diff options
Diffstat (limited to 'kernel')
56 files changed, 3158 insertions, 2164 deletions
diff --git a/kernel/Makefile b/kernel/Makefile index 470839d1a30e..35ef1185e359 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -2,7 +2,7 @@ # Makefile for the linux kernel. # -obj-y = fork.o exec_domain.o panic.o printk.o \ +obj-y = fork.o exec_domain.o panic.o \ cpu.o exit.o itimer.o time.o softirq.o resource.o \ sysctl.o sysctl_binary.o capability.o ptrace.o timer.o user.o \ signal.o sys.o kmod.o workqueue.o pid.o task_work.o \ @@ -24,6 +24,7 @@ endif obj-y += sched/ obj-y += power/ +obj-y += printk/ obj-y += cpu/ obj-$(CONFIG_CHECKPOINT_RESTORE) += kcmp.o diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 0e0b20b8c5db..e0aeb32415ff 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -81,7 +81,7 @@ */ #ifdef CONFIG_PROVE_RCU DEFINE_MUTEX(cgroup_mutex); -EXPORT_SYMBOL_GPL(cgroup_mutex); /* only for task_subsys_state_check() */ +EXPORT_SYMBOL_GPL(cgroup_mutex); /* only for lockdep */ #else static DEFINE_MUTEX(cgroup_mutex); #endif @@ -117,6 +117,7 @@ struct cfent { struct list_head node; struct dentry *dentry; struct cftype *type; + struct cgroup_subsys_state *css; /* file xattrs */ struct simple_xattrs xattrs; @@ -159,9 +160,9 @@ struct css_id { */ struct cgroup_event { /* - * Cgroup which the event belongs to. + * css which the event belongs to. */ - struct cgroup *cgrp; + struct cgroup_subsys_state *css; /* * Control file which the event associated. */ @@ -215,10 +216,33 @@ static u64 cgroup_serial_nr_next = 1; */ static int need_forkexit_callback __read_mostly; -static void cgroup_offline_fn(struct work_struct *work); +static struct cftype cgroup_base_files[]; + +static void cgroup_destroy_css_killed(struct cgroup *cgrp); static int cgroup_destroy_locked(struct cgroup *cgrp); -static int cgroup_addrm_files(struct cgroup *cgrp, struct cgroup_subsys *subsys, - struct cftype cfts[], bool is_add); +static int cgroup_addrm_files(struct cgroup *cgrp, struct cftype cfts[], + bool is_add); + +/** + * cgroup_css - obtain a cgroup's css for the specified subsystem + * @cgrp: the cgroup of interest + * @ss: the subsystem of interest (%NULL returns the dummy_css) + * + * Return @cgrp's css (cgroup_subsys_state) associated with @ss. This + * function must be called either under cgroup_mutex or rcu_read_lock() and + * the caller is responsible for pinning the returned css if it wants to + * keep accessing it outside the said locks. This function may return + * %NULL if @cgrp doesn't have @subsys_id enabled. + */ +static struct cgroup_subsys_state *cgroup_css(struct cgroup *cgrp, + struct cgroup_subsys *ss) +{ + if (ss) + return rcu_dereference_check(cgrp->subsys[ss->subsys_id], + lockdep_is_held(&cgroup_mutex)); + else + return &cgrp->dummy_css; +} /* convenient tests for these bits */ static inline bool cgroup_is_dead(const struct cgroup *cgrp) @@ -365,9 +389,11 @@ static struct cgrp_cset_link init_cgrp_cset_link; static int cgroup_init_idr(struct cgroup_subsys *ss, struct cgroup_subsys_state *css); -/* css_set_lock protects the list of css_set objects, and the - * chain of tasks off each css_set. Nests outside task->alloc_lock - * due to cgroup_iter_start() */ +/* + * css_set_lock protects the list of css_set objects, and the chain of + * tasks off each css_set. Nests outside task->alloc_lock due to + * css_task_iter_start(). + */ static DEFINE_RWLOCK(css_set_lock); static int css_set_count; @@ -392,10 +418,12 @@ static unsigned long css_set_hash(struct cgroup_subsys_state *css[]) return key; } -/* We don't maintain the lists running through each css_set to its - * task until after the first call to cgroup_iter_start(). This - * reduces the fork()/exit() overhead for people who have cgroups - * compiled into their kernel but not actually in use */ +/* + * We don't maintain the lists running through each css_set to its task + * until after the first call to css_task_iter_start(). This reduces the + * fork()/exit() overhead for people who have cgroups compiled into their + * kernel but not actually in use. + */ static int use_task_css_set_links __read_mostly; static void __put_css_set(struct css_set *cset, int taskexit) @@ -464,7 +492,7 @@ static inline void put_css_set_taskexit(struct css_set *cset) * @new_cgrp: cgroup that's being entered by the task * @template: desired set of css pointers in css_set (pre-calculated) * - * Returns true if "cg" matches "old_cg" except for the hierarchy + * Returns true if "cset" matches "old_cset" except for the hierarchy * which "new_cgrp" belongs to, for which it should match "new_cgrp". */ static bool compare_css_sets(struct css_set *cset, @@ -555,7 +583,7 @@ static struct css_set *find_existing_css_set(struct css_set *old_cset, /* Subsystem is in this hierarchy. So we want * the subsystem state from the new * cgroup */ - template[i] = cgrp->subsys[i]; + template[i] = cgroup_css(cgrp, ss); } else { /* Subsystem is not in this hierarchy, so we * don't want to change the subsystem state */ @@ -803,8 +831,7 @@ static struct cgroup *task_cgroup_from_root(struct task_struct *task, static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry); -static int cgroup_populate_dir(struct cgroup *cgrp, bool base_files, - unsigned long subsys_mask); +static int cgroup_populate_dir(struct cgroup *cgrp, unsigned long subsys_mask); static const struct inode_operations cgroup_dir_inode_operations; static const struct file_operations proc_cgroupstats_operations; @@ -813,8 +840,7 @@ static struct backing_dev_info cgroup_backing_dev_info = { .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, }; -static int alloc_css_id(struct cgroup_subsys *ss, - struct cgroup *parent, struct cgroup *child); +static int alloc_css_id(struct cgroup_subsys_state *child_css); static struct inode *cgroup_new_inode(umode_t mode, struct super_block *sb) { @@ -845,15 +871,8 @@ static struct cgroup_name *cgroup_alloc_name(struct dentry *dentry) static void cgroup_free_fn(struct work_struct *work) { struct cgroup *cgrp = container_of(work, struct cgroup, destroy_work); - struct cgroup_subsys *ss; mutex_lock(&cgroup_mutex); - /* - * Release the subsystem state objects. - */ - for_each_root_subsys(cgrp->root, ss) - ss->css_free(cgrp); - cgrp->root->number_of_cgroups--; mutex_unlock(&cgroup_mutex); @@ -864,8 +883,6 @@ static void cgroup_free_fn(struct work_struct *work) */ dput(cgrp->parent->dentry); - ida_simple_remove(&cgrp->root->cgroup_ida, cgrp->id); - /* * Drop the active superblock reference that we took when we * created the cgroup. This will free cgrp->root, if we are @@ -956,27 +973,22 @@ static void cgroup_rm_file(struct cgroup *cgrp, const struct cftype *cft) } /** - * cgroup_clear_directory - selective removal of base and subsystem files - * @dir: directory containing the files - * @base_files: true if the base files should be removed + * cgroup_clear_dir - remove subsys files in a cgroup directory + * @cgrp: target cgroup * @subsys_mask: mask of the subsystem ids whose files should be removed */ -static void cgroup_clear_directory(struct dentry *dir, bool base_files, - unsigned long subsys_mask) +static void cgroup_clear_dir(struct cgroup *cgrp, unsigned long subsys_mask) { - struct cgroup *cgrp = __d_cgrp(dir); struct cgroup_subsys *ss; + int i; - for_each_root_subsys(cgrp->root, ss) { + for_each_subsys(ss, i) { struct cftype_set *set; - if (!test_bit(ss->subsys_id, &subsys_mask)) + + if (!test_bit(i, &subsys_mask)) continue; list_for_each_entry(set, &ss->cftsets, node) - cgroup_addrm_files(cgrp, NULL, set->cfts, false); - } - if (base_files) { - while (!list_empty(&cgrp->files)) - cgroup_rm_file(cgrp, NULL); + cgroup_addrm_files(cgrp, set->cfts, false); } } @@ -986,9 +998,6 @@ static void cgroup_clear_directory(struct dentry *dir, bool base_files, static void cgroup_d_remove_dir(struct dentry *dentry) { struct dentry *parent; - struct cgroupfs_root *root = dentry->d_sb->s_fs_info; - - cgroup_clear_directory(dentry, true, root->subsys_mask); parent = dentry->d_parent; spin_lock(&parent->d_lock); @@ -1009,79 +1018,84 @@ static int rebind_subsystems(struct cgroupfs_root *root, { struct cgroup *cgrp = &root->top_cgroup; struct cgroup_subsys *ss; - int i; + unsigned long pinned = 0; + int i, ret; BUG_ON(!mutex_is_locked(&cgroup_mutex)); BUG_ON(!mutex_is_locked(&cgroup_root_mutex)); /* Check that any added subsystems are currently free */ for_each_subsys(ss, i) { - unsigned long bit = 1UL << i; - - if (!(bit & added_mask)) + if (!(added_mask & (1 << i))) continue; + /* is the subsystem mounted elsewhere? */ if (ss->root != &cgroup_dummy_root) { - /* Subsystem isn't free */ - return -EBUSY; + ret = -EBUSY; + goto out_put; } + + /* pin the module */ + if (!try_module_get(ss->module)) { + ret = -ENOENT; + goto out_put; + } + pinned |= 1 << i; } - /* Currently we don't handle adding/removing subsystems when - * any child cgroups exist. This is theoretically supportable - * but involves complex error handling, so it's being left until - * later */ - if (root->number_of_cgroups > 1) - return -EBUSY; + /* subsys could be missing if unloaded between parsing and here */ + if (added_mask != pinned) { + ret = -ENOENT; + goto out_put; + } + + ret = cgroup_populate_dir(cgrp, added_mask); + if (ret) + goto out_put; + + /* + * Nothing can fail from this point on. Remove files for the + * removed subsystems and rebind each subsystem. + */ + cgroup_clear_dir(cgrp, removed_mask); - /* Process each subsystem */ for_each_subsys(ss, i) { unsigned long bit = 1UL << i; if (bit & added_mask) { /* We're binding this subsystem to this hierarchy */ - BUG_ON(cgrp->subsys[i]); - BUG_ON(!cgroup_dummy_top->subsys[i]); - BUG_ON(cgroup_dummy_top->subsys[i]->cgroup != cgroup_dummy_top); + BUG_ON(cgroup_css(cgrp, ss)); + BUG_ON(!cgroup_css(cgroup_dummy_top, ss)); + BUG_ON(cgroup_css(cgroup_dummy_top, ss)->cgroup != cgroup_dummy_top); + + rcu_assign_pointer(cgrp->subsys[i], + cgroup_css(cgroup_dummy_top, ss)); + cgroup_css(cgrp, ss)->cgroup = cgrp; - cgrp->subsys[i] = cgroup_dummy_top->subsys[i]; - cgrp->subsys[i]->cgroup = cgrp; list_move(&ss->sibling, &root->subsys_list); ss->root = root; if (ss->bind) - ss->bind(cgrp); + ss->bind(cgroup_css(cgrp, ss)); /* refcount was already taken, and we're keeping it */ root->subsys_mask |= bit; } else if (bit & removed_mask) { /* We're removing this subsystem */ - BUG_ON(cgrp->subsys[i] != cgroup_dummy_top->subsys[i]); - BUG_ON(cgrp->subsys[i]->cgroup != cgrp); + BUG_ON(cgroup_css(cgrp, ss) != cgroup_css(cgroup_dummy_top, ss)); + BUG_ON(cgroup_css(cgrp, ss)->cgroup != cgrp); if (ss->bind) - ss->bind(cgroup_dummy_top); - cgroup_dummy_top->subsys[i]->cgroup = cgroup_dummy_top; - cgrp->subsys[i] = NULL; + ss->bind(cgroup_css(cgroup_dummy_top, ss)); + + cgroup_css(cgroup_dummy_top, ss)->cgroup = cgroup_dummy_top; + RCU_INIT_POINTER(cgrp->subsys[i], NULL); + cgroup_subsys[i]->root = &cgroup_dummy_root; list_move(&ss->sibling, &cgroup_dummy_root.subsys_list); /* subsystem is now free - drop reference on module */ module_put(ss->module); root->subsys_mask &= ~bit; - } else if (bit & root->subsys_mask) { - /* Subsystem state should already exist */ - BUG_ON(!cgrp->subsys[i]); - /* - * a refcount was taken, but we already had one, so - * drop the extra reference. - */ - module_put(ss->module); -#ifdef CONFIG_MODULE_UNLOAD - BUG_ON(ss->module && !module_refcount(ss->module)); -#endif - } else { - /* Subsystem state shouldn't exist */ - BUG_ON(cgrp->subsys[i]); } } @@ -1092,6 +1106,12 @@ static int rebind_subsystems(struct cgroupfs_root *root, root->flags |= CGRP_ROOT_SUBSYS_BOUND; return 0; + +out_put: + for_each_subsys(ss, i) + if (pinned & (1 << i)) + module_put(ss->module); + return ret; } static int cgroup_show_options(struct seq_file *seq, struct dentry *dentry) @@ -1142,7 +1162,6 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) char *token, *o = data; bool all_ss = false, one_ss = false; unsigned long mask = (unsigned long)-1; - bool module_pin_failed = false; struct cgroup_subsys *ss; int i; @@ -1285,52 +1304,9 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) if (!opts->subsys_mask && !opts->name) return -EINVAL; - /* - * Grab references on all the modules we'll need, so the subsystems - * don't dance around before rebind_subsystems attaches them. This may - * take duplicate reference counts on a subsystem that's already used, - * but rebind_subsystems handles this case. - */ - for_each_subsys(ss, i) { - if (!(opts->subsys_mask & (1UL << i))) - continue; - if (!try_module_get(cgroup_subsys[i]->module)) { - module_pin_failed = true; - break; - } - } - if (module_pin_failed) { - /* - * oops, one of the modules was going away. this means that we - * raced with a module_delete call, and to the user this is - * essentially a "subsystem doesn't exist" case. - */ - for (i--; i >= 0; i--) { - /* drop refcounts only on the ones we took */ - unsigned long bit = 1UL << i; - - if (!(bit & opts->subsys_mask)) - continue; - module_put(cgroup_subsys[i]->module); - } - return -ENOENT; - } - return 0; } -static void drop_parsed_module_refcounts(unsigned long subsys_mask) -{ - struct cgroup_subsys *ss; - int i; - - mutex_lock(&cgroup_mutex); - for_each_subsys(ss, i) - if (subsys_mask & (1UL << i)) - module_put(cgroup_subsys[i]->module); - mutex_unlock(&cgroup_mutex); -} - static int cgroup_remount(struct super_block *sb, int *flags, char *data) { int ret = 0; @@ -1370,22 +1346,15 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) goto out_unlock; } - /* - * Clear out the files of subsystems that should be removed, do - * this before rebind_subsystems, since rebind_subsystems may - * change this hierarchy's subsys_list. - */ - cgroup_clear_directory(cgrp->dentry, false, removed_mask); - - ret = rebind_subsystems(root, added_mask, removed_mask); - if (ret) { - /* rebind_subsystems failed, re-populate the removed files */ - cgroup_populate_dir(cgrp, false, removed_mask); + /* remounting is not allowed for populated hierarchies */ + if (root->number_of_cgroups > 1) { + ret = -EBUSY; goto out_unlock; } - /* re-populate subsystem files */ - cgroup_populate_dir(cgrp, false, added_mask); + ret = rebind_subsystems(root, added_mask, removed_mask); + if (ret) + goto out_unlock; if (opts.release_agent) strcpy(root->release_agent_path, opts.release_agent); @@ -1395,8 +1364,6 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) mutex_unlock(&cgroup_root_mutex); mutex_unlock(&cgroup_mutex); mutex_unlock(&cgrp->dentry->d_inode->i_mutex); - if (ret) - drop_parsed_module_refcounts(opts.subsys_mask); return ret; } @@ -1416,6 +1383,7 @@ static void init_cgroup_housekeeping(struct cgroup *cgrp) INIT_LIST_HEAD(&cgrp->release_list); INIT_LIST_HEAD(&cgrp->pidlists); mutex_init(&cgrp->pidlist_mutex); + cgrp->dummy_css.cgroup = cgrp; INIT_LIST_HEAD(&cgrp->event_list); spin_lock_init(&cgrp->event_list_lock); simple_xattrs_init(&cgrp->xattrs); @@ -1431,6 +1399,7 @@ static void init_cgroup_root(struct cgroupfs_root *root) cgrp->root = root; RCU_INIT_POINTER(cgrp->name, &root_cgroup_name); init_cgroup_housekeeping(cgrp); + idr_init(&root->cgroup_idr); } static int cgroup_init_root_id(struct cgroupfs_root *root, int start, int end) @@ -1503,7 +1472,6 @@ static struct cgroupfs_root *cgroup_root_from_opts(struct cgroup_sb_opts *opts) */ root->subsys_mask = opts->subsys_mask; root->flags = opts->flags; - ida_init(&root->cgroup_ida); if (opts->release_agent) strcpy(root->release_agent_path, opts->release_agent); if (opts->name) @@ -1519,7 +1487,7 @@ static void cgroup_free_root(struct cgroupfs_root *root) /* hierarhcy ID shoulid already have been released */ WARN_ON_ONCE(root->hierarchy_id); - ida_destroy(&root->cgroup_ida); + idr_destroy(&root->cgroup_idr); kfree(root); } } @@ -1584,7 +1552,9 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, int ret = 0; struct super_block *sb; struct cgroupfs_root *new_root; + struct list_head tmp_links; struct inode *inode; + const struct cred *cred; /* First find the desired set of subsystems */ mutex_lock(&cgroup_mutex); @@ -1600,7 +1570,7 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, new_root = cgroup_root_from_opts(&opts); if (IS_ERR(new_root)) { ret = PTR_ERR(new_root); - goto drop_modules; + goto out_err; } opts.new_root = new_root; @@ -1609,17 +1579,15 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, if (IS_ERR(sb)) { ret = PTR_ERR(sb); cgroup_free_root(opts.new_root); - goto drop_modules; + goto out_err; } root = sb->s_fs_info; BUG_ON(!root); if (root == opts.new_root) { /* We used the new root structure, so this is a new hierarchy */ - struct list_head tmp_links; struct cgroup *root_cgrp = &root->top_cgroup; struct cgroupfs_root *existing_root; - const struct cred *cred; int i; struct css_set *cset; @@ -1634,6 +1602,11 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, mutex_lock(&cgroup_mutex); mutex_lock(&cgroup_root_mutex); + root_cgrp->id = idr_alloc(&root->cgroup_idr, root_cgrp, + 0, 1, GFP_KERNEL); + if (root_cgrp->id < 0) + goto unlock_drop; + /* Check for name clashes with existing mounts */ ret = -EBUSY; if (strlen(root->name)) @@ -1657,26 +1630,37 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, if (ret) goto unlock_drop; + sb->s_root->d_fsdata = root_cgrp; + root_cgrp->dentry = sb->s_root; + + /* + * We're inside get_sb() and will call lookup_one_len() to + * create the root files, which doesn't work if SELinux is + * in use. The following cred dancing somehow works around + * it. See 2ce9738ba ("cgroupfs: use init_cred when + * populating new cgroupfs mount") for more details. + */ + cred = override_creds(&init_cred); + + ret = cgroup_addrm_files(root_cgrp, cgroup_base_files, true); + if (ret) + goto rm_base_files; + ret = rebind_subsystems(root, root->subsys_mask, 0); - if (ret == -EBUSY) { - free_cgrp_cset_links(&tmp_links); - goto unlock_drop; - } + if (ret) + goto rm_base_files; + + revert_creds(cred); + /* * There must be no failure case after here, since rebinding * takes care of subsystems' refcounts, which are explicitly * dropped in the failure exit path. */ - /* EBUSY should be the only error here */ - BUG_ON(ret); - list_add(&root->root_list, &cgroup_roots); cgroup_root_count++; - sb->s_root->d_fsdata = root_cgrp; - root->top_cgroup.dentry = sb->s_root; - /* Link the top cgroup in this hierarchy into all * the css_set objects */ write_lock(&css_set_lock); @@ -1689,9 +1673,6 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, BUG_ON(!list_empty(&root_cgrp->children)); BUG_ON(root->number_of_cgroups != 1); - cred = override_creds(&init_cred); - cgroup_populate_dir(root_cgrp, true, root->subsys_mask); - revert_creds(cred); mutex_unlock(&cgroup_root_mutex); mutex_unlock(&cgroup_mutex); mutex_unlock(&inode->i_mutex); @@ -1711,15 +1692,16 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, pr_warning("cgroup: new mount options do not match the existing superblock, will be ignored\n"); } } - - /* no subsys rebinding, so refcounts don't change */ - drop_parsed_module_refcounts(opts.subsys_mask); } kfree(opts.release_agent); kfree(opts.name); return dget(sb->s_root); + rm_base_files: + free_cgrp_cset_links(&tmp_links); + cgroup_addrm_files(&root->top_cgroup, cgroup_base_files, false); + revert_creds(cred); unlock_drop: cgroup_exit_root_id(root); mutex_unlock(&cgroup_root_mutex); @@ -1727,8 +1709,6 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, mutex_unlock(&inode->i_mutex); drop_new_super: deactivate_locked_super(sb); - drop_modules: - drop_parsed_module_refcounts(opts.subsys_mask); out_err: kfree(opts.release_agent); kfree(opts.name); @@ -1746,6 +1726,7 @@ static void cgroup_kill_sb(struct super_block *sb) { BUG_ON(root->number_of_cgroups != 1); BUG_ON(!list_empty(&cgrp->children)); + mutex_lock(&cgrp->dentry->d_inode->i_mutex); mutex_lock(&cgroup_mutex); mutex_lock(&cgroup_root_mutex); @@ -1778,6 +1759,7 @@ static void cgroup_kill_sb(struct super_block *sb) { mutex_unlock(&cgroup_root_mutex); mutex_unlock(&cgroup_mutex); + mutex_unlock(&cgrp->dentry->d_inode->i_mutex); simple_xattrs_free(&cgrp->xattrs); @@ -1845,36 +1827,43 @@ out: EXPORT_SYMBOL_GPL(cgroup_path); /** - * task_cgroup_path_from_hierarchy - cgroup path of a task on a hierarchy + * task_cgroup_path - cgroup path of a task in the first cgroup hierarchy * @task: target task - * @hierarchy_id: the hierarchy to look up @task's cgroup from * @buf: the buffer to write the path into * @buflen: the length of the buffer * - * Determine @task's cgroup on the hierarchy specified by @hierarchy_id and - * copy its path into @buf. This function grabs cgroup_mutex and shouldn't - * be used inside locks used by cgroup controller callbacks. + * Determine @task's cgroup on the first (the one with the lowest non-zero + * hierarchy_id) cgroup hierarchy and copy its path into @buf. This + * function grabs cgroup_mutex and shouldn't be used inside locks used by + * cgroup controller callbacks. + * + * Returns 0 on success, fails with -%ENAMETOOLONG if @buflen is too short. */ -int task_cgroup_path_from_hierarchy(struct task_struct *task, int hierarchy_id, - char *buf, size_t buflen) +int task_cgroup_path(struct task_struct *task, char *buf, size_t buflen) { struct cgroupfs_root *root; - struct cgroup *cgrp = NULL; - int ret = -ENOENT; + struct cgroup *cgrp; + int hierarchy_id = 1, ret = 0; + + if (buflen < 2) + return -ENAMETOOLONG; mutex_lock(&cgroup_mutex); - root = idr_find(&cgroup_hierarchy_idr, hierarchy_id); + root = idr_get_next(&cgroup_hierarchy_idr, &hierarchy_id); + if (root) { cgrp = task_cgroup_from_root(task, root); ret = cgroup_path(cgrp, buf, buflen); + } else { + /* if no hierarchy exists, everyone is in "/" */ + memcpy(buf, "/", 2); } mutex_unlock(&cgroup_mutex); - return ret; } -EXPORT_SYMBOL_GPL(task_cgroup_path_from_hierarchy); +EXPORT_SYMBOL_GPL(task_cgroup_path); /* * Control Group taskset @@ -1882,7 +1871,7 @@ EXPORT_SYMBOL_GPL(task_cgroup_path_from_hierarchy); struct task_and_cgroup { struct task_struct *task; struct cgroup *cgrp; - struct css_set *cg; + struct css_set *cset; }; struct cgroup_taskset { @@ -1932,18 +1921,20 @@ struct task_struct *cgroup_taskset_next(struct cgroup_taskset *tset) EXPORT_SYMBOL_GPL(cgroup_taskset_next); /** - * cgroup_taskset_cur_cgroup - return the matching cgroup for the current task + * cgroup_taskset_cur_css - return the matching css for the current task * @tset: taskset of interest + * @subsys_id: the ID of the target subsystem * - * Return the cgroup for the current (last returned) task of @tset. This - * function must be preceded by either cgroup_taskset_first() or - * cgroup_taskset_next(). + * Return the css for the current (last returned) task of @tset for + * subsystem specified by @subsys_id. This function must be preceded by + * either cgroup_taskset_first() or cgroup_taskset_next(). */ -struct cgroup *cgroup_taskset_cur_cgroup(struct cgroup_taskset *tset) +struct cgroup_subsys_state *cgroup_taskset_cur_css(struct cgroup_taskset *tset, + int subsys_id) { - return tset->cur_cgrp; + return cgroup_css(tset->cur_cgrp, cgroup_subsys[subsys_id]); } -EXPORT_SYMBOL_GPL(cgroup_taskset_cur_cgroup); +EXPORT_SYMBOL_GPL(cgroup_taskset_cur_css); /** * cgroup_taskset_size - return the number of tasks in taskset @@ -2082,8 +2073,10 @@ static int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk, * step 1: check that we can legitimately attach to the cgroup. */ for_each_root_subsys(root, ss) { + struct cgroup_subsys_state *css = cgroup_css(cgrp, ss); + if (ss->can_attach) { - retval = ss->can_attach(cgrp, &tset); + retval = ss->can_attach(css, &tset); if (retval) { failed_ss = ss; goto out_cancel_attach; @@ -2100,8 +2093,8 @@ static int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk, tc = flex_array_get(group, i); old_cset = task_css_set(tc->task); - tc->cg = find_css_set(old_cset, cgrp); - if (!tc->cg) { + tc->cset = find_css_set(old_cset, cgrp); + if (!tc->cset) { retval = -ENOMEM; goto out_put_css_set_refs; } @@ -2114,7 +2107,7 @@ static int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk, */ for (i = 0; i < group_size; i++) { tc = flex_array_get(group, i); - cgroup_task_migrate(tc->cgrp, tc->task, tc->cg); + cgroup_task_migrate(tc->cgrp, tc->task, tc->cset); } /* nothing is sensitive to fork() after this point. */ @@ -2122,8 +2115,10 @@ static int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk, * step 4: do subsystem attach callbacks. */ for_each_root_subsys(root, ss) { + struct cgroup_subsys_state *css = cgroup_css(cgrp, ss); + if (ss->attach) - ss->attach(cgrp, &tset); + ss->attach(css, &tset); } /* @@ -2134,18 +2129,20 @@ out_put_css_set_refs: if (retval) { for (i = 0; i < group_size; i++) { tc = flex_array_get(group, i); - if (!tc->cg) + if (!tc->cset) break; - put_css_set(tc->cg); + put_css_set(tc->cset); } } out_cancel_attach: if (retval) { for_each_root_subsys(root, ss) { + struct cgroup_subsys_state *css = cgroup_css(cgrp, ss); + if (ss == failed_ss) break; if (ss->cancel_attach) - ss->cancel_attach(cgrp, &tset); + ss->cancel_attach(css, &tset); } } out_free_group_list: @@ -2246,9 +2243,9 @@ int cgroup_attach_task_all(struct task_struct *from, struct task_struct *tsk) mutex_lock(&cgroup_mutex); for_each_active_root(root) { - struct cgroup *from_cg = task_cgroup_from_root(from, root); + struct cgroup *from_cgrp = task_cgroup_from_root(from, root); - retval = cgroup_attach_task(from_cg, tsk, false); + retval = cgroup_attach_task(from_cgrp, tsk, false); if (retval) break; } @@ -2258,34 +2255,38 @@ int cgroup_attach_task_all(struct task_struct *from, struct task_struct *tsk) } EXPORT_SYMBOL_GPL(cgroup_attach_task_all); -static int cgroup_tasks_write(struct cgroup *cgrp, struct cftype *cft, u64 pid) +static int cgroup_tasks_write(struct cgroup_subsys_state *css, + struct cftype *cft, u64 pid) { - return attach_task_by_pid(cgrp, pid, false); + return attach_task_by_pid(css->cgroup, pid, false); } -static int cgroup_procs_write(struct cgroup *cgrp, struct cftype *cft, u64 tgid) +static int cgroup_procs_write(struct cgroup_subsys_state *css, + struct cftype *cft, u64 tgid) { - return attach_task_by_pid(cgrp, tgid, true); + return attach_task_by_pid(css->cgroup, tgid, true); } -static int cgroup_release_agent_write(struct cgroup *cgrp, struct cftype *cft, - const char *buffer) +static int cgroup_release_agent_write(struct cgroup_subsys_state *css, + struct cftype *cft, const char *buffer) { - BUILD_BUG_ON(sizeof(cgrp->root->release_agent_path) < PATH_MAX); + BUILD_BUG_ON(sizeof(css->cgroup->root->release_agent_path) < PATH_MAX); if (strlen(buffer) >= PATH_MAX) return -EINVAL; - if (!cgroup_lock_live_group(cgrp)) + if (!cgroup_lock_live_group(css->cgroup)) return -ENODEV; mutex_lock(&cgroup_root_mutex); - strcpy(cgrp->root->release_agent_path, buffer); + strcpy(css->cgroup->root->release_agent_path, buffer); mutex_unlock(&cgroup_root_mutex); mutex_unlock(&cgroup_mutex); return 0; } -static int cgroup_release_agent_show(struct cgroup *cgrp, struct cftype *cft, - struct seq_file *seq) +static int cgroup_release_agent_show(struct cgroup_subsys_state *css, + struct cftype *cft, struct seq_file *seq) { + struct cgroup *cgrp = css->cgroup; + if (!cgroup_lock_live_group(cgrp)) return -ENODEV; seq_puts(seq, cgrp->root->release_agent_path); @@ -2294,20 +2295,20 @@ static int cgroup_release_agent_show(struct cgroup *cgrp, struct cftype *cft, return 0; } -static int cgroup_sane_behavior_show(struct cgroup *cgrp, struct cftype *cft, - struct seq_file *seq) +static int cgroup_sane_behavior_show(struct cgroup_subsys_state *css, + struct cftype *cft, struct seq_file *seq) { - seq_printf(seq, "%d\n", cgroup_sane_behavior(cgrp)); + seq_printf(seq, "%d\n", cgroup_sane_behavior(css->cgroup)); return 0; } /* A buffer size big enough for numbers or short strings */ #define CGROUP_LOCAL_BUFFER_SIZE 64 -static ssize_t cgroup_write_X64(struct cgroup *cgrp, struct cftype *cft, - struct file *file, - const char __user *userbuf, - size_t nbytes, loff_t *unused_ppos) +static ssize_t cgroup_write_X64(struct cgroup_subsys_state *css, + struct cftype *cft, struct file *file, + const char __user *userbuf, size_t nbytes, + loff_t *unused_ppos) { char buffer[CGROUP_LOCAL_BUFFER_SIZE]; int retval = 0; @@ -2325,22 +2326,22 @@ static ssize_t cgroup_write_X64(struct cgroup *cgrp, struct cftype *cft, u64 val = simple_strtoull(strstrip(buffer), &end, 0); if (*end) return -EINVAL; - retval = cft->write_u64(cgrp, cft, val); + retval = cft->write_u64(css, cft, val); } else { s64 val = simple_strtoll(strstrip(buffer), &end, 0); if (*end) return -EINVAL; - retval = cft->write_s64(cgrp, cft, val); + retval = cft->write_s64(css, cft, val); } if (!retval) retval = nbytes; return retval; } -static ssize_t cgroup_write_string(struct cgroup *cgrp, struct cftype *cft, - struct file *file, - const char __user *userbuf, - size_t nbytes, loff_t *unused_ppos) +static ssize_t cgroup_write_string(struct cgroup_subsys_state *css, + struct cftype *cft, struct file *file, + const char __user *userbuf, size_t nbytes, + loff_t *unused_ppos) { char local_buffer[CGROUP_LOCAL_BUFFER_SIZE]; int retval = 0; @@ -2363,7 +2364,7 @@ static ssize_t cgroup_write_string(struct cgroup *cgrp, struct cftype *cft, } buffer[nbytes] = 0; /* nul-terminate */ - retval = cft->write_string(cgrp, cft, strstrip(buffer)); + retval = cft->write_string(css, cft, strstrip(buffer)); if (!retval) retval = nbytes; out: @@ -2373,65 +2374,60 @@ out: } static ssize_t cgroup_file_write(struct file *file, const char __user *buf, - size_t nbytes, loff_t *ppos) + size_t nbytes, loff_t *ppos) { + struct cfent *cfe = __d_cfe(file->f_dentry); struct cftype *cft = __d_cft(file->f_dentry); - struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent); + struct cgroup_subsys_state *css = cfe->css; - if (cgroup_is_dead(cgrp)) - return -ENODEV; if (cft->write) - return cft->write(cgrp, cft, file, buf, nbytes, ppos); + return cft->write(css, cft, file, buf, nbytes, ppos); if (cft->write_u64 || cft->write_s64) - return cgroup_write_X64(cgrp, cft, file, buf, nbytes, ppos); + return cgroup_write_X64(css, cft, file, buf, nbytes, ppos); if (cft->write_string) - return cgroup_write_string(cgrp, cft, file, buf, nbytes, ppos); + return cgroup_write_string(css, cft, file, buf, nbytes, ppos); if (cft->trigger) { - int ret = cft->trigger(cgrp, (unsigned int)cft->private); + int ret = cft->trigger(css, (unsigned int)cft->private); return ret ? ret : nbytes; } return -EINVAL; } -static ssize_t cgroup_read_u64(struct cgroup *cgrp, struct cftype *cft, - struct file *file, - char __user *buf, size_t nbytes, - loff_t *ppos) +static ssize_t cgroup_read_u64(struct cgroup_subsys_state *css, + struct cftype *cft, struct file *file, + char __user *buf, size_t nbytes, loff_t *ppos) { char tmp[CGROUP_LOCAL_BUFFER_SIZE]; - u64 val = cft->read_u64(cgrp, cft); + u64 val = cft->read_u64(css, cft); int len = sprintf(tmp, "%llu\n", (unsigned long long) val); return simple_read_from_buffer(buf, nbytes, ppos, tmp, len); } -static ssize_t cgroup_read_s64(struct cgroup *cgrp, struct cftype *cft, - struct file *file, - char __user *buf, size_t nbytes, - loff_t *ppos) +static ssize_t cgroup_read_s64(struct cgroup_subsys_state *css, + struct cftype *cft, struct file *file, + char __user *buf, size_t nbytes, loff_t *ppos) { char tmp[CGROUP_LOCAL_BUFFER_SIZE]; - s64 val = cft->read_s64(cgrp, cft); + s64 val = cft->read_s64(css, cft); int len = sprintf(tmp, "%lld\n", (long long) val); return simple_read_from_buffer(buf, nbytes, ppos, tmp, len); } static ssize_t cgroup_file_read(struct file *file, char __user *buf, - size_t nbytes, loff_t *ppos) + size_t nbytes, loff_t *ppos) { + struct cfent *cfe = __d_cfe(file->f_dentry); struct cftype *cft = __d_cft(file->f_dentry); - struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent); - - if (cgroup_is_dead(cgrp)) - return -ENODEV; + struct cgroup_subsys_state *css = cfe->css; if (cft->read) - return cft->read(cgrp, cft, file, buf, nbytes, ppos); + return cft->read(css, cft, file, buf, nbytes, ppos); if (cft->read_u64) - return cgroup_read_u64(cgrp, cft, file, buf, nbytes, ppos); + return cgroup_read_u64(css, cft, file, buf, nbytes, ppos); if (cft->read_s64) - return cgroup_read_s64(cgrp, cft, file, buf, nbytes, ppos); + return cgroup_read_s64(css, cft, file, buf, nbytes, ppos); return -EINVAL; } @@ -2440,11 +2436,6 @@ static ssize_t cgroup_file_read(struct file *file, char __user *buf, * supports string->u64 maps, but can be extended in future. */ -struct cgroup_seqfile_state { - struct cftype *cft; - struct cgroup *cgroup; -}; - static int cgroup_map_add(struct cgroup_map_cb *cb, const char *key, u64 value) { struct seq_file *sf = cb->state; @@ -2453,69 +2444,86 @@ static int cgroup_map_add(struct cgroup_map_cb *cb, const char *key, u64 value) static int cgroup_seqfile_show(struct seq_file *m, void *arg) { - struct cgroup_seqfile_state *state = m->private; - struct cftype *cft = state->cft; + struct cfent *cfe = m->private; + struct cftype *cft = cfe->type; + struct cgroup_subsys_state *css = cfe->css; + if (cft->read_map) { struct cgroup_map_cb cb = { .fill = cgroup_map_add, .state = m, }; - return cft->read_map(state->cgroup, cft, &cb); + return cft->read_map(css, cft, &cb); } - return cft->read_seq_string(state->cgroup, cft, m); -} - -static int cgroup_seqfile_release(struct inode *inode, struct file *file) -{ - struct seq_file *seq = file->private_data; - kfree(seq->private); - return single_release(inode, file); + return cft->read_seq_string(css, cft, m); } static const struct file_operations cgroup_seqfile_operations = { .read = seq_read, .write = cgroup_file_write, .llseek = seq_lseek, - .release = cgroup_seqfile_release, + .release = single_release, }; static int cgroup_file_open(struct inode *inode, struct file *file) { + struct cfent *cfe = __d_cfe(file->f_dentry); + struct cftype *cft = __d_cft(file->f_dentry); + struct cgroup *cgrp = __d_cgrp(cfe->dentry->d_parent); + struct cgroup_subsys_state *css; int err; - struct cftype *cft; err = generic_file_open(inode, file); if (err) return err; - cft = __d_cft(file->f_dentry); - if (cft->read_map || cft->read_seq_string) { - struct cgroup_seqfile_state *state; + /* + * If the file belongs to a subsystem, pin the css. Will be + * unpinned either on open failure or release. This ensures that + * @css stays alive for all file operations. + */ + rcu_read_lock(); + css = cgroup_css(cgrp, cft->ss); + if (cft->ss && !css_tryget(css)) + css = NULL; + rcu_read_unlock(); - state = kzalloc(sizeof(*state), GFP_USER); - if (!state) - return -ENOMEM; + if (!css) + return -ENODEV; + + /* + * @cfe->css is used by read/write/close to determine the + * associated css. @file->private_data would be a better place but + * that's already used by seqfile. Multiple accessors may use it + * simultaneously which is okay as the association never changes. + */ + WARN_ON_ONCE(cfe->css && cfe->css != css); + cfe->css = css; - state->cft = cft; - state->cgroup = __d_cgrp(file->f_dentry->d_parent); + if (cft->read_map || cft->read_seq_string) { file->f_op = &cgroup_seqfile_operations; - err = single_open(file, cgroup_seqfile_show, state); - if (err < 0) - kfree(state); - } else if (cft->open) + err = single_open(file, cgroup_seqfile_show, cfe); + } else if (cft->open) { err = cft->open(inode, file); - else - err = 0; + } + if (css->ss && err) + css_put(css); return err; } static int cgroup_file_release(struct inode *inode, struct file *file) { + struct cfent *cfe = __d_cfe(file->f_dentry); struct cftype *cft = __d_cft(file->f_dentry); + struct cgroup_subsys_state *css = cfe->css; + int ret = 0; + if (cft->release) - return cft->release(inode, file); - return 0; + ret = cft->release(inode, file); + if (css->ss) + css_put(css); + return ret; } /* @@ -2729,8 +2737,7 @@ static umode_t cgroup_file_mode(const struct cftype *cft) return mode; } -static int cgroup_add_file(struct cgroup *cgrp, struct cgroup_subsys *subsys, - struct cftype *cft) +static int cgroup_add_file(struct cgroup *cgrp, struct cftype *cft) { struct dentry *dir = cgrp->dentry; struct cgroup *parent = __d_cgrp(dir); @@ -2740,8 +2747,9 @@ static int cgroup_add_file(struct cgroup *cgrp, struct cgroup_subsys *subsys, umode_t mode; char name[MAX_CGROUP_TYPE_NAMELEN + MAX_CFTYPE_NAME + 2] = { 0 }; - if (subsys && !(cgrp->root->flags & CGRP_ROOT_NOPREFIX)) { - strcpy(name, subsys->name); + if (cft->ss && !(cft->flags & CFTYPE_NO_PREFIX) && + !(cgrp->root->flags & CGRP_ROOT_NOPREFIX)) { + strcpy(name, cft->ss->name); strcat(name, "."); } strcat(name, cft->name); @@ -2775,11 +2783,25 @@ out: return error; } -static int cgroup_addrm_files(struct cgroup *cgrp, struct cgroup_subsys *subsys, - struct cftype cfts[], bool is_add) +/** + * cgroup_addrm_files - add or remove files to a cgroup directory + * @cgrp: the target cgroup + * @cfts: array of cftypes to be added + * @is_add: whether to add or remove + * + * Depending on @is_add, add or remove files defined by @cfts on @cgrp. + * For removals, this function never fails. If addition fails, this + * function doesn't remove files already added. The caller is responsible + * for cleaning up. + */ +static int cgroup_addrm_files(struct cgroup *cgrp, struct cftype cfts[], + bool is_add) { struct cftype *cft; - int err, ret = 0; + int ret; + + lockdep_assert_held(&cgrp->dentry->d_inode->i_mutex); + lockdep_assert_held(&cgroup_mutex); for (cft = cfts; cft->name[0] != '\0'; cft++) { /* does cft->flags tell us to skip this file on @cgrp? */ @@ -2791,16 +2813,17 @@ static int cgroup_addrm_files(struct cgroup *cgrp, struct cgroup_subsys *subsys, continue; if (is_add) { - err = cgroup_add_file(cgrp, subsys, cft); - if (err) + ret = cgroup_add_file(cgrp, cft); + if (ret) { pr_warn("cgroup_addrm_files: failed to add %s, err=%d\n", - cft->name, err); - ret = err; + cft->name, ret); + return ret; + } } else { cgroup_rm_file(cgrp, cft); } } - return ret; + return 0; } static void cgroup_cfts_prepare(void) @@ -2809,28 +2832,30 @@ static void cgroup_cfts_prepare(void) /* * Thanks to the entanglement with vfs inode locking, we can't walk * the existing cgroups under cgroup_mutex and create files. - * Instead, we use cgroup_for_each_descendant_pre() and drop RCU - * read lock before calling cgroup_addrm_files(). + * Instead, we use css_for_each_descendant_pre() and drop RCU read + * lock before calling cgroup_addrm_files(). */ mutex_lock(&cgroup_mutex); } -static void cgroup_cfts_commit(struct cgroup_subsys *ss, - struct cftype *cfts, bool is_add) +static int cgroup_cfts_commit(struct cftype *cfts, bool is_add) __releases(&cgroup_mutex) { LIST_HEAD(pending); - struct cgroup *cgrp, *root = &ss->root->top_cgroup; + struct cgroup_subsys *ss = cfts[0].ss; + struct cgroup *root = &ss->root->top_cgroup; struct super_block *sb = ss->root->sb; struct dentry *prev = NULL; struct inode *inode; + struct cgroup_subsys_state *css; u64 update_before; + int ret = 0; /* %NULL @cfts indicates abort and don't bother if @ss isn't attached */ if (!cfts || ss->root == &cgroup_dummy_root || !atomic_inc_not_zero(&sb->s_active)) { mutex_unlock(&cgroup_mutex); - return; + return 0; } /* @@ -2842,17 +2867,11 @@ static void cgroup_cfts_commit(struct cgroup_subsys *ss, mutex_unlock(&cgroup_mutex); - /* @root always needs to be updated */ - inode = root->dentry->d_inode; - mutex_lock(&inode->i_mutex); - mutex_lock(&cgroup_mutex); - cgroup_addrm_files(root, ss, cfts, is_add); - mutex_unlock(&cgroup_mutex); - mutex_unlock(&inode->i_mutex); - /* add/rm files for all cgroups created before */ rcu_read_lock(); - cgroup_for_each_descendant_pre(cgrp, root) { + css_for_each_descendant_pre(css, cgroup_css(root, ss)) { + struct cgroup *cgrp = css->cgroup; + if (cgroup_is_dead(cgrp)) continue; @@ -2866,15 +2885,18 @@ static void cgroup_cfts_commit(struct cgroup_subsys *ss, mutex_lock(&inode->i_mutex); mutex_lock(&cgroup_mutex); if (cgrp->serial_nr < update_before && !cgroup_is_dead(cgrp)) - cgroup_addrm_files(cgrp, ss, cfts, is_add); + ret = cgroup_addrm_files(cgrp, cfts, is_add); mutex_unlock(&cgroup_mutex); mutex_unlock(&inode->i_mutex); rcu_read_lock(); + if (ret) + break; } rcu_read_unlock(); dput(prev); deactivate_super(sb); + return ret; } /** @@ -2894,49 +2916,56 @@ static void cgroup_cfts_commit(struct cgroup_subsys *ss, int cgroup_add_cftypes(struct cgroup_subsys *ss, struct cftype *cfts) { struct cftype_set *set; + struct cftype *cft; + int ret; set = kzalloc(sizeof(*set), GFP_KERNEL); if (!set) return -ENOMEM; + for (cft = cfts; cft->name[0] != '\0'; cft++) + cft->ss = ss; + cgroup_cfts_prepare(); set->cfts = cfts; list_add_tail(&set->node, &ss->cftsets); - cgroup_cfts_commit(ss, cfts, true); - - return 0; + ret = cgroup_cfts_commit(cfts, true); + if (ret) + cgroup_rm_cftypes(cfts); + return ret; } EXPORT_SYMBOL_GPL(cgroup_add_cftypes); /** * cgroup_rm_cftypes - remove an array of cftypes from a subsystem - * @ss: target cgroup subsystem * @cfts: zero-length name terminated array of cftypes * - * Unregister @cfts from @ss. Files described by @cfts are removed from - * all existing cgroups to which @ss is attached and all future cgroups - * won't have them either. This function can be called anytime whether @ss - * is attached or not. + * Unregister @cfts. Files described by @cfts are removed from all + * existing cgroups and all future cgroups won't have them either. This + * function can be called anytime whether @cfts' subsys is attached or not. * * Returns 0 on successful unregistration, -ENOENT if @cfts is not - * registered with @ss. + * registered. */ -int cgroup_rm_cftypes(struct cgroup_subsys *ss, struct cftype *cfts) +int cgroup_rm_cftypes(struct cftype *cfts) { struct cftype_set *set; + if (!cfts || !cfts[0].ss) + return -ENOENT; + cgroup_cfts_prepare(); - list_for_each_entry(set, &ss->cftsets, node) { + list_for_each_entry(set, &cfts[0].ss->cftsets, node) { if (set->cfts == cfts) { list_del(&set->node); kfree(set); - cgroup_cfts_commit(ss, cfts, false); + cgroup_cfts_commit(cfts, false); return 0; } } - cgroup_cfts_commit(ss, NULL, false); + cgroup_cfts_commit(NULL, false); return -ENOENT; } @@ -2959,34 +2988,10 @@ int cgroup_task_count(const struct cgroup *cgrp) } /* - * Advance a list_head iterator. The iterator should be positioned at - * the start of a css_set - */ -static void cgroup_advance_iter(struct cgroup *cgrp, struct cgroup_iter *it) -{ - struct list_head *l = it->cset_link; - struct cgrp_cset_link *link; - struct css_set *cset; - - /* Advance to the next non-empty css_set */ - do { - l = l->next; - if (l == &cgrp->cset_links) { - it->cset_link = NULL; - return; - } - link = list_entry(l, struct cgrp_cset_link, cset_link); - cset = link->cset; - } while (list_empty(&cset->tasks)); - it->cset_link = l; - it->task = cset->tasks.next; -} - -/* - * To reduce the fork() overhead for systems that are not actually - * using their cgroups capability, we don't maintain the lists running - * through each css_set to its tasks until we see the list actually - * used - in other words after the first call to cgroup_iter_start(). + * To reduce the fork() overhead for systems that are not actually using + * their cgroups capability, we don't maintain the lists running through + * each css_set to its tasks until we see the list actually used - in other + * words after the first call to css_task_iter_start(). */ static void cgroup_enable_task_cg_lists(void) { @@ -3017,16 +3022,21 @@ static void cgroup_enable_task_cg_lists(void) } /** - * cgroup_next_sibling - find the next sibling of a given cgroup - * @pos: the current cgroup + * css_next_child - find the next child of a given css + * @pos_css: the current position (%NULL to initiate traversal) + * @parent_css: css whose children to walk * - * This function returns the next sibling of @pos and should be called - * under RCU read lock. The only requirement is that @pos is accessible. - * The next sibling is guaranteed to be returned regardless of @pos's - * state. + * This function returns the next child of @parent_css and should be called + * under RCU read lock. The only requirement is that @parent_css and + * @pos_css are accessible. The next sibling is guaranteed to be returned + * regardless of their states. */ -struct cgroup *cgroup_next_sibling(struct cgroup *pos) +struct cgroup_subsys_state * +css_next_child(struct cgroup_subsys_state *pos_css, + struct cgroup_subsys_state *parent_css) { + struct cgroup *pos = pos_css ? pos_css->cgroup : NULL; + struct cgroup *cgrp = parent_css->cgroup; struct cgroup *next; WARN_ON_ONCE(!rcu_read_lock_held()); @@ -3041,78 +3051,81 @@ struct cgroup *cgroup_next_sibling(struct cgroup *pos) * safe to dereference from this RCU critical section. If * ->sibling.next is inaccessible, cgroup_is_dead() is guaranteed * to be visible as %true here. + * + * If @pos is dead, its next pointer can't be dereferenced; + * however, as each cgroup is given a monotonically increasing + * unique serial number and always appended to the sibling list, + * the next one can be found by walking the parent's children until + * we see a cgroup with higher serial number than @pos's. While + * this path can be slower, it's taken only when either the current + * cgroup is removed or iteration and removal race. */ - if (likely(!cgroup_is_dead(pos))) { + if (!pos) { + next = list_entry_rcu(cgrp->children.next, struct cgroup, sibling); + } else if (likely(!cgroup_is_dead(pos))) { next = list_entry_rcu(pos->sibling.next, struct cgroup, sibling); - if (&next->sibling != &pos->parent->children) - return next; - return NULL; + } else { + list_for_each_entry_rcu(next, &cgrp->children, sibling) + if (next->serial_nr > pos->serial_nr) + break; } - /* - * Can't dereference the next pointer. Each cgroup is given a - * monotonically increasing unique serial number and always - * appended to the sibling list, so the next one can be found by - * walking the parent's children until we see a cgroup with higher - * serial number than @pos's. - * - * While this path can be slow, it's taken only when either the - * current cgroup is removed or iteration and removal race. - */ - list_for_each_entry_rcu(next, &pos->parent->children, sibling) - if (next->serial_nr > pos->serial_nr) - return next; - return NULL; + if (&next->sibling == &cgrp->children) + return NULL; + + return cgroup_css(next, parent_css->ss); } -EXPORT_SYMBOL_GPL(cgroup_next_sibling); +EXPORT_SYMBOL_GPL(css_next_child); /** - * cgroup_next_descendant_pre - find the next descendant for pre-order walk + * css_next_descendant_pre - find the next descendant for pre-order walk * @pos: the current position (%NULL to initiate traversal) - * @cgroup: cgroup whose descendants to walk + * @root: css whose descendants to walk * - * To be used by cgroup_for_each_descendant_pre(). Find the next - * descendant to visit for pre-order traversal of @cgroup's descendants. + * To be used by css_for_each_descendant_pre(). Find the next descendant + * to visit for pre-order traversal of @root's descendants. @root is + * included in the iteration and the first node to be visited. * * While this function requires RCU read locking, it doesn't require the * whole traversal to be contained in a single RCU critical section. This * function will return the correct next descendant as long as both @pos - * and @cgroup are accessible and @pos is a descendant of @cgroup. + * and @root are accessible and @pos is a descendant of @root. */ -struct cgroup *cgroup_next_descendant_pre(struct cgroup *pos, - struct cgroup *cgroup) +struct cgroup_subsys_state * +css_next_descendant_pre(struct cgroup_subsys_state *pos, + struct cgroup_subsys_state *root) { - struct cgroup *next; + struct cgroup_subsys_state *next; WARN_ON_ONCE(!rcu_read_lock_held()); - /* if first iteration, pretend we just visited @cgroup */ + /* if first iteration, visit @root */ if (!pos) - pos = cgroup; + return root; /* visit the first child if exists */ - next = list_first_or_null_rcu(&pos->children, struct cgroup, sibling); + next = css_next_child(NULL, pos); if (next) return next; /* no child, visit my or the closest ancestor's next sibling */ - while (pos != cgroup) { - next = cgroup_next_sibling(pos); + while (pos != root) { + next = css_next_child(pos, css_parent(pos)); if (next) return next; - pos = pos->parent; + pos = css_parent(pos); } return NULL; } -EXPORT_SYMBOL_GPL(cgroup_next_descendant_pre); +EXPORT_SYMBOL_GPL(css_next_descendant_pre); /** - * cgroup_rightmost_descendant - return the rightmost descendant of a cgroup - * @pos: cgroup of interest + * css_rightmost_descendant - return the rightmost descendant of a css + * @pos: css of interest * - * Return the rightmost descendant of @pos. If there's no descendant, - * @pos is returned. This can be used during pre-order traversal to skip + * Return the rightmost descendant of @pos. If there's no descendant, @pos + * is returned. This can be used during pre-order traversal to skip * subtree of @pos. * * While this function requires RCU read locking, it doesn't require the @@ -3120,9 +3133,10 @@ EXPORT_SYMBOL_GPL(cgroup_next_descendant_pre); * function will return the correct rightmost descendant as long as @pos is * accessible. */ -struct cgroup *cgroup_rightmost_descendant(struct cgroup *pos) +struct cgroup_subsys_state * +css_rightmost_descendant(struct cgroup_subsys_state *pos) { - struct cgroup *last, *tmp; + struct cgroup_subsys_state *last, *tmp; WARN_ON_ONCE(!rcu_read_lock_held()); @@ -3130,82 +3144,138 @@ struct cgroup *cgroup_rightmost_descendant(struct cgroup *pos) last = pos; /* ->prev isn't RCU safe, walk ->next till the end */ pos = NULL; - list_for_each_entry_rcu(tmp, &last->children, sibling) + css_for_each_child(tmp, last) pos = tmp; } while (pos); return last; } -EXPORT_SYMBOL_GPL(cgroup_rightmost_descendant); +EXPORT_SYMBOL_GPL(css_rightmost_descendant); -static struct cgroup *cgroup_leftmost_descendant(struct cgroup *pos) +static struct cgroup_subsys_state * +css_leftmost_descendant(struct cgroup_subsys_state *pos) { - struct cgroup *last; + struct cgroup_subsys_state *last; do { last = pos; - pos = list_first_or_null_rcu(&pos->children, struct cgroup, - sibling); + pos = css_next_child(NULL, pos); } while (pos); return last; } /** - * cgroup_next_descendant_post - find the next descendant for post-order walk + * css_next_descendant_post - find the next descendant for post-order walk * @pos: the current position (%NULL to initiate traversal) - * @cgroup: cgroup whose descendants to walk + * @root: css whose descendants to walk * - * To be used by cgroup_for_each_descendant_post(). Find the next - * descendant to visit for post-order traversal of @cgroup's descendants. + * To be used by css_for_each_descendant_post(). Find the next descendant + * to visit for post-order traversal of @root's descendants. @root is + * included in the iteration and the last node to be visited. * * While this function requires RCU read locking, it doesn't require the * whole traversal to be contained in a single RCU critical section. This * function will return the correct next descendant as long as both @pos * and @cgroup are accessible and @pos is a descendant of @cgroup. */ -struct cgroup *cgroup_next_descendant_post(struct cgroup *pos, - struct cgroup *cgroup) +struct cgroup_subsys_state * +css_next_descendant_post(struct cgroup_subsys_state *pos, + struct cgroup_subsys_state *root) { - struct cgroup *next; + struct cgroup_subsys_state *next; WARN_ON_ONCE(!rcu_read_lock_held()); /* if first iteration, visit the leftmost descendant */ if (!pos) { - next = cgroup_leftmost_descendant(cgroup); - return next != cgroup ? next : NULL; + next = css_leftmost_descendant(root); + return next != root ? next : NULL; } + /* if we visited @root, we're done */ + if (pos == root) + return NULL; + /* if there's an unvisited sibling, visit its leftmost descendant */ - next = cgroup_next_sibling(pos); + next = css_next_child(pos, css_parent(pos)); if (next) - return cgroup_leftmost_descendant(next); + return css_leftmost_descendant(next); /* no sibling left, visit parent */ - next = pos->parent; - return next != cgroup ? next : NULL; + return css_parent(pos); +} +EXPORT_SYMBOL_GPL(css_next_descendant_post); + +/** + * css_advance_task_iter - advance a task itererator to the next css_set + * @it: the iterator to advance + * + * Advance @it to the next css_set to walk. + */ +static void css_advance_task_iter(struct css_task_iter *it) +{ + struct list_head *l = it->cset_link; + struct cgrp_cset_link *link; + struct css_set *cset; + + /* Advance to the next non-empty css_set */ + do { + l = l->next; + if (l == &it->origin_css->cgroup->cset_links) { + it->cset_link = NULL; + return; + } + link = list_entry(l, struct cgrp_cset_link, cset_link); + cset = link->cset; + } while (list_empty(&cset->tasks)); + it->cset_link = l; + it->task = cset->tasks.next; } -EXPORT_SYMBOL_GPL(cgroup_next_descendant_post); -void cgroup_iter_start(struct cgroup *cgrp, struct cgroup_iter *it) +/** + * css_task_iter_start - initiate task iteration + * @css: the css to walk tasks of + * @it: the task iterator to use + * + * Initiate iteration through the tasks of @css. The caller can call + * css_task_iter_next() to walk through the tasks until the function + * returns NULL. On completion of iteration, css_task_iter_end() must be + * called. + * + * Note that this function acquires a lock which is released when the + * iteration finishes. The caller can't sleep while iteration is in + * progress. + */ +void css_task_iter_start(struct cgroup_subsys_state *css, + struct css_task_iter *it) __acquires(css_set_lock) { /* - * The first time anyone tries to iterate across a cgroup, - * we need to enable the list linking each css_set to its - * tasks, and fix up all existing tasks. + * The first time anyone tries to iterate across a css, we need to + * enable the list linking each css_set to its tasks, and fix up + * all existing tasks. */ if (!use_task_css_set_links) cgroup_enable_task_cg_lists(); read_lock(&css_set_lock); - it->cset_link = &cgrp->cset_links; - cgroup_advance_iter(cgrp, it); + + it->origin_css = css; + it->cset_link = &css->cgroup->cset_links; + + css_advance_task_iter(it); } -struct task_struct *cgroup_iter_next(struct cgroup *cgrp, - struct cgroup_iter *it) +/** + * css_task_iter_next - return the next task for the iterator + * @it: the task iterator being iterated + * + * The "next" function for task iteration. @it should have been + * initialized via css_task_iter_start(). Returns NULL when the iteration + * reaches the end. + */ +struct task_struct *css_task_iter_next(struct css_task_iter *it) { struct task_struct *res; struct list_head *l = it->task; @@ -3219,16 +3289,24 @@ struct task_struct *cgroup_iter_next(struct cgroup *cgrp, l = l->next; link = list_entry(it->cset_link, struct cgrp_cset_link, cset_link); if (l == &link->cset->tasks) { - /* We reached the end of this task list - move on to - * the next cg_cgroup_link */ - cgroup_advance_iter(cgrp, it); + /* + * We reached the end of this task list - move on to the + * next cgrp_cset_link. + */ + css_advance_task_iter(it); } else { it->task = l; } return res; } -void cgroup_iter_end(struct cgroup *cgrp, struct cgroup_iter *it) +/** + * css_task_iter_end - finish task iteration + * @it: the task iterator to finish + * + * Finish task iteration started by css_task_iter_start(). + */ +void css_task_iter_end(struct css_task_iter *it) __releases(css_set_lock) { read_unlock(&css_set_lock); @@ -3269,46 +3347,49 @@ static inline int started_after(void *p1, void *p2) } /** - * cgroup_scan_tasks - iterate though all the tasks in a cgroup - * @scan: struct cgroup_scanner containing arguments for the scan + * css_scan_tasks - iterate though all the tasks in a css + * @css: the css to iterate tasks of + * @test: optional test callback + * @process: process callback + * @data: data passed to @test and @process + * @heap: optional pre-allocated heap used for task iteration + * + * Iterate through all the tasks in @css, calling @test for each, and if it + * returns %true, call @process for it also. * - * Arguments include pointers to callback functions test_task() and - * process_task(). - * Iterate through all the tasks in a cgroup, calling test_task() for each, - * and if it returns true, call process_task() for it also. - * The test_task pointer may be NULL, meaning always true (select all tasks). - * Effectively duplicates cgroup_iter_{start,next,end}() - * but does not lock css_set_lock for the call to process_task(). - * The struct cgroup_scanner may be embedded in any structure of the caller's - * creation. - * It is guaranteed that process_task() will act on every task that - * is a member of the cgroup for the duration of this call. This - * function may or may not call process_task() for tasks that exit - * or move to a different cgroup during the call, or are forked or - * move into the cgroup during the call. + * @test may be NULL, meaning always true (select all tasks), which + * effectively duplicates css_task_iter_{start,next,end}() but does not + * lock css_set_lock for the call to @process. * - * Note that test_task() may be called with locks held, and may in some - * situations be called multiple times for the same task, so it should - * be cheap. - * If the heap pointer in the struct cgroup_scanner is non-NULL, a heap has been - * pre-allocated and will be used for heap operations (and its "gt" member will - * be overwritten), else a temporary heap will be used (allocation of which - * may cause this function to fail). + * It is guaranteed that @process will act on every task that is a member + * of @css for the duration of this call. This function may or may not + * call @process for tasks that exit or move to a different css during the + * call, or are forked or move into the css during the call. + * + * Note that @test may be called with locks held, and may in some + * situations be called multiple times for the same task, so it should be + * cheap. + * + * If @heap is non-NULL, a heap has been pre-allocated and will be used for + * heap operations (and its "gt" member will be overwritten), else a + * temporary heap will be used (allocation of which may cause this function + * to fail). */ -int cgroup_scan_tasks(struct cgroup_scanner *scan) +int css_scan_tasks(struct cgroup_subsys_state *css, + bool (*test)(struct task_struct *, void *), + void (*process)(struct task_struct *, void *), + void *data, struct ptr_heap *heap) { int retval, i; - struct cgroup_iter it; + struct css_task_iter it; struct task_struct *p, *dropped; /* Never dereference latest_task, since it's not refcounted */ struct task_struct *latest_task = NULL; struct ptr_heap tmp_heap; - struct ptr_heap *heap; struct timespec latest_time = { 0, 0 }; - if (scan->heap) { + if (heap) { /* The caller supplied our heap and pre-allocated its memory */ - heap = scan->heap; heap->gt = &started_after; } else { /* We need to allocate our own heap memory */ @@ -3321,25 +3402,24 @@ int cgroup_scan_tasks(struct cgroup_scanner *scan) again: /* - * Scan tasks in the cgroup, using the scanner's "test_task" callback - * to determine which are of interest, and using the scanner's - * "process_task" callback to process any of them that need an update. - * Since we don't want to hold any locks during the task updates, - * gather tasks to be processed in a heap structure. - * The heap is sorted by descending task start time. - * If the statically-sized heap fills up, we overflow tasks that - * started later, and in future iterations only consider tasks that - * started after the latest task in the previous pass. This + * Scan tasks in the css, using the @test callback to determine + * which are of interest, and invoking @process callback on the + * ones which need an update. Since we don't want to hold any + * locks during the task updates, gather tasks to be processed in a + * heap structure. The heap is sorted by descending task start + * time. If the statically-sized heap fills up, we overflow tasks + * that started later, and in future iterations only consider tasks + * that started after the latest task in the previous pass. This * guarantees forward progress and that we don't miss any tasks. */ heap->size = 0; - cgroup_iter_start(scan->cg, &it); - while ((p = cgroup_iter_next(scan->cg, &it))) { + css_task_iter_start(css, &it); + while ((p = css_task_iter_next(&it))) { /* * Only affect tasks that qualify per the caller's callback, * if he provided one */ - if (scan->test_task && !scan->test_task(p, scan)) + if (test && !test(p, data)) continue; /* * Only process tasks that started after the last task @@ -3367,7 +3447,7 @@ int cgroup_scan_tasks(struct cgroup_scanner *scan) * the heap and wasn't inserted */ } - cgroup_iter_end(scan->cg, &it); + css_task_iter_end(&it); if (heap->size) { for (i = 0; i < heap->size; i++) { @@ -3377,7 +3457,7 @@ int cgroup_scan_tasks(struct cgroup_scanner *scan) latest_task = q; } /* Process the task per the caller's callback */ - scan->process_task(q, scan); + process(q, data); put_task_struct(q); } /* @@ -3394,10 +3474,9 @@ int cgroup_scan_tasks(struct cgroup_scanner *scan) return 0; } -static void cgroup_transfer_one_task(struct task_struct *task, - struct cgroup_scanner *scan) +static void cgroup_transfer_one_task(struct task_struct *task, void *data) { - struct cgroup *new_cgroup = scan->data; + struct cgroup *new_cgroup = data; mutex_lock(&cgroup_mutex); cgroup_attach_task(new_cgroup, task, false); @@ -3411,15 +3490,8 @@ static void cgroup_transfer_one_task(struct task_struct *task, */ int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from) { - struct cgroup_scanner scan; - - scan.cg = from; - scan.test_task = NULL; /* select all tasks in cgroup */ - scan.process_task = cgroup_transfer_one_task; - scan.heap = NULL; - scan.data = to; - - return cgroup_scan_tasks(&scan); + return css_scan_tasks(&from->dummy_css, NULL, cgroup_transfer_one_task, + to, NULL); } /* @@ -3461,7 +3533,7 @@ struct cgroup_pidlist { /* pointer to the cgroup we belong to, for list removal purposes */ struct cgroup *owner; /* protects the other fields */ - struct rw_semaphore mutex; + struct rw_semaphore rwsem; }; /* @@ -3534,7 +3606,7 @@ static struct cgroup_pidlist *cgroup_pidlist_find(struct cgroup *cgrp, struct pid_namespace *ns = task_active_pid_ns(current); /* - * We can't drop the pidlist_mutex before taking the l->mutex in case + * We can't drop the pidlist_mutex before taking the l->rwsem in case * the last ref-holder is trying to remove l from the list at the same * time. Holding the pidlist_mutex precludes somebody taking whichever * list we find out from under us - compare release_pid_array(). @@ -3543,7 +3615,7 @@ static struct cgroup_pidlist *cgroup_pidlist_find(struct cgroup *cgrp, list_for_each_entry(l, &cgrp->pidlists, links) { if (l->key.type == type && l->key.ns == ns) { /* make sure l doesn't vanish out from under us */ - down_write(&l->mutex); + down_write(&l->rwsem); mutex_unlock(&cgrp->pidlist_mutex); return l; } @@ -3554,8 +3626,8 @@ static struct cgroup_pidlist *cgroup_pidlist_find(struct cgroup *cgrp, mutex_unlock(&cgrp->pidlist_mutex); return l; } - init_rwsem(&l->mutex); - down_write(&l->mutex); + init_rwsem(&l->rwsem); + down_write(&l->rwsem); l->key.type = type; l->key.ns = get_pid_ns(ns); l->owner = cgrp; @@ -3573,7 +3645,7 @@ static int pidlist_array_load(struct cgroup *cgrp, enum cgroup_filetype type, pid_t *array; int length; int pid, n = 0; /* used for populating the array */ - struct cgroup_iter it; + struct css_task_iter it; struct task_struct *tsk; struct cgroup_pidlist *l; @@ -3588,8 +3660,8 @@ static int pidlist_array_load(struct cgroup *cgrp, enum cgroup_filetype type, if (!array) return -ENOMEM; /* now, populate the array */ - cgroup_iter_start(cgrp, &it); - while ((tsk = cgroup_iter_next(cgrp, &it))) { + css_task_iter_start(&cgrp->dummy_css, &it); + while ((tsk = css_task_iter_next(&it))) { if (unlikely(n == length)) break; /* get tgid or pid for procs or tasks file respectively */ @@ -3600,7 +3672,7 @@ static int pidlist_array_load(struct cgroup *cgrp, enum cgroup_filetype type, if (pid > 0) /* make sure to only use valid results */ array[n++] = pid; } - cgroup_iter_end(cgrp, &it); + css_task_iter_end(&it); length = n; /* now sort & (if procs) strip out duplicates */ sort(array, length, sizeof(pid_t), cmppid, NULL); @@ -3616,7 +3688,7 @@ static int pidlist_array_load(struct cgroup *cgrp, enum cgroup_filetype type, l->list = array; l->length = length; l->use_count++; - up_write(&l->mutex); + up_write(&l->rwsem); *lp = l; return 0; } @@ -3634,7 +3706,7 @@ int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry) { int ret = -EINVAL; struct cgroup *cgrp; - struct cgroup_iter it; + struct css_task_iter it; struct task_struct *tsk; /* @@ -3648,8 +3720,8 @@ int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry) ret = 0; cgrp = dentry->d_fsdata; - cgroup_iter_start(cgrp, &it); - while ((tsk = cgroup_iter_next(cgrp, &it))) { + css_task_iter_start(&cgrp->dummy_css, &it); + while ((tsk = css_task_iter_next(&it))) { switch (tsk->state) { case TASK_RUNNING: stats->nr_running++; @@ -3669,7 +3741,7 @@ int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry) break; } } - cgroup_iter_end(cgrp, &it); + css_task_iter_end(&it); err: return ret; @@ -3694,7 +3766,7 @@ static void *cgroup_pidlist_start(struct seq_file *s, loff_t *pos) int index = 0, pid = *pos; int *iter; - down_read(&l->mutex); + down_read(&l->rwsem); if (pid) { int end = l->length; @@ -3721,7 +3793,7 @@ static void *cgroup_pidlist_start(struct seq_file *s, loff_t *pos) static void cgroup_pidlist_stop(struct seq_file *s, void *v) { struct cgroup_pidlist *l = s->private; - up_read(&l->mutex); + up_read(&l->rwsem); } static void *cgroup_pidlist_next(struct seq_file *s, void *v, loff_t *pos) @@ -3767,7 +3839,7 @@ static void cgroup_release_pid_array(struct cgroup_pidlist *l) * pidlist_mutex, we have to take pidlist_mutex first. */ mutex_lock(&l->owner->pidlist_mutex); - down_write(&l->mutex); + down_write(&l->rwsem); BUG_ON(!l->use_count); if (!--l->use_count) { /* we're the last user if refcount is 0; remove and free */ @@ -3775,12 +3847,12 @@ static void cgroup_release_pid_array(struct cgroup_pidlist *l) mutex_unlock(&l->owner->pidlist_mutex); pidlist_free(l->list); put_pid_ns(l->key.ns); - up_write(&l->mutex); + up_write(&l->rwsem); kfree(l); return; } mutex_unlock(&l->owner->pidlist_mutex); - up_write(&l->mutex); + up_write(&l->rwsem); } static int cgroup_pidlist_release(struct inode *inode, struct file *file) @@ -3844,21 +3916,20 @@ static int cgroup_procs_open(struct inode *unused, struct file *file) return cgroup_pidlist_open(file, CGROUP_FILE_PROCS); } -static u64 cgroup_read_notify_on_release(struct cgroup *cgrp, - struct cftype *cft) +static u64 cgroup_read_notify_on_release(struct cgroup_subsys_state *css, + struct cftype *cft) { - return notify_on_release(cgrp); + return notify_on_release(css->cgroup); } -static int cgroup_write_notify_on_release(struct cgroup *cgrp, - struct cftype *cft, - u64 val) +static int cgroup_write_notify_on_release(struct cgroup_subsys_state *css, + struct cftype *cft, u64 val) { - clear_bit(CGRP_RELEASABLE, &cgrp->flags); + clear_bit(CGRP_RELEASABLE, &css->cgroup->flags); if (val) - set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); + set_bit(CGRP_NOTIFY_ON_RELEASE, &css->cgroup->flags); else - clear_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); + clear_bit(CGRP_NOTIFY_ON_RELEASE, &css->cgroup->flags); return 0; } @@ -3888,18 +3959,18 @@ static void cgroup_event_remove(struct work_struct *work) { struct cgroup_event *event = container_of(work, struct cgroup_event, remove); - struct cgroup *cgrp = event->cgrp; + struct cgroup_subsys_state *css = event->css; remove_wait_queue(event->wqh, &event->wait); - event->cft->unregister_event(cgrp, event->cft, event->eventfd); + event->cft->unregister_event(css, event->cft, event->eventfd); /* Notify userspace the event is going away. */ eventfd_signal(event->eventfd, 1); eventfd_ctx_put(event->eventfd); kfree(event); - cgroup_dput(cgrp); + css_put(css); } /* @@ -3912,7 +3983,7 @@ static int cgroup_event_wake(wait_queue_t *wait, unsigned mode, { struct cgroup_event *event = container_of(wait, struct cgroup_event, wait); - struct cgroup *cgrp = event->cgrp; + struct cgroup *cgrp = event->css->cgroup; unsigned long flags = (unsigned long)key; if (flags & POLLHUP) { @@ -3956,14 +4027,15 @@ static void cgroup_event_ptable_queue_proc(struct file *file, * Input must be in format '<event_fd> <control_fd> <args>'. * Interpretation of args is defined by control file implementation. */ -static int cgroup_write_event_control(struct cgroup *cgrp, struct cftype *cft, - const char *buffer) +static int cgroup_write_event_control(struct cgroup_subsys_state *dummy_css, + struct cftype *cft, const char *buffer) { - struct cgroup_event *event = NULL; - struct cgroup *cgrp_cfile; + struct cgroup *cgrp = dummy_css->cgroup; + struct cgroup_event *event; + struct cgroup_subsys_state *cfile_css; unsigned int efd, cfd; - struct file *efile = NULL; - struct file *cfile = NULL; + struct file *efile; + struct file *cfile; char *endp; int ret; @@ -3980,7 +4052,7 @@ static int cgroup_write_event_control(struct cgroup *cgrp, struct cftype *cft, event = kzalloc(sizeof(*event), GFP_KERNEL); if (!event) return -ENOMEM; - event->cgrp = cgrp; + INIT_LIST_HEAD(&event->list); init_poll_funcptr(&event->pt, cgroup_event_ptable_queue_proc); init_waitqueue_func_entry(&event->wait, cgroup_event_wake); @@ -3989,62 +4061,68 @@ static int cgroup_write_event_control(struct cgroup *cgrp, struct cftype *cft, efile = eventfd_fget(efd); if (IS_ERR(efile)) { ret = PTR_ERR(efile); - goto fail; + goto out_kfree; } event->eventfd = eventfd_ctx_fileget(efile); if (IS_ERR(event->eventfd)) { ret = PTR_ERR(event->eventfd); - goto fail; + goto out_put_efile; } cfile = fget(cfd); if (!cfile) { ret = -EBADF; - goto fail; + goto out_put_eventfd; } /* the process need read permission on control file */ /* AV: shouldn't we check that it's been opened for read instead? */ ret = inode_permission(file_inode(cfile), MAY_READ); if (ret < 0) - goto fail; + goto out_put_cfile; event->cft = __file_cft(cfile); if (IS_ERR(event->cft)) { ret = PTR_ERR(event->cft); - goto fail; + goto out_put_cfile; + } + + if (!event->cft->ss) { + ret = -EBADF; + goto out_put_cfile; } /* - * The file to be monitored must be in the same cgroup as - * cgroup.event_control is. + * Determine the css of @cfile, verify it belongs to the same + * cgroup as cgroup.event_control, and associate @event with it. + * Remaining events are automatically removed on cgroup destruction + * but the removal is asynchronous, so take an extra ref. */ - cgrp_cfile = __d_cgrp(cfile->f_dentry->d_parent); - if (cgrp_cfile != cgrp) { - ret = -EINVAL; - goto fail; - } + rcu_read_lock(); + + ret = -EINVAL; + event->css = cgroup_css(cgrp, event->cft->ss); + cfile_css = css_from_dir(cfile->f_dentry->d_parent, event->cft->ss); + if (event->css && event->css == cfile_css && css_tryget(event->css)) + ret = 0; + + rcu_read_unlock(); + if (ret) + goto out_put_cfile; if (!event->cft->register_event || !event->cft->unregister_event) { ret = -EINVAL; - goto fail; + goto out_put_css; } - ret = event->cft->register_event(cgrp, event->cft, + ret = event->cft->register_event(event->css, event->cft, event->eventfd, buffer); if (ret) - goto fail; + goto out_put_css; efile->f_op->poll(efile, &event->pt); - /* - * Events should be removed after rmdir of cgroup directory, but before - * destroying subsystem state objects. Let's take reference to cgroup - * directory dentry to do that. - */ - dget(cgrp->dentry); - spin_lock(&cgrp->event_list_lock); list_add(&event->list, &cgrp->event_list); spin_unlock(&cgrp->event_list_lock); @@ -4054,35 +4132,33 @@ static int cgroup_write_event_control(struct cgroup *cgrp, struct cftype *cft, return 0; -fail: - if (cfile) - fput(cfile); - - if (event && event->eventfd && !IS_ERR(event->eventfd)) - eventfd_ctx_put(event->eventfd); - - if (!IS_ERR_OR_NULL(efile)) - fput(efile); - +out_put_css: + css_put(event->css); +out_put_cfile: + fput(cfile); +out_put_eventfd: + eventfd_ctx_put(event->eventfd); +out_put_efile: + fput(efile); +out_kfree: kfree(event); return ret; } -static u64 cgroup_clone_children_read(struct cgroup *cgrp, - struct cftype *cft) +static u64 cgroup_clone_children_read(struct cgroup_subsys_state *css, + struct cftype *cft) { - return test_bit(CGRP_CPUSET_CLONE_CHILDREN, &cgrp->flags); + return test_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags); } -static int cgroup_clone_children_write(struct cgroup *cgrp, - struct cftype *cft, - u64 val) +static int cgroup_clone_children_write(struct cgroup_subsys_state *css, + struct cftype *cft, u64 val) { if (val) - set_bit(CGRP_CPUSET_CLONE_CHILDREN, &cgrp->flags); + set_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags); else - clear_bit(CGRP_CPUSET_CLONE_CHILDREN, &cgrp->flags); + clear_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags); return 0; } @@ -4141,36 +4217,34 @@ static struct cftype cgroup_base_files[] = { }; /** - * cgroup_populate_dir - selectively creation of files in a directory + * cgroup_populate_dir - create subsys files in a cgroup directory * @cgrp: target cgroup - * @base_files: true if the base files should be added * @subsys_mask: mask of the subsystem ids whose files should be added + * + * On failure, no file is added. */ -static int cgroup_populate_dir(struct cgroup *cgrp, bool base_files, - unsigned long subsys_mask) +static int cgroup_populate_dir(struct cgroup *cgrp, unsigned long subsys_mask) { - int err; struct cgroup_subsys *ss; - - if (base_files) { - err = cgroup_addrm_files(cgrp, NULL, cgroup_base_files, true); - if (err < 0) - return err; - } + int i, ret = 0; /* process cftsets of each subsystem */ - for_each_root_subsys(cgrp->root, ss) { + for_each_subsys(ss, i) { struct cftype_set *set; - if (!test_bit(ss->subsys_id, &subsys_mask)) + + if (!test_bit(i, &subsys_mask)) continue; - list_for_each_entry(set, &ss->cftsets, node) - cgroup_addrm_files(cgrp, ss, set->cfts, true); + list_for_each_entry(set, &ss->cftsets, node) { + ret = cgroup_addrm_files(cgrp, set->cfts, true); + if (ret < 0) + goto err; + } } /* This cgroup is ready now */ for_each_root_subsys(cgrp->root, ss) { - struct cgroup_subsys_state *css = cgrp->subsys[ss->subsys_id]; + struct cgroup_subsys_state *css = cgroup_css(cgrp, ss); struct css_id *id = rcu_dereference_protected(css->id, true); /* @@ -4183,14 +4257,57 @@ static int cgroup_populate_dir(struct cgroup *cgrp, bool base_files, } return 0; +err: + cgroup_clear_dir(cgrp, subsys_mask); + return ret; } -static void css_dput_fn(struct work_struct *work) +/* + * css destruction is four-stage process. + * + * 1. Destruction starts. Killing of the percpu_ref is initiated. + * Implemented in kill_css(). + * + * 2. When the percpu_ref is confirmed to be visible as killed on all CPUs + * and thus css_tryget() is guaranteed to fail, the css can be offlined + * by invoking offline_css(). After offlining, the base ref is put. + * Implemented in css_killed_work_fn(). + * + * 3. When the percpu_ref reaches zero, the only possible remaining + * accessors are inside RCU read sections. css_release() schedules the + * RCU callback. + * + * 4. After the grace period, the css can be freed. Implemented in + * css_free_work_fn(). + * + * It is actually hairier because both step 2 and 4 require process context + * and thus involve punting to css->destroy_work adding two additional + * steps to the already complex sequence. + */ +static void css_free_work_fn(struct work_struct *work) { struct cgroup_subsys_state *css = - container_of(work, struct cgroup_subsys_state, dput_work); + container_of(work, struct cgroup_subsys_state, destroy_work); + struct cgroup *cgrp = css->cgroup; - cgroup_dput(css->cgroup); + if (css->parent) + css_put(css->parent); + + css->ss->css_free(css); + cgroup_dput(cgrp); +} + +static void css_free_rcu_fn(struct rcu_head *rcu_head) +{ + struct cgroup_subsys_state *css = + container_of(rcu_head, struct cgroup_subsys_state, rcu_head); + + /* + * css holds an extra ref to @cgrp->dentry which is put on the last + * css_put(). dput() requires process context which we don't have. + */ + INIT_WORK(&css->destroy_work, css_free_work_fn); + schedule_work(&css->destroy_work); } static void css_release(struct percpu_ref *ref) @@ -4198,49 +4315,47 @@ static void css_release(struct percpu_ref *ref) struct cgroup_subsys_state *css = container_of(ref, struct cgroup_subsys_state, refcnt); - schedule_work(&css->dput_work); + call_rcu(&css->rcu_head, css_free_rcu_fn); } -static void init_cgroup_css(struct cgroup_subsys_state *css, - struct cgroup_subsys *ss, - struct cgroup *cgrp) +static void init_css(struct cgroup_subsys_state *css, struct cgroup_subsys *ss, + struct cgroup *cgrp) { css->cgroup = cgrp; + css->ss = ss; css->flags = 0; css->id = NULL; - if (cgrp == cgroup_dummy_top) + + if (cgrp->parent) + css->parent = cgroup_css(cgrp->parent, ss); + else css->flags |= CSS_ROOT; - BUG_ON(cgrp->subsys[ss->subsys_id]); - cgrp->subsys[ss->subsys_id] = css; - /* - * css holds an extra ref to @cgrp->dentry which is put on the last - * css_put(). dput() requires process context, which css_put() may - * be called without. @css->dput_work will be used to invoke - * dput() asynchronously from css_put(). - */ - INIT_WORK(&css->dput_work, css_dput_fn); + BUG_ON(cgroup_css(cgrp, ss)); } -/* invoke ->post_create() on a new CSS and mark it online if successful */ -static int online_css(struct cgroup_subsys *ss, struct cgroup *cgrp) +/* invoke ->css_online() on a new CSS and mark it online if successful */ +static int online_css(struct cgroup_subsys_state *css) { + struct cgroup_subsys *ss = css->ss; int ret = 0; lockdep_assert_held(&cgroup_mutex); if (ss->css_online) - ret = ss->css_online(cgrp); - if (!ret) - cgrp->subsys[ss->subsys_id]->flags |= CSS_ONLINE; + ret = ss->css_online(css); + if (!ret) { + css->flags |= CSS_ONLINE; + css->cgroup->nr_css++; + rcu_assign_pointer(css->cgroup->subsys[ss->subsys_id], css); + } return ret; } -/* if the CSS is online, invoke ->pre_destory() on it and mark it offline */ -static void offline_css(struct cgroup_subsys *ss, struct cgroup *cgrp) - __releases(&cgroup_mutex) __acquires(&cgroup_mutex) +/* if the CSS is online, invoke ->css_offline() on it and mark it offline */ +static void offline_css(struct cgroup_subsys_state *css) { - struct cgroup_subsys_state *css = cgrp->subsys[ss->subsys_id]; + struct cgroup_subsys *ss = css->ss; lockdep_assert_held(&cgroup_mutex); @@ -4248,9 +4363,11 @@ static void offline_css(struct cgroup_subsys *ss, struct cgroup *cgrp) return; if (ss->css_offline) - ss->css_offline(cgrp); + ss->css_offline(css); - cgrp->subsys[ss->subsys_id]->flags &= ~CSS_ONLINE; + css->flags &= ~CSS_ONLINE; + css->cgroup->nr_css--; + RCU_INIT_POINTER(css->cgroup->subsys[ss->subsys_id], css); } /* @@ -4264,6 +4381,7 @@ static void offline_css(struct cgroup_subsys *ss, struct cgroup *cgrp) static long cgroup_create(struct cgroup *parent, struct dentry *dentry, umode_t mode) { + struct cgroup_subsys_state *css_ar[CGROUP_SUBSYS_COUNT] = { }; struct cgroup *cgrp; struct cgroup_name *name; struct cgroupfs_root *root = parent->root; @@ -4281,7 +4399,11 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, goto err_free_cgrp; rcu_assign_pointer(cgrp->name, name); - cgrp->id = ida_simple_get(&root->cgroup_ida, 1, 0, GFP_KERNEL); + /* + * Temporarily set the pointer to NULL, so idr_find() won't return + * a half-baked cgroup. + */ + cgrp->id = idr_alloc(&root->cgroup_idr, NULL, 1, 0, GFP_KERNEL); if (cgrp->id < 0) goto err_free_name; @@ -4310,6 +4432,7 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, cgrp->dentry = dentry; cgrp->parent = parent; + cgrp->dummy_css.parent = &parent->dummy_css; cgrp->root = parent->root; if (notify_on_release(parent)) @@ -4321,20 +4444,21 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, for_each_root_subsys(root, ss) { struct cgroup_subsys_state *css; - css = ss->css_alloc(cgrp); + css = ss->css_alloc(cgroup_css(parent, ss)); if (IS_ERR(css)) { err = PTR_ERR(css); goto err_free_all; } + css_ar[ss->subsys_id] = css; err = percpu_ref_init(&css->refcnt, css_release); if (err) goto err_free_all; - init_cgroup_css(css, ss, cgrp); + init_css(css, ss, cgrp); if (ss->use_id) { - err = alloc_css_id(ss, parent, cgrp); + err = alloc_css_id(css); if (err) goto err_free_all; } @@ -4356,16 +4480,22 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, list_add_tail_rcu(&cgrp->sibling, &cgrp->parent->children); root->number_of_cgroups++; - /* each css holds a ref to the cgroup's dentry */ - for_each_root_subsys(root, ss) + /* each css holds a ref to the cgroup's dentry and the parent css */ + for_each_root_subsys(root, ss) { + struct cgroup_subsys_state *css = css_ar[ss->subsys_id]; + dget(dentry); + css_get(css->parent); + } /* hold a ref to the parent's dentry */ dget(parent->dentry); /* creation succeeded, notify subsystems */ for_each_root_subsys(root, ss) { - err = online_css(ss, cgrp); + struct cgroup_subsys_state *css = css_ar[ss->subsys_id]; + + err = online_css(css); if (err) goto err_destroy; @@ -4379,7 +4509,13 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, } } - err = cgroup_populate_dir(cgrp, true, root->subsys_mask); + idr_replace(&root->cgroup_idr, cgrp, cgrp->id); + + err = cgroup_addrm_files(cgrp, cgroup_base_files, true); + if (err) + goto err_destroy; + + err = cgroup_populate_dir(cgrp, root->subsys_mask); if (err) goto err_destroy; @@ -4390,18 +4526,18 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, err_free_all: for_each_root_subsys(root, ss) { - struct cgroup_subsys_state *css = cgrp->subsys[ss->subsys_id]; + struct cgroup_subsys_state *css = css_ar[ss->subsys_id]; if (css) { percpu_ref_cancel_init(&css->refcnt); - ss->css_free(cgrp); + ss->css_free(css); } } mutex_unlock(&cgroup_mutex); /* Release the reference count that we took on the superblock */ deactivate_super(sb); err_free_id: - ida_simple_remove(&root->cgroup_ida, cgrp->id); + idr_remove(&root->cgroup_idr, cgrp->id); err_free_name: kfree(rcu_dereference_raw(cgrp->name)); err_free_cgrp: @@ -4423,22 +4559,84 @@ static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) return cgroup_create(c_parent, dentry, mode | S_IFDIR); } -static void cgroup_css_killed(struct cgroup *cgrp) +/* + * This is called when the refcnt of a css is confirmed to be killed. + * css_tryget() is now guaranteed to fail. + */ +static void css_killed_work_fn(struct work_struct *work) { - if (!atomic_dec_and_test(&cgrp->css_kill_cnt)) - return; + struct cgroup_subsys_state *css = + container_of(work, struct cgroup_subsys_state, destroy_work); + struct cgroup *cgrp = css->cgroup; - /* percpu ref's of all css's are killed, kick off the next step */ - INIT_WORK(&cgrp->destroy_work, cgroup_offline_fn); - schedule_work(&cgrp->destroy_work); + mutex_lock(&cgroup_mutex); + + /* + * css_tryget() is guaranteed to fail now. Tell subsystems to + * initate destruction. + */ + offline_css(css); + + /* + * If @cgrp is marked dead, it's waiting for refs of all css's to + * be disabled before proceeding to the second phase of cgroup + * destruction. If we are the last one, kick it off. + */ + if (!cgrp->nr_css && cgroup_is_dead(cgrp)) + cgroup_destroy_css_killed(cgrp); + + mutex_unlock(&cgroup_mutex); + + /* + * Put the css refs from kill_css(). Each css holds an extra + * reference to the cgroup's dentry and cgroup removal proceeds + * regardless of css refs. On the last put of each css, whenever + * that may be, the extra dentry ref is put so that dentry + * destruction happens only after all css's are released. + */ + css_put(css); } -static void css_ref_killed_fn(struct percpu_ref *ref) +/* css kill confirmation processing requires process context, bounce */ +static void css_killed_ref_fn(struct percpu_ref *ref) { struct cgroup_subsys_state *css = container_of(ref, struct cgroup_subsys_state, refcnt); - cgroup_css_killed(css->cgroup); + INIT_WORK(&css->destroy_work, css_killed_work_fn); + schedule_work(&css->destroy_work); +} + +/** + * kill_css - destroy a css + * @css: css to destroy + * + * This function initiates destruction of @css by removing cgroup interface + * files and putting its base reference. ->css_offline() will be invoked + * asynchronously once css_tryget() is guaranteed to fail and when the + * reference count reaches zero, @css will be released. + */ +static void kill_css(struct cgroup_subsys_state *css) +{ + cgroup_clear_dir(css->cgroup, 1 << css->ss->subsys_id); + + /* + * Killing would put the base ref, but we need to keep it alive + * until after ->css_offline(). + */ + css_get(css); + + /* + * cgroup core guarantees that, by the time ->css_offline() is + * invoked, no new css reference will be given out via + * css_tryget(). We can't simply call percpu_ref_kill() and + * proceed to offlining css's because percpu_ref_kill() doesn't + * guarantee that the ref is seen as killed on all CPUs on return. + * + * Use percpu_ref_kill_and_confirm() to get notifications as each + * css is confirmed to be seen as killed on all CPUs. + */ + percpu_ref_kill_and_confirm(&css->refcnt, css_killed_ref_fn); } /** @@ -4471,6 +4669,7 @@ static int cgroup_destroy_locked(struct cgroup *cgrp) struct dentry *d = cgrp->dentry; struct cgroup_event *event, *tmp; struct cgroup_subsys *ss; + struct cgroup *child; bool empty; lockdep_assert_held(&d->d_inode->i_mutex); @@ -4481,47 +4680,41 @@ static int cgroup_destroy_locked(struct cgroup *cgrp) * @cgrp from being removed while __put_css_set() is in progress. */ read_lock(&css_set_lock); - empty = list_empty(&cgrp->cset_links) && list_empty(&cgrp->children); + empty = list_empty(&cgrp->cset_links); read_unlock(&css_set_lock); if (!empty) return -EBUSY; /* - * Block new css_tryget() by killing css refcnts. cgroup core - * guarantees that, by the time ->css_offline() is invoked, no new - * css reference will be given out via css_tryget(). We can't - * simply call percpu_ref_kill() and proceed to offlining css's - * because percpu_ref_kill() doesn't guarantee that the ref is seen - * as killed on all CPUs on return. - * - * Use percpu_ref_kill_and_confirm() to get notifications as each - * css is confirmed to be seen as killed on all CPUs. The - * notification callback keeps track of the number of css's to be - * killed and schedules cgroup_offline_fn() to perform the rest of - * destruction once the percpu refs of all css's are confirmed to - * be killed. + * Make sure there's no live children. We can't test ->children + * emptiness as dead children linger on it while being destroyed; + * otherwise, "rmdir parent/child parent" may fail with -EBUSY. */ - atomic_set(&cgrp->css_kill_cnt, 1); - for_each_root_subsys(cgrp->root, ss) { - struct cgroup_subsys_state *css = cgrp->subsys[ss->subsys_id]; - - /* - * Killing would put the base ref, but we need to keep it - * alive until after ->css_offline. - */ - percpu_ref_get(&css->refcnt); - - atomic_inc(&cgrp->css_kill_cnt); - percpu_ref_kill_and_confirm(&css->refcnt, css_ref_killed_fn); + empty = true; + rcu_read_lock(); + list_for_each_entry_rcu(child, &cgrp->children, sibling) { + empty = cgroup_is_dead(child); + if (!empty) + break; } - cgroup_css_killed(cgrp); + rcu_read_unlock(); + if (!empty) + return -EBUSY; + + /* + * Initiate massacre of all css's. cgroup_destroy_css_killed() + * will be invoked to perform the rest of destruction once the + * percpu refs of all css's are confirmed to be killed. + */ + for_each_root_subsys(cgrp->root, ss) + kill_css(cgroup_css(cgrp, ss)); /* * Mark @cgrp dead. This prevents further task migration and child * creation by disabling cgroup_lock_live_group(). Note that - * CGRP_DEAD assertion is depended upon by cgroup_next_sibling() to + * CGRP_DEAD assertion is depended upon by css_next_child() to * resume iteration after dropping RCU read lock. See - * cgroup_next_sibling() for details. + * css_next_child() for details. */ set_bit(CGRP_DEAD, &cgrp->flags); @@ -4532,9 +4725,20 @@ static int cgroup_destroy_locked(struct cgroup *cgrp) raw_spin_unlock(&release_list_lock); /* - * Remove @cgrp directory. The removal puts the base ref but we - * aren't quite done with @cgrp yet, so hold onto it. + * If @cgrp has css's attached, the second stage of cgroup + * destruction is kicked off from css_killed_work_fn() after the + * refs of all attached css's are killed. If @cgrp doesn't have + * any css, we kick it off here. */ + if (!cgrp->nr_css) + cgroup_destroy_css_killed(cgrp); + + /* + * Clear the base files and remove @cgrp directory. The removal + * puts the base ref but we aren't quite done with @cgrp yet, so + * hold onto it. + */ + cgroup_addrm_files(cgrp, cgroup_base_files, false); dget(d); cgroup_d_remove_dir(d); @@ -4554,50 +4758,36 @@ static int cgroup_destroy_locked(struct cgroup *cgrp) }; /** - * cgroup_offline_fn - the second step of cgroup destruction + * cgroup_destroy_css_killed - the second step of cgroup destruction * @work: cgroup->destroy_free_work * * This function is invoked from a work item for a cgroup which is being - * destroyed after the percpu refcnts of all css's are guaranteed to be - * seen as killed on all CPUs, and performs the rest of destruction. This - * is the second step of destruction described in the comment above - * cgroup_destroy_locked(). + * destroyed after all css's are offlined and performs the rest of + * destruction. This is the second step of destruction described in the + * comment above cgroup_destroy_locked(). */ -static void cgroup_offline_fn(struct work_struct *work) +static void cgroup_destroy_css_killed(struct cgroup *cgrp) { - struct cgroup *cgrp = container_of(work, struct cgroup, destroy_work); struct cgroup *parent = cgrp->parent; struct dentry *d = cgrp->dentry; - struct cgroup_subsys *ss; - mutex_lock(&cgroup_mutex); + lockdep_assert_held(&cgroup_mutex); - /* - * css_tryget() is guaranteed to fail now. Tell subsystems to - * initate destruction. - */ - for_each_root_subsys(cgrp->root, ss) - offline_css(ss, cgrp); + /* delete this cgroup from parent->children */ + list_del_rcu(&cgrp->sibling); /* - * Put the css refs from cgroup_destroy_locked(). Each css holds - * an extra reference to the cgroup's dentry and cgroup removal - * proceeds regardless of css refs. On the last put of each css, - * whenever that may be, the extra dentry ref is put so that dentry - * destruction happens only after all css's are released. + * We should remove the cgroup object from idr before its grace + * period starts, so we won't be looking up a cgroup while the + * cgroup is being freed. */ - for_each_root_subsys(cgrp->root, ss) - css_put(cgrp->subsys[ss->subsys_id]); - - /* delete this cgroup from parent->children */ - list_del_rcu(&cgrp->sibling); + idr_remove(&cgrp->root->cgroup_idr, cgrp->id); + cgrp->id = -1; dput(d); set_bit(CGRP_RELEASABLE, &parent->flags); check_for_release(parent); - - mutex_unlock(&cgroup_mutex); } static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry) @@ -4620,6 +4810,11 @@ static void __init_or_module cgroup_init_cftsets(struct cgroup_subsys *ss) * deregistration. */ if (ss->base_cftypes) { + struct cftype *cft; + + for (cft = ss->base_cftypes; cft->name[0] != '\0'; cft++) + cft->ss = ss; + ss->base_cftset.cfts = ss->base_cftypes; list_add_tail(&ss->base_cftset.node, &ss->cftsets); } @@ -4639,10 +4834,10 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss) /* Create the top cgroup state for this subsystem */ list_add(&ss->sibling, &cgroup_dummy_root.subsys_list); ss->root = &cgroup_dummy_root; - css = ss->css_alloc(cgroup_dummy_top); + css = ss->css_alloc(cgroup_css(cgroup_dummy_top, ss)); /* We don't handle early failures gracefully */ BUG_ON(IS_ERR(css)); - init_cgroup_css(css, ss, cgroup_dummy_top); + init_css(css, ss, cgroup_dummy_top); /* Update the init_css_set to contain a subsys * pointer to this state - since the subsystem is @@ -4657,7 +4852,7 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss) * need