ -379,6 +299,8 @@ static inline unsigned long __cmpxchg_u64(volatile int * m, unsigned long old, local_irq_restore(flags); /* implies memory barrier */ } + smp_mb(); + return retval; } #else @@ -406,31 +328,21 @@ static inline unsigned long __cmpxchg(volatile void * ptr, unsigned long old, #define cmpxchg(ptr,old,new) ((__typeof__(*(ptr)))__cmpxchg((ptr), (unsigned long)(old), (unsigned long)(new),sizeof(*(ptr)))) +extern void set_handler (unsigned long offset, void *addr, unsigned long len); +extern void set_uncached_handler (unsigned long offset, void *addr, unsigned long len); +extern void *set_vi_handler (int n, void *addr); extern void *set_except_vector(int n, void *addr); +extern unsigned long ebase; extern void per_cpu_trap_init(void); -extern NORET_TYPE void __die(const char *, struct pt_regs *, const char *file, - const char *func, unsigned long line); -extern void __die_if_kernel(const char *, struct pt_regs *, const char *file, - const char *func, unsigned long line); - -#define die(msg, regs) \ - __die(msg, regs, __FILE__ ":", __FUNCTION__, __LINE__) -#define die_if_kernel(msg, regs) \ - __die_if_kernel(msg, regs, __FILE__ ":", __FUNCTION__, __LINE__) - extern int stop_a_enabled; /* - * Taken from include/asm-ia64/system.h; prevents deadlock on SMP + * See include/asm-ia64/system.h; prevents deadlock on SMP * systems. */ -#define prepare_arch_switch(rq, next) \ -do { \ - spin_lock(&(next)->switch_lock); \ - spin_unlock(&(rq)->lock); \ -} while (0) -#define finish_arch_switch(rq, prev) spin_unlock_irq(&(prev)->switch_lock) -#define task_running(rq, p) ((rq)->curr == (p) || spin_is_locked(&(p)->switch_lock)) +#define __ARCH_WANT_UNLOCKED_CTXSW + +#define arch_align_stack(x) (x) #endif /* _ASM_SYSTEM_H */