diff options
Diffstat (limited to 'kernel')
36 files changed, 1143 insertions, 1465 deletions
diff --git a/kernel/Makefile b/kernel/Makefile index 188c43223f52..1c9938addb9d 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -9,7 +9,7 @@ obj-y = sched.o fork.o exec_domain.o panic.o printk.o profile.o \ rcupdate.o extable.o params.o posix-timers.o \ kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \ hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \ - notifier.o ksysfs.o pm_qos_params.o + notifier.o ksysfs.o pm_qos_params.o sched_clock.o obj-$(CONFIG_SYSCTL_SYSCALL_CHECK) += sysctl_check.o obj-$(CONFIG_STACKTRACE) += stacktrace.o diff --git a/kernel/audit.c b/kernel/audit.c index b7d3709cc452..e092f1c0ce30 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -572,16 +572,17 @@ void audit_send_reply(int pid, int seq, int type, int done, int multi, skb = audit_make_reply(pid, seq, type, done, multi, payload, size); if (!skb) - return; + goto out; reply->pid = pid; reply->skb = skb; tsk = kthread_run(audit_send_reply_thread, reply, "audit_send_reply"); - if (IS_ERR(tsk)) { - kfree(reply); - kfree_skb(skb); - } + if (!IS_ERR(tsk)) + return; + kfree_skb(skb); +out: + kfree(reply); } /* @@ -737,7 +738,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) if (!audit_enabled && msg_type != AUDIT_USER_AVC) return 0; - err = audit_filter_user(&NETLINK_CB(skb), msg_type); + err = audit_filter_user(&NETLINK_CB(skb)); if (err == 1) { err = 0; if (msg_type == AUDIT_USER_TTY) { @@ -778,7 +779,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) } /* fallthrough */ case AUDIT_LIST: - err = audit_receive_filter(nlh->nlmsg_type, NETLINK_CB(skb).pid, + err = audit_receive_filter(msg_type, NETLINK_CB(skb).pid, uid, seq, data, nlmsg_len(nlh), loginuid, sessionid, sid); break; @@ -797,7 +798,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) } /* fallthrough */ case AUDIT_LIST_RULES: - err = audit_receive_filter(nlh->nlmsg_type, NETLINK_CB(skb).pid, + err = audit_receive_filter(msg_type, NETLINK_CB(skb).pid, uid, seq, data, nlmsg_len(nlh), loginuid, sessionid, sid); break; diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 9ef5e0aacc3c..f7921a2ecf16 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -172,10 +172,9 @@ static void insert_hash(struct audit_chunk *chunk) struct audit_chunk *audit_tree_lookup(const struct inode *inode) { struct list_head *list = chunk_hash(inode); - struct list_head *pos; + struct audit_chunk *p; - list_for_each_rcu(pos, list) { - struct audit_chunk *p = container_of(pos, struct audit_chunk, hash); + list_for_each_entry_rcu(p, list, hash) { if (p->watch.inode == inode) { get_inotify_watch(&p->watch); return p; diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c index 0e0bd27e6512..98c50cc671bb 100644 --- a/kernel/auditfilter.c +++ b/kernel/auditfilter.c @@ -1544,6 +1544,7 @@ static void audit_log_rule_change(uid_t loginuid, u32 sessionid, u32 sid, * @data: payload data * @datasz: size of payload data * @loginuid: loginuid of sender + * @sessionid: sessionid for netlink audit message * @sid: SE Linux Security ID of sender */ int audit_receive_filter(int type, int pid, int uid, int seq, void *data, @@ -1720,7 +1721,7 @@ static int audit_filter_user_rules(struct netlink_skb_parms *cb, return 1; } -int audit_filter_user(struct netlink_skb_parms *cb, int type) +int audit_filter_user(struct netlink_skb_parms *cb) { enum audit_state state = AUDIT_DISABLED; struct audit_entry *e; diff --git a/kernel/capability.c b/kernel/capability.c index 39e8193b41ea..901e0fdc3fff 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -53,11 +53,95 @@ static void warn_legacy_capability_use(void) } /* + * Version 2 capabilities worked fine, but the linux/capability.h file + * that accompanied their introduction encouraged their use without + * the necessary user-space source code changes. As such, we have + * created a version 3 with equivalent functionality to version 2, but + * with a header change to protect legacy source code from using + * version 2 when it wanted to use version 1. If your system has code + * that trips the following warning, it is using version 2 specific + * capabilities and may be doing so insecurely. + * + * The remedy is to either upgrade your version of libcap (to 2.10+, + * if the application is linked against it), or recompile your + * application with modern kernel headers and this warning will go + * away. + */ + +static void warn_deprecated_v2(void) +{ + static int warned; + + if (!warned) { + char name[sizeof(current->comm)]; + + printk(KERN_INFO "warning: `%s' uses deprecated v2" + " capabilities in a way that may be insecure.\n", + get_task_comm(name, current)); + warned = 1; + } +} + +/* + * Version check. Return the number of u32s in each capability flag + * array, or a negative value on error. + */ +static int cap_validate_magic(cap_user_header_t header, unsigned *tocopy) +{ + __u32 version; + + if (get_user(version, &header->version)) + return -EFAULT; + + switch (version) { + case _LINUX_CAPABILITY_VERSION_1: + warn_legacy_capability_use(); + *tocopy = _LINUX_CAPABILITY_U32S_1; + break; + case _LINUX_CAPABILITY_VERSION_2: + warn_deprecated_v2(); + /* + * fall through - v3 is otherwise equivalent to v2. + */ + case _LINUX_CAPABILITY_VERSION_3: + *tocopy = _LINUX_CAPABILITY_U32S_3; + break; + default: + if (put_user((u32)_KERNEL_CAPABILITY_VERSION, &header->version)) + return -EFAULT; + return -EINVAL; + } + + return 0; +} + +/* * For sys_getproccap() and sys_setproccap(), any of the three * capability set pointers may be NULL -- indicating that that set is * uninteresting and/or not to be changed. */ +/* + * Atomically modify the effective capabilities returning the original + * value. No permission check is performed here - it is assumed that the + * caller is permitted to set the desired effective capabilities. + */ +kernel_cap_t cap_set_effective(const kernel_cap_t pE_new) +{ + kernel_cap_t pE_old; + + spin_lock(&task_capability_lock); + + pE_old = current->cap_effective; + current->cap_effective = pE_new; + + spin_unlock(&task_capability_lock); + + return pE_old; +} + +EXPORT_SYMBOL(cap_set_effective); + /** * sys_capget - get the capabilities of a given process. * @header: pointer to struct that contains capability version and @@ -71,27 +155,13 @@ asmlinkage long sys_capget(cap_user_header_t header, cap_user_data_t dataptr) { int ret = 0; pid_t pid; - __u32 version; struct task_struct *target; unsigned tocopy; kernel_cap_t pE, pI, pP; - if (get_user(version, &header->version)) - return -EFAULT; - - switch (version) { - case _LINUX_CAPABILITY_VERSION_1: - warn_legacy_capability_use(); - tocopy = _LINUX_CAPABILITY_U32S_1; - break; - case _LINUX_CAPABILITY_VERSION_2: - tocopy = _LINUX_CAPABILITY_U32S_2; - break; - default: - if (put_user(_LINUX_CAPABILITY_VERSION, &header->version)) - return -EFAULT; - return -EINVAL; - } + ret = cap_validate_magic(header, &tocopy); + if (ret != 0) + return ret; if (get_user(pid, &header->pid)) return -EFAULT; @@ -118,7 +188,7 @@ out: spin_unlock(&task_capability_lock); if (!ret) { - struct __user_cap_data_struct kdata[_LINUX_CAPABILITY_U32S]; + struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S]; unsigned i; for (i = 0; i < tocopy; i++) { @@ -128,7 +198,7 @@ out: } /* - * Note, in the case, tocopy < _LINUX_CAPABILITY_U32S, + * Note, in the case, tocopy < _KERNEL_CAPABILITY_U32S, * we silently drop the upper capabilities here. This * has the effect of making older libcap * implementations implicitly drop upper capability @@ -240,30 +310,16 @@ static inline int cap_set_all(kernel_cap_t *effective, */ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data) { - struct __user_cap_data_struct kdata[_LINUX_CAPABILITY_U32S]; + struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S]; unsigned i, tocopy; kernel_cap_t inheritable, permitted, effective; - __u32 version; struct task_struct *target; int ret; pid_t pid; - if (get_user(version, &header->version)) - return -EFAULT; - - switch (version) { - case _LINUX_CAPABILITY_VERSION_1: - warn_legacy_capability_use(); - tocopy = _LINUX_CAPABILITY_U32S_1; - break; - case _LINUX_CAPABILITY_VERSION_2: - tocopy = _LINUX_CAPABILITY_U32S_2; - break; - default: - if (put_user(_LINUX_CAPABILITY_VERSION, &header->version)) - return -EFAULT; - return -EINVAL; - } + ret = cap_validate_magic(header, &tocopy); + if (ret != 0) + return ret; if (get_user(pid, &header->pid)) return -EFAULT; @@ -281,7 +337,7 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data) permitted.cap[i] = kdata[i].permitted; inheritable.cap[i] = kdata[i].inheritable; } - while (i < _LINUX_CAPABILITY_U32S) { + while (i < _KERNEL_CAPABILITY_U32S) { effective.cap[i] = 0; permitted.cap[i] = 0; inheritable.cap[i] = 0; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index fbc6fc8949b4..15ac0e1e4f4d 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -2903,7 +2903,7 @@ int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys) cg = tsk->cgroups; parent = task_cgroup(tsk, subsys->subsys_id); - snprintf(nodename, MAX_CGROUP_TYPE_NAMELEN, "node_%d", tsk->pid); + snprintf(nodename, MAX_CGROUP_TYPE_NAMELEN, "%d", tsk->pid); /* Pin the hierarchy */ atomic_inc(&parent->root->sb->s_active); diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 8da627d33804..9fceb97e989c 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -797,8 +797,10 @@ static int update_cpumask(struct cpuset *cs, char *buf) retval = cpulist_parse(buf, trialcs.cpus_allowed); if (retval < 0) return retval; + + if (!cpus_subset(trialcs.cpus_allowed, cpu_online_map)) + return -EINVAL; } - cpus_and(trialcs.cpus_allowed, trialcs.cpus_allowed, cpu_online_map); retval = validate_change(cs, &trialcs); if (retval < 0) return retval; @@ -932,9 +934,11 @@ static int update_nodemask(struct cpuset *cs, char *buf) retval = nodelist_parse(buf, trialcs.mems_allowed); if (retval < 0) goto done; + + if (!nodes_subset(trialcs.mems_allowed, + node_states[N_HIGH_MEMORY])) + return -EINVAL; } - nodes_and(trialcs.mems_allowed, trialcs.mems_allowed, - node_states[N_HIGH_MEMORY]); oldmem = cs->mems_allowed; if (nodes_equal(oldmem, trialcs.mems_allowed)) { retval = 0; /* Too easy - nothing to do */ @@ -1031,12 +1035,10 @@ int current_cpuset_is_being_rebound(void) return task_cs(current) == cpuset_being_rebound; } -static int update_relax_domain_level(struct cpuset *cs, char *buf) +static int update_relax_domain_level(struct cpuset *cs, s64 val) { - int val = simple_strtol(buf, NULL, 10); - - if (val < 0) - val = -1; + if (val < -1 || val >= SD_LV_MAX) + return -EINVAL; if (val != cs->relax_domain_level) { cs->relax_domain_level = val; @@ -1280,9 +1282,6 @@ static ssize_t cpuset_common_file_write(struct cgroup *cont, case FILE_MEMLIST: retval = update_nodemask(cs, buffer); break; - case FILE_SCHED_RELAX_DOMAIN_LEVEL: - retval = update_relax_domain_level(cs, buffer); - break; default: retval = -EINVAL; goto out2; @@ -1348,6 +1347,30 @@ static int cpuset_write_u64(struct cgroup *cgrp, struct cftype *cft, u64 val) return retval; } +static int cpuset_write_s64(struct cgroup *cgrp, struct cftype *cft, s64 val) +{ + int retval = 0; + struct cpuset *cs = cgroup_cs(cgrp); + cpuset_filetype_t type = cft->private; + + cgroup_lock(); + + if (cgroup_is_removed(cgrp)) { + cgroup_unlock(); + return -ENODEV; + } + switch (type) { + case FILE_SCHED_RELAX_DOMAIN_LEVEL: + retval = update_relax_domain_level(cs, val); + break; + default: + retval = -EINVAL; + break; + } + cgroup_unlock(); + return retval; +} + /* * These ascii lists should be read in a single call, by using a user * buffer large enough to hold the entire map. If read in smaller @@ -1406,9 +1429,6 @@ static ssize_t cpuset_common_file_read(struct cgroup *cont, case FILE_MEMLIST: s += cpuset_sprintf_memlist(s, cs); break; - case FILE_SCHED_RELAX_DOMAIN_LEVEL: - s += sprintf(s, "%d", cs->relax_domain_level); - break; default: retval = -EINVAL; goto out; @@ -1449,6 +1469,18 @@ static u64 cpuset_read_u64(struct cgroup *cont, struct cftype *cft) } } +static s64 cpuset_read_s64(struct cgroup *cont, struct cftype *cft) +{ + struct cpuset *cs = cgroup_cs(cont); + cpuset_filetype_t type = cft->private; + switch (type) { + case FILE_SCHED_RELAX_DOMAIN_LEVEL: + return cs->relax_domain_level; + default: + BUG(); + } +} + /* * for the common functions, 'private' gives the type of file @@ -1499,8 +1531,8 @@ static struct cftype files[] = { { .name = "sched_relax_domain_level", - .read_u64 = cpuset_read_u64, - .write_u64 = cpuset_write_u64, + .read_s64 = cpuset_read_s64, + .write_s64 = cpuset_write_s64, .private = FILE_SCHED_RELAX_DOMAIN_LEVEL, }, @@ -1858,6 +1890,12 @@ static void common_cpu_mem_hotplug_unplug(void) top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; scan_for_empty_cpusets(&top_cpuset); + /* + * Scheduler destroys domains on hotplug events. + * Rebuild them based on the current settings. + */ + rebuild_sched_domains(); + cgroup_unlock(); } diff --git a/kernel/exit.c b/kernel/exit.c index d3ad54677f9c..8f6185e69b69 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -19,6 +19,7 @@ #include <linux/acct.h> #include <linux/tsacct_kern.h> #include <linux/file.h> +#include <linux/fdtable.h> #include <linux/binfmts.h> #include <linux/nsproxy.h> #include <linux/pid_namespace.h> @@ -125,6 +126,12 @@ static void __exit_signal(struct task_struct *tsk) __unhash_process(tsk); + /* + * Do this under ->siglock, we can race with another thread + * doing sigqueue_free() if we have SIGQUEUE_PREALLOC signals. + */ + flush_sigqueue(&tsk->pending); + tsk->signal = NULL; tsk->sighand = NULL; spin_unlock(&sighand->siglock); @@ -132,7 +139,6 @@ static void __exit_signal(struct task_struct *tsk) __cleanup_sighand(sighand); clear_tsk_thread_flag(tsk,TIF_SIGPENDING); - flush_sigqueue(&tsk->pending); if (sig) { flush_sigqueue(&sig->shared_pending); taskstats_tgid_free(sig); diff --git a/kernel/fork.c b/kernel/fork.c index 2bb675af4de3..19908b26cf80 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -22,6 +22,7 @@ #include <linux/mempolicy.h> #include <linux/sem.h> #include <linux/file.h> +#include <linux/fdtable.h> #include <linux/key.h> #include <linux/binfmts.h> #include <linux/mman.h> @@ -659,136 +660,6 @@ static int copy_fs(unsigned long clone_flags, struct task_struct *tsk) return 0; } -static int count_open_files(struct fdtable *fdt) -{ - int size = fdt->max_fds; - int i; - - /* Find the last open fd */ - for (i = size/(8*sizeof(long)); i > 0; ) { - if (fdt->open_fds->fds_bits[--i]) - break; - } - i = (i+1) * 8 * sizeof(long); - return i; -} - -static struct files_struct *alloc_files(void) -{ - struct files_struct *newf; - struct fdtable *fdt; - - newf = kmem_cache_alloc(files_cachep, GFP_KERNEL); - if (!newf) - goto out; - - atomic_set(&newf->count, 1); - - spin_lock_init(&newf->file_lock); - newf->next_fd = 0; - fdt = &newf->fdtab; - fdt->max_fds = NR_OPEN_DEFAULT; - fdt->close_on_exec = (fd_set *)&newf->close_on_exec_init; - fdt->open_fds = (fd_set *)&newf->open_fds_init; - fdt->fd = &newf->fd_array[0]; - INIT_RCU_HEAD(&fdt->rcu); - fdt->next = NULL; - rcu_assign_pointer(newf->fdt, fdt); -out: - return newf; -} - -/* - * Allocate a new files structure and copy contents from the - * passed in files structure. - * errorp will be valid only when the returned files_struct is NULL. - */ -static struct files_struct *dup_fd(struct files_struct *oldf, int *errorp) -{ - struct files_struct *newf; - struct file **old_fds, **new_fds; - int open_files, size, i; - struct fdtable *old_fdt, *new_fdt; - - *errorp = -ENOMEM; - newf = alloc_files(); - if (!newf) - goto out; - - spin_lock(&oldf->file_lock); - old_fdt = files_fdtable(oldf); - new_fdt = files_fdtable(newf); - open_files = count_open_files(old_fdt); - - /* - * Check whether we need to allocate a larger fd array and fd set. - * Note: we're not a clone task, so the open count won't change. - */ - if (open_files > new_fdt->max_fds) { - new_fdt->max_fds = 0; - spin_unlock(&oldf->file_lock); - spin_lock(&newf->file_lock); - *errorp = expand_files(newf, open_files-1); - spin_unlock(&newf->file_lock); - if (*errorp < 0) - goto out_release; - new_fdt = files_fdtable(newf); - /* - * Reacquire the oldf lock and a pointer to its fd table - * who knows it may have a new bigger fd table. We need - * the latest pointer. - */ - spin_lock(&oldf->file_lock); - old_fdt = files_fdtable(oldf); - } - - old_fds = old_fdt->fd; - new_fds = new_fdt->fd; - - memcpy(new_fdt->open_fds->fds_bits, - old_fdt->open_fds->fds_bits, open_files/8); - memcpy(new_fdt->close_on_exec->fds_bits, - old_fdt->close_on_exec->fds_bits, open_files/8); - - for (i = open_files; i != 0; i--) { - struct file *f = *old_fds++; - if (f) { - get_file(f); - } else { - /* - * The fd may be claimed in the fd bitmap but not yet - * instantiated in the files array if a sibling thread - * is partway through open(). So make sure that this - * fd is available to the new process. - */ - FD_CLR(open_files - i, new_fdt->open_fds); - } - rcu_assign_pointer(*new_fds++, f); - } - spin_unlock(&oldf->file_lock); - - /* compute the remainder to be cleared */ - size = (new_fdt->max_fds - open_files) * sizeof(struct file *); - - /* This is long word aligned thus could use a optimized version */ - memset(new_fds, 0, size); - - if (new_fdt->max_fds > open_files) { - int left = (new_fdt->max_fds-open_files)/8; - int start = open_files / (8 * sizeof(unsigned long)); - - memset(&new_fdt->open_fds->fds_bits[start], 0, left); - memset(&new_fdt->close_on_exec->fds_bits[start], 0, left); - } - - return newf; - -out_release: - kmem_cache_free(files_cachep, newf); -out: - return NULL; -} - static int copy_files(unsigned long clone_flags, struct task_struct * tsk) { struct files_struct *oldf, *newf; diff --git a/kernel/futex.c b/kernel/futex.c index 98092c9817f4..7d1136e97c14 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -104,10 +104,6 @@ struct futex_q { /* Key which the futex is hashed on: */ union futex_key key; - /* For fd, sigio sent using these: */ - int fd; - struct file *filp; - /* Optional priority inheritance state: */ struct futex_pi_state *pi_state; struct task_struct *task; @@ -126,9 +122,6 @@ struct futex_hash_bucket { static struct futex_hash_bucket futex_queues[1<<FUTEX_HASHBITS]; -/* Futex-fs vfsmount entry: */ -static struct vfsmount *futex_mnt; - /* * Take mm->mmap_sem, when futex is shared */ @@ -610,8 +603,6 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, static void wake_futex(struct futex_q *q) { plist_del(&q->list, &q->list.plist); - if (q->filp) - send_sigio(&q->filp->f_owner, q->fd, POLL_IN); /* * The lock in wake_up_all() is a crucial memory barrier after the * plist_del() and also before assigning to q->lock_ptr. @@ -988,14 +979,10 @@ out: } /* The key must be already stored in q->key. */ -static inline struct futex_hash_bucket * -queue_lock(struct futex_q *q, int fd, struct file *filp) +static inline struct futex_hash_bucket *queue_lock(struct futex_q *q) { struct futex_hash_bucket *hb; - q->fd = fd; - q->filp = filp; - init_waitqueue_head(&q->waiters); get_futex_key_refs(&q->key); @@ -1006,7 +993,7 @@ queue_lock(struct futex_q *q, int fd, struct file *filp) return hb; } -static inline void __queue_me(struct futex_q *q, struct futex_hash_bucket *hb) +static inline void queue_me(struct futex_q *q, struct futex_hash_bucket *hb) { int prio; @@ -1041,15 +1028,6 @@ queue_unlock(struct futex_q *q, struct futex_hash_bucket *hb) * exactly once. They are called with the hashed spinlock held. */ -/* The key must be already stored in q->key. */ -static void queue_me(struct futex_q *q, int fd, struct file *filp) -{ - struct futex_hash_bucket *hb; - - hb = queue_lock(q, fd, filp); - __queue_me(q, hb); -} - /* Return 1 if we were still queued (ie. 0 means we were woken) */ static int unqueue_me(struct futex_q *q) { @@ -1118,21 +1096,64 @@ static void unqueue_me_pi(struct futex_q *q) * private futexes. */ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, - struct task_struct *newowner) + struct task_struct *newowner, + struct rw_semaphore *fshared) { u32 newtid = task_pid_vnr(newowner) | FUTEX_WAITERS; struct futex_pi_state *pi_state = q->pi_state; + struct task_struct *oldowner = pi_state->owner; u32 uval, curval, newval; - int ret; + int ret, attempt = 0; /* Owner died? */ + if (!pi_state->owner) + newtid |= FUTEX_OWNER_DIED; + + /* + * We are here either because we stole the rtmutex from the + * pending owner or we are the pending owner which failed to + * get the rtmutex. We have to replace the pending owner TID + * in the user space variable. This must be atomic as we have + * to preserve the owner died bit here. + * + * Note: We write the user space value _before_ changing the + * pi_state because we can fault here. Imagine swapped out + * pages or a fork, which was running right before we acquired + * mmap_sem, that marked all the anonymous memory readonly for + * cow. + * + * Modifying pi_state _before_ the user space value would + * leave the pi_state in an inconsistent state when we fault + * here, because we need to drop the hash bucket lock to + * handle the fault. This might be observed in the PID check + * in lookup_pi_state. + */ +retry: + if (get_futex_value_locked(&uval, uaddr)) + goto handle_fault; + + while (1) { + newval = (uval & FUTEX_OWNER_DIED) | newtid; + + curval = cmpxchg_futex_value_locked(uaddr, uval, newval); + + if (curval == -EFAULT) + goto handle_fault; + if (curval == uval) + break; + uval = curval; + } + + /* + * We fixed up user space. Now we need to fix the pi_state + * itself. + */ if (pi_state->owner != NULL) { spin_lock_irq(&pi_state->owner->pi_lock); WARN_ON(list_empty(&pi_state->list)); list_del_init(&pi_state->list); spin_unlock_irq(&pi_state->owner->pi_lock); - } else - newtid |= FUTEX_OWNER_DIED; + } pi_state->owner = newowner; @@ -1140,26 +1161,35 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, WARN_ON(!list_empty(&pi_state->list)); list_add(&pi_state->list, &newowner->pi_state_list); spin_unlock_irq(&newowner->pi_lock); + return 0; /* - * We own it, so we have to replace the pending owner - * TID. This must be atomic as we have preserve the - * owner died bit here. + * To handle the page fault we need to drop the hash bucket + * lock here. That gives the other task (either the pending + * owner itself or the task which stole the rtmutex) the + * chance to try the fixup of the pi_state. So once we are + * back from handling the fault we need to check the pi_state + * after reacquiring the hash bucket lock and before trying to + * do another fixup. When the fixup has been done already we + * simply return. */ - ret = get_futex_value_locked(&uval, uaddr); +handle_fault: + spin_unlock(q->lock_ptr); - while (!ret) { - newval = (uval & FUTEX_OWNER_DIED) | newtid; + ret = futex_handle_fault((unsigned long)uaddr, fshared, attempt++); - curval = cmpxchg_futex_value_locked(uaddr, uval, newval); + spin_lock(q->lock_ptr); - if (curval == -EFAULT) - ret = -EFAULT; - if (curval == uval) - break; - uval = curval; - } - return ret; + /* + * Check if someone else fixed it for us: + */ + if (pi_state->owner != oldowner) + return 0; + + if (ret) + return ret; + + goto retry; } /* @@ -1194,7 +1224,7 @@ static int futex_wait(u32 __user *uaddr, struct rw_semaphore *fshared, if (unlikely(ret != 0)) goto out_release_sem; - hb = queue_lock(&q, -1, NULL); + hb = queue_lock(&q); /* * Access the page AFTER the futex is queued. @@ -1238,7 +1268,7 @@ static int futex_wait(u32 __user *uaddr, struct rw_semaphore *fshared, goto out_unlock_release_sem; /* Only actually queue if *uaddr contained val. */ - __queue_me(&q, hb); + queue_me(&q, hb); /* * Now the futex is queued and we have checked the data, we @@ -1386,7 +1416,7 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, goto out_release_sem; retry_unlocked: - hb = queue_lock(&q, -1, NULL); + hb = queue_lock(&q); retry_locked: ret = lock_taken = 0; @@ -1499,7 +1529,7 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, /* * Only actually queue now that the atomic ops are done: */ - __queue_me(&q, hb); + queue_me(&q, hb); /* * Now the futex is queued and we have checked the data, we @@ -1529,7 +1559,7 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, * that case: */ if (q.pi_state->owner != curr) - ret = fixup_pi_state_owner(uaddr, &q, curr); + ret = fixup_pi_state_owner(uaddr, &q, curr, fshared); } else { /* * Catch the rare case, where the lock was released @@ -1561,7 +1591,8 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, int res; owner = rt_mutex_owner(&q.pi_state->pi_mutex); - res = fixup_pi_state_owner(uaddr, &q, owner); + res = fixup_pi_state_owner(uaddr, &q, owner, + fshared); /* propagate -EFAULT, if the fixup failed */ if (res) @@ -1746,121 +1777,6 @@ pi_faulted: return ret; } -static int futex_close(struct inode *inode, struct file *filp) -{ - struct futex_q *q = filp->private_data; - - unqueue_me(q); - kfree(q); - - return 0; -} - -/* This is one-shot: once it's gone off you need a new fd */ -static unsigned int futex_poll(struct file *filp, - struct poll_table_struct *wait) -{ - struct futex_q *q = filp->private_data; - int ret = 0; - - poll_wait(filp, &q->waiters, wait); - - /* - * plist_node_empty() is safe here without any lock. - * q->lock_ptr != 0 is not safe, because of ordering against wakeup. - */ - if (plist_node_empty(&q->list)) - ret = POLLIN | POLLRDNORM; - - return ret; -} - -static const struct file_operations futex_fops = { - .release = futex_close, - .poll = futex_poll, -}; - -/* - * Signal allows caller to avoid the race which would occur if they - * set the sigio stuff up afterwards. - */ -static int futex_fd(u32 __user *uaddr, int signal) -{ - struct futex_q *q; - struct file *filp; - int ret, err; - struct rw_semaphore *fshared; - static unsigned long printk_interval; - - if (printk_timed_ratelimit(&printk_interval, 60 * 60 * 1000)) { - printk(KERN_WARNING "Process `%s' used FUTEX_FD, which " - "will be removed from the kernel in June 2007\n", - current->comm); - } - - ret = -EINVAL; - if (!valid_signal(signal)) - goto out; - - ret = get_unused_fd(); - if (ret < 0) - goto out; - filp = get_empty_filp(); - if (!filp) { - put_unused_fd(ret); - ret = -ENFILE; - goto out; - } - filp->f_op = &futex_fops; - filp->f_path.mnt = mntget(futex_mnt); - filp->f_path.dentry = dget(futex_mnt->mnt_root); - filp->f_mapping = filp->f_path.dentry->d_inode->i_mapping; - - if (signal) { - err = __f_setown(filp, task_pid(current), PIDTYPE_PID, 1); - if (err < 0) { - goto error; - } - filp->f_owner.signum = signal; - } - - q = kmalloc(sizeof(*q), GFP_KERNEL); - if (!q) { - err = -ENOMEM; - goto error; - } - q->pi_state = NULL; - - fshared = ¤t->mm->mmap_sem; - down_read(fshared); - err = get_futex_key(uaddr, fshared, &q->key); - - if (unlikely(err != 0)) { - up_read(fshared); - kfree(q); - goto error; - } - - /* - * queue_me() must be called before releasing mmap_sem, because - * key->shared.inode needs to be referenced while holding it. - */ - filp->private_data = q; - - queue_me(q, ret, filp); - up_read(fshared); - - /* Now we map fd to filp, so userspace can access it */ - fd_install(ret, filp); -out: - return ret; -error: - put_unused_fd(ret); - put_filp(filp); - ret = err; - goto out; -} - /* * Support for robust futexes: the kernel cleans up held futexes at * thread exit time. @@ -2092,10 +2008,6 @@ long do_futex(u32 __user *uaddr, int op, u32 val, ktime_t *timeout, case FUTEX_WAKE_BITSET: ret = futex_wake(uaddr, fshared, val, val3); break; - case FUTEX_FD: - /* non-zero val means F_SETOWN(getpid()) & F_SETSIG(val) */ - ret = futex_fd(uaddr, val); - break; case FUTEX_REQUEUE: ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, NULL); break; @@ -2156,19 +2068,6 @@ asmlinkage long sys_futex(u32 __user *uaddr, int op, u32 val, return do_futex(uaddr, op, val, tp, uaddr2, val2, val3); } -static int futexfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, - struct vfsmount *mnt) -{ - return get_sb_pseudo(fs_type, "futex", NULL, FUTEXFS_SUPER_MAGIC, mnt); -} - -static struct file_system_type futex_fs_type = { - .name = "futexfs", - .get_sb = futexfs_get_sb, - .kill_sb = kill_anon_super, -}; - static int __init futex_init(void) { u32 curval; @@ -2193,16 +2092,6 @@ static int __init futex_init(void) spin_lock_init(&futex_queues[i].lock); } - i = register_filesystem(&futex_fs_type); - if (i) - return i; - - futex_mnt = kern_mount(&futex_fs_type); - if (IS_ERR(futex_mnt)) { - unregister_filesystem(&futex_fs_type); - return PTR_ERR(futex_mnt); - } - return 0; } __initcall(futex_init); diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index 9af1d6a8095e..ab80515008f4 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -154,15 +154,6 @@ static void hrtimer_get_softirq_time(struct hrtimer_cpu_base *base) } /* - * Helper function to check, whether the timer is running the callback - * function - */ -static inline int hrtimer_callback_running(struct hrtimer *timer) -{ - return timer->state & HRTIMER_STATE_CALLBACK; -} - -/* * Functions and macros which are different for UP/SMP systems are kept in a * single place */ @@ -1012,10 +1003,18 @@ hrtimer_start(struct hrtimer *timer, ktime_t tim, const enum hrtimer_mode mode) */ raise = timer->state == HRTIMER_STATE_PENDING; + /* + * We use preempt_disable to prevent this task from migrating after + * setting up the softirq and raising it. Otherwise, if me migrate + * we will raise the softirq on the wrong CPU. + */ + preempt_disable(); + unlock_hrtimer_base(timer, &flags); if (raise) hrtimer_raise_softirq(); + preempt_enable(); return ret; } diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 46e4ad1723f0..46d6611a33bb 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -150,6 +150,26 @@ void disable_irq(unsigned int irq) } EXPORT_SYMBOL(disable_irq); +static void __enable_irq(struct irq_desc *desc, unsigned int irq) +{ + switch (desc->depth) { + case 0: + printk(KERN_WARNING "Unbalanced enable for IRQ %d\n", irq); + WARN_ON(1); + break; + case 1: { + unsigned int status = desc->status & ~IRQ_DISABLED; + + /* Prevent probing on this irq: */ + desc->status = status | IRQ_NOPROBE; + check_irq_resend(desc, irq); + /* fall-through */ + } + default: + desc->depth--; + } +} + /** * enable_irq - enable handling of an irq * @irq: Interrupt to enable @@ -169,22 +189,7 @@ void enable_irq(unsigned int irq) return; spin_lock_irqsave(&desc->lock, flags); - switch (desc->depth) { - case 0: - printk(KERN_WARNING "Unbalanced enable for IRQ %d\n", irq); - WARN_ON(1); - break; - case 1: { - unsigned int status = desc->status & ~IRQ_DISABLED; - - /* Prevent probing on this irq: */ - desc->status = status | IRQ_NOPROBE; - check_irq_resend(desc, irq); - /* fall-through */ - } - default: - desc->depth--; - } + __enable_irq(desc, irq); spin_unlock_irqrestore(&desc->lock, flags); } EXPORT_SYMBOL(enable_irq); @@ -365,7 +370,7 @@ int setup_irq(unsigned int irq, struct irqaction *new) compat_irq_chip_set_default_handler(desc); desc->status &= ~(IRQ_AUTODETECT | IRQ_WAITING | - IRQ_INPROGRESS); + IRQ_INPROGRESS | IRQ_SPURIOUS_DISABLED); if (!(desc->status & IRQ_NOAUTOEN)) { desc->depth = 0; @@ -381,6 +386,16 @@ int setup_irq(unsigned int irq, struct irqaction *new) /* Reset broken irq detection when installing new handler */ desc->irq_count = 0; desc->irqs_unhandled = 0; + + /* + * Check whether we disabled the irq via the spurious handler + * before. Reenable it and give it another chance. + */ + if (shared && (desc->status & IRQ_SPURIOUS_DISABLED)) { + desc->status &= ~IRQ_SPURIOUS_DISABLED; + __enable_irq(desc, irq); + } + spin_unlock_irqrestore(&desc->lock, flags); new->irq = irq; diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index 088dabbf2d6a..c66d3f10e853 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -209,8 +209,8 @@ void note_interrupt(unsigned int irq, struct irq_desc *desc, * Now kill the IRQ */ printk(KERN_EMERG "Disabling IRQ #%d\n", irq); - desc->status |= IRQ_DISABLED; - desc->depth = 1; + desc->status |= IRQ_DISABLED | IRQ_SPURIOUS_DISABLED; + desc->depth++; desc->chip->disable(irq); } desc->irqs_unhandled = 0; diff --git a/kernel/kgdb.c b/kernel/kgdb.c index 1bd0ec1c80b2..3ec23c3ec97f 100644 --- a/kernel/kgdb.c +++ b/kernel/kgdb.c @@ -52,6 +52,7 @@ #include <asm/byteorder.h> #include <asm/atomic.h> #include <asm/system.h> +#include <asm/unaligned.h> static int kgdb_break_asap; @@ -61,7 +62,7 @@ struct kgdb_state { int err_code; int cpu; int pass_exception; - long threadid; + unsigned long threadid; long kgdb_usethreadid; struct pt_regs *linux_regs; }; @@ -146,7 +147,7 @@ atomic_t kgdb_cpu_doing_single_step = ATOMIC_INIT(-1); * the other CPUs might interfere with your debugging context, so * use this with care: */ -int kgdb_do_roundup = 1; +static int kgdb_do_roundup = 1; static int __init opt_nokgdbroundup(char *str) { @@ -227,8 +228,6 @@ void __weak kgdb_disable_hw_debug(struct pt_regs *regs) * GDB remote protocol parser: */ -static const char hexchars[] = "0123456789abcdef"; - static int hex(char ch) { if ((ch >= 'a') && (ch <= 'f')) @@ -316,8 +315,8 @@ static void put_packet(char *buffer) } kgdb_io_ops->write_char('#'); - kgdb_io_ops->write_char(hexchars[checksum >> 4]); - kgdb_io_ops->write_char(hexchars[checksum & 0xf]); + kgdb_io_ops->write_char(hex_asc_hi(checksum)); + kgdb_io_ops->write_char(hex_asc_lo(checksum)); if (kgdb_io_ops->flush) kgdb_io_ops->flush(); @@ -346,14 +345,6 @@ static void put_packet(char *buffer) } } -static char *pack_hex_byte(char *pkt, u8 byte) -{ - *pkt++ = hexchars[byte >> 4]; - *pkt++ = hexchars[byte & 0xf]; - - return pkt; -} - /* * Convert the memory pointed to by mem into hex, placing result in buf. * Return a pointer to the last char put in buf (null). May return an error. @@ -438,7 +429,7 @@ int kgdb_hex2mem(char *buf, char *mem, int count) * While we find nice hex chars, build a long_val. * Return number of chars processed. */ -int kgdb_hex2long(char **ptr, long *long_val) +int kgdb_hex2long(char **ptr, unsigned long *long_val) { int hex_val; int num = 0; @@ -486,8 +477,8 @@ static void error_packet(char *pkt, int error) { error = -error; pkt[0] = 'E'; - pkt[1] = hexchars[(error / 10)]; - pkt[2] = hexchars[(error % 10)]; + pkt[1] = hex_asc[(error / 10)]; + pkt[2] = hex_asc[(error % 10)]; pkt[3] = '\0'; } @@ -518,10 +509,7 @@ static void int_to_threadref(unsigned char *id, int value) scan = (unsigned char *)id; while (i--) *scan++ = 0; - *scan++ = (value >> 24) & 0xff; - *scan++ = (value >> 16) & 0xff; - *scan++ = (value >> 8) & 0xff; - *scan++ = (value & 0xff); + put_unaligned_be32(value, scan); } static struct task_struct *getthread(struct pt_regs *regs, int tid) @@ -709,7 +697,7 @@ int kgdb_isremovedbreak(unsigned long addr) return 0; } -int remove_all_break(void) +static int remove_all_break(void) { unsigned long addr; int error; @@ -1511,7 +1499,8 @@ int kgdb_nmicallback(int cpu, void *regs) return 1; } -void kgdb_console_write(struct console *co, const char *s, unsigned count) +static void kgdb_console_write(struct console *co, const char *s, + unsigned count) { unsigned long flags; diff --git a/kernel/kmod.c b/kernel/kmod.c index e2764047ec03..8df97d3dfda8 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -27,6 +27,7 @@ #include <linux/mnt_namespace.h> #include <linux/completion.h> #include <linux/file.h> +#include <linux/fdtable.h> #include <linux/workqueue.h> #include <linux/security.h> #include <linux/mount.h> diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 1e0250cb9486..1485ca8d0e00 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -79,7 +79,7 @@ static DEFINE_PER_CPU(struct kprobe *, kprobe_instance) = NULL; * * For such cases, we now have a blacklist */ -struct kprobe_blackpoint kprobe_blacklist[] = { +static struct kprobe_blackpoint kprobe_blacklist[] = { {"preempt_schedule",}, {NULL} /* Terminator */ }; @@ -699,8 +699,9 @@ static int __register_kprobes(struct kprobe **kps, int num, return -EINVAL; for (i = 0; i < num; i++) { ret = __register_kprobe(kps[i], called_from); - if (ret < 0 && i > 0) { - unregister_kprobes(kps, i); + if (ret < 0) { + if (i > 0) + unregister_kprobes(kps, i); break; } } @@ -776,8 +777,9 @@ static int __register_jprobes(struct jprobe **jps, int num, jp->kp.break_handler = longjmp_break_handler; ret = __register_kprobe(&jp->kp, called_from); } - if (ret < 0 && i > 0) { - unregister_jprobes(jps, i); + if (ret < 0) { + if (i > 0) + unregister_jprobes(jps, i); break; } } @@ -920,8 +922,9 @@ static int __register_kretprobes(struct kretprobe **rps, int num, return -EINVAL; for (i = 0; i < num; i++) { ret = __register_kretprobe(rps[i], called_from); - if (ret < 0 && i > 0) { - unregister_kretprobes(rps, i); + if (ret < 0) { + if (i > 0) + unregister_kretprobes(rps, i); break; } } diff --git a/kernel/module.c b/kernel/module.c index 8674a390a2e8..5f80478b746d 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -890,6 +890,19 @@ static struct module_attribute *modinfo_attrs[] = { static const char vermagic[] = VERMAGIC_STRING; +static int try_to_force_load(struct module *mod, const char *symname) +{ +#ifdef CONFIG_MODULE_FORCE_LOAD + if (!(tainted & TAINT_FORCED_MODULE)) + printk("%s: no version for \"%s\" found: kernel tainted.\n", + mod->name, symname); + add_taint_module(mod, TAINT_FORCED_MODULE); + return 0; +#else + return -ENOEXEC; +#endif +} + #ifdef CONFIG_MODVERSIONS static int check_version(Elf_Shdr *sechdrs, unsigned int versindex, @@ -904,6 +917,10 @@ static int check_version(Elf_Shdr *sechdrs, if (!crc) return 1; + /* No versions at all? modprobe --force does this. */ + if (versindex == 0) + return try_to_force_load(mod, symname) == 0; + versions = (void *) sechdrs[versindex].sh_addr; num_versions = sechdrs[versindex].sh_size / sizeof(struct modversion_info); @@ -914,18 +931,19 @@ static int check_version(Elf_Shdr *sechdrs, if (versions[i].crc == *crc) return 1; - printk("%s: disagrees about version of symbol %s\n", - mod->name, symname); DEBUGP("Found checksum %lX vs module %lX\n", *crc, versions[i].crc); - return 0; + goto bad_version; } - /* Not in module's version table. OK, but that taints the kernel. */ - if (!(tainted & TAINT_FORCED_MODULE)) - printk("%s: no version for \"%s\" found: kernel tainted.\n", - mod->name, symname); - add_taint_module(mod, TAINT_FORCED_MODULE); - return 1; + + printk(KERN_WARNING "%s: no symbol version for %s\n", + mod->name, symname); + return 0; + +bad_version: + printk("%s: disagrees about version of symbol %s\n", + mod->name, symname); + return 0; } static inline int check_modstruct_version(Elf_Shdr *sechdrs, @@ -939,11 +957,14 @@ static inline int check_modstruct_version(Elf_Shdr *sechdrs, return check_version(sechdrs, versindex, "struct_module", mod, crc); } -/* First part is kernel version, which we ignore. */ -static inline int same_magic(const char *amagic, const char *bmagic) +/* First part is kernel version, which we ignore if module has crcs. */ +static inline int same_magic(const char *amagic, const char *bmagic, + bool has_crcs) { - amagic += strcspn(amagic, " "); - bmagic += strcspn(bmagic, " "); + if (has_crcs) { + amagic += strcspn(amagic, " "); + bmagic += strcspn(bmagic, " "); + } return strcmp(amagic, bmagic) == 0; } #else @@ -963,7 +984,8 @@ static inline int check_modstruct_version(Elf_Shdr *sechdrs, return 1; } -static inline int same_magic(const char *amagic, const char *bmagic) +static inline int same_magic(const char *amagic, const char *bmagic, + bool has_crcs) { return strcmp(amagic, bmagic) == 0; } @@ -1315,7 +1337,19 @@ out_unreg: kobject_put(&mod->mkobj.kobj); return err; } -#endif + +static void mod_sysfs_fini(struct module *mod) +{ + kobject_put(&mod->mkobj.kobj); +} + +#else /* CONFIG_SYSFS */ + +static void mod_sysfs_fini(struct module *mod) +{ +} + +#endif /* CONFIG_SYSFS */ static void mod_kobject_remove(struct module *mod) { @@ -1323,7 +1357,7 @@ static void mod_kobject_remove(struct module *mod) module_param_sysfs_remove(mod); kobject_put(mod->mkobj.drivers_dir); kobject_put(mod->holders_dir); - kobject_put(&mod->mkobj.kobj); + mod_sysfs_fini(mod); } /* @@ -1758,7 +1792,7 @@ static struct module *load_module(void __user *umod, /* Sanity checks against insmoding binaries or wrong arch, weird elf version */ - if (memcmp(hdr->e_ident, ELFMAG, 4) != 0 + if (memcmp(hdr->e_ident, ELFMAG, SELFMAG) != 0 || hdr->e_type != ET_REL || !elf_check_arch(hdr) || hdr->e_shentsize != sizeof(*sechdrs)) { @@ -1853,10 +1887,10 @@ static struct module *load_module(void __user *umod, modmagic = get_modinfo(sechdrs, infoindex, "vermagic"); /* This is allowed: modprobe --force will invalidate it. */ if (!modmagic) { - add_taint_module(mod, TAINT_FORCED_MODULE); - printk(KERN_WARNING "%s: no version magic, tainting kernel.\n", - mod->name); - } else if (!same_magic(modmagic, vermagic)) { + err = try_to_force_load(mod, "magic"); + if (err) + goto free_hdr; + } else if (!same_magic(modmagic, vermagic, versindex)) { printk(KERN_ERR "%s: version magic '%s' should be '%s'\n", mod->name, modmagic, vermagic); err = -ENOEXEC; @@ -2006,9 +2040,10 @@ static struct module *load_module(void __user *umod, (mod->num_gpl_future_syms && !gplfuturecrcindex) || (mod->num_unused_syms && !unusedcrcindex) || (mod->num_unused_gpl_syms && !unusedgplcrcindex)) { - printk(KERN_WARNING "%s: No versions for exported symbols." - " Tainting kernel.\n", mod->name); - add_taint_module(mod, TAINT_FORCED_MODULE); + printk(KERN_WARNING "%s: No versions for exported symbols.\n", mod->name); + err = try_to_force_load(mod, "nocrc"); + if (err) + goto cleanup; } #endif markersindex = find_sec(hdr, sechdrs, secstrings, "__markers"); diff --git a/kernel/printk.c b/kernel/printk.c index 8fb01c32aa3b..e2129e83fd75 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -666,7 +666,7 @@ static int acquire_console_semaphore_for_printk(unsigned int cpu) return retval; } -const char printk_recursion_bug_msg [] = +static const char printk_recursion_bug_msg [] = KERN_CRIT "BUG: recent printk recursion!\n"; static int printk_recursion_bug; diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c index f4ffbd0f306f..a38895a5b8e2 100644 --- a/kernel/rcuclassic.c +++ b/kernel/rcuclassic.c @@ -89,8 +89,22 @@ static void force_quiescent_state(struct rcu_data *rdp, /* * Don't send IPI to itself. With irqs disabled, * rdp->cpu is the current cpu. + * + * cpu_online_map is updated by the _cpu_down() + * using stop_machine_run(). Since we're in irqs disabled + * section, stop_machine_run() is not exectuting, hence + * the cpu_online_map is stable. + * + * However, a cpu might have been offlined _just_ before + * we disabled irqs while entering here. + * And rcu subsystem might not yet have handled the CPU_DEAD + * notification, leading to the offlined cpu's bit + * being set in the rcp->cpumask. + * + * Hence cpumask = (rcp->cpumask & cpu_online_map) to prevent + * sending smp_reschedule() to an offlined CPU. */ - cpumask = rcp->cpumask; + cpus_and(cpumask, rcp->cpumask, cpu_online_map); cpu_clear(rdp->cpu, cpumask); for_each_cpu_mask(cpu, cpumask) smp_send_reschedule(cpu); diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c index e1cdf196a515..41d275a81df5 100644 --- a/kernel/rcupreempt.c +++ b/kernel/rcupreempt.c @@ -217,8 +217,6 @@ long rcu_batches_completed(void) } EXPORT_SYMBOL_GPL(rcu_batches_completed); -EXPORT_SYMBOL_GPL(rcu_batches_completed_bh); - void __rcu_read_lock(void) { int idx; @@ -927,26 +925,22 @@ void rcu_offline_cpu(int cpu) spin_unlock_irqrestore(&rdp->lock, flags); } -void __devinit rcu_online_cpu(int cpu) -{ - unsigned long flags; - - spin_lock_irqsave(&rcu_ctrlblk.fliplock, flags); - cpu_set(cpu, rcu_cpu_online_map); - spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags); -} - #else /* #ifdef CONFIG_HOTPLUG_CPU */ void rcu_offline_cpu(int cpu) { } -void __devinit rcu_online_cpu(int cpu) +#endif /* #else #ifdef CONFIG_HOTPLUG_CPU */ + +void __cpuinit rcu_online_cpu(int cpu) { -} + unsigned long flags; -#endif /* #else #ifdef CONFIG_HOTPLUG_CPU */ + spin_lock_irqsave(&rcu_ctrlblk.fliplock, flags); + cpu_set(cpu, rcu_cpu_online_map); + spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags); +} static void rcu_process_callbacks(struct softirq_action *unused) { diff --git a/kernel/sched.c b/kernel/sched.c index 34bcc5bc120e..4e2f60335656 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -75,16 +75,6 @@ #include <asm/irq_regs.h> /* - * Scheduler clock - returns current time in nanosec units. - * This is default implementation. - * Architectures and sub-architectures can override this. - */ -unsigned long long __attribute__((weak)) sched_clock(void) -{ - return (unsigned long long)jiffies * (NSEC_PER_SEC / HZ); -} - -/* * Convert user-nice values [ -20 ... 0 ... 19 ] * to static priority [ MAX_RT_PRIO..MAX_PRIO-1 ], * and back. @@ -146,7 +136,7 @@ static inline void sg_inc_cpu_power(struct sched_group *sg, u32 val) static inline int rt_policy(int policy) { - if (unlikely(policy == SCHED_FIFO) || unlikely(policy == SCHED_RR)) + if (unlikely(policy == SCHED_FIFO || policy == SCHED_RR)) return 1; return 0; } @@ -242,6 +232,12 @@ static void destroy_rt_bandwidth(struct rt_bandwidth *rt_b) } #endif +/* + * sched_domains_mutex serializes calls to arch_init_sched_domains, + * detach_destroy_domains and partition_sched_domains. + */ +static DEFINE_MUTEX(sched_domains_mutex); + #ifdef CONFIG_GROUP_SCHED #include <linux/cgroup.h> @@ -308,9 +304,6 @@ static DEFINE_PER_CPU(struct rt_rq, init_rt_rq) ____cacheline_aligned_in_smp; */ static DEFINE_SPINLOCK(task_group_lock); -/* doms_cur_mutex serializes access to doms_cur[] array */ -static DEFINE_MUTEX(doms_cur_mutex); - #ifdef CONFIG_FAIR_GROUP_SCHED #ifdef CONFIG_USER_SCHED # define INIT_TASK_GROUP_LOAD (2*NICE_0_LOAD) @@ -318,7 +311,16 @@ static DEFINE_MUTEX(doms_cur_mutex); # define INIT_TASK_GROUP_LOAD NICE_0_LOAD #endif +/* + * A weight of 0 or 1 can cause arithmetics problems. + * A weight of a cfs_rq is the sum of weights of which entities + * are queued on this cfs_rq, so a weight of a entity should not be + * too large, so as the shares value of a task group. + * (The default weight is 1024 - so there's no practical + * limitation from this.) + */ #define MIN_SHARES 2 +#define MAX_SHARES (1UL << 18) static int init_task_group_load = INIT_TASK_GROUP_LOAD; #endif @@ -358,21 +360,9 @@ static inline void set_task_rq(struct task_struct *p, unsigned int cpu) #endif } -static inline void lock_doms_cur(void) -{ - mutex_lock(&doms_cur_mutex); -} - -static inline void unlock_doms_cur(void) -{ - mutex_unlock(&doms_cur_mutex); -} - #else static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { } -static inline void lock_doms_cur(void) { } -static inline void unlock_doms_cur(void) { } #endif /* CONFIG_GROUP_SCHED */ @@ -411,43 +401,6 @@ struct cfs_rq { */ struct list_head leaf_cfs_rq_list; struct task_group *tg; /* group that "owns" this runqueue */ - -#ifdef CONFIG_SMP - unsigned long task_weight; - unsigned long shares; - /* - * We need space to build a sched_domain wide view of the full task - * group tree, in order to avoid depending on dynamic memory allocation - * during the load balancing we place this in the per cpu task group - * hierarchy. This limits the load balancing to one instance per cpu, - * but more should not be needed anyway. - */ - struct aggregate_struct { - /* - * load = weight(cpus) * f(tg) - * - * Where f(tg) is the recursive weight fraction assigned to - * this group. - */ - unsigned long load; - - /* - * part of the group weight distributed to this span. - */ - unsigned long shares; - - /* - * The sum of all runqueue weights within this span. - */ - unsigned long rq_weight; - - /* - * Weight contributed by tasks; this is the part we can - * influence by moving tasks around. - */ - unsigned long task_weight; - } aggregate; -#endif #endif }; @@ -560,13 +513,7 @@ struct rq { unsigned long next_balance; struct mm_struct *prev_mm; - u64 clock, prev_clock_raw; - s64 clock_max_delta; - - unsigned int clock_warps, clock_overflows, clock_underflows; - u64 idle_clock; - unsigned int clock_deep_idle_events; - u64 tick_timestamp; + u64 clock; atomic_t nr_iowait; @@ -631,82 +578,6 @@ static inline int cpu_of(struct rq *rq) #endif } -#ifdef CONFIG_NO_HZ -static inline bool nohz_on(int cpu) -{ - return tick_get_tick_sched(cpu)->nohz_mode != NOHZ_MODE_INACTIVE; -} - -static inline u64 max_skipped_ticks(struct rq *rq) -{ - return nohz_on(cpu_of(rq)) ? jiffies - rq->last_tick_seen + 2 : 1; -} - -static inline void update_last_tick_seen(struct rq *rq) -{ - rq->last_tick_seen = jiffies; -} -#else -static inline u64 max_skipped_ticks(struct rq *rq) -{ - return 1; -} - -static inline void update_last_tick_seen(struct rq *rq) -{ -} -#endif - -/* - * Update the per-runqueue clock, as finegrained as the platform can give - * us, but without assuming monotonicity, etc.: - */ -static void __update_rq_clock(struct rq *rq) -{ - u64 prev_raw = rq->prev_clock_raw; - u64 now = sched_clock(); - s64 delta = now - prev_raw; - u64 clock = rq->clock; - -#ifdef CONFIG_SCHED_DEBUG - WARN_ON_ONCE(cpu_of(rq) != smp_processor_id()); -#endif - /* - * Protect against sched_clock() occasionally going backwards: - */ - if (unlikely(delta < 0)) { - clock++; - rq->clock_warps++; - } else { - /* - * Catch too large forward jumps too: - */ - u64 max_jump = max_skipped_ticks(rq) * TICK_NSEC; - u64 max_time = rq->tick_timestamp + max_jump; - - if (unlikely(clock + delta > max_time)) { - if (clock < max_time) - clock = max_time; - else - clock++; - rq->clock_overflows++; - } else { - if (unlikely(delta > rq->clock_max_delta)) - rq->clock_max_delta = delta; - clock += delta; - } - } - - rq->prev_clock_raw = now; - rq->clock = clock; -} - -static void update_rq_clock(struct rq *rq) -{ - if (likely(smp_processor_id() == cpu_of(rq))) - __update_rq_clock(rq); -} - /* * The domain tree (rq->sd) is protected by RCU's quiescent state transition. * See detach_destroy_domains: synchronize_sched for details. @@ -722,6 +593,11 @@ static void update_rq_clock(struct rq *rq) #define task_rq(p) cpu_rq(task_cpu(p)) #define cpu_curr(cpu) (cpu_rq(cpu)->curr) +static inline void update_rq_clock(struct rq *rq) +{ + rq->clock = sched_clock_cpu(cpu_of(rq)); +} + /* * Tunables that become constants when CONFIG_SCHED_DEBUG is off: */ @@ -757,14 +633,14 @@ const_debug unsigned int sysctl_sched_features = #define SCHED_FEAT(name, enabled) \ #name , -__read_mostly char *sched_feat_names[] = { +static __read_mostly char *sched_feat_names[] = { #include "sched_features.h" NULL }; #undef SCHED_FEAT -int sched_feat_open(struct inode *inode, struct file *filp) +static int sched_feat_open(struct inode *inode, struct file *filp) { filp->private_data = inode->i_private; return 0; @@ -899,7 +775,7 @@ static inline u64 global_rt_runtime(void) return (u64)sysctl_sched_rt_runtime * NSEC_PER_USEC; } -static const unsigned long long time_sync_thresh = 100000; +unsigned long long time_sync_thresh = 100000; static DEFINE_PER_CPU(unsigned long long, time_offset); static DEFINE_PER_CPU(unsigned long long, prev_cpu_time); @@ -913,11 +789,14 @@ static DEFINE_PER_CPU(unsigned long long, prev_cpu_time); static DEFINE_SPINLOCK(time_sync_lock); static unsigned long long prev_global_time; -static unsigned long long __sync_cpu_clock(cycles_t time, int cpu) +static unsigned long long __sync_cpu_clock(unsigned long long time, int cpu) { - unsigned long flags; - - spin_lock_irqsave(&time_sync_lock, flags); + /* + * We want this inlined, to not get tracer function calls + * in this critical section: + */ + spin_acquire(&time_sync_lock.dep_map, 0, 0, _THIS_IP_); + __raw_spin_lock(&time_sync_lock.raw_lock); if (time < prev_global_time) { per_cpu(time_offset, cpu) += prev_global_time - time; @@ -926,7 +805,8 @@ static unsigned long long __sync_cpu_clock(cycles_t time, int cpu) prev_global_time = time; } - spin_unlock_irqrestore(&time_sync_lock, flags); + __raw_spin_unlock(&time_sync_lock.raw_lock); + spin_release(&time_sync_lock.dep_map, 1, _THIS_IP_); return time; } @@ -934,8 +814,6 @@ static unsigned long long __sync_cpu_clock(cycles_t time, int cpu) static unsigned long long __cpu_clock(int cpu) { unsigned long long now; - unsigned long flags; - struct rq *rq; /* * Only call sched_clock() if the scheduler has already been @@ -944,11 +822,7 @@ static unsigned long long __cpu_clock(int cpu) if (unlikely(!scheduler_running)) return 0; - local_irq_save(flags); - rq = cpu_rq(cpu); - update_rq_clock(rq); - now = rq->clock; - local_irq_restore(flags); + now = sched_clock_cpu(cpu); return now; } @@ -960,13 +834,18 @@ static unsigned long long __cpu_clock(int cpu) unsigned long long cpu_clock(int cpu) { unsigned long long prev_cpu_time, time, delta_time; + unsigned long flags; + local_irq_save(flags); prev_cpu_time = per_cpu(prev_cpu_time, cpu); time = __cpu_clock(cpu) + per_cpu(time_offset, cpu); delta_time = time-prev_cpu_time; - if (unlikely(delta_time > time_sync_thresh)) + if (unlikely(delta_time > time_sync_thresh)) { time = __sync_cpu_clock(time, cpu); + per_cpu(prev_cpu_time, cpu) = time; + } + local_irq_restore(flags); return time; } @@ -1117,43 +996,6 @@ static struct rq *this_rq_lock(void) return rq; } -/* - * We are going deep-idle (irqs are disabled): - */ -void sched_clock_idle_sleep_event(void) -{ - struct rq *rq = cpu_rq(smp_processor_id()); - - spin_lock(&rq->lock); - __update_rq_clock(rq); - spin_unlock(&rq->lock); - rq->clock_deep_idle_events++; -} -EXPORT_SYMBOL_GPL(sched_clock_idle_sleep_event); - -/* - * We just idled delta nanoseconds (called with irqs disabled): - */ -void sched_clock_idle_wakeup_event(u64 delta_ns) -{ - struct rq *rq = cpu_rq(smp_processor_id()); - u64 now = sched_clock(); - - rq->idle_clock += delta_ns; - /* - * Override the previous timestamp and ignore all - * sched_clock() deltas that occured while we idled, - * and use the PM-provided delta_ns to advance the - * rq clock: - */ - spin_lock(&rq->lock); - rq->prev_clock_raw = now; - rq->clock += delta_ns; - spin_unlock(&rq->lock); - touch_softlockup_watchdog(); -} -EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event); - static void __resched_task(struct task_struct *p, int tif_bit); static inline void resched_task(struct task_struct *p) @@ -1189,6 +1031,7 @@ static inline void resched_rq(struct rq *rq) enum { HRTICK_SET, /* re-programm hrtick_timer */ HRTICK_RESET, /* not a new slice */ + HRTICK_BLOCK, /* stop hrtick operations */ }; /* @@ -1200,6 +1043,8 @@ static inline int hrtick_enabled(struct rq *rq) { if (!sched_feat(HRTICK)) return 0; + if (unlikely(test_bit(HRTICK_BLOCK, &rq->hrtick_flags))) + return 0; return hrtimer_is_hres_active(&rq->hrtick_timer); } @@ -1275,14 +1120,72 @@ static enum hrtimer_restart hrtick(struct hrtimer *timer) WARN_ON_ONCE(cpu_of(rq) != smp_processor_id()); spin_lock(&rq->lock); - __update_rq_clock(rq); + update_rq_clock(rq); rq->curr->sched_class->task_tick(rq, rq->curr, 1); spin_unlock(&rq->lock); return HRTIMER_NORESTART; } -static inline void init_rq_hrtick(struct rq *rq) +#ifdef CONFIG_SMP +static void hotplug_hrtick_disable(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long flags; + + spin_lock_irqsave(&rq->lock, flags); + rq->hrtick_flags = 0; + __set_bit(HRTICK_BLOCK, &rq->hrtick_flags); + spin_unlock_irqrestore(&rq->lock, flags); + + hrtick_clear(rq); +} + +static void hotplug_hrtick_enable(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long flags; + + spin_lock_irqsave(&rq->lock, flags); + __clear_bit(HRTICK_BLOCK, &rq->hrtick_flags); + spin_unlock_irqrestore(&rq->lock, flags); +} + +static int +hotplug_hrtick(struct notifier_block *nfb, unsigned long action, void *hcpu) +{ + int cpu = (int)(long)hcpu; + + switch (action) { + case CPU_UP_CANCELED: + case CPU_UP_CANCELED_FROZEN: + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + case CPU_DEAD: + case CPU_DEAD_FROZEN: + hotplug_hrtick_disable(cpu); + return NOTIFY_OK; + + case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: + case CPU_DOWN_FAILED: + case CPU_DOWN_FAILED_FROZEN: + case CPU_ONLINE: + case CPU_ONLINE_FROZEN: + hotplug_hrtick_enable(cpu); + return NOTIFY_OK; + } + + return NOTIFY_DONE; +} + +static void init_hrtick(void) +{ + hotcpu_notifier(hotplug_hrtick, 0); +} +#endif /* CONFIG_SMP */ + +static void init_rq_hrtick(struct rq *rq) { rq->hrtick_flags = 0; hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); @@ -1319,6 +1222,10 @@ static inline void init_rq_hrtick(struct rq *rq) void hrtick_resched(void) { } + +static inline void init_hrtick(void) +{ +} #endif /* @@ -1429,17 +1336,19 @@ static void __resched_task(struct task_struct *p, int tif_bit) */ #define SRR(x, y) (((x) + (1UL << ((y) - 1))) >> (y)) -/* - * delta *= weight / lw - */ static unsigned long calc_delta_mine(unsigned long delta_exec, unsigned long weight, struct load_weight *lw) { u64 tmp; - if (unlikely(!lw->inv_weight)) - lw->inv_weight = (WMULT_CONST-lw->weight/2) / (lw->weight+1); + if (!lw->inv_weight) { + if (BITS_PER_LONG > 32 && unlikely(lw->weight >= WMULT_CONST)) + lw->inv_weight = 1; + else + lw->inv_weight = 1 + (WMULT_CONST-lw->weight/2) + / (lw->weight+1); + } tmp = (u64)delta_exec * weight; /* @@ -1454,6 +1363,12 @@ calc_delta_mine(unsigned long delta_exec, unsigned long weight, return (unsigned long)min(tmp, (u64)(unsigned long)LONG_MAX); } +static inline unsigned long +calc_delta_fair(unsigned long delta_exec, struct load_weight *lw) +{ + return calc_delta_mine(delta_exec, NICE_0_LOAD, lw); +} + static inline void update_load_add(struct load_weight *lw, unsigned long inc) { lw->weight += inc; @@ -1566,324 +1481,6 @@ static unsigned long source_load(int cpu, int type); static unsigned long target_load(int cpu, int type); static unsigned long cpu_avg_load_per_task(int cpu); static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd); - -#ifdef CONFIG_FAIR_GROUP_SCHED - -/* - * Group load balancing. - * - * We calculate a few balance domain wide aggregate numbers; load and weight. - * Given the pictures below, and assuming each item has equal weight: - * - * root 1 - thread - * / | \ A - group - * A 1 B - * /|\ / \ - * C 2 D 3 4 - * | | - * 5 6 - * - * load: - * A and B get 1/3-rd of the total load. C and D get 1/3-rd of A's 1/3-rd, - * which equals 1/9-th of the total load. - * - * shares: - * The weight of this group on the selected cpus. - * - * rq_weight: - * Direct sum of all the cpu's their rq weight, e.g. A would get 3 while - * B would get 2. - * - * task_weight: - * Part of the rq_weight contributed by tasks; all groups except B would - * get 1, B gets 2. - */ - -static inline struct aggregate_struct * -aggregate(struct task_group *tg, struct sched_domain *sd) -{ - return &tg->cfs_rq[sd->first_cpu]->aggregate; -} - -typedef void (*aggregate_func)(struct task_group *, struct sched_domain *); - -/* - * Iterate the full tree, calling @down when first entering a node and @up when - * leaving it for the final time. - */ -static -void aggregate_walk_tree(aggregate_func down, aggregate_func up, - struct sched_domain *sd) -{ - struct task_group *parent, *child; - - rcu_read_lock(); - parent = &root_task_group; -down: - (*down)(parent, sd); - list_for_each_entry_rcu(child, &parent->children, siblings) { - parent = child; - goto down; - -up: - continue; - } - (*up)(parent, sd); - - child = parent; - parent = parent->parent; - if (parent) - goto up; - rcu_read_unlock(); -} - -/* - * Calculate the aggregate runqueue weight. - */ -static -void aggregate_group_weight(struct task_group *tg, struct sched_domain *sd) -{ - unsigned long rq_weight = 0; - unsigned long task_weight = 0; - int i; - - for_each_cpu_mask(i, sd->span) { - rq_weight += tg->cfs_rq[i]->load.weight; - task_weight += tg->cfs_rq[i]->task_weight; - } - - aggregate(tg, sd)->rq_weight = rq_weight; - aggregate(tg, sd)->task_weight = task_weight; -} - -/* - * Compute the weight of this group on the given cpus. - */ -static -void aggregate_group_shares(struct task_group *tg, struct sched_domain *sd) -{ - unsigned long shares = 0; - int i; - - for_each_cpu_mask(i, sd->span) - shares += tg->cfs_rq[i]->shares; - - if ((!shares && aggregate(tg, sd)->rq_weight) || shares > tg->shares) - shares = tg->shares; - - aggregate(tg, sd)->shares = shares; -} - -/* - * Compute the load fraction assigned to this group, relies on the aggregate - * weight and this group's parent's load, i.e. top-down. - */ -static -void aggregate_group_load(struct task_group *tg, struct sched_domain *sd) -{ - unsigned long load; - - if (!tg->parent) { - int i; - - load = 0; - for_each_cpu_mask(i, sd->span) - load += cpu_rq(i)->load.weight; - - } else { - load = aggregate(tg->parent, sd)->load; - - /* - * shares is our weight in the parent's rq so - * shares/parent->rq_weight gives our fraction of the load - */ - load *= aggregate(tg, sd)->shares; - load /= aggregate(tg->parent, sd)->rq_weight + 1; - } - - aggregate(tg, sd)->load = load; -} - -static void __set_se_shares(struct sched_entity *se, unsigned long shares); - -/* - * Calculate and set the cpu's group shares. - */ -static void -__update_group_shares_cpu(struct task_group *tg, struct sched_domain *sd, - int tcpu) -{ - int boost = 0; - unsigned long shares; - unsigned long rq_weight; - - if (!tg->se[tcpu]) - return; - - rq_weight = tg->cfs_rq[tcpu]->load.weight; - - /* - * If there are currently no tasks on the cpu pretend there is one of - * average load so that when a new task gets to run here it will not - * get delayed by group starvation. - */ - if (!rq_weight) { - boost = 1; - rq_weight = NICE_0_LOAD; - } - - /* - * \Sum shares * rq_weight - * shares = ----------------------- - * \Sum rq_weight - * - */ - shares = aggregate(tg, sd)->shares * rq_weight; - shares /= aggregate(tg, sd)->rq_weight + 1; - - /* - * record the actual number of shares, not the boosted amount. - */ - tg->cfs_rq[tcpu]->shares = boost ? 0 : shares; - - if (shares < MIN_SHARES) - shares = MIN_SHARES; - - __set_se_shares(tg->se[tcpu], shares); -} - -/* - * Re-adjust the weights on the cpu the task came from and on the cpu the - * task went to. - */ -static void -__move_group_shares(struct task_group *tg, struct sched_domain *sd, - int scpu, int dcpu) -{ - unsigned long shares; - - shares = tg->cfs_rq[scpu]->shares + tg->cfs_rq[dcpu]->shares; - - __update_group_shares_cpu(tg, sd, scpu); - __update_group_shares_cpu(tg, sd, dcpu); - - /* - * ensure we never loose shares due to rounding errors in the - * above redistribution. - */ - shares -= tg->cfs_rq[scpu]->shares + tg->cfs_rq[dcpu]->shares; - if (shares) - tg->cfs_rq[dcpu]->shares += shares; -} - -/* - * Because changing a group's shares changes the weight of the super-group - * we need to walk up the tree and change all shares until we hit the root. - */ -static void -move_group_shares(struct task_group *tg, struct sched_domain *sd, - int scpu, int dcpu) -{ - while (tg) { - __move_group_shares(tg, sd, scpu, dcpu); - tg = tg->parent; - } -} - -static -void aggregate_group_set_shares(struct task_group *tg, struct sched_domain *sd) -{ - unsigned long shares = aggregate(tg, sd)->shares; - int i; - - for_each_cpu_mask(i, sd->span) { - struct rq *rq = cpu_rq(i); - unsigned long flags; - - spin_lock_irqsave(&rq->lock, flags); - __update_group_shares_cpu(tg, sd, i); - spin_unlock_irqrestore(&rq->lock, flags); - } - - aggregate_group_shares(tg, sd); - - /* - * ensure we never loose shares due to rounding errors in the - * above redistribution. - */ - shares -= aggregate(tg, sd)->shares; - if (shares) { - tg->cfs_rq[sd->first_cpu]->shares += shares; - aggregate(tg, sd)->shares += shares; - } -} - -/* - * Calculate the accumulative weight and recursive load of each task group - * while walking down the tree. - */ -static -void aggregate_get_down(struct task_group *tg, struct sched_domain *sd) -{ - aggregate_group_weight(tg, sd); - aggregate_group_shares(tg, sd); - aggregate_group_load(tg, sd); -} - -/* - * Rebalance the cpu shares while walking back up the tree. - */ -static -void aggregate_get_up(struct task_group *tg, struct sched_domain *sd) -{ - aggregate_group_set_shares(tg, sd); -} - -static DEFINE_PER_CPU(spinlock_t, aggregate_lock); - -static void __init init_aggregate(void) -{ - int i; - - for_each_possible_cpu(i) - spin_lock_init(&per_cpu(aggregate_lock, i)); -} - -static int get_aggregate(struct sched_domain *sd) -{ - if (!spin_trylock(&per_cpu(aggregate_lock, sd->first_cpu))) - return 0; - - aggregate_walk_tree(aggregate_get_down, aggregate_get_up, sd); - return 1; -} - -static void put_aggregate(struct sched_domain *sd) -{ - spin_unlock(&per_cpu(aggregate_lock, sd->first_cpu)); -} - -static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares) -{ - cfs_rq->shares = shares; -} - -#else - -static inline void init_aggregate(void) -{ -} - -static inline int get_aggregate(struct sched_domain *sd) -{ - return 0; -} - -static inline void put_aggregate(struct sched_domain *sd) -{ -} -#endif - #else /* CONFIG_SMP */ #ifdef CONFIG_FAIR_GROUP_SCHED @@ -1904,14 +1501,26 @@ static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares) #define sched_class_highest (&rt_sched_class) -static void inc_nr_running(struct rq *rq) +static inline void inc_load(struct rq *rq, const struct task_struct *p) +{ + update_load_add(&rq->load, p->se.load.weight); +} + +static inline void dec_load(struct rq *rq, const struct task_struct *p) +{ + update_load_sub(&rq->load, p->se.load.weight); +} + +static void inc_nr_running(struct task_struct *p, struct rq *rq) { rq->nr_running++; + inc_load(rq, p); } -static void dec_nr_running(struct rq *rq) +static void dec_nr_running(struct task_struct *p, struct rq *rq) { rq->nr_running--; + dec_load(rq, p); } static void set_load_weight(struct task_struct *p) @@ -2003,7 +1612,7 @@ static void activate_task(struct rq *rq, struct task_struct *p, int wakeup) rq->nr_uninterruptible--; enqueue_task(rq, p, wakeup); - inc_nr_running(rq); + inc_nr_running(p, rq); } /* @@ -2015,7 +1624,7 @@ static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep) rq->nr_uninterruptible++; dequeue_task(rq, p, sleep); - dec_nr_running(rq); + dec_nr_running(p, rq); } /** @@ -2668,7 +2277,7 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags) * management (if any): */ p->sched_class->task_new(rq, p); - inc_nr_running(rq); + inc_nr_running(p, rq); } check_preempt_curr(rq, p); #ifdef CONFIG_SMP @@ -3659,12 +3268,9 @@ static int load_balance(int this_cpu, struct rq *this_rq, unsigned long imbalance; struct rq *busiest; unsigned long flags; - int unlock_aggregate; cpus_setall(*cpus); - unlock_aggregate = get_aggregate(sd); - /* * When power savings policy is enabled for the parent domain, idle * sibling can pick up load irrespective of busy siblings. In this case, @@ -3780,9 +3386,8 @@ redo: if (!ld_moved && !sd_idle && sd->flags & SD_SHARE_CPUPOWER && !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE)) - ld_moved = -1; - - goto out; + return -1; + return ld_moved; out_balanced: schedstat_inc(sd, lb_balanced[idle]); @@ -3797,13 +3402,8 @@ out_one_pinned: if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER && !test_sd_parent(sd, SD_POWERSAVINGS_BALANCE)) - ld_moved = -1; - else - ld_moved = 0; -out: - if (unlock_aggregate) - put_aggregate(sd); - return ld_moved; + return -1; + return 0; } /* @@ -4339,8 +3939,10 @@ void account_system_time(struct task_struct *p, int hardirq_offset, struct rq *rq = this_rq(); cputime64_t tmp; - if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) - return account_guest_time(p, cputime); + if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) { + account_guest_time(p, cputime); + return; + } p->stime = cputime_add(p->stime, cputime); @@ -4404,19 +4006,11 @@ void scheduler_tick(void) int cpu = smp_processor_id(); struct rq *rq = cpu_rq(cpu); struct task_struct *curr = rq->curr; - u64 next_tick = rq->tick_timestamp + TICK_NSEC; + + sched_clock_tick(); spin_lock(&rq->lock); - __update_rq_clock(rq); - /* - * Let rq->clock advance by at least TICK_NSEC: - */ - if (unlikely(rq->clock < next_tick)) { - rq->clock = next_tick; - rq->clock_underflows++; - } - rq->tick_timestamp = rq->clock; - update_last_tick_seen(rq); + update_rq_clock(rq); update_cpu_load(rq); curr->sched_class->task_tick(rq, curr, 0); spin_unlock(&rq->lock); @@ -4495,7 +4089,7 @@ static inline void schedule_debug(struct task_struct *prev) * schedule() atomically, we ignore that path for now. * Otherwise, whine if we are scheduling when we should not be. */ - if (unlikely(in_atomic_preempt_off()) && unlikely(!prev->exit_state)) + if (unlikely(in_atomic_preempt_off() && !prev->exit_state)) __schedule_bug(prev); profile_hit(SCHED_PROFILING, __builtin_return_address(0)); @@ -4570,17 +4164,15 @@ need_resched_nonpreemptible: * Do the rq-clock update outside the rq lock: */ local_irq_disable(); - __update_rq_clock(rq); + update_rq_clock(rq); spin_lock(&rq->lock); clear_tsk_need_resched(prev); if (prev->state && !(preempt_count() & PREEMPT_ACTIVE)) { - if (unlikely((prev->state & TASK_INTERRUPTIBLE) && - signal_pending(prev))) { + if (unlikely(signal_pending_state(prev->state, prev))) prev->state = TASK_RUNNING; - } else { + else deactivate_task(rq, prev, 1); - } switch_count = &prev->nvcsw; } @@ -4595,9 +4187,9 @@ need_resched_nonpreemptible: prev->sched_class->put_prev_task(rq, prev); next = pick_next_task(rq, prev); - sched_info_switch(prev, next); - if (likely(prev != next)) { + sched_info_switch(prev, next); + rq->nr_switches++; rq->curr = next; ++*switch_count; @@ -4632,8 +4224,6 @@ EXPORT_SYMBOL(schedule); asmlinkage void __sched preempt_schedule(void) { struct thread_info *ti = current_thread_info(); - struct task_struct *task = current; - int saved_lock_depth; /* * If there is a non-zero preempt_count or interrupts are disabled, @@ -4644,16 +4234,7 @@ asmlinkage void __sched preempt_schedule(void) do { add_preempt_count(PREEMPT_ACTIVE); - - /* - * We keep the big kernel semaphore locked, but we - * clear ->lock_depth so that schedule() doesnt - * auto-release the semaphore: - */ - saved_lock_depth = task->lock_depth; - task->lock_depth = -1; schedule(); - task->lock_depth = saved_lock_depth; sub_preempt_count(PREEMPT_ACTIVE); /* @@ -4674,26 +4255,15 @@ EXPORT_SYMBOL(preempt_schedule); asmlinkage void __sched preempt_schedule_irq(void) { struct thread_info *ti = current_thread_info(); - struct task_struct *task = current; - int saved_lock_depth; /* Catch callers which need to be fixed */ BUG_ON(ti->preempt_count || !irqs_disabled()); do { add_preempt_count(PREEMPT_ACTIVE); - - /* - * We keep the big kernel semaphore locked, but we - * clear ->lock_depth so that schedule() doesnt - * auto-release the semaphore: - */ - saved_lock_depth = task->lock_depth; - task->lock_depth = -1; local_irq_enable(); schedule(); local_irq_disable(); - task->lock_depth = saved_lock_depth; sub_preempt_count(PREEMPT_ACTIVE); /* @@ -4828,22 +4398,20 @@ do_wait_for_common(struct completion *x, long timeout, int state) signal_pending(current)) || (state == TASK_KILLABLE && fatal_signal_pending(current))) { - __remove_wait_queue(&x->wait, &wait); - return -ERESTARTSYS; + timeout = -ERESTARTSYS; + break; } __set_current_state(state); spin_unlock_irq(&x->wait.lock); timeout = schedule_timeout(timeout); spin_lock_irq(&x->wait.lock); - if (!timeout) { - __remove_wait_queue(&x->wait, &wait); - return timeout; - } - } while (!x->done); + } while (!x->done && timeout); __remove_wait_queue(&x->wait, &wait); + if (!x->done) + return timeout; } x->done--; - return timeout; + return timeout ?: 1; } static long __sched @@ -5018,8 +4586,10 @@ void set_user_nice(struct task_struct *p, long nice) goto out_unlock; } on_rq = p->se.on_rq; - if (on_rq) + if (on_rq) { dequeue_task(rq, p, 0); + dec_load(rq, p); + } p->static_prio = NICE_TO_PRIO(nice); set_load_weight(p); @@ -5029,6 +4599,7 @@ void set_user_nice(struct task_struct *p, long nice) if (on_rq) { enqueue_task(rq, p, 0); + inc_load(rq, p); /* * If the task increased its priority or is running and * lowered its priority, then reschedule its CPU: @@ -5612,7 +5183,6 @@ static void __cond_resched(void) } while (need_resched()); } -#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PREEMPT_VOLUNTARY) int __sched _cond_resched(void) { if (need_resched() && !(preempt_count() & PREEMPT_ACTIVE) && @@ -5623,7 +5193,6 @@ int __sched _cond_resched(void) return 0; } EXPORT_SYMBOL(_cond_resched); -#endif /* * cond_resched_lock() - if a reschedule is pending, drop the given lock, @@ -5918,8 +5487,11 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) spin_unlock_irqrestore(&rq->lock, flags); /* Set the preempt count _outside_ the spinlocks! */ +#if defined(CONFIG_PREEMPT) + task_thread_info(idle)->preempt_count = (idle->lock_depth >= 0); +#else task_thread_info(idle)->preempt_count = 0; - +#endif /* * The idle tasks have their own, simple scheduling class: */ @@ -6050,10 +5622,10 @@ static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu) double_rq_lock(rq_src, rq_dest); /* Already moved. */ if (task_cpu(p) != src_cpu) - goto out; + goto done; /* Affinity changed (again). */ if (!cpu_isset(dest_cpu, p->cpus_allowed)) - goto out; + goto fail; on_rq = p->se.on_rq; if (on_rq) @@ -6064,8 +5636,9 @@ static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu) activate_task(rq_dest, p, 0); check_preempt_curr(rq_dest, p); } +done: ret = 1; -out: +fail: double_rq_unlock(rq_src, rq_dest); return ret; } @@ -6315,6 +5888,7 @@ static void migrate_dead_tasks(unsigned int dead_cpu) next = pick_next_task(rq, rq->curr); if (!next) break; + next->sched_class->put_prev_task(rq, next); migrate_dead(dead_cpu, next); } @@ -7305,7 +6879,12 @@ static int default_relax_domain_level = -1; static int __init setup_relax_domain_level(char *str) { - default_relax_domain_level = simple_strtoul(str, NULL, 0); + unsigned long val; + + val = simple_strtoul(str, NULL, 0); + if (val < SD_LV_MAX) + default_relax_domain_level = val; + return 1; } __setup("relax_domain_level=", setup_relax_domain_level); @@ -7402,7 +6981,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map, SD_INIT(sd, ALLNODES); set_domain_attribute(sd, attr); sd->span = *cpu_map; - sd->first_cpu = first_cpu(sd->span); cpu_to_allnodes_group(i, cpu_map, &sd->groups, tmpmask); p = sd; sd_allnodes = 1; @@ -7413,7 +6991,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map, SD_INIT(sd, NODE); set_domain_attribute(sd, attr); sched_domain_node_span(cpu_to_node(i), &sd->span); - sd->first_cpu = first_cpu(sd->span); sd->parent = p; if (p) p->child = sd; @@ -7425,7 +7002,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map, SD_INIT(sd, CPU); set_domain_attribute(sd, attr); sd->span = *nodemask; - sd->first_cpu = first_cpu(sd->span); sd->parent = p; if (p) p->child = sd; @@ -7437,7 +7013,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map, SD_INIT(sd, MC); set_domain_attribute(sd, attr); sd->span = cpu_coregroup_map(i); - sd->first_cpu = first_cpu(sd->span); cpus_and(sd->span, sd->span, *cpu_map); sd->parent = p; p->child = sd; @@ -7450,7 +7025,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map, SD_INIT(sd, SIBLING); set_domain_attribute(sd, attr); sd->span = per_cpu(cpu_sibling_map, i); - sd->first_cpu = first_cpu(sd->span); cpus_and(sd->span, sd->span, *cpu_map); sd->parent = p; p->child = sd; @@ -7654,8 +7228,8 @@ static int build_sched_domains(const cpumask_t *cpu_map) static cpumask_t *doms_cur; /* current sched domains */ static int ndoms_cur; /* number of sched domains in 'doms_cur' */ -static struct sched_domain_attr *dattr_cur; /* attribues of custom domains - in 'doms_cur' */ +static struct sched_domain_attr *dattr_cur; + /* attribues of custom domains in 'doms_cur' */ /* * Special case: If a kmalloc of a doms_cur partition (array of @@ -7669,6 +7243,18 @@ void __attribute__((weak)) arch_update_cpu_topology(void) } /* + * Free current domain masks. + * Called after all cpus are attached to NULL domain. + */ +static void free_sched_domains(void) +{ + ndoms_cur = 0; + if (doms_cur != &fallback_doms) + kfree(doms_cur); + doms_cur = &fallback_doms; +} + +/* * Set up scheduler domains and groups. Callers must hold the hotplug lock. * For now this just excludes isolated cpus, but could be used to * exclude other special cases in the future. @@ -7755,7 +7341,7 @@ void partition_sched_domains(int ndoms_new, cpumask_t *doms_new, { int i, j; - lock_doms_cur(); + mutex_lock(&sched_domains_mutex); /* always unregister in case we don't destroy any domains */ unregister_sched_domain_sysctl(); @@ -7804,7 +7390,7 @@ match2: register_sched_domain_sysctl(); - unlock_doms_cur(); + mutex_unlock(&sched_domains_mutex); } #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) @@ -7813,8 +7399,11 @@ int arch_reinit_sched_domains(void) int err; get_online_cpus(); + mutex_lock(&sched_domains_mutex); detach_destroy_domains(&cpu_online_map); + free_sched_domains(); err = arch_init_sched_domains(&cpu_online_map); + mutex_unlock(&sched_domains_mutex); put_online_cpus(); return err; @@ -7898,6 +7487,7 @@ static int update_sched_domains(struct notifier_block *nfb, case CPU_DOWN_PREPARE: case CPU_DOWN_PREPARE_FROZEN: detach_destroy_domains(&cpu_online_map); + free_sched_domains(); return NOTIFY_OK; case CPU_UP_CANCELED: @@ -7916,8 +7506,16 @@ static int update_sched_domains(struct notifier_block *nfb, return NOTIFY_DONE; } +#ifndef CONFIG_CPUSETS + /* + * Create default domain partitioning if cpusets are disabled. + * Otherwise we let cpusets rebuild the domains based on the + * current setup. + */ + /* The hotplug lock is already held by cpu_up/cpu_down */ arch_init_sched_domains(&cpu_online_map); +#endif return NOTIFY_OK; } @@ -7932,13 +7530,16 @@ void __init sched_init_smp(void) BUG_ON(sched_group_nodes_bycpu == NULL); #endif get_online_cpus(); + mutex_lock(&sched_domains_mutex); arch_init_sched_domains(&cpu_online_map); cpus_andnot(non_isolated_cpus, cpu_possible_map, cpu_isolated_map); if (cpus_empty(non_isolated_cpus)) cpu_set(smp_processor_id(), non_isolated_cpus); + mutex_unlock(&sched_domains_mutex); put_online_cpus(); /* XXX: Theoretical race here - CPU may be hotplugged now */ hotcpu_notifier(update_sched_domains, 0); + init_hrtick(); /* Move init over to a non-isolated CPU */ if (set_cpus_allowed_ptr(current, &non_isolated_cpus) < 0) @@ -8025,7 +7626,7 @@ static void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq, se->my_q = cfs_rq; se->load.weight = tg->shares; - se->load.inv_weight = div64_u64(1ULL<<32, se->load.weight); + se->load.inv_weight = 0; se->parent = parent; } #endif @@ -8054,7 +7655,6 @@ static void init_tg_rt_entry(struct task_group *tg, struct rt_rq *rt_rq, else rt_se->rt_rq = parent->my_q; - rt_se->rt_rq = &rq->rt; rt_se->my_q = rt_rq; rt_se->parent = parent; INIT_LIST_HEAD(&rt_se->run_list); @@ -8115,7 +7715,6 @@ void __init sched_init(void) } #ifdef CONFIG_SMP - init_aggregate(); init_defrootdomain(); #endif @@ -8149,8 +7748,6 @@ void __init sched_init(void) spin_lock_init(&rq->lock); lockdep_set_class(&rq->lock, &rq->rq_lock_key); rq->nr_running = 0; - rq->clock = 1; - update_last_tick_seen(rq); init_cfs_rq(&rq->cfs, rq); init_rt_rq(&rq->rt, rq); #ifdef CONFIG_FAIR_GROUP_SCHED @@ -8294,6 +7891,7 @@ EXPORT_SYMBOL(__might_sleep); static void normalize_task(struct rq *rq, struct task_struct *p) { int on_rq; + update_rq_clock(rq); on_rq = p->se.on_rq; if (on_rq) @@ -8325,7 +7923,6 @@ void normalize_rt_tasks(void) p->se.sleep_start = 0; p->se.block_start = 0; #endif - task_rq(p)->clock = 0; if (!rt_task(p)) { /* @@ -8682,31 +8279,25 @@ void sched_move_task(struct task_struct *tsk) #endif #ifdef CONFIG_FAIR_GROUP_SCHED -static void __set_se_shares(struct sched_entity *se, unsigned long shares) +static void set_se_shares(struct sched_entity *se, unsigned long shares) { struct cfs_rq *cfs_rq = se->cfs_rq; + struct rq *rq = cfs_rq->rq; int on_rq; + spin_lock_irq(&rq->lock); + on_rq = se->on_rq; if (on_rq) dequeue_entity(cfs_rq, se, 0); se->load.weight = shares; - se->load.inv_weight = div64_u64((1ULL<<32), shares); + se->load.inv_weight = 0; if (on_rq) enqueue_entity(cfs_rq, se, 0); -} - -static void set_se_shares(struct sched_entity *se, unsigned long shares) -{ - struct cfs_rq *cfs_rq = se->cfs_rq; - struct rq *rq = cfs_rq->rq; - unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); - __set_se_shares(se, shares); - spin_unlock_irqrestore(&rq->lock, flags); + spin_unlock_irq(&rq->lock); } static DEFINE_MUTEX(shares_mutex); @@ -8722,13 +8313,10 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares) if (!tg->se[0]) return -EINVAL; - /* - * A weight of 0 or 1 can cause arithmetics problems. - * (The default weight is 1024 - so there's no practical - * limitation from this.) - */ if (shares < MIN_SHARES) shares = MIN_SHARES; + else if (shares > MAX_SHARES) + shares = MAX_SHARES; mutex_lock(&shares_mutex); if (tg->shares == shares) @@ -8748,13 +8336,8 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares) * w/o tripping rebalance_share or load_balance_fair. */ tg->shares = shares; - for_each_possible_cpu(i) { - /* - * force a rebalance - */ - cfs_rq_set_shares(tg->cfs_rq[i], 0); - set_se_shares(tg->se[i], shares/nr_cpu_ids); - } + for_each_possible_cpu(i) + set_se_shares(tg->se[i], shares); /* * Enable load balance activity on this group, by inserting it back on @@ -8793,7 +8376,7 @@ static unsigned long to_ratio(u64 period, u64 runtime) #ifdef CONFIG_CGROUP_SCHED static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime) { - struct task_group *tgi, *parent = tg->parent; + struct task_group *tgi, *parent = tg ? tg->parent : NULL; unsigned long total = 0; if (!parent) { @@ -8920,6 +8503,9 @@ int sched_group_set_rt_period(struct task_group *tg, long rt_period_us) rt_period = (u64)rt_period_us * NSEC_PER_USEC; rt_runtime = tg->rt_bandwidth.rt_runtime; + if (rt_period == 0) + return -EINVAL; + return tg_set_bandwidth(tg, rt_period, rt_runtime); } @@ -9072,7 +8658,7 @@ static u64 cpu_shares_read_u64(struct cgroup *cgrp, struct cftype *cft) #endif #ifdef CONFIG_RT_GROUP_SCHED -static ssize_t cpu_rt_runtime_write(struct cgroup *cgrp, struct cftype *cft, +static int cpu_rt_runtime_write(struct cgroup *cgrp, struct cftype *cft, s64 val) { return sched_group_set_rt_runtime(cgroup_tg(cgrp), val); diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c new file mode 100644 index 000000000000..ce05271219ab --- /dev/null +++ b/kernel/sched_clock.c @@ -0,0 +1,246 @@ +/* + * sched_clock for unstable cpu clocks + * + * Copyright (C) 2008 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> + * + * Based on code by: + * Ingo Molnar <mingo@redhat.com> + * Guillaume Chazarain <guichaz@gmail.com> + * + * Create a semi stable clock from a mixture of other events, including: + * - gtod + * - jiffies + * - sched_clock() + * - explicit idle events + * + * We use gtod as base and the unstable clock deltas. The deltas are filtered, + * making it monotonic and keeping it within an expected window. This window + * is set up using jiffies. + * + * Furthermore, explicit sleep and wakeup hooks allow us to account for time + * that is otherwise invisible (TSC gets stopped). + * + * The clock: sched_clock_cpu() is monotonic per cpu, and should be somewhat + * consistent between cpus (never more than 1 jiffies difference). + */ +#include <linux/sched.h> +#include <linux/percpu.h> +#include <linux/spinlock.h> +#include <linux/ktime.h> +#include <linux/module.h> + + +#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK + +struct sched_clock_data { + /* + * Raw spinlock - this is a special case: this might be called + * from within instrumentation code so we dont want to do any + * instrumentation ourselves. + */ + raw_spinlock_t lock; + + unsigned long prev_jiffies; + u64 prev_raw; + u64 tick_raw; + u64 tick_gtod; + u64 clock; +}; + +static DEFINE_PER_CPU_SHARED_ALIGNED(struct sched_clock_data, sched_clock_data); + +static inline struct sched_clock_data *this_scd(void) +{ + return &__get_cpu_var(sched_clock_data); +} + +static inline struct sched_clock_data *cpu_sdc(int cpu) +{ + return &per_cpu(sched_clock_data, cpu); +} + +static __read_mostly int sched_clock_running; + +void sched_clock_init(void) +{ + u64 ktime_now = ktime_to_ns(ktime_get()); + unsigned long now_jiffies = jiffies; + int cpu; + + for_each_possible_cpu(cpu) { + struct sched_clock_data *scd = cpu_sdc(cpu); + + scd->lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; + scd->prev_jiffies = now_jiffies; + scd->prev_raw = 0; + scd->tick_raw = 0; + scd->tick_gtod = ktime_now; + scd->clock = ktime_now; + } + + sched_clock_running = 1; +} + +/* + * update the percpu scd from the raw @now value + * + * - filter out backward motion + * - use jiffies to generate a min,max window to clip the raw values + */ +static void __update_sched_clock(struct sched_clock_data *scd, u64 now) +{ + unsigned long now_jiffies = jiffies; + long delta_jiffies = now_jiffies - scd->prev_jiffies; + u64 clock = scd->clock; + u64 min_clock, max_clock; + s64 delta = now - scd->prev_raw; + + WARN_ON_ONCE(!irqs_disabled()); + min_clock = scd->tick_gtod + delta_jiffies * TICK_NSEC; + + if (unlikely(delta < 0)) { + clock++; + goto out; + } + + max_clock = min_clock + TICK_NSEC; + + if (unlikely(clock + delta > max_clock)) { + if (clock < max_clock) + clock = max_clock; + else + clock++; + } else { + clock += delta; + } + + out: + if (unlikely(clock < min_clock)) + clock = min_clock; + + scd->prev_raw = now; + scd->prev_jiffies = now_jiffies; + scd->clock = clock; +} + +static void lock_double_clock(struct sched_clock_data *data1, + struct sched_clock_data *data2) +{ + if (data1 < data2) { + __raw_spin_lock(&data1->lock); + __raw_spin_lock(&data2->lock); + } else { + __raw_spin_lock(&data2->lock); + __raw_spin_lock(&data1->lock); + } +} + +u64 sched_clock_cpu(int cpu) +{ + struct sched_clock_data *scd = cpu_sdc(cpu); + u64 now, clock; + + if (unlikely(!sched_clock_running)) + return 0ull; + + WARN_ON_ONCE(!irqs_disabled()); + now = sched_clock(); + + if (cpu != raw_smp_processor_id()) { + /* + * in order to update a remote cpu's clock based on our + * unstable raw time rebase it against: + * tick_raw (offset between raw counters) + * tick_gotd (tick offset between cpus) + */ + struct sched_clock_data *my_scd = this_scd(); + + lock_double_clock(scd, my_scd); + + now -= my_scd->tick_raw; + now += scd->tick_raw; + + now -= my_scd->tick_gtod; + now += scd->tick_gtod; + + __raw_spin_unlock(&my_scd->lock); + } else { + __raw_spin_lock(&scd->lock); + } + + __update_sched_clock(scd, now); + clock = scd->clock; + + __raw_spin_unlock(&scd->lock); + + return clock; +} + +void sched_clock_tick(void) +{ + struct sched_clock_data *scd = this_scd(); + u64 now, now_gtod; + + if (unlikely(!sched_clock_running)) + return; + + WARN_ON_ONCE(!irqs_disabled()); + + now = sched_clock(); + now_gtod = ktime_to_ns(ktime_get()); + + __raw_spin_lock(&scd->lock); + __update_sched_clock(scd, now); + /* + * update tick_gtod after __update_sched_clock() because that will + * already observe 1 new jiffy; adding a new tick_gtod to that would + * increase the clock 2 jiffies. + */ + scd->tick_raw = now; + scd->tick_gtod = now_gtod; + __raw_spin_unlock(&scd->lock); +} + +/* + * We are going deep-idle (irqs are disabled): + */ +void sched_clock_idle_sleep_event(void) +{ + sched_clock_cpu(smp_processor_id()); +} +EXPORT_SYMBOL_GPL(sched_clock_idle_sleep_event); + +/* + * We just idled delta nanoseconds (called with irqs disabled): + */ +void sched_clock_idle_wakeup_event(u64 delta_ns) +{ + struct sched_clock_data *scd = this_scd(); + u64 now = sched_clock(); + + /* + * Override the previous timestamp and ignore all + * sched_clock() deltas that occured while we idled, + * and use the PM-provided delta_ns to advance the + * rq clock: + */ + __raw_spin_lock(&scd->lock); + scd->prev_raw = now; + scd->clock += delta_ns; + __raw_spin_unlock(&scd->lock); + + touch_softlockup_watchdog(); +} +EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event); + +#endif + +/* + * Scheduler clock - returns current time in nanosec units. + * This is default implementation. + * Architectures and sub-architectures can override this. + */ +unsigned long long __attribute__((weak)) sched_clock(void) +{ + return (unsigned long long)jiffies * (NSEC_PER_SEC / HZ); +} diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 6b4a12558e88..8bb713040ac9 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -167,11 +167,6 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) #endif SEQ_printf(m, " .%-30s: %ld\n", "nr_spread_over", cfs_rq->nr_spread_over); -#ifdef CONFIG_FAIR_GROUP_SCHED -#ifdef CONFIG_SMP - SEQ_printf(m, " .%-30s: %lu\n", "shares", cfs_rq->shares); -#endif -#endif } static void print_cpu(struct seq_file *m, int cpu) @@ -204,13 +199,6 @@ static void print_cpu(struct seq_file *m, int cpu) PN(next_balance); P(curr->pid); PN(clock); - PN(idle_clock); - PN(prev_clock_raw); - P(clock_warps); - P(clock_overflows); - P(clock_underflows); - P(clock_deep_idle_events); - PN(clock_max_delta); P(cpu_load[0]); P(cpu_load[1]); P(cpu_load[2]); diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 89fa32b4edf2..08ae848b71d4 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -334,34 +334,6 @@ int sched_nr_latency_handler(struct ctl_table *table, int write, #endif /* - * delta *= w / rw - */ -static inline unsigned long -calc_delta_weight(unsigned long delta, struct sched_entity *se) -{ - for_each_sched_entity(se) { - delta = calc_delta_mine(delta, - se->load.weight, &cfs_rq_of(se)->load); - } - - return delta; -} - -/* - * delta *= rw / w - */ -static inline unsigned long -calc_delta_fair(unsigned long delta, struct sched_entity *se) -{ - for_each_sched_entity(se) { - delta = calc_delta_mine(delta, - cfs_rq_of(se)->load.weight, &se->load); - } - - return delta; -} - -/* * The idea is to set a period in which each task runs once. * * When there are too many tasks (sysctl_sched_nr_latency) we have to stretch @@ -390,54 +362,47 @@ static u64 __sched_period(unsigned long nr_running) */ static u64 sched_slice(struct cfs_rq *cfs_rq, struct sched_entity *se) { - return calc_delta_weight(__sched_period(cfs_rq->nr_running), se); + u64 slice = __sched_period(cfs_rq->nr_running); + + for_each_sched_entity(se) { + cfs_rq = cfs_rq_of(se); + + slice *= se->load.weight; + do_div(slice, cfs_rq->load.weight); + } + + + return slice; } /* * We calculate the vruntime slice of a to be inserted task * - * vs = s*rw/w = p + * vs = s/w = p/rw */ static u64 sched_vslice_add(struct cfs_rq *cfs_rq, struct sched_entity *se) { unsigned long nr_running = cfs_rq->nr_running; + unsigned long weight; + u64 vslice; if (!se->on_rq) nr_running++; - return __sched_period(nr_running); -} - -/* - * The goal of calc_delta_asym() is to be asymmetrically around NICE_0_LOAD, in - * that it favours >=0 over <0. - * - * -20 | - * | - * 0 --------+------- - * .' - * 19 .' - * - */ -static unsigned long -calc_delta_asym(unsigned long delta, struct sched_entity *se) -{ - struct load_weight lw = { - .weight = NICE_0_LOAD, - .inv_weight = 1UL << (WMULT_SHIFT-NICE_0_SHIFT) - }; + vslice = __sched_period(nr_running); for_each_sched_entity(se) { - struct load_weight *se_lw = &se->load; + cfs_rq = cfs_rq_of(se); - if (se->load.weight < NICE_0_LOAD) - se_lw = &lw; + weight = cfs_rq->load.weight; + if (!se->on_rq) + weight += se->load.weight; - delta = calc_delta_mine(delta, - cfs_rq_of(se)->load.weight, se_lw); + vslice *= NICE_0_LOAD; + do_div(vslice, weight); } - return delta; + return vslice; } /* @@ -454,7 +419,11 @@ __update_curr(struct cfs_rq *cfs_rq, struct sched_entity *curr, curr->sum_exec_runtime += delta_exec; schedstat_add(cfs_rq, exec_clock, delta_exec); - delta_exec_weighted = calc_delta_fair(delta_exec, curr); + delta_exec_weighted = delta_exec; + if (unlikely(curr->load.weight != NICE_0_LOAD)) { + delta_exec_weighted = calc_delta_fair(delta_exec_weighted, + &curr->load); + } curr->vruntime += delta_exec_weighted; } @@ -541,27 +510,10 @@ update_stats_curr_start(struct cfs_rq *cfs_rq, struct sched_entity *se) * Scheduling class queueing methods: */ -#if defined CONFIG_SMP && defined CONFIG_FAIR_GROUP_SCHED -static void -add_cfs_task_weight(struct cfs_rq *cfs_rq, unsigned long weight) -{ - cfs_rq->task_weight += weight; -} -#else -static inline void -add_cfs_task_weight(struct cfs_rq *cfs_rq, unsigned long weight) -{ -} -#endif - static void account_entity_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se) { update_load_add(&cfs_rq->load, se->load.weight); - if (!parent_entity(se)) - inc_cpu_load(rq_of(cfs_rq), se->load.weight); - if (entity_is_task(se)) - add_cfs_task_weight(cfs_rq, se->load.weight); cfs_rq->nr_running++; se->on_rq = 1; list_add(&se->group_node, &cfs_rq->tasks); @@ -571,10 +523,6 @@ static void account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) { update_load_sub(&cfs_rq->load, se->load.weight); - if (!parent_entity(se)) - dec_cpu_load(rq_of(cfs_rq), se->load.weight); - if (entity_is_task(se)) - add_cfs_task_weight(cfs_rq, -se->load.weight); cfs_rq->nr_running--; se->on_rq = 0; list_del_init(&se->group_node); @@ -661,12 +609,8 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) if (!initial) { /* sleeps upto a single latency don't count. */ - if (sched_feat(NEW_FAIR_SLEEPERS)) { - if (sched_feat(NORMALIZED_SLEEPER)) - vruntime -= calc_delta_weight(sysctl_sched_latency, se); - else - vruntime -= sysctl_sched_latency; - } + if (sched_feat(NEW_FAIR_SLEEPERS)) + vruntime -= sysctl_sched_latency; /* ensure we never gain time by being placed backwards. */ vruntime = max_vruntime(se->vruntime, vruntime); @@ -682,6 +626,7 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int wakeup) * Update run-time statistics of the 'current'. */ update_curr(cfs_rq); + account_entity_enqueue(cfs_rq, se); if (wakeup) { place_entity(cfs_rq, se, 0); @@ -692,7 +637,6 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int wakeup) check_spread(cfs_rq, se); if (se != cfs_rq->curr) __enqueue_entity(cfs_rq, se); - account_entity_enqueue(cfs_rq, se); } static void update_avg(u64 *avg, u64 sample) @@ -841,8 +785,10 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) * queued ticks are scheduled to match the slice, so don't bother * validating it and just reschedule. */ - if (queued) - return resched_task(rq_of(cfs_rq)->curr); + if (queued) { + resched_task(rq_of(cfs_rq)->curr); + return; + } /* * don't let the period tick interfere with the hrtick preemption */ @@ -957,7 +903,7 @@ static void yield_task_fair(struct rq *rq) return; if (likely(!sysctl_sched_compat_yield) && curr->policy != SCHED_BATCH) { - __update_rq_clock(rq); + update_rq_clock(rq); /* * Update run-time statistics of the 'current'. */ @@ -1007,7 +953,7 @@ static int wake_idle(int cpu, struct task_struct *p) * sibling runqueue info. This will avoid the checks and cache miss * penalities associated with that. */ - if (idle_cpu(cpu) || cpu_rq(cpu)->nr_running > 1) + if (idle_cpu(cpu) || cpu_rq(cpu)->cfs.nr_running > 1) return cpu; for_each_domain(cpu, sd) { @@ -1050,16 +996,27 @@ wake_affine(struct rq *rq, struct sched_domain *this_sd, struct rq *this_rq, struct task_struct *curr = this_rq->curr; unsigned long tl = this_load; unsigned long tl_per_task; + int balanced; - if (!(this_sd->flags & SD_WAKE_AFFINE)) + if (!(this_sd->flags & SD_WAKE_AFFINE) || !sched_feat(AFFINE_WAKEUPS)) return 0; /* + * If sync wakeup then subtract the (maximum possible) + * effect of the currently running task from the load + * of the current CPU: + */ + if (sync) + tl -= current->se.load.weight; + + balanced = 100*(tl + p->se.load.weight) <= imbalance*load; + + /* * If the currently running task will sleep within * a reasonable amount of time then attract this newly * woken task: */ - if (sync && curr->sched_class == &fair_sched_class) { + if (sync && balanced && curr->sched_class == &fair_sched_class) { if (curr->se.avg_overlap < sysctl_sched_migration_cost && p->se.avg_overlap < sysctl_sched_migration_cost) return 1; @@ -1068,16 +1025,8 @@ wake_affine(struct rq *rq, struct sched_domain *this_sd, struct rq *this_rq, schedstat_inc(p, se.nr_wakeups_affine_attempts); tl_per_task = cpu_avg_load_per_task(this_cpu); - /* - * If sync wakeup then subtract the (maximum possible) - * effect of the currently running task from the load - * of the current CPU: - */ - if (sync) - tl -= current->se.load.weight; - if ((tl <= load && tl + target_load(prev_cpu, idx) <= tl_per_task) || - 100*(tl + p->se.load.weight) <= imbalance*load) { + balanced) { /* * This domain has SD_WAKE_AFFINE and * p is cache cold in this domain, and @@ -1162,10 +1111,11 @@ static unsigned long wakeup_gran(struct sched_entity *se) unsigned long gran = sysctl_sched_wakeup_granularity; /* - * More easily preempt - nice tasks, while not making it harder for - * + nice tasks. + * More easily preempt - nice tasks, while not making + * it harder for + nice tasks. */ - gran = calc_delta_asym(sysctl_sched_wakeup_granularity, se); + if (unlikely(se->load.weight > NICE_0_LOAD)) + gran = calc_delta_fair(gran, &se->load); return gran; } @@ -1359,90 +1309,75 @@ static struct task_struct *load_balance_next_fair(void *arg) return __load_balance_iterator(cfs_rq, cfs_rq->balance_iterator); } -static unsigned long -__load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, - unsigned long max_load_move, struct sched_domain *sd, - enum cpu_idle_type idle, int *all_pinned, int *this_best_prio, - struct cfs_rq *cfs_rq) +#ifdef CONFIG_FAIR_GROUP_SCHED +static int cfs_rq_best_prio(struct cfs_rq *cfs_rq) { - struct rq_iterator cfs_rq_iterator; + struct sched_entity *curr; + struct task_struct *p; - cfs_rq_iterator.start = load_balance_start_fair; - cfs_rq_iterator.next = load_balance_next_fair; - cfs_rq_iterator.arg = cfs_rq; + if (!cfs_rq->nr_running || !first_fair(cfs_rq)) + return MAX_PRIO; - return balance_tasks(this_rq, this_cpu, busiest, - max_load_move, sd, idle, all_pinned, - this_best_prio, &cfs_rq_iterator); + curr = cfs_rq->curr; + if (!curr) + curr = __pick_next_entity(cfs_rq); + + p = task_of(curr); + + return p->prio; } +#endif -#ifdef CONFIG_FAIR_GROUP_SCHED static unsigned long load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, unsigned long max_load_move, struct sched_domain *sd, enum cpu_idle_type idle, int *all_pinned, int *this_best_prio) { + struct cfs_rq *busy_cfs_rq; long rem_load_move = max_load_move; - int busiest_cpu = cpu_of(busiest); - struct task_group *tg; - - rcu_read_lock(); - list_for_each_entry(tg, &task_groups, list) { - long imbalance; - unsigned long this_weight, busiest_weight; - long rem_load, max_load, moved_load; - - /* - * empty group - */ - if (!aggregate(tg, sd)->task_weight) - continue; - - rem_load = rem_load_move * aggregate(tg, sd)->rq_weight; - rem_load /= aggregate(tg, sd)->load + 1; - - this_weight = tg->cfs_rq[this_cpu]->task_weight; - busiest_weight = tg->cfs_rq[busiest_cpu]->task_weight; + struct rq_iterator cfs_rq_iterator; - imbalance = (busiest_weight - this_weight) / 2; + cfs_rq_iterator.start = load_balance_start_fair; + cfs_rq_iterator.next = load_balance_next_fair; - if (imbalance < 0) - imbalance = busiest_weight; + for_each_leaf_cfs_rq(busiest, busy_cfs_rq) { +#ifdef CONFIG_FAIR_GROUP_SCHED + struct cfs_rq *this_cfs_rq; + long imbalance; + unsigned long maxload; - max_load = max(rem_load, imbalance); - moved_load = __load_balance_fair(this_rq, this_cpu, busiest, - max_load, sd, idle, all_pinned, this_best_prio, - tg->cfs_rq[busiest_cpu]); + this_cfs_rq = cpu_cfs_rq(busy_cfs_rq, this_cpu); - if (!moved_load) + imbalance = busy_cfs_rq->load.weight - this_cfs_rq->load.weight; + /* Don't pull if this_cfs_rq has more load than busy_cfs_rq */ + if (imbalance <= 0) continue; - move_group_shares(tg, sd, busiest_cpu, this_cpu); + /* Don't pull more than imbalance/2 */ + imbalance /= 2; + maxload = min(rem_load_move, imbalance); - moved_load *= aggregate(tg, sd)->load; - moved_load /= aggregate(tg, sd)->rq_weight + 1; + *this_best_prio = cfs_rq_best_prio(this_cfs_rq); +#else +# define maxload rem_load_move +#endif + /* + * pass busy_cfs_rq argument into + * load_balance_[start|next]_fair iterators + */ + cfs_rq_iterator.arg = busy_cfs_rq; + rem_load_move -= balance_tasks(this_rq, this_cpu, busiest, + maxload, sd, idle, all_pinned, + this_best_prio, + &cfs_rq_iterator); - rem_load_move -= moved_load; - if (rem_load_move < 0) + if (rem_load_move <= 0) break; } - rcu_read_unlock(); return max_load_move - rem_load_move; } -#else -static unsigned long -load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, - unsigned long max_load_move, - struct sched_domain *sd, enum cpu_idle_type idle, - int *all_pinned, int *this_best_prio) -{ - return __load_balance_fair(this_rq, this_cpu, busiest, - max_load_move, sd, idle, all_pinned, - this_best_prio, &busiest->cfs); -} -#endif static int move_one_task_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, @@ -1611,30 +1546,6 @@ static const struct sched_class fair_sched_class = { }; #ifdef CONFIG_SCHED_DEBUG -static void -print_cfs_rq_tasks(struct seq_file *m, struct cfs_rq *cfs_rq, int depth) -{ - struct sched_entity *se; - - if (!cfs_rq) - return; - - list_for_each_entry_rcu(se, &cfs_rq->tasks, group_node) { - int i; - - for (i = depth; i; i--) - seq_puts(m, " "); - - seq_printf(m, "%lu %s %lu\n", - se->load.weight, - entity_is_task(se) ? "T" : "G", - calc_delta_weight(SCHED_LOAD_SCALE, se) - ); - if (!entity_is_task(se)) - print_cfs_rq_tasks(m, group_cfs_rq(se), depth + 1); - } -} - static void print_cfs_stats(struct seq_file *m, int cpu) { struct cfs_rq *cfs_rq; @@ -1642,9 +1553,6 @@ static void print_cfs_stats(struct seq_file *m, int cpu) rcu_read_lock(); for_each_leaf_cfs_rq(cpu_rq(cpu), cfs_rq) print_cfs_rq(m, cpu, cfs_rq); - - seq_printf(m, "\nWeight tree:\n"); - print_cfs_rq_tasks(m, &cpu_rq(cpu)->cfs, 1); rcu_read_unlock(); } #endif diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c index 2bcafa375633..3a4f92dbbe66 100644 --- a/kernel/sched_idletask.c +++ b/kernel/sched_idletask.c @@ -99,7 +99,7 @@ static void prio_changed_idle(struct rq *rq, struct task_struct *p, /* * Simple, special scheduling class for the per-CPU idle tasks: */ -const struct sched_class idle_sched_class = { +static const struct sched_class idle_sched_class = { /* .next is NULL */ /* no enqueue/yield_task for idle tasks */ diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index c2730a5a4f05..0f3c19197fa4 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -250,7 +250,8 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) if (rt_rq->rt_time || rt_rq->rt_nr_running) idle = 0; spin_unlock(&rt_rq->rt_runtime_lock); - } + } else if (rt_rq->rt_nr_running) + idle = 0; if (enqueue) sched_rt_rq_enqueue(rt_rq); @@ -449,13 +450,19 @@ void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) #endif } -static void enqueue_rt_entity(struct sched_rt_entity *rt_se) +static void __enqueue_rt_entity(struct sched_rt_entity *rt_se) { struct rt_rq *rt_rq = rt_rq_of_se(rt_se); struct rt_prio_array *array = &rt_rq->active; struct rt_rq *group_rq = group_rt_rq(rt_se); - if (group_rq && rt_rq_throttled(group_rq)) + /* + * Don't enqueue the group if its throttled, or when empty. + * The latter is a consequence of the former when a child group + * get throttled and the current group doesn't have any other + * active members. + */ + if (group_rq && (rt_rq_throttled(group_rq) || !group_rq->rt_nr_running)) return; list_add_tail(&rt_se->run_list, array->queue + rt_se_prio(rt_se)); @@ -464,7 +471,7 @@ static void enqueue_rt_entity(struct sched_rt_entity *rt_se) inc_rt_tasks(rt_se, rt_rq); } -static void dequeue_rt_entity(struct sched_rt_entity *rt_se) +static void __dequeue_rt_entity(struct sched_rt_entity *rt_se) { struct rt_rq *rt_rq = rt_rq_of_se(rt_se); struct rt_prio_array *array = &rt_rq->active; @@ -480,11 +487,10 @@ static void dequeue_rt_entity(struct sched_rt_entity *rt_se) * Because the prio of an upper entry depends on the lower * entries, we must remove entries top - down. */ -static void dequeue_rt_stack(struct task_struct *p) +static void dequeue_rt_stack(struct sched_rt_entity *rt_se) { - struct sched_rt_entity *rt_se, *back = NULL; + struct sched_rt_entity *back = NULL; - rt_se = &p->rt; for_each_sched_rt_entity(rt_se) { rt_se->back = back; back = rt_se; @@ -492,7 +498,26 @@ static void dequeue_rt_stack(struct task_struct *p) for (rt_se = back; rt_se; rt_se = rt_se->back) { if (on_rt_rq(rt_se)) - dequeue_rt_entity(rt_se); + __dequeue_rt_entity(rt_se); + } +} + +static void enqueue_rt_entity(struct sched_rt_entity *rt_se) +{ + dequeue_rt_stack(rt_se); + for_each_sched_rt_entity(rt_se) + __enqueue_rt_entity(rt_se); +} + +static void dequeue_rt_entity(struct sched_rt_entity *rt_se) +{ + dequeue_rt_stack(rt_se); + + for_each_sched_rt_entity(rt_se) { + struct rt_rq *rt_rq = group_rt_rq(rt_se); + + if (rt_rq && rt_rq->rt_nr_running) + __enqueue_rt_entity(rt_se); } } @@ -506,36 +531,15 @@ static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup) if (wakeup) rt_se->timeout = 0; - dequeue_rt_stack(p); - - /* - * enqueue everybody, bottom - up. - */ - for_each_sched_rt_entity(rt_se) - enqueue_rt_entity(rt_se); - - inc_cpu_load(rq, p->se.load.weight); + enqueue_rt_entity(rt_se); } static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep) { struct sched_rt_entity *rt_se = &p->rt; - struct rt_rq *rt_rq; update_curr_rt(rq); - - dequeue_rt_stack(p); - - /* - * re-enqueue all non-empty rt_rq entities. - */ - for_each_sched_rt_entity(rt_se) { - rt_rq = group_rt_rq(rt_se); - if (rt_rq && rt_rq->rt_nr_running) - enqueue_rt_entity(rt_se); - } - - dec_cpu_load(rq, p->se.load.weight); + dequeue_rt_entity(rt_se); } /* @@ -546,8 +550,10 @@ static void requeue_rt_entity(struct rt_rq *rt_rq, struct sched_rt_entity *rt_se) { struct rt_prio_array *array = &rt_rq->active; + struct list_head *queue = array->queue + rt_se_prio(rt_se); - list_move_tail(&rt_se->run_list, array->queue + rt_se_prio(rt_se)); + if (on_rt_rq(rt_se)) + list_move_tail(&rt_se->run_list, queue); } static void requeue_task_rt(struct rq *rq, struct task_struct *p) @@ -1098,11 +1104,14 @@ static void post_schedule_rt(struct rq *rq) } } - +/* + * If we are not running and we are not going to reschedule soon, we should + * try to push tasks away now + */ static void task_wake_up_rt(struct rq *rq, struct task_struct *p) { if (!task_running(rq, p) && - (p->prio >= rq->rt.highest_prio) && + !test_tsk_need_resched(rq->curr) && rq->rt.overloaded) push_rt_tasks(rq); } @@ -1309,7 +1318,7 @@ static void set_curr_task_rt(struct rq *rq) p->se.exec_start = rq->clock; } -const struct sched_class rt_sched_class = { +static const struct sched_class rt_sched_class = { .next = &fair_sched_class, .enqueue_task = enqueue_task_rt, .dequeue_task = dequeue_task_rt, diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h index 5bae2e0c3ff2..80179ef7450e 100644 --- a/kernel/sched_stats.h +++ b/kernel/sched_stats.h @@ -67,6 +67,7 @@ static int show_schedstat(struct seq_file *seq, void *v) preempt_enable(); #endif } + kfree(mask_str); return 0; } @@ -197,6 +198,9 @@ static inline void sched_info_queued(struct task_struct *t) /* * Called when a process ceases being the active-running process, either * voluntarily or involuntarily. Now we can calculate how long we ran. + * Also, if the process is still in the TASK_RUNNING state, call + * sched_info_queued() to mark that it has now again started waiting on + * the runqueue. */ static inline void sched_info_depart(struct task_struct *t) { @@ -205,6 +209,9 @@ static inline void sched_info_depart(struct task_struct *t) t->sched_info.cpu_time += delta; rq_sched_info_depart(task_rq(t), delta); + + if (t->state == TASK_RUNNING) + sched_info_queued(t); } /* diff --git a/kernel/signal.c b/kernel/signal.c index 72bb4f51f963..6c0958e52ea7 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -231,6 +231,40 @@ void flush_signals(struct task_struct *t) spin_unlock_irqrestore(&t->sighand->siglock, flags); } +static void __flush_itimer_signals(struct sigpending *pending) +{ + sigset_t signal, retain; + struct sigqueue *q, *n; + + signal = pending->signal; + sigemptyset(&retain); + + list_for_each_entry_safe(q, n, &pending->list, list) { + int sig = q->info.si_signo; + + if (likely(q->info.si_code != SI_TIMER)) { + sigaddset(&retain, sig); + } else { + sigdelset(&signal, sig); + list_del_init(&q->list); + __sigqueue_free(q); + } + } + + sigorsets(&pending->signal, &signal, &retain); +} + +void flush_itimer_signals(void) +{ + struct task_struct *tsk = current; + unsigned long flags; + + spin_lock_irqsave(&tsk->sighand->siglock, flags); + __flush_itimer_signals(&tsk->pending); + __flush_itimer_signals(&tsk->signal->shared_pending); + spin_unlock_irqrestore(&tsk->sighand->siglock, flags); +} + void ignore_signals(struct task_struct *t) { int i; @@ -1240,17 +1274,22 @@ void sigqueue_free(struct sigqueue *q) BUG_ON(!