X-Git-Url: http://ftp.safe.ca/?p=safe%2Fjmp%2Flinux-2.6;a=blobdiff_plain;f=kernel%2Ftaskstats.c;h=11281d5792bd5b4d0eb7d5b86034ca4b2a4d256a;hp=4f3f0e48c8458fcbf1006ab930c0ebf411d11341;hb=b054b747a694927879c94dd11af54d04346aed7d;hpb=339bf98ffc6a8d8eb16fc532ac57ffbced2f8a68 diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 4f3f0e4..11281d5 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -20,9 +20,13 @@ #include #include #include -#include #include #include +#include +#include +#include +#include +#include #include #include @@ -32,9 +36,9 @@ */ #define TASKSTATS_CPUMASK_MAXLEN (100+6*NR_CPUS) -static DEFINE_PER_CPU(__u32, taskstats_seqnum) = { 0 }; +static DEFINE_PER_CPU(__u32, taskstats_seqnum); static int family_registered; -kmem_cache_t *taskstats_cache; +struct kmem_cache *taskstats_cache; static struct genl_family family = { .id = GENL_ID_GENERATE, @@ -43,13 +47,16 @@ static struct genl_family family = { .maxattr = TASKSTATS_CMD_ATTR_MAX, }; -static struct nla_policy taskstats_cmd_get_policy[TASKSTATS_CMD_ATTR_MAX+1] -__read_mostly = { +static const struct nla_policy taskstats_cmd_get_policy[TASKSTATS_CMD_ATTR_MAX+1] = { [TASKSTATS_CMD_ATTR_PID] = { .type = NLA_U32 }, [TASKSTATS_CMD_ATTR_TGID] = { .type = NLA_U32 }, [TASKSTATS_CMD_ATTR_REGISTER_CPUMASK] = { .type = NLA_STRING }, [TASKSTATS_CMD_ATTR_DEREGISTER_CPUMASK] = { .type = NLA_STRING },}; +static const struct nla_policy cgroupstats_cmd_get_policy[CGROUPSTATS_CMD_ATTR_MAX+1] = { + [CGROUPSTATS_CMD_ATTR_FD] = { .type = NLA_U32 }, +}; + struct listener { struct list_head list; pid_t pid; @@ -69,7 +76,7 @@ enum actions { }; static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, - void **replyp, size_t size) + size_t size) { struct sk_buff *skb; void *reply; @@ -77,7 +84,7 @@ static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, /* * If new attributes are added, please revisit this allocation */ - skb = nlmsg_new(genlmsg_total_size(size), GFP_KERNEL); + skb = genlmsg_new(size, GFP_KERNEL); if (!skb) return -ENOMEM; @@ -85,29 +92,24 @@ static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, int seq = get_cpu_var(taskstats_seqnum)++; put_cpu_var(taskstats_seqnum); - reply = genlmsg_put(skb, 0, seq, - family.id, 0, 0, - cmd, family.version); + reply = genlmsg_put(skb, 0, seq, &family, 0, cmd); } else - reply = genlmsg_put(skb, info->snd_pid, info->snd_seq, - family.id, 0, 0, - cmd, family.version); + reply = genlmsg_put_reply(skb, info, &family, 0, cmd); if (reply == NULL) { nlmsg_free(skb); return -EINVAL; } *skbp = skb; - *replyp = reply; return 0; } /* * Send taskstats data in @skb to listener with nl_pid @pid */ -static int send_reply(struct sk_buff *skb, pid_t pid) +static int send_reply(struct sk_buff *skb, struct genl_info *info) { - struct genlmsghdr *genlhdr = nlmsg_data((struct nlmsghdr *)skb->data); + struct genlmsghdr *genlhdr = nlmsg_data(nlmsg_hdr(skb)); void *reply = genlmsg_data(genlhdr); int rc; @@ -117,16 +119,16 @@ static int send_reply(struct sk_buff *skb, pid_t pid) return rc; } - return genlmsg_unicast(skb, pid); + return genlmsg_reply(skb, info); } /* * Send taskstats data in @skb to listeners registered for @cpu's exit data */ -static void send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) +static void send_cpu_listeners(struct sk_buff *skb, + struct listener_list *listeners) { - struct genlmsghdr *genlhdr = nlmsg_data((struct nlmsghdr *)skb->data); - struct listener_list *listeners; + struct genlmsghdr *genlhdr = nlmsg_data(nlmsg_hdr(skb)); struct listener *s, *tmp; struct sk_buff *skb_next, *skb_cur = skb; void *reply = genlmsg_data(genlhdr); @@ -139,7 +141,6 @@ static void send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) } rc = 0; - listeners = &per_cpu(listener_array, cpu); down_read(&listeners->sem); list_for_each_entry(s, &listeners->list, list) { skb_next = NULL; @@ -148,7 +149,7 @@ static void send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) if (!skb_next) break; } - rc = genlmsg_unicast(skb_cur, s->pid); + rc = genlmsg_unicast(&init_net, skb_cur, s->pid); if (rc == -ECONNREFUSED) { s->valid = 0; delcount++; @@ -181,7 +182,7 @@ static int fill_pid(pid_t pid, struct task_struct *tsk, if (!tsk) { rcu_read_lock(); - tsk = find_task_by_pid(pid); + tsk = find_task_by_vpid(pid); if (tsk) get_task_struct(tsk); rcu_read_unlock(); @@ -190,6 +191,7 @@ static int fill_pid(pid_t pid, struct task_struct *tsk, } else get_task_struct(tsk); + memset(stats, 0, sizeof(*stats)); /* * Each accounting subsystem adds calls to its functions to * fill in relevant parts of struct taskstsats as follows @@ -201,6 +203,8 @@ static int fill_pid(pid_t pid, struct task_struct *tsk, /* fill in basic acct fields */ stats->version = TASKSTATS_VERSION; + stats->nvcsw = tsk->nvcsw; + stats->nivcsw = tsk->nivcsw; bacct_add_tsk(stats, tsk); /* fill in extended acct fields */ @@ -225,13 +229,15 @@ static int fill_tgid(pid_t tgid, struct task_struct *first, */ rcu_read_lock(); if (!first) - first = find_task_by_pid(tgid); + first = find_task_by_vpid(tgid); if (!first || !lock_task_sighand(first, &flags)) goto out; if (first->signal->stats) memcpy(stats, first->signal->stats, sizeof(*stats)); + else + memset(stats, 0, sizeof(*stats)); tsk = first; do { @@ -245,6 +251,8 @@ static int fill_tgid(pid_t tgid, struct task_struct *first, */ delayacct_add_tsk(stats, tsk); + stats->nvcsw += tsk->nvcsw; + stats->nivcsw += tsk->nivcsw; } while_each_thread(first, tsk); unlock_task_sighand(first, &flags); @@ -254,7 +262,7 @@ out: stats->version = TASKSTATS_VERSION; /* - * Accounting subsytems can also add calls here to modify + * Accounting subsystems can also add calls here to modify * fields of taskstats. */ return rc; @@ -281,18 +289,17 @@ ret: return; } -static int add_del_listener(pid_t pid, cpumask_t *maskp, int isadd) +static int add_del_listener(pid_t pid, const struct cpumask *mask, int isadd) { struct listener_list *listeners; struct listener *s, *tmp; unsigned int cpu; - cpumask_t mask = *maskp; - if (!cpus_subset(mask, cpu_possible_map)) + if (!cpumask_subset(mask, cpu_possible_mask)) return -EINVAL; if (isadd == REGISTER) { - for_each_cpu_mask(cpu, mask) { + for_each_cpu(cpu, mask) { s = kmalloc_node(sizeof(struct listener), GFP_KERNEL, cpu_to_node(cpu)); if (!s) @@ -311,7 +318,7 @@ static int add_del_listener(pid_t pid, cpumask_t *maskp, int isadd) /* Deregister or cleanup */ cleanup: - for_each_cpu_mask(cpu, mask) { + for_each_cpu(cpu, mask) { listeners = &per_cpu(listener_array, cpu); down_write(&listeners->sem); list_for_each_entry_safe(s, tmp, &listeners->list, list) { @@ -326,7 +333,7 @@ cleanup: return 0; } -static int parse(struct nlattr *na, cpumask_t *mask) +static int parse(struct nlattr *na, struct cpumask *mask) { char *data; int len; @@ -343,32 +350,109 @@ static int parse(struct nlattr *na, cpumask_t *mask) if (!data) return -ENOMEM; nla_strlcpy(data, na, len); - ret = cpulist_parse(data, *mask); + ret = cpulist_parse(data, mask); kfree(data); return ret; } -static int taskstats_user_cmd(struct sk_buff *skb, struct genl_info *info) +static struct taskstats *mk_reply(struct sk_buff *skb, int type, u32 pid) +{ + struct nlattr *na, *ret; + int aggr; + + aggr = (type == TASKSTATS_TYPE_PID) + ? TASKSTATS_TYPE_AGGR_PID + : TASKSTATS_TYPE_AGGR_TGID; + + na = nla_nest_start(skb, aggr); + if (!na) + goto err; + if (nla_put(skb, type, sizeof(pid), &pid) < 0) + goto err; + ret = nla_reserve(skb, TASKSTATS_TYPE_STATS, sizeof(struct taskstats)); + if (!ret) + goto err; + nla_nest_end(skb, na); + + return nla_data(ret); +err: + return NULL; +} + +static int cgroupstats_user_cmd(struct sk_buff *skb, struct genl_info *info) { int rc = 0; struct sk_buff *rep_skb; - struct taskstats stats; - void *reply; - size_t size; + struct cgroupstats *stats; struct nlattr *na; - cpumask_t mask; + size_t size; + u32 fd; + struct file *file; + int fput_needed; + + na = info->attrs[CGROUPSTATS_CMD_ATTR_FD]; + if (!na) + return -EINVAL; + + fd = nla_get_u32(info->attrs[CGROUPSTATS_CMD_ATTR_FD]); + file = fget_light(fd, &fput_needed); + if (!file) + return 0; - rc = parse(info->attrs[TASKSTATS_CMD_ATTR_REGISTER_CPUMASK], &mask); + size = nla_total_size(sizeof(struct cgroupstats)); + + rc = prepare_reply(info, CGROUPSTATS_CMD_NEW, &rep_skb, + size); if (rc < 0) - return rc; - if (rc == 0) - return add_del_listener(info->snd_pid, &mask, REGISTER); + goto err; + + na = nla_reserve(rep_skb, CGROUPSTATS_TYPE_CGROUP_STATS, + sizeof(struct cgroupstats)); + stats = nla_data(na); + memset(stats, 0, sizeof(*stats)); + + rc = cgroupstats_build(stats, file->f_dentry); + if (rc < 0) { + nlmsg_free(rep_skb); + goto err; + } + + rc = send_reply(rep_skb, info); + +err: + fput_light(file, fput_needed); + return rc; +} + +static int taskstats_user_cmd(struct sk_buff *skb, struct genl_info *info) +{ + int rc; + struct sk_buff *rep_skb; + struct taskstats *stats; + size_t size; + cpumask_var_t mask; + + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) + return -ENOMEM; + + rc = parse(info->attrs[TASKSTATS_CMD_ATTR_REGISTER_CPUMASK], mask); + if (rc < 0) + goto free_return_rc; + if (rc == 0) { + rc = add_del_listener(info->snd_pid, mask, REGISTER); + goto free_return_rc; + } - rc = parse(info->attrs[TASKSTATS_CMD_ATTR_DEREGISTER_CPUMASK], &mask); + rc = parse(info->attrs[TASKSTATS_CMD_ATTR_DEREGISTER_CPUMASK], mask); if (rc < 0) + goto free_return_rc; + if (rc == 0) { + rc = add_del_listener(info->snd_pid, mask, DEREGISTER); +free_return_rc: + free_cpumask_var(mask); return rc; - if (rc == 0) - return add_del_listener(info->snd_pid, &mask, DEREGISTER); + } + free_cpumask_var(mask); /* * Size