to invoke fork callbacks here. */ BUG_ON(!list_empty(&init_task.tasks)); - BUG_ON(online_css(ss, cgroup_dummy_top)); + BUG_ON(online_css(css)); mutex_unlock(&cgroup_mutex); @@ -4718,7 +4913,7 @@ int __init_or_module cgroup_load_subsys(struct cgroup_subsys *ss) * struct, so this can happen first (i.e. before the dummy root * attachment). */ - css = ss->css_alloc(cgroup_dummy_top); + css = ss->css_alloc(cgroup_css(cgroup_dummy_top, ss)); if (IS_ERR(css)) { /* failure case - need to deassign the cgroup_subsys[] slot. */ cgroup_subsys[ss->subsys_id] = NULL; @@ -4730,8 +4925,8 @@ int __init_or_module cgroup_load_subsys(struct cgroup_subsys *ss) ss->root = &cgroup_dummy_root; /* our new subsystem will be attached to the dummy hierarchy. */ - init_cgroup_css(css, ss, cgroup_dummy_top); - /* init_idr must be after init_cgroup_css because it sets css->id. */ + init_css(css, ss, cgroup_dummy_top); + /* init_idr must be after init_css() because it sets css->id. */ if (ss->use_id) { ret = cgroup_init_idr(ss, css); if (ret) @@ -4761,7 +4956,7 @@ int __init_or_module cgroup_load_subsys(struct cgroup_subsys *ss) } write_unlock(&css_set_lock); - ret = online_css(ss, cgroup_dummy_top); + ret = online_css(css); if (ret) goto err_unload; @@ -4793,14 +4988,14 @@ void cgroup_unload_subsys(struct cgroup_subsys *ss) /* * we shouldn't be called if the subsystem is in use, and the use of - * try_module_get in parse_cgroupfs_options should ensure that it + * try_module_get() in rebind_subsystems() should ensure that it * doesn't start being used while we're killing it off. */ BUG_ON(ss->root != &cgroup_dummy_root); mutex_lock(&cgroup_mutex); - offline_css(ss, cgroup_dummy_top); + offline_css(cgroup_css(cgroup_dummy_top, ss)); if (ss->use_id) idr_destroy(&ss->idr); @@ -4834,8 +5029,8 @@ void cgroup_unload_subsys(struct cgroup_subsys *ss) * the cgrp->subsys pointer to find their state. note that this * also takes care of freeing the css_id. */ - ss->css_free(cgroup_dummy_top); - cgroup_dummy_top->subsys[ss->subsys_id] = NULL; + ss->css_free(cgroup_css(cgroup_dummy_top, ss)); + RCU_INIT_POINTER(cgroup_dummy_top->subsys[ss->subsys_id], NULL); mutex_unlock(&cgroup_mutex); } @@ -4917,6 +5112,10 @@ int __init cgroup_init(void) BUG_ON(cgroup_init_root_id(&cgroup_dummy_root, 0, 1)); + err = idr_alloc(&cgroup_dummy_root.cgroup_idr, cgroup_dummy_top, + 0, 1, GFP_KERNEL); + BUG_ON(err < 0); + mutex_unlock(&cgroup_root_mutex); mutex_unlock(&cgroup_mutex); @@ -5073,7 +5272,7 @@ void cgroup_fork(struct task_struct *child) * Adds the task to the list running through its css_set if necessary and * call the subsystem fork() callbacks. Has to be after the task is * visible on the task list in case we race with the first call to - * cgroup_iter_start() - to guarantee that the new task ends up on its + * cgroup_task_iter_start() - to guarantee that the new task ends up on its * list. */ void cgroup_post_fork(struct task_struct *child) @@ -5186,10 +5385,10 @@ void cgroup_exit(struct task_struct *tsk, int run_callbacks) */ for_each_builtin_subsys(ss, i) { if (ss->exit) { - struct cgroup *old_cgrp = cset->subsys[i]->cgroup; - struct cgroup *cgrp = task_cgroup(tsk, i); + struct cgroup_subsys_state *old_css = cset->subsys[i]; + struct cgroup_subsys_state *css = task_css(tsk, i); - ss->exit(cgrp, old_cgrp, tsk); + ss->exit(css, old_css, tsk); } } } @@ -5448,20 +5647,16 @@ static int __init_or_module cgroup_init_idr(struct cgroup_subsys *ss, return 0; } -static int alloc_css_id(struct cgroup_subsys *ss, struct cgroup *parent, - struct cgroup *child) +static int alloc_css_id(struct cgroup_subsys_state *child_css) { - int subsys_id, i, depth = 0; - struct cgroup_subsys_state *parent_css, *child_css; + struct cgroup_subsys_state *parent_css = css_parent(child_css); struct css_id *child_id, *parent_id; + int i, depth; - subsys_id = ss->subsys_id; - parent_css = parent->subsys[subsys_id]; - child_css = child->subsys[subsys_id]; parent_id = rcu_dereference_protected(parent_css->id, true); depth = parent_id->depth + 1; - child_id = get_new_cssid(ss, depth); + child_id = get_new_cssid(child_css->ss, depth); if (IS_ERR(child_id)) return PTR_ERR(child_id); @@ -5499,31 +5694,56 @@ struct cgroup_subsys_state *css_lookup(struct cgroup_subsys *ss, int id) } EXPORT_SYMBOL_GPL(css_lookup); -/* - * get corresponding css from file open on cgroupfs directory +/** + * css_from_dir - get corresponding css from the dentry of a cgroup dir + * @dentry: directory dentry of interest + * @ss: subsystem of interest + * + * Must be called under RCU read lock. The caller is responsible for + * pinning the returned css if it needs to be accessed outside the RCU + * critical section. */ -struct cgroup_subsys_state *cgroup_css_from_dir(struct file *f, int id) +struct cgroup_subsys_state *css_from_dir(struct dentry *dentry, + struct cgroup_subsys *ss) { struct cgroup *cgrp; - struct inode *inode; - struct cgroup_subsys_state *css; - inode = file_inode(f); - /* check in cgroup filesystem dir */ - if (inode->i_op != &cgroup_dir_inode_operations) + WARN_ON_ONCE(!rcu_read_lock_held()); + + /* is @dentry a cgroup dir? */ + if (!dentry->d_inode || + dentry->d_inode->i_op != &cgroup_dir_inode_operations) return ERR_PTR(-EBADF); - if (id < 0 || id >= CGROUP_SUBSYS_COUNT) - return ERR_PTR(-EINVAL); + cgrp = __d_cgrp(dentry); + return cgroup_css(cgrp, ss) ?: ERR_PTR(-ENOENT); +} - /* get cgroup */ - cgrp = __d_cgrp(f->f_dentry); - css = cgrp->subsys[id]; - return css ? css : ERR_PTR(-ENOENT); +/** + * css_from_id - lookup css by id + * @id: the cgroup id + * @ss: cgroup subsys to be looked into + * + * Returns the css if there's valid one with @id, otherwise returns NULL. + * Should be called under rcu_read_lock(). + */ +struct cgroup_subsys_state *css_from_id(int id, struct cgroup_subsys *ss) +{ + struct cgroup *cgrp; + + rcu_lockdep_assert(rcu_read_lock_held() || + lockdep_is_held(&cgroup_mutex), + "css_from_id() needs proper protection"); + + cgrp = idr_find(&ss->root->cgroup_idr, id); + if (cgrp) + return cgroup_css(cgrp, ss); + return NULL; } #ifdef CONFIG_CGROUP_DEBUG -static struct cgroup_subsys_state *debug_css_alloc(struct cgroup *cgrp) +static struct cgroup_subsys_state * +debug_css_alloc(struct cgroup_subsys_state *parent_css) { struct cgroup_subsys_state *css = kzalloc(sizeof(*css), GFP_KERNEL); @@ -5533,22 +5753,24 @@ static struct cgroup_subsys_state *debug_css_alloc(struct cgroup *cgrp) return css; } -static void debug_css_free(struct cgroup *cgrp) +static void debug_css_free(struct cgroup_subsys_state *css) { - kfree(cgrp->subsys[debug_subsys_id]); + kfree(css); } -static u64 debug_taskcount_read(struct cgroup *cgrp, struct cftype *cft) +static u64 debug_taskcount_read(struct cgroup_subsys_state *css, + struct cftype *cft) { - return cgroup_task_count(cgrp); + return cgroup_task_count(css->cgroup); } -static u64 current_css_set_read(struct cgroup *cgrp, struct cftype *cft) +static u64 current_css_set_read(struct cgroup_subsys_state *css, + struct cftype *cft) { return (u64)(unsigned long)current->cgroups; } -static u64 current_css_set_refcount_read(struct cgroup *cgrp, +static u64 current_css_set_refcount_read(struct cgroup_subsys_state *css, struct cftype *cft) { u64 count; @@ -5559,7 +5781,7 @@ static u64 current_css_set_refcount_read(struct cgroup *cgrp, return count; } -static int current_css_set_cg_links_read(struct cgroup *cgrp, +static int current_css_set_cg_links_read(struct cgroup_subsys_state *css, struct cftype *cft, struct seq_file *seq) { @@ -5586,14 +5808,13 @@ static int current_css_set_cg_links_read(struct cgroup *cgrp, } #define MAX_TASKS_SHOWN_PER_CSS 25 -static int cgroup_css_links_read(struct cgroup *cgrp, - struct cftype *cft, - struct seq_file *seq) +static int cgroup_css_links_read(struct cgroup_subsys_state *css, + struct cftype *cft, struct seq_file *seq) { struct cgrp_cset_link *link; read_lock(&css_set_lock); - list_for_each_entry(link, &cgrp->cset_links, cset_link) { + list_for_each_entry(link, &css->cgroup->cset_links, cset_link) { struct css_set *cset = link->cset; struct task_struct *task; int count = 0; @@ -5612,9 +5833,9 @@ static int cgroup_css_links_read(struct cgroup *cgrp, return 0; } -static u64 releasable_read(struct cgroup *cgrp, struct cftype *cft) +static u64 releasable_read(struct cgroup_subsys_state *css, struct cftype *cft) { - return test_bit(CGRP_RELEASABLE, &cgrp->flags); + return test_bit(CGRP_RELEASABLE, &css->cgroup->flags); } static struct cftype debug_files[] = { diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup_freezer.c index 75dda1ea5026..f0ff64d0ebaa 100644 --- a/kernel/cgroup_freezer.c +++ b/kernel/cgroup_freezer.c @@ -45,25 +45,19 @@ struct freezer { spinlock_t lock; }; -static inline struct freezer *cgroup_freezer(struct cgroup *cgroup) +static inline struct freezer *css_freezer(struct cgroup_subsys_state *css) { - return container_of(cgroup_subsys_state(cgroup, freezer_subsys_id), - struct freezer, css); + return css ? container_of(css, struct freezer, css) : NULL; } static inline struct freezer *task_freezer(struct task_struct *task) { - return container_of(task_subsys_state(task, freezer_subsys_id), - struct freezer, css); + return css_freezer(task_css(task, freezer_subsys_id)); } static struct freezer *parent_freezer(struct freezer *freezer) { - struct cgroup *pcg = freezer->css.cgroup->parent; - - if (pcg) - return cgroup_freezer(pcg); - return NULL; + return css_freezer(css_parent(&freezer->css)); } bool cgroup_freezing(struct task_struct *task) @@ -92,7 +86,8 @@ static const char *freezer_state_strs(unsigned int state) struct cgroup_subsys freezer_subsys; -static struct cgroup_subsys_state *freezer_css_alloc(struct cgroup *cgroup) +static struct cgroup_subsys_state * +freezer_css_alloc(struct cgroup_subsys_state *parent_css) { struct freezer *freezer; @@ -105,22 +100,22 @@ static struct cgroup_subsys_state *freezer_css_alloc(struct cgroup *cgroup) } /** - * freezer_css_online - commit creation of a freezer cgroup - * @cgroup: cgroup being created + * freezer_css_online - commit creation of a freezer css + * @css: css being created * - * We're committing to creation of @cgroup. Mark it online and inherit + * We're committing to creation of @css. Mark it online and inherit * parent's freezing state while holding both parent's and our * freezer->lock. */ -static int freezer_css_online(struct cgroup *cgroup) +static int freezer_css_online(struct cgroup_subsys_state *css) { - struct freezer *freezer = cgroup_freezer(cgroup); + struct freezer *freezer = css_freezer(css); struct freezer *parent = parent_freezer(freezer); /* * The following double locking and freezing state inheritance * guarantee that @cgroup can never escape ancestors' freezing - * states. See cgroup_for_each_descendant_pre() for details. + * states. See css_for_each_descendant_pre() for details. */ if (parent) spin_lock_irq(&parent->lock); @@ -141,15 +136,15 @@ static int freezer_css_online(struct cgroup *cgroup) } /** - * freezer_css_offline - initiate destruction of @cgroup - * @cgroup: cgroup being destroyed + * freezer_css_offline - initiate destruction of a freezer css + * @css: css being destroyed * - * @cgroup is going away. Mark it dead and decrement system_freezing_count - * if it was holding one. + * @css is going away. Mark it dead and decrement system_freezing_count if + * it was holding one. */ -static void freezer_css_offline(struct cgroup *cgroup) +static void freezer_css_offline(struct cgroup_subsys_state *css) { - struct freezer *freezer = cgroup_freezer(cgroup); + struct freezer *freezer = css_freezer(css); spin_lock_irq(&freezer->lock); @@ -161,9 +156,9 @@ static void freezer_css_offline(struct cgroup *cgroup) spin_unlock_irq(&freezer->lock); } -static void freezer_css_free(struct cgroup *cgroup) +static void freezer_css_free(struct cgroup_subsys_state *css) { - kfree(cgroup_freezer(cgroup)); + kfree(css_freezer(css)); } /* @@ -175,25 +170,26 @@ static void freezer_css_free(struct cgroup *cgroup) * @freezer->lock. freezer_attach() makes the new tasks conform to the * current state and all following state changes can see the new tasks. */ -static void freezer_attach(struct cgroup *new_cgrp, struct cgroup_taskset *tset) +static void freezer_attach(struct cgroup_subsys_state *new_css, + struct cgroup_taskset *tset) { - struct freezer *freezer = cgroup_freezer(new_cgrp); + struct freezer *freezer = css_freezer(new_css); struct task_struct *task; bool clear_frozen = false; spin_lock_irq(&freezer->lock); /* - * Make the new tasks conform to the current state of @new_cgrp. + * Make the new tasks conform to the current state of @new_css. * For simplicity, when migrating any task to a FROZEN cgroup, we * revert it to FREEZING and let update_if_frozen() determine the * correct state later. * - * Tasks in @tset are on @new_cgrp but may not conform to its + * Tasks in @tset are on @new_css but may not conform to its * current state before executing the following - !frozen tasks may * be visible in a FROZEN cgroup and frozen tasks in a THAWED one. */ - cgroup_taskset_for_each(task, new_cgrp, tset) { + cgroup_taskset_for_each(task, new_css, tset) { if (!(freezer->state & CGROUP_FREEZING)) { __thaw_task(task); } else { @@ -231,7 +227,7 @@ static void freezer_fork(struct task_struct *task) * The root cgroup is non-freezable, so we can skip the * following check. */ - if (!freezer->css.cgroup->parent) + if (!parent_freezer(freezer)) goto out; spin_lock_irq(&freezer->lock); @@ -244,7 +240,7 @@ out: /** * update_if_frozen - update whether a cgroup finished freezing - * @cgroup: cgroup of interest + * @css: css of interest * * Once FREEZING is initiated, transition to FROZEN is lazily updated by * calling this function. If the current state is FREEZING but not FROZEN, @@ -255,14 +251,14 @@ out: * update_if_frozen() on all descendants prior to invoking this function. * * Task states and freezer state might disagree while tasks are being - * migrated into or out of @cgroup, so we can't verify task states against + * migrated into or out of @css, so we can't verify task states against * @freezer state here. See freezer_attach() for details. */ -static void update_if_frozen(struct cgroup *cgroup) +static void update_if_frozen(struct cgroup_subsys_state *css) { - struct freezer *freezer = cgroup_freezer(cgroup); - struct cgroup *pos; - struct cgroup_iter it; + struct freezer *freezer = css_freezer(css); + struct cgroup_subsys_state *pos; + struct css_task_iter it; struct task_struct *task; WARN_ON_ONCE(!rcu_read_lock_held()); @@ -274,8 +270,8 @@ static void update_if_frozen(struct cgroup *cgroup) goto out_unlock; /* are all (live) children frozen? */ - cgroup_for_each_child(pos, cgroup) { - struct freezer *child = cgroup_freezer(pos); + css_for_each_child(pos, css) { + struct freezer *child = css_freezer(pos); if ((child->state & CGROUP_FREEZER_ONLINE) && !