(q->flags & SIGQUEUE_PREALLOC)); /* - * If the signal is still pending remove it from the - * pending queue. We must hold ->siglock while testing - * q->list to serialize with collect_signal(). + * We must hold ->siglock while testing q->list + * to serialize with collect_signal() or with + * __exit_signal()->flush_sigqueue(). */ spin_lock_irqsave(lock, flags); + q->flags &= ~SIGQUEUE_PREALLOC; + /* + * If it is queued it will be freed when dequeued, + * like the "regular" sigqueue. + */ if (!list_empty(&q->list)) - list_del_init(&q->list); + q = NULL; spin_unlock_irqrestore(lock, flags); - q->flags &= ~SIGQUEUE_PREALLOC; - __sigqueue_free(q); + if (q) + __sigqueue_free(q); } int send_sigqueue(struct sigqueue *q, struct task_struct *t, int group) diff --git a/kernel/softlockup.c b/kernel/softlockup.c index 01b6522fd92b..a272d78185eb 100644 --- a/kernel/softlockup.c +++ b/kernel/softlockup.c @@ -49,12 +49,17 @@ static unsigned long get_timestamp(int this_cpu) return cpu_clock(this_cpu) >> 30LL; /* 2^30 ~= 10^9 */ } -void touch_softlockup_watchdog(void) +static void __touch_softlockup_watchdog(void) { int this_cpu = raw_smp_processor_id(); __raw_get_cpu_var(touch_timestamp) = get_timestamp(this_cpu); } + +void touch_softlockup_watchdog(void) +{ + __raw_get_cpu_var(touch_timestamp) = 0; +} EXPORT_SYMBOL(touch_softlockup_watchdog); void touch_all_softlockup_watchdogs(void) @@ -80,7 +85,7 @@ void softlockup_tick(void) unsigned long now; if (touch_timestamp == 0) { - touch_softlockup_watchdog(); + __touch_softlockup_watchdog(); return; } @@ -95,7 +100,7 @@ void softlockup_tick(void) /* do not print during early bootup: */ if (unlikely(system_state != SYSTEM_RUNNING)) { - touch_softlockup_watchdog(); + __touch_softlockup_watchdog(); return; } @@ -115,6 +120,7 @@ void softlockup_tick(void) printk(KERN_ERR "BUG: soft lockup - CPU#%d stuck for %lus! [%s:%d]\n", this_cpu, now - touch_timestamp, current->comm, task_pid_nr(current)); + print_modules(); if (regs) show_regs(regs); else @@ -214,7 +220,7 @@ static int watchdog(void *__bind_cpu) sched_setscheduler(current, SCHED_FIFO, ¶m); /* initialize timestamp */ - touch_softlockup_watchdog(); + __touch_softlockup_watchdog(); set_current_state(TASK_INTERRUPTIBLE); /* @@ -223,7 +229,7 @@ static int watchdog(void *__bind_cpu) * debug-printout triggers in softlockup_tick(). */ while (!kthread_should_stop()) { - touch_softlockup_watchdog(); + __touch_softlockup_watchdog(); schedule(); if (kthread_should_stop()) diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index 0101aeef7ed7..b7350bbfb076 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -62,8 +62,7 @@ static int stopmachine(void *cpu) * help our sisters onto their CPUs. */ if (!prepared && !irqs_disabled) yield(); - else - cpu_relax(); + cpu_relax(); } /* Ack: we are exiting. */ @@ -106,8 +105,10 @@ static int stop_machine(void) } /* Wait for them all to come to life. */ - while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads) + while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads) { yield(); + cpu_relax(); + } /* If some failed, kill them all. */ if (ret < 0) { diff --git a/kernel/sys.c b/kernel/sys.c index 895d2d4c9493..14e97282eb6c 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1652,7 +1652,7 @@ asmlinkage long sys_umask(int mask) asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, unsigned long arg4, unsigned long arg5) { - long uninitialized_var(error); + long error = 0; if (security_task_prctl(option, arg2, arg3, arg4, arg5, &error)) return error; @@ -1701,9 +1701,7 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, error = PR_TIMING_STATISTICAL; break; case PR_SET_TIMING: - if (arg2 == PR_TIMING_STATISTICAL) - error = 0; - else + if (arg2 != PR_TIMING_STATISTICAL) error = -EINVAL; break; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index d7ffdc59816a..29116652dca8 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -81,6 +81,7 @@ extern int compat_log; extern int maps_protect; extern int sysctl_stat_interval; extern int latencytop_enabled; +extern int sysctl_nr_open_min, sysctl_nr_open_max; /* Constants used for minimum and maximum */ #if defined(CONFIG_DETECT_SOFTLOCKUP) || defined(CONFIG_HIGHMEM) @@ -1190,7 +1191,9 @@ static struct ctl_table fs_table[] = { .data = &sysctl_nr_open, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec, + .proc_handler = &proc_dointvec_minmax, + .extra1 = &sysctl_nr_open_min, + .extra2 = &sysctl_nr_open_max, }, { .ctl_name = FS_DENTRY, diff --git a/kernel/time.c b/kernel/time.c index cbe0d5a222ff..6a08660b4fac 100644 --- a/kernel/time.c +++ b/kernel/time.c @@ -246,7 +246,7 @@ unsigned int inline jiffies_to_msecs(const unsigned long j) return (j + (HZ / MSEC_PER_SEC) - 1)/(HZ / MSEC_PER_SEC); #else # if BITS_PER_LONG == 32 - return ((u64)HZ_TO_MSEC_MUL32 * j) >> HZ_TO_MSEC_SHR32; + return (HZ_TO_MSEC_MUL32 * j) >> HZ_TO_MSEC_SHR32; # else return (j * HZ_TO_MSEC_NUM) / HZ_TO_MSEC_DEN; # endif @@ -262,7 +262,7 @@ unsigned int inline jiffies_to_usecs(const unsigned long j) return (j + (HZ / USEC_PER_SEC) - 1)/(HZ / USEC_PER_SEC); #else # if BITS_PER_LONG == 32 - return ((u64)HZ_TO_USEC_MUL32 * j) >> HZ_TO_USEC_SHR32; + return (HZ_TO_USEC_MUL32 * j) >> HZ_TO_USEC_SHR32; # else return (j * HZ_TO_USEC_NUM) / HZ_TO_USEC_DEN; # endif @@ -476,7 +476,7 @@ unsigned long msecs_to_jiffies(const unsigned int m) if (HZ > MSEC_PER_SEC && m > jiffies_to_msecs(MAX_JIFFY_OFFSET)) return MAX_JIFFY_OFFSET; - return ((u64)MSEC_TO_HZ_MUL32 * m + MSEC_TO_HZ_ADJ32) + return (MSEC_TO_HZ_MUL32 * m + MSEC_TO_HZ_ADJ32) >> MSEC_TO_HZ_SHR32; #endif } @@ -491,7 +491,7 @@ unsigned long usecs_to_jiffies(const unsigned int u) #elif HZ > USEC_PER_SEC && !(HZ % USEC_PER_SEC) return u * (HZ / USEC_PER_SEC); #else - return ((u64)USEC_TO_HZ_MUL32 * u + USEC_TO_HZ_ADJ32) + return (USEC_TO_HZ_MUL32 * u + USEC_TO_HZ_ADJ32) >> USEC_TO_HZ_SHR32; #endif } diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 73961f35fdc8..dadde5361f32 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -471,10 +471,10 @@ sysfs_show_available_clocksources(struct sys_device *dev, char *buf) /* * Sysfs setup bits: */ -static SYSDEV_ATTR(current_clocksource, 0600, sysfs_show_current_clocksources, +static SYSDEV_ATTR(current_clocksource, 0644, sysfs_show_current_clocksources, sysfs_override_clocksource); -static SYSDEV_ATTR(available_clocksource, 0600, +static SYSDEV_ATTR(available_clocksource, 0444, sysfs_show_available_clocksources, NULL); static struct sysdev_class clocksource_sysclass = { diff --git a/kernel/timeconst.pl b/kernel/timeconst.pl index 41468035473c..eb51d76e058a 100644 --- a/kernel/timeconst.pl +++ b/kernel/timeconst.pl @@ -1,7 +1,7 @@ #!/usr/bin/perl # ----------------------------------------------------------------------- # -# Copyright 2007 rPath, Inc. - All Rights Reserved +# Copyright 2007-2008 rPath, Inc. - All Rights Reserved # # This file is part of the Linux kernel, and is made available under # the terms of the GNU General Public License version 2 or (at your @@ -20,198 +20,138 @@ %canned_values = ( 24 => [ '0xa6aaaaab','0x2aaaaaa',26, - '0xa6aaaaaaaaaaaaab','0x2aaaaaaaaaaaaaa',58, 125,3, '0xc49ba5e4','0x1fbe76c8b4',37, - '0xc49ba5e353f7ceda','0x1fbe76c8b439581062',69, 3,125, '0xa2c2aaab','0xaaaa',16, - '0xa2c2aaaaaaaaaaab','0xaaaaaaaaaaaa',48, 125000,3, '0xc9539b89','0x7fffbce4217d',47, - '0xc9539b8887229e91','0x7fffbce4217d2849cb25',79, 3,125000, ], 32 => [ '0xfa000000','0x6000000',27, - '0xfa00000000000000','0x600000000000000',59, 125,4, '0x83126e98','0xfdf3b645a',36, - '0x83126e978d4fdf3c','0xfdf3b645a1cac0831',68, 4,125, '0xf4240000','0x0',17, - '0xf424000000000000','0x0',49, 31250,1, '0x8637bd06','0x3fff79c842fa',46, - '0x8637bd05af6c69b6','0x3fff79c842fa5093964a',78, 1,31250, ], 48 => [ '0xa6aaaaab','0x6aaaaaa',27, - '0xa6aaaaaaaaaaaaab','0x6aaaaaaaaaaaaaa',59, 125,6, '0xc49ba5e4','0xfdf3b645a',36, - '0xc49ba5e353f7ceda','0xfdf3b645a1cac0831',68, 6,125, '0xa2c2aaab','0x15555',17, - '0xa2c2aaaaaaaaaaab','0x1555555555555',49, 62500,3, '0xc9539b89','0x3fffbce4217d',46, - '0xc9539b8887229e91','0x3fffbce4217d2849cb25',78, 3,62500, ], 64 => [ '0xfa000000','0xe000000',28, - '0xfa00000000000000','0xe00000000000000',60, 125,8, '0x83126e98','0x7ef9db22d',35, - '0x83126e978d4fdf3c','0x7ef9db22d0e560418',67, 8,125, '0xf4240000','0x0',18, - '0xf424000000000000','0x0',50, 15625,1, '0x8637bd06','0x1fff79c842fa',45, - '0x8637bd05af6c69b6','0x1fff79c842fa5093964a',77, 1,15625, ], 100 => [ '0xa0000000','0x0',28, - '0xa000000000000000','0x0',60, 10,1, '0xcccccccd','0x733333333',35, - '0xcccccccccccccccd','0x73333333333333333',67, 1,10, '0x9c400000','0x0',18, - '0x9c40000000000000','0x0',50, 10000,1, '0xd1b71759','0x1fff2e48e8a7',45, - '0xd1b71758e219652c','0x1fff2e48e8a71de69ad4',77, 1,10000, ], 122 => [ '0x8325c53f','0xfbcda3a',28, - '0x8325c53ef368eb05','0xfbcda3ac10c9714',60, 500,61, '0xf9db22d1','0x7fbe76c8b',35, - '0xf9db22d0e560418a','0x7fbe76c8b43958106',67, 61,500, '0x8012e2a0','0x3ef36',18, - '0x8012e29f79b47583','0x3ef368eb04325',50, 500000,61, '0xffda4053','0x1ffffbce4217',45, - '0xffda4052d666a983','0x1ffffbce4217d2849cb2',77, 61,500000, ], 128 => [ '0xfa000000','0x1e000000',29, - '0xfa00000000000000','0x1e00000000000000',61, 125,16, '0x83126e98','0x3f7ced916',34, - '0x83126e978d4fdf3c','0x3f7ced916872b020c',66, 16,125, '0xf4240000','0x40000',19, - '0xf424000000000000','0x4000000000000',51, 15625,2, '0x8637bd06','0xfffbce4217d',44, - '0x8637bd05af6c69b6','0xfffbce4217d2849cb25',76, 2,15625, ], 200 => [ '0xa0000000','0x0',29, - '0xa000000000000000','0x0',61, 5,1, '0xcccccccd','0x333333333',34, - '0xcccccccccccccccd','0x33333333333333333',66, 1,5, '0x9c400000','0x0',19, - '0x9c40000000000000','0x0',51, 5000,1, '0xd1b71759','0xfff2e48e8a7',44, - '0xd1b71758e219652c','0xfff2e48e8a71de69ad4',76, 1,5000, ], 250 => [ '0x80000000','0x0',29, - '0x8000000000000000','0x0',61, 4,1, '0x80000000','0x180000000',33, - '0x8000000000000000','0x18000000000000000',65, 1,4, '0xfa000000','0x0',20, - '0xfa00000000000000','0x0',52, 4000,1, '0x83126e98','0x7ff7ced9168',43, - '0x83126e978d4fdf3c','0x7ff7ced916872b020c4',75, 1,4000, ], 256 => [ '0xfa000000','0x3e000000',30, - '0xfa00000000000000','0x3e00000000000000',62, 125,32, '0x83126e98','0x1fbe76c8b',33, - '0x83126e978d4fdf3c','0x1fbe76c8b43958106',65, 32,125, '0xf4240000','0xc0000',20, - '0xf424000000000000','0xc000000000000',52, 15625,4, '0x8637bd06','0x7ffde7210be',43, - '0x8637bd05af6c69b6','0x7ffde7210be9424e592',75, 4,15625, ], 300 => [ '0xd5555556','0x2aaaaaaa',30, - '0xd555555555555556','0x2aaaaaaaaaaaaaaa',62, 10,3, '0x9999999a','0x1cccccccc',33, - '0x999999999999999a','0x1cccccccccccccccc',65, 3,10, '0xd0555556','0xaaaaa',20, - '0xd055555555555556','0xaaaaaaaaaaaaa',52, 10000,3, '0x9d495183','0x7ffcb923a29',43, - '0x9d495182a9930be1','0x7ffcb923a29c779a6b5',75, 3,10000, ], 512 => [ '0xfa000000','0x7e000000',31, - '0xfa00000000000000','0x7e00000000000000',63, 125,64, '0x83126e98','0xfdf3b645',32, - '0x83126e978d4fdf3c','0xfdf3b645a1cac083',64, 64,125, '0xf4240000','0x1c0000',21, - '0xf424000000000000','0x1c000000000000',53, 15625,8, '0x8637bd06','0x3ffef39085f',42, - '0x8637bd05af6c69b6','0x3ffef39085f4a1272c9',74, 8,15625, ], 1000 => [ '0x80000000','0x0',31, - '0x8000000000000000','0x0',63, 1,1, '0x80000000','0x0',31, - '0x8000000000000000','0x0',63, 1,1, '0xfa000000','0x0',22, - '0xfa00000000000000','0x0',54, 1000,1, '0x83126e98','0x1ff7ced9168',41, - '0x83126e978d4fdf3c','0x1ff7ced916872b020c4',73, 1,1000, ], 1024 => [ '0xfa000000','0xfe000000',32, - '0xfa00000000000000','0xfe00000000000000',64, 125,128, '0x83126e98','0x7ef9db22',31, - '0x83126e978d4fdf3c','0x7ef9db22d0e56041',63, 128,125, '0xf4240000','0x3c0000',22, - '0xf424000000000000','0x3c000000000000',54, 15625,16, '0x8637bd06','0x1fff79c842f',41, - '0x8637bd05af6c69b6','0x1fff79c842fa5093964',73, 16,15625, ], 1200 => [ '0xd5555556','0xd5555555',32, - '0xd555555555555556','0xd555555555555555',64, 5,6, '0x9999999a','0x66666666',31, - '0x999999999999999a','0x6666666666666666',63, 6,5, '0xd0555556','0x2aaaaa',22, - '0xd055555555555556','0x2aaaaaaaaaaaaa',54, 2500,3, '0x9d495183','0x1ffcb923a29',41, - '0x9d495182a9930be1','0x1ffcb923a29c779a6b5',73, 3,2500, ] ); @@ -264,6 +204,15 @@ sub fmuls($$$) { return 0; } +# Generate a hex value if the result fits in 64 bits; +# otherwise skip. +sub bignum_hex($) { + my($x) = @_; + my $s = $x->as_hex(); + + return (length($s) > 18) ? undef : $s; +} + # Provides mul, adj, and shr factors for a specific # (bit, time, hz) combination sub muladj($$$) { @@ -271,7 +220,7 @@ sub muladj($$$) { my $s = fmuls($b, $t, $hz); my $m = fmul($s, $t, $hz); my $a = fadj($s, $t, $hz); - return ($m->as_hex(), $a->as_hex(), $s); + return (bignum_hex($m), bignum_hex($a), $s); } # Provides numerator, denominator values @@ -288,12 +237,10 @@ sub conversions($$) { # HZ_TO_xx push(@val, muladj(32, $t, $hz)); - push(@val, muladj(64, $t, $hz)); push(@val, numden($t, $hz)); # xx_TO_HZ push(@val, muladj(32, $hz, $t)); - push(@val, muladj(64, $hz, $t)); push(@val, numden($hz, $t)); return @val; @@ -318,6 +265,19 @@ sub compute_values($) { return @val; } +sub outputval($$) +{ + my($name, $val) = @_; + my $csuf; + + if (defined($val)) { + if ($name !~ /SHR/) { + $val = "U64_C($val)"; + } + printf "#define %-23s %s\n", $name.$csuf, $val.$csuf; + } +} + sub output($@) { my($hz, @val) = @_; @@ -331,6 +291,7 @@ sub output($@) print "\n"; print "#include <linux/param.h>\n"; + print "#include <linux/types.h>\n"; print "\n"; print "#if HZ != $hz\n"; @@ -340,15 +301,13 @@ sub output($@) foreach $pfx ('HZ_TO_MSEC','MSEC_TO_HZ', 'HZ_TO_USEC','USEC_TO_HZ') { - foreach $bit (32, 64) { + foreach $bit (32) { foreach $suf ('MUL', 'ADJ', 'SHR') { - printf "#define %-23s %s\n", - "${pfx}_$suf$bit", shift(@val); + outputval("${pfx}_$suf$bit", shift(@val)); } } foreach $suf ('NUM', 'DEN') { - printf "#define %-23s %s\n", - "${pfx}_$suf", shift(@val); + outputval("${pfx}_$suf", shift(@val)); } } @@ -356,6 +315,23 @@ sub output($@) print "#endif /* KERNEL_TIMECONST_H */\n"; } +# Pretty-print Perl values +sub perlvals(@) { + my $v; + my @l = (); + + foreach $v (@_) { + if (!defined($v)) { + push(@l, 'undef'); + } elsif ($v =~ /^0x/) { + push(@l, "\'".$v."\'"); + } else { + push(@l, $v.''); + } + } + return join(',', @l); +} + ($hz) = @ARGV; # Use this to generate the %canned_values structure @@ -373,15 +349,15 @@ if ($hz eq '--can') { print "$pf$hz => [\n"; while (scalar(@values)) { my $bit; - foreach $bit (32, 64) { + foreach $bit (32) { my $m = shift(@values); my $a = shift(@values); my $s = shift(@values); - print "\t\t\'",$m,"\',\'",$a,"\',",$s,",\n"; + print "\t\t", perlvals($m,$a,$s), ",\n"; } my $n = shift(@values); my $d = shift(@values); - print "\t\t",$n,',',$d,",\n"; + print "\t\t", perlvals($n,$d), ",\n"; } print "\t]"; $pf = ', '; diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 29fc39f1029c..ce7799540c91 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -13,7 +13,7 @@ * Kai Petzke <wpp@marie.physik.tu-berlin.de> * Theodore Ts'o <tytso@mit.edu> * - * Made to use alloc_percpu by Christoph Lameter <clameter@sgi.com>. + * Made to use alloc_percpu by Christoph Lameter. */ #include <linux/module.h> |