includes space for nested attributes @@ -376,83 +460,71 @@ static int taskstats_user_cmd(struct sk_buff *skb, struct genl_info *info) size = nla_total_size(sizeof(u32)) + nla_total_size(sizeof(struct taskstats)) + nla_total_size(0); - memset(&stats, 0, sizeof(stats)); - rc = prepare_reply(info, TASKSTATS_CMD_NEW, &rep_skb, &reply, size); + rc = prepare_reply(info, TASKSTATS_CMD_NEW, &rep_skb, size); if (rc < 0) return rc; + rc = -EINVAL; if (info->attrs[TASKSTATS_CMD_ATTR_PID]) { u32 pid = nla_get_u32(info->attrs[TASKSTATS_CMD_ATTR_PID]); - rc = fill_pid(pid, NULL, &stats); - if (rc < 0) + stats = mk_reply(rep_skb, TASKSTATS_TYPE_PID, pid); + if (!stats) goto err; - na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_PID); - NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_PID, pid); - NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS, - stats); + rc = fill_pid(pid, NULL, stats); + if (rc < 0) + goto err; } else if (info->attrs[TASKSTATS_CMD_ATTR_TGID]) { u32 tgid = nla_get_u32(info->attrs[TASKSTATS_CMD_ATTR_TGID]); - rc = fill_tgid(tgid, NULL, &stats); - if (rc < 0) + stats = mk_reply(rep_skb, TASKSTATS_TYPE_TGID, tgid); + if (!stats) goto err; - na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_TGID); - NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_TGID, tgid); - NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS, - stats); - } else { - rc = -EINVAL; + rc = fill_tgid(tgid, NULL, stats); + if (rc < 0) + goto err; + } else goto err; - } - nla_nest_end(rep_skb, na); - - return send_reply(rep_skb, info->snd_pid); - -nla_put_failure: - rc = genlmsg_cancel(rep_skb, reply); + return send_reply(rep_skb, info); err: nlmsg_free(rep_skb); return rc; } -void taskstats_exit_alloc(struct taskstats **ptidstats, unsigned int *mycpu) +static struct taskstats *taskstats_tgid_alloc(struct task_struct *tsk) { - struct listener_list *listeners; - struct taskstats *tmp; - /* - * This is the cpu on which the task is exiting currently and will - * be the one for which the exit event is sent, even if the cpu - * on which this function is running changes later. - */ - *mycpu = raw_smp_processor_id(); + struct signal_struct *sig = tsk->signal; + struct taskstats *stats; - *ptidstats = NULL; - tmp = kmem_cache_zalloc(taskstats_cache, SLAB_KERNEL); - if (!tmp) - return; + if (sig->stats || thread_group_empty(tsk)) + goto ret; - listeners = &per_cpu(listener_array, *mycpu); - down_read(&listeners->sem); - if (!list_empty(&listeners->list)) { - *ptidstats = tmp; - tmp = NULL; + /* No problem if kmem_cache_zalloc() fails */ + stats = kmem_cache_zalloc(taskstats_cache, GFP_KERNEL); + + spin_lock_irq(&tsk->sighand->siglock); + if (!sig->stats) { + sig->stats = stats; + stats = NULL; } - up_read(&listeners->sem); - kfree(tmp); + spin_unlock_irq(&tsk->sighand->siglock); + + if (stats) + kmem_cache_free(taskstats_cache, stats); +ret: + return sig->stats; } /* Send pid data out on exit */ -void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, - int group_dead, unsigned int mycpu) +void taskstats_exit(struct task_struct *tsk, int group_dead) { int rc; + struct listener_list *listeners; + struct taskstats *stats; struct sk_buff *rep_skb; - void *reply; size_t size; int is_thread_group; - struct nlattr *na; if (!family_registered) return; @@ -463,7 +535,7 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, size = nla_total_size(sizeof(u32)) + nla_total_size(sizeof(struct taskstats)) + nla_total_size(0); - is_thread_group = (tsk->signal->stats != NULL); + is_thread_group = !!taskstats_tgid_alloc(tsk); if (is_thread_group) { /* PID + STATS + TGID + STATS */ size = 2 * size; @@ -471,49 +543,39 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, fill_tgid_exit(tsk); } - if (!tidstats) + listeners = &__raw_get_cpu_var(listener_array); + if (list_empty(&listeners->list)) return; - rc = prepare_reply(NULL, TASKSTATS_CMD_NEW, &rep_skb, &reply, size); + rc = prepare_reply(NULL, TASKSTATS_CMD_NEW, &rep_skb, size); if (rc < 0) - goto ret; - - rc = fill_pid(tsk->pid, tsk, tidstats); - if (rc < 0) - goto err_skb; + return; - na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_PID); - NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_PID, (u32)tsk->pid); - NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS, - *tidstats); - nla_nest_end(rep_skb, na); + stats = mk_reply(rep_skb, TASKSTATS_TYPE_PID, tsk->pid); + if (!stats) + goto err; - if (!is_thread_group) - goto send; + rc = fill_pid(-1, tsk, stats); + if (rc < 0) + goto err; /* * Doesn't matter if tsk is the leader or the last group member leaving */ - if (!group_dead) + if (!is_thread_group || !group_dead) goto send; - na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_TGID); - NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_TGID, (u32)tsk->tgid); - /* No locking needed for tsk->signal->stats since group is dead */ - NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS, - *tsk->signal->stats); - nla_nest_end(rep_skb, na); + stats = mk_reply(rep_skb, TASKSTATS_TYPE_TGID, tsk->tgid); + if (!stats) + goto err; + + memcpy(stats, tsk->signal->stats, sizeof(*stats)); send: - send_cpu_listeners(rep_skb, mycpu); + send_cpu_listeners(rep_skb, listeners); return; - -nla_put_failure: - genlmsg_cancel(rep_skb, reply); -err_skb: +err: nlmsg_free(rep_skb); -ret: - return; } static struct genl_ops taskstats_ops = { @@ -522,14 +584,18 @@ static struct genl_ops taskstats_ops = { .policy = taskstats_cmd_get_policy, }; +static struct genl_ops cgroupstats_ops = { + .cmd = CGROUPSTATS_CMD_GET, + .doit = cgroupstats_user_cmd, + .policy = cgroupstats_cmd_get_policy, +}; + /* Needed early in initialization */ void __init taskstats_init_early(void) { unsigned int i; - taskstats_cache = kmem_cache_create("taskstats_cache", - sizeof(struct taskstats), - 0, SLAB_PANIC, NULL, NULL); + taskstats_cache = KMEM_CACHE(taskstats, SLAB_PANIC); for_each_possible_cpu(i) { INIT_LIST_HEAD(&(per_cpu(listener_array, i).list)); init_rwsem(&(per_cpu(listener_array, i).sem)); @@ -548,8 +614,15 @@ static int __init taskstats_init(void) if (rc < 0) goto err; + rc = genl_register_ops(&family, &cgroupstats_ops); + if (rc < 0) + goto err_cgroup_ops; + family_registered = 1; + printk("registered taskstats version %d\n", TASKSTATS_GENL_VERSION); return 0; +err_cgroup_ops: + genl_unregister_ops(&family, &taskstats_ops); err: genl_unregister_family(&family); return rc;