(child->state & CGROUP_FROZEN)) @@ -283,9 +279,9 @@ static void update_if_frozen(struct cgroup *cgroup) } /* are all tasks frozen? */ - cgroup_iter_start(cgroup, &it); + css_task_iter_start(css, &it); - while ((task = cgroup_iter_next(cgroup, &it))) { + while ((task = css_task_iter_next(&it))) { if (freezing(task)) { /* * freezer_should_skip() indicates that the task @@ -300,52 +296,49 @@ static void update_if_frozen(struct cgroup *cgroup) freezer->state |= CGROUP_FROZEN; out_iter_end: - cgroup_iter_end(cgroup, &it); + css_task_iter_end(&it); out_unlock: spin_unlock_irq(&freezer->lock); } -static int freezer_read(struct cgroup *cgroup, struct cftype *cft, +static int freezer_read(struct cgroup_subsys_state *css, struct cftype *cft, struct seq_file *m) { - struct cgroup *pos; + struct cgroup_subsys_state *pos; rcu_read_lock(); /* update states bottom-up */ - cgroup_for_each_descendant_post(pos, cgroup) + css_for_each_descendant_post(pos, css) update_if_frozen(pos); - update_if_frozen(cgroup); rcu_read_unlock(); - seq_puts(m, freezer_state_strs(cgroup_freezer(cgroup)->state)); + seq_puts(m, freezer_state_strs(css_freezer(css)->state)); seq_putc(m, '\n'); return 0; } static void freeze_cgroup(struct freezer *freezer) { - struct cgroup *cgroup = freezer->css.cgroup; - struct cgroup_iter it; + struct css_task_iter it; struct task_struct *task; - cgroup_iter_start(cgroup, &it); - while ((task = cgroup_iter_next(cgroup, &it))) + css_task_iter_start(&freezer->css, &it); + while ((task = css_task_iter_next(&it))) freeze_task(task); - cgroup_iter_end(cgroup, &it); + css_task_iter_end(&it); } static void unfreeze_cgroup(struct freezer *freezer) { - struct cgroup *cgroup = freezer->css.cgroup; - struct cgroup_iter it; + struct css_task_iter it; struct task_struct *task; - cgroup_iter_start(cgroup, &it); - while ((task = cgroup_iter_next(cgroup, &it))) + css_task_iter_start(&freezer->css, &it); + while ((task = css_task_iter_next(&it))) __thaw_task(task); - cgroup_iter_end(cgroup, &it); + css_task_iter_end(&it); } /** @@ -395,12 +388,7 @@ static void freezer_apply_state(struct freezer *freezer, bool freeze, */ static void freezer_change_state(struct freezer *freezer, bool freeze) { - struct cgroup *pos; - - /* update @freezer */ - spin_lock_irq(&freezer->lock); - freezer_apply_state(freezer, freeze, CGROUP_FREEZING_SELF); - spin_unlock_irq(&freezer->lock); + struct cgroup_subsys_state *pos; /* * Update all its descendants in pre-order traversal. Each @@ -408,24 +396,33 @@ static void freezer_change_state(struct freezer *freezer, bool freeze) * CGROUP_FREEZING_PARENT. */ rcu_read_lock(); - cgroup_for_each_descendant_pre(pos, freezer->css.cgroup) { - struct freezer *pos_f = cgroup_freezer(pos); + css_for_each_descendant_pre(pos, &freezer->css) { + struct freezer *pos_f = css_freezer(pos); struct freezer *parent = parent_freezer(pos_f); - /* - * Our update to @parent->state is already visible which is - * all we need. No need to lock @parent. For more info on - * synchronization, see freezer_post_create(). - */ spin_lock_irq(&pos_f->lock); - freezer_apply_state(pos_f, parent->state & CGROUP_FREEZING, - CGROUP_FREEZING_PARENT); + + if (pos_f == freezer) { + freezer_apply_state(pos_f, freeze, + CGROUP_FREEZING_SELF); + } else { + /* + * Our update to @parent->state is already visible + * which is all we need. No need to lock @parent. + * For more info on synchronization, see + * freezer_post_create(). + */ + freezer_apply_state(pos_f, + parent->state & CGROUP_FREEZING, + CGROUP_FREEZING_PARENT); + } + spin_unlock_irq(&pos_f->lock); } rcu_read_unlock(); } -static int freezer_write(struct cgroup *cgroup, struct cftype *cft, +static int freezer_write(struct cgroup_subsys_state *css, struct cftype *cft, const char *buffer) { bool freeze; @@ -437,20 +434,22 @@ static int freezer_write(struct cgroup *cgroup, struct cftype *cft, else return -EINVAL; - freezer_change_state(cgroup_freezer(cgroup), freeze); + freezer_change_state(css_freezer(css), freeze); return 0; } -static u64 freezer_self_freezing_read(struct cgroup *cgroup, struct cftype *cft) +static u64 freezer_self_freezing_read(struct cgroup_subsys_state *css, + struct cftype *cft) { - struct freezer *freezer = cgroup_freezer(cgroup); + struct freezer *freezer = css_freezer(css); return (bool)(freezer->state & CGROUP_FREEZING_SELF); } -static u64 freezer_parent_freezing_read(struct cgroup *cgroup, struct cftype *cft) +static u64 freezer_parent_freezing_read(struct cgroup_subsys_state *css, + struct cftype *cft) { - struct freezer *freezer = cgroup_freezer(cgroup); + struct freezer *freezer = css_freezer(css); return (bool)(freezer->state & CGROUP_FREEZING_PARENT); } diff --git a/kernel/cpu.c b/kernel/cpu.c index b2b227b82123..d7f07a2da5a6 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -113,7 +113,7 @@ EXPORT_SYMBOL_GPL(put_online_cpus); * get_online_cpus() not an api which is called all that often. * */ -static void cpu_hotplug_begin(void) +void cpu_hotplug_begin(void) { cpu_hotplug.active_writer = current; @@ -127,7 +127,7 @@ static void cpu_hotplug_begin(void) } } -static void cpu_hotplug_done(void) +void cpu_hotplug_done(void) { cpu_hotplug.active_writer = NULL; mutex_unlock(&cpu_hotplug.lock); @@ -154,10 +154,7 @@ void cpu_hotplug_enable(void) cpu_maps_update_done(); } -#else /* #if CONFIG_HOTPLUG_CPU */ -static void cpu_hotplug_begin(void) {} -static void cpu_hotplug_done(void) {} -#endif /* #else #if CONFIG_HOTPLUG_CPU */ +#endif /* CONFIG_HOTPLUG_CPU */ /* Need to know about CPUs going up/down? */ int __ref register_cpu_notifier(struct notifier_block *nb) diff --git a/kernel/cpuset.c b/kernel/cpuset.c index e5657788fedd..6bf981e13c43 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -68,10 +68,6 @@ */ int number_of_cpusets __read_mostly; -/* Forward declare cgroup structures */ -struct cgroup_subsys cpuset_subsys; -struct cpuset; - /* See "Frequency meter" comments, below. */ struct fmeter { @@ -115,27 +111,20 @@ struct cpuset { int relax_domain_level; }; -/* Retrieve the cpuset for a cgroup */ -static inline struct cpuset *cgroup_cs(struct cgroup *cgrp) +static inline struct cpuset *css_cs(struct cgroup_subsys_state *css) { - return container_of(cgroup_subsys_state(cgrp, cpuset_subsys_id), - struct cpuset, css); + return css ? container_of(css, struct cpuset, css) : NULL; } /* Retrieve the cpuset for a task */ static inline struct cpuset *task_cs(struct task_struct *task) { - return container_of(task_subsys_state(task, cpuset_subsys_id), - struct cpuset, css); + return css_cs(task_css(task, cpuset_subsys_id)); } -static inline struct cpuset *parent_cs(const struct cpuset *cs) +static inline struct cpuset *parent_cs(struct cpuset *cs) { - struct cgroup *pcgrp = cs->css.cgroup->parent; - - if (pcgrp) - return cgroup_cs(pcgrp); - return NULL; + return css_cs(css_parent(&cs->css)); } #ifdef CONFIG_NUMA @@ -212,29 +201,30 @@ static struct cpuset top_cpuset = { /** * cpuset_for_each_child - traverse online children of a cpuset * @child_cs: loop cursor pointing to the current child - * @pos_cgrp: used for iteration + * @pos_css: used for iteration * @parent_cs: target cpuset to walk children of * * Walk @child_cs through the online children of @parent_cs. Must be used * with RCU read locked. */ -#define cpuset_for_each_child(child_cs, pos_cgrp, parent_cs) \ - cgroup_for_each_child((pos_cgrp), (parent_cs)->css.cgroup) \ - if (is_cpuset_online(((child_cs) = cgroup_cs((pos_cgrp))))) +#define cpuset_for_each_child(child_cs, pos_css, parent_cs) \ + css_for_each_child((pos_css), &(parent_cs)->css) \ + if (is_cpuset_online(((child_cs) = css_cs((pos_css))))) /** * cpuset_for_each_descendant_pre - pre-order walk of a cpuset's descendants * @des_cs: loop cursor pointing to the current descendant - * @pos_cgrp: used for iteration + * @pos_css: used for iteration * @root_cs: target cpuset to walk ancestor of * * Walk @des_cs through the online descendants of @root_cs. Must be used - * with RCU read locked. The caller may modify @pos_cgrp by calling - * cgroup_rightmost_descendant() to skip subtree. + * with RCU read locked. The caller may modify @pos_css by calling + * css_rightmost_descendant() to skip subtree. @root_cs is included in the + * iteration and the first node to be visited. */ -#define cpuset_for_each_descendant_pre(des_cs, pos_cgrp, root_cs) \ - cgroup_for_each_descendant_pre((pos_cgrp), (root_cs)->css.cgroup) \ - if (is_cpuset_online(((des_cs) = cgroup_cs((pos_cgrp))))) +#define cpuset_for_each_descendant_pre(des_cs, pos_css, root_cs) \ + css_for_each_descendant_pre((pos_css), &(root_cs)->css) \ + if (is_cpuset_online(((des_cs) = css_cs((pos_css))))) /* * There are two global mutexes guarding cpuset structures - cpuset_mutex @@ -320,8 +310,7 @@ static struct file_system_type cpuset_fs_type = { * * Call with callback_mutex held. */ -static void guarantee_online_cpus(const struct cpuset *cs, - struct cpumask *pmask) +static void guarantee_online_cpus(struct cpuset *cs, struct cpumask *pmask) { while (!cpumask_intersects(cs->cpus_allowed, cpu_online_mask)) cs = parent_cs(cs); @@ -339,7 +328,7 @@ static void guarantee_online_cpus(const struct cpuset *cs, * * Call with callback_mutex held. */ -static void guarantee_online_mems(const struct cpuset *cs, nodemask_t *pmask) +static void guarantee_online_mems(struct cpuset *cs, nodemask_t *pmask) { while (!nodes_intersects(cs->mems_allowed, node_states[N_MEMORY])) cs = parent_cs(cs); @@ -384,7 +373,7 @@ static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q) * alloc_trial_cpuset - allocate a trial cpuset * @cs: the cpuset that the trial cpuset duplicates */ -static struct cpuset *alloc_trial_cpuset(const struct cpuset *cs) +static struct cpuset *alloc_trial_cpuset(struct cpuset *cs) { struct cpuset *trial; @@ -431,9 +420,9 @@ static void free_trial_cpuset(struct cpuset *trial) * Return 0 if valid, -errno if not. */ -static int validate_change(const struct cpuset *cur, const struct cpuset *trial) +static int validate_change(struct cpuset *cur, struct cpuset *trial) { - struct cgroup *cgrp; + struct cgroup_subsys_state *css; struct cpuset *c, *par; int ret; @@ -441,7 +430,7 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) /* Each of our child cpusets must be a subset of us */ ret = -EBUSY; - cpuset_for_each_child(c, cgrp, cur) + cpuset_for_each_child(c, css, cur) if (!is_cpuset_subset(c, trial)) goto out; @@ -462,7 +451,7 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) * overlap */ ret = -EINVAL; - cpuset_for_each_child(c, cgrp, par) { + cpuset_for_each_child(c, css, par) { if ((is_cpu_exclusive(trial) || is_cpu_exclusive(c)) && c != cur && cpumask_intersects(trial->cpus_allowed, c->cpus_allowed)) @@ -475,13 +464,17 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) /* * Cpusets with tasks - existing or newly being attached - can't - * have empty cpus_allowed or mems_allowed. + * be changed to have empty cpus_allowed or mems_allowed. */ ret = -ENOSPC; - if ((cgroup_task_count(cur->css.cgroup) || cur->attach_in_progress) && - (cpumask_empty(trial->cpus_allowed) && - nodes_empty(trial->mems_allowed))) - goto out; + if ((cgroup_task_count(cur->css.cgroup) || cur->attach_in_progress)) { + if (!cpumask_empty(cur->cpus_allowed) && + cpumask_empty(trial->cpus_allowed)) + goto out; + if (!nodes_empty(cur->mems_allowed) && + nodes_empty(trial->mems_allowed)) + goto out; + } ret = 0; out: @@ -511,13 +504,16 @@ static void update_domain_attr_tree(struct sched_domain_attr *dattr, struct cpuset *root_cs) { struct cpuset *cp; - struct cgroup *pos_cgrp; + struct cgroup_subsys_state *pos_css; rcu_read_lock(); - cpuset_for_each_descendant_pre(cp, pos_cgrp, root_cs) { + cpuset_for_each_descendant_pre(cp, pos_css, root_cs) { + if (cp == root_cs) + continue; + /* skip the whole subtree if @cp doesn't have any CPU */ if (cpumask_empty(cp->cpus_allowed)) { - pos_cgrp = cgroup_rightmost_descendant(pos_cgrp); + pos_css = css_rightmost_descendant(pos_css); continue; } @@ -592,7 +588,7 @@ static int generate_sched_domains(cpumask_var_t **domains, struct sched_domain_attr *dattr; /* attributes for custom domains */ int ndoms = 0; /* number of sched domains in result */ int nslot; /* next empty doms[] struct cpumask slot */ - struct cgroup *pos_cgrp; + struct cgroup_subsys_state *pos_css; doms = NULL; dattr = NULL; @@ -621,7 +617,9 @@ static int generate_sched_domains(cpumask_var_t **domains, csn = 0; rcu_read_lock(); - cpuset_for_each_descendant_pre(cp, pos_cgrp, &top_cpuset) { + cpuset_for_each_descendant_pre(cp, pos_css, &top_cpuset) { + if (cp == &top_cpuset) + continue; /* * Continue traversing beyond @cp iff @cp has some CPUs and * isn't load balancing. The former is obvious. The @@ -638,7 +636,7 @@ static int generate_sched_domains(cpumask_var_t **domains, csa[csn++] = cp; /* skip @cp's subtree */ - pos_cgrp = cgroup_rightmost_descendant(pos_cgrp); + pos_css = css_rightmost_descendant(pos_css); } rcu_read_unlock(); @@ -833,52 +831,45 @@ static struct cpuset *effective_nodemask_cpuset(struct cpuset *cs) /** * cpuset_change_cpumask - make a task's cpus_allowed the same as its cpuset's * @tsk: task to test - * @scan: struct cgroup_scanner containing the cgroup of the task + * @data: cpuset to @tsk belongs to * - * Called by cgroup_scan_tasks() for each task in a cgroup whose - * cpus_allowed mask needs to be changed. + * Called by css_scan_tasks() for each task in a cgroup whose cpus_allowed + * mask needs to be changed. * * We don't need to re-check for the cgroup/cpuset membership, since we're * holding cpuset_mutex at this point. */ -static void cpuset_change_cpumask(struct task_struct *tsk, - struct cgroup_scanner *scan) +static void cpuset_change_cpumask(struct task_struct *tsk, void *data) { - struct cpuset *cpus_cs; + struct cpuset *cs = data; + struct cpuset *cpus_cs = effective_cpumask_cpuset(cs); - cpus_cs = effective_cpumask_cpuset(cgroup_cs(scan->cg)); set_cpus_allowed_ptr(tsk, cpus_cs->cpus_allowed); } /** * update_tasks_cpumask - Update the cpumasks of tasks in the cpuset. * @cs: the cpuset in which each task's cpus_allowed mask needs to be changed - * @heap: if NULL, defer allocating heap memory to cgroup_scan_tasks() + * @heap: if NULL, defer allocating heap memory to css_scan_tasks() * * Called with cpuset_mutex held * - * The cgroup_scan_tasks() function will scan all the tasks in a cgroup, + * The css_scan_tasks() function will scan all the tasks in a cgroup, * calling callback functions for each. * - * No return value. It's guaranteed that cgroup_scan_tasks() always returns 0 + * No return value. It's guaranteed that css_scan_tasks() always returns 0 * if @heap != NULL. */ static void update_tasks_cpumask(struct cpuset *cs, struct ptr_heap *heap) { - struct cgroup_scanner scan; - - scan.cg = cs->css.cgroup; - scan.test_task = NULL; - scan.process_task = cpuset_change_cpumask; - scan.heap = heap; - cgroup_scan_tasks(&scan); + css_scan_tasks(&cs->css, NULL, cpuset_change_cpumask, cs, heap); } /* * update_tasks_cpumask_hier - Update the cpumasks of tasks in the hierarchy. * @root_cs: the root cpuset of the hierarchy * @update_root: update root cpuset or not? - * @heap: the heap used by cgroup_scan_tasks() + * @heap: the heap used by css_scan_tasks() * * This will update cpumasks of tasks in @root_cs and all other empty cpusets * which take on cpumask of @root_cs. @@ -889,17 +880,19 @@ static void update_tasks_cpumask_hier(struct cpuset *root_cs, bool update_root, struct ptr_heap *heap) { struct cpuset *cp; - struct cgroup *pos_cgrp; - - if (update_root) - update_tasks_cpumask(root_cs, heap); + struct cgroup_subsys_state *pos_css; rcu_read_lock(); - cpuset_for_each_descendant_pre(cp, pos_cgrp, root_cs) { - /* skip the whole subtree if @cp have some CPU */ - if (!cpumask_empty(cp->cpus_allowed)) { - pos_cgrp = cgroup_rightmost_descