I've tested this now with making cpumask_t a struct, and it works fine
(at the moment it's unsigned long for every arch, no change).
First tiny patch adds a "migrate_to_cpu(cpu)" helper to handle that
(common) case simply.
Hot-unplug cpu patch does more cpu mask manip, so I want this fixed FIRST.
Thanks,
Rusty.
-- Anyone who quotes me in their sig is an idiot. -- Rusty Russell.Name: migrate_to_cpu patch Author: Rusty Russell Status: Trivial
D: This patch adds an easy "migrate_to_cpu()" call for simple D: in-kernel affinity, in preparation for generic cpu maps.
diff -urpN -I \$.*\$ --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal linux-2.5.29.9350/Documentation/sched-coding.txt linux-2.5.29.9350.updated/Documentation/sched-coding.txt --- linux-2.5.29.9350/Documentation/sched-coding.txt Thu May 30 10:00:45 2002 +++ linux-2.5.29.9350.updated/Documentation/sched-coding.txt Sat Jul 27 17:14:47 2002 @@ -103,7 +103,10 @@ void set_user_nice(task_t *p, long nice) Sets the "nice" value of task p to the given value. int setscheduler(pid_t pid, int policy, struct sched_param *param) Sets the scheduling policy and parameters for the given pid. -void set_cpus_allowed(task_t *p, unsigned long new_mask) +void migrate_to_cpu(unsigned int cpu) + Migrates the current task to the cpu and nails it there. This + is a more convenient form of: +void set_cpus_allowed(task_t *p, cpu_mask_t new_mask) Sets a given task's CPU affinity and migrates it to a proper cpu. Callers must have a valid reference to the task and assure the task not exit prematurely. No locks can be held during the call. diff -urpN -I '\$.*\$' --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .5600-linux-2.5.30/arch/ppc64/kernel/rtasd.c .5600-linux-2.5.30.updated/arch/ppc64/kernel/rtasd.c --- .5600-linux-2.5.30/arch/ppc64/kernel/rtasd.c 2002-07-25 10:13:04.000000000 +1000 +++ .5600-linux-2.5.30.updated/arch/ppc64/kernel/rtasd.c 2002-08-06 10:56:10.000000000 +1000 @@ -226,7 +226,7 @@ repeat: continue; DEBUG("scheduling on %d\n", cpu); - set_cpus_allowed(current, 1UL << cpu); + migrate_to_cpu(cpu); DEBUG("watchdog scheduled on cpu %d\n", smp_processor_id()); do { diff -urpN -I \$.*\$ --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal linux-2.5.29.9350/include/linux/sched.h linux-2.5.29.9350.updated/include/linux/sched.h --- linux-2.5.29.9350/include/linux/sched.h Sat Jul 27 15:24:39 2002 +++ linux-2.5.29.9350.updated/include/linux/sched.h Sat Jul 27 17:14:47 2002 @@ -411,8 +411,16 @@ do { if (atomic_dec_and_test(&(tsk)->usa #if CONFIG_SMP extern void set_cpus_allowed(task_t *p, unsigned long new_mask); +static inline void migrate_to_cpu(unsigned int cpu) +{ + unsigned long mask = 0; + BUG_ON(!cpu_online(cpu)); + __set_bit(cpu, &mask); + set_cpus_allowed(current, mask); +} #else -# define set_cpus_allowed(p, new_mask) do { } while (0) +#define set_cpus_allowed(p, new_mask) do { } while (0) +#define migrate_to_cpu(cpu) do { BUG_ON((cpu) != 0); } while(0) #endif extern void set_user_nice(task_t *p, long nice); diff -urpN -I \$.*\$ --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal linux-2.5.29.9350/kernel/softirq.c linux-2.5.29.9350.updated/kernel/softirq.c --- linux-2.5.29.9350/kernel/softirq.c Sat Jul 27 16:07:05 2002 +++ linux-2.5.29.9350.updated/kernel/softirq.c Sat Jul 27 17:14:47 2002 @@ -361,8 +361,7 @@ static int ksoftirqd(void * __bind_cpu) current->flags |= PF_IOTHREAD; sigfillset(¤t->blocked); - /* Migrate to the right CPU */ - set_cpus_allowed(current, 1UL << cpu); + migrate_to_cpu(cpu); if (smp_processor_id() != cpu) BUG(); diff -urpN -I '\$.*\$' --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .5600-linux-2.5.30/kernel/sched.c .5600-linux-2.5.30.updated/kernel/sched.c --- .5600-linux-2.5.30/kernel/sched.c 2002-08-02 11:15:10.000000000 +1000 +++ .5600-linux-2.5.30.updated/kernel/sched.c 2002-08-06 10:56:45.000000000 +1000 @@ -1972,7 +1973,7 @@ static int migration_thread(void * data) sigfillset(¤t->blocked); set_fs(KERNEL_DS); - set_cpus_allowed(current, 1UL << cpu); + migrate_to_cpu(cpu); /* * Migration can happen without a migration thread on the Name: CPU mask patch Author: Rusty Russell Status: Trivial Depends: Hotcpu/migrate-to-cpu.patch.gz
D: This patch adds a cpumask_t for easy future expansion of cpu masks. D: It also changes any_online_cpu to return NR_CPUS on failure, like D: find_first_bit.
diff -urpN --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .25506-2.5.30-cpumask.pre/include/asm-i386/smp.h .25506-2.5.30-cpumask/include/asm-i386/smp.h --- .25506-2.5.30-cpumask.pre/include/asm-i386/smp.h 2002-08-02 11:15:10.000000000 +1000 +++ .25506-2.5.30-cpumask/include/asm-i386/smp.h 2002-08-08 17:27:18.000000000 +1000 @@ -90,17 +90,26 @@ extern volatile unsigned long cpu_callou #define cpu_possible(cpu) (cpu_callout_map & (1<<(cpu))) #define cpu_online(cpu) (cpu_online_map & (1<<(cpu))) -extern inline unsigned int num_online_cpus(void) +/* This has to be bitop capable, so for most archs, an unsigned long */ +typedef unsigned long cpumask_t; +#define CPU_MASK_ALL (~0UL) +#define CPU_MASK_NONE (0UL) + +static inline void cpu_online_mask(cpumask_t *res, const cpumask_t *mask) { - return hweight32(cpu_online_map); + *res = (*mask & cpu_online_map); } -extern inline int any_online_cpu(unsigned int mask) +static inline unsigned int any_online_cpu(const cpumask_t *mask) { - if (mask & cpu_online_map) - return __ffs(mask & cpu_online_map); + if (*mask & cpu_online_map) + return __ffs(*mask & cpu_online_map); + return NR_CPUS; +} - return -1; +static inline unsigned int num_online_cpus(void) +{ + return hweight32(cpu_online_map); } static __inline int hard_smp_processor_id(void) diff -urpN --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .25506-2.5.30-cpumask.pre/include/asm-ia64/smp.h .25506-2.5.30-cpumask/include/asm-ia64/smp.h --- .25506-2.5.30-cpumask.pre/include/asm-ia64/smp.h 2002-06-20 01:28:51.000000000 +1000 +++ .25506-2.5.30-cpumask/include/asm-ia64/smp.h 2002-08-08 17:27:18.000000000 +1000 @@ -51,12 +51,22 @@ extern inline unsigned int num_online_cp return hweight64(cpu_online_map); } -extern inline int any_online_cpu(unsigned int mask) +/* This has to be bitop capable, so for most archs, an unsigned long */ +typedef unsigned long cpumask_t; +#define CPU_MASK_ALL (~0UL) +#define CPU_MASK_NONE (0UL) + +static inline void cpu_online_mask(cpumask_t *res, const cpumask_t *mask) { - if (mask & cpu_online_map) - return __ffs(mask & cpu_online_map); + *res = (*mask & cpu_online_map); +} - return -1; +static inline unsigned int any_online_cpu(const cpumask_t *mask) +{ + if (*mask & cpu_online_map) + return __ffs(*mask & cpu_online_map); + + return NR_CPUS; } /* diff -urpN --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .25506-2.5.30-cpumask.pre/include/asm-ppc/smp.h .25506-2.5.30-cpumask/include/asm-ppc/smp.h --- .25506-2.5.30-cpumask.pre/include/asm-ppc/smp.h 2002-07-27 15:24:39.000000000 +1000 +++ .25506-2.5.30-cpumask/include/asm-ppc/smp.h 2002-08-08 17:27:18.000000000 +1000 @@ -56,12 +56,20 @@ extern inline unsigned int num_online_cp return hweight32(cpu_online_map); } -extern inline int any_online_cpu(unsigned int mask) +typedef unsigned long cpumask_t; +#define CPU_MASK_ALL (~0UL) +#define CPU_MASK_NONE (0UL) + +static inline void cpu_online_mask(cpumask_t *res, const cpumask_t *mask) { - if (mask & cpu_online_map) - return __ffs(mask & cpu_online_map); + *res = (*mask & cpu_online_map); +} - return -1; +static inline unsigned int any_online_cpu(const cpumask_t *mask) +{ + if (*mask & cpu_online_map) + return __ffs(*mask & cpu_online_map); + return NR_CPUS; } extern int __cpu_up(unsigned int cpu); diff -urpN --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .25506-2.5.30-cpumask.pre/include/asm-ppc64/smp.h .25506-2.5.30-cpumask/include/asm-ppc64/smp.h --- .25506-2.5.30-cpumask.pre/include/asm-ppc64/smp.h 2002-07-25 10:13:17.000000000 +1000 +++ .25506-2.5.30-cpumask/include/asm-ppc64/smp.h 2002-08-08 17:27:18.000000000 +1000 @@ -53,6 +53,23 @@ static inline int num_online_cpus(void) return nr; } +typedef unsigned long cpumask_t; +#define CPU_MASK_ALL (~0UL) +#define CPU_MASK_NONE (0UL) + +static inline void cpu_online_mask(cpumask_t *res, const cpumask_t *mask) +{ + *res = (mask & cpu_online_map); +} + +static inline unsigned int any_online_cpu(const cpumask_t *mask) +{ + if (*mask & cpu_online_map) + return __ffs(*mask & cpu_online_map); + + return NR_CPUS; +} + extern volatile unsigned long cpu_callin_map[NR_CPUS]; #define smp_processor_id() (get_paca()->xPacaIndex) diff -urpN --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .25506-2.5.30-cpumask.pre/include/asm-sparc64/smp.h .25506-2.5.30-cpumask/include/asm-sparc64/smp.h --- .25506-2.5.30-cpumask.pre/include/asm-sparc64/smp.h 2002-06-21 09:41:55.000000000 +1000 +++ .25506-2.5.30-cpumask/include/asm-sparc64/smp.h 2002-08-08 17:27:18.000000000 +1000 @@ -70,11 +70,20 @@ extern unsigned long cpu_online_map; extern atomic_t sparc64_num_cpus_online; #define num_online_cpus() (atomic_read(&sparc64_num_cpus_online)) -static inline int any_online_cpu(unsigned long mask) +typedef unsigned long cpumask_t; +#define CPU_MASK_ALL (~0UL) +#define CPU_MASK_NONE (0UL) + +static inline void cpu_online_mask(cpumask_t *res, const cpumask_t *mask) { - if ((mask &= cpu_online_map) != 0UL) - return __ffs(mask); - return -1; + *res = (*mask & cpu_online_map); +} + +static inline int any_online_cpu(const cpumask_t *mask) +{ + if (*mask & cpu_online_map) != 0UL) + return __ffs(*mask & cpu_online_map); + return NR_CPUS; } /* diff -urpN --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .25506-2.5.30-cpumask.pre/include/linux/init_task.h .25506-2.5.30-cpumask/include/linux/init_task.h --- .25506-2.5.30-cpumask.pre/include/linux/init_task.h 2002-07-25 10:13:18.000000000 +1000 +++ .25506-2.5.30-cpumask/include/linux/init_task.h 2002-08-08 17:27:18.000000000 +1000 @@ -48,7 +48,7 @@ prio: MAX_PRIO-20, \ static_prio: MAX_PRIO-20, \ policy: SCHED_NORMAL, \ - cpus_allowed: -1, \ + cpus_allowed: CPU_MASK_ALL, \ mm: NULL, \ active_mm: &init_mm, \ run_list: LIST_HEAD_INIT(tsk.run_list), \ diff -urpN --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .25506-2.5.30-cpumask.pre/include/linux/sched.h .25506-2.5.30-cpumask/include/linux/sched.h --- .25506-2.5.30-cpumask.pre/include/linux/sched.h 2002-08-08 17:26:59.000000000 +1000 +++ .25506-2.5.30-cpumask/include/linux/sched.h 2002-08-08 17:27:18.000000000 +1000 @@ -20,6 +20,7 @@ extern unsigned long event; #include <asm/page.h> #include <asm/ptrace.h> #include <asm/mmu.h> +#include <asm/current.h> #include <linux/smp.h> #include <linux/sem.h> @@ -262,7 +263,7 @@ struct task_struct { unsigned long sleep_timestamp; unsigned long policy; - unsigned long cpus_allowed; + cpumask_t cpus_allowed; unsigned int time_slice, first_time_slice; struct list_head tasks; @@ -410,13 +411,13 @@ do { if (atomic_dec_and_test(&(tsk)->usa #define _STK_LIM (8*1024*1024) #if CONFIG_SMP -extern void set_cpus_allowed(task_t *p, unsigned long new_mask); +extern void set_cpus_allowed(task_t *p, cpumask_t *new_mask); static inline void migrate_to_cpu(unsigned int cpu) { - unsigned long mask = 0; + cpumask_t mask = CPU_MASK_NONE; BUG_ON(!cpu_online(cpu)); - __set_bit(cpu, &mask); - set_cpus_allowed(current, mask); + __set_bit(cpu, (unsigned long *)&mask); + set_cpus_allowed(current, &mask); } #else #define set_cpus_allowed(p, new_mask) do { } while (0) @@ -487,8 +488,6 @@ static inline struct task_struct *find_t extern struct user_struct * alloc_uid(uid_t); extern void free_uid(struct user_struct *); -#include <asm/current.h> - extern unsigned long itimer_ticks; extern unsigned long itimer_next; extern void do_timer(struct pt_regs *); diff -urpN --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .25506-2.5.30-cpumask.pre/include/linux/smp.h .25506-2.5.30-cpumask/include/linux/smp.h --- .25506-2.5.30-cpumask.pre/include/linux/smp.h 2002-08-02 11:15:10.000000000 +1000 +++ .25506-2.5.30-cpumask/include/linux/smp.h 2002-08-08 17:27:18.000000000 +1000 @@ -93,13 +93,18 @@ int cpu_up(unsigned int cpu); #define smp_call_function(func,info,retry,wait) ({ 0; }) static inline void smp_send_reschedule(int cpu) { } static inline void smp_send_reschedule_all(void) { } -#define cpu_online_map 1 #define cpu_online(cpu) ({ cpu; 1; }) #define num_online_cpus() 1 #define __per_cpu_data #define per_cpu(var, cpu) var #define this_cpu(var) var +typedef unsigned long cpumask_t; +#define CPU_MASK_ALL 1 +#define CPU_MASK_NONE 0 +#define cpu_online_mask(res,mask) do { *(res) = (*(mask) & 1); } while(0) +#define any_online_cpu(mask) ((*(mask) & 1) ? 0 : 1) + /* Need to know about CPUs going up/down? */ static inline int register_cpu_notifier(struct notifier_block *nb) { diff -urpN --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .25506-2.5.30-cpumask.pre/kernel/sched.c .25506-2.5.30-cpumask/kernel/sched.c --- .25506-2.5.30-cpumask.pre/kernel/sched.c 2002-08-08 17:26:59.000000000 +1000 +++ .25506-2.5.30-cpumask/kernel/sched.c 2002-08-08 17:27:18.000000000 +1000 @@ -415,7 +415,8 @@ repeat_lock_task: */ if (unlikely(sync && !task_running(rq, p) && (task_cpu(p) != smp_processor_id()) && - (p->cpus_allowed & (1UL << smp_processor_id())))) { + test_bit(smp_processor_id(), + (unsigned long *)&p->cpus_allowed))) { set_task_cpu(p, smp_processor_id()); task_rq_unlock(rq, &flags); @@ -789,7 +790,7 @@ skip_queue: #define CAN_MIGRATE_TASK(p,rq,this_cpu) \ ((jiffies - (p)->sleep_timestamp > cache_decay_ticks) && \ !task_running(rq, p) && \ - ((p)->cpus_allowed & (1UL << (this_cpu)))) + (test_bit((this_cpu), (unsigned long *)&(p)->cpus_allowed))) curr = curr->prev; @@ -1542,7 +1543,7 @@ out_unlock: asmlinkage int sys_sched_setaffinity(pid_t pid, unsigned int len, unsigned long *user_mask_ptr) { - unsigned long new_mask; + cpumask_t online_mask, new_mask; int retval; task_t *p; @@ -1552,8 +1553,8 @@ asmlinkage int sys_sched_setaffinity(pid if (copy_from_user(&new_mask, user_mask_ptr, sizeof(new_mask))) return -EFAULT; - new_mask &= cpu_online_map; - if (!new_mask) + cpu_online_mask(&online_mask, &new_mask); + if (find_first_bit((unsigned long *)&online_mask, NR_CPUS) == NR_CPUS) return -EINVAL; read_lock(&tasklist_lock); @@ -1578,7 +1579,7 @@ asmlinkage int sys_sched_setaffinity(pid goto out_unlock; retval = 0; - set_cpus_allowed(p, new_mask); + set_cpus_allowed(p, &online_mask); out_unlock: put_task_struct(p); @@ -1595,7 +1596,7 @@ asmlinkage int sys_sched_getaffinity(pid unsigned long *user_mask_ptr) { unsigned int real_len; - unsigned long mask; + cpumask_t mask; int retval; task_t *p; @@ -1611,7 +1612,7 @@ asmlinkage int sys_sched_getaffinity(pid goto out_unlock; retval = 0; - mask = p->cpus_allowed & cpu_online_map; + cpu_online_mask(&mask, &p->cpus_allowed); out_unlock: read_unlock(&tasklist_lock); @@ -1914,7 +1915,7 @@ typedef struct { * task must not exit() & deallocate itself prematurely. The * call is not atomic; no spinlocks may be held. */ -void set_cpus_allowed(task_t *p, unsigned long new_mask) +void set_cpus_allowed(task_t *p, cpumask_t *new_mask) { unsigned long flags; migration_req_t req; @@ -1928,12 +1929,12 @@ void set_cpus_allowed(task_t *p, unsigne preempt_disable(); rq = task_rq_lock(p, &flags); - p->cpus_allowed = new_mask; + p->cpus_allowed = *new_mask; /* * Can the task run on the task's current CPU? If not then * migrate the thread off to a proper CPU. */ - if (new_mask & (1UL << task_cpu(p))) { + if (test_bit(task_cpu(p), (unsigned long *)new_mask)) { task_rq_unlock(rq, &flags); goto out; } @@ -1942,7 +1943,8 @@ void set_cpus_allowed(task_t *p, unsigne * it is sufficient to simply update the task's cpu field. */ if (!p->array && !task_running(rq, p)) { - set_task_cpu(p, __ffs(p->cpus_allowed)); + set_task_cpu(p, any_online_cpu(&p->cpus_allowed)); + BUG_ON(task_cpu(p) >= NR_CPUS); task_rq_unlock(rq, &flags); goto out; } @@ -2010,7 +2012,7 @@ static int migration_thread(void * data) spin_unlock_irqrestore(&rq->lock, flags); p = req->task; - cpu_dest = __ffs(p->cpus_allowed); + cpu_dest = any_online_cpu(&p->cpus_allowed); rq_dest = cpu_rq(cpu_dest); repeat: cpu_src = task_cpu(p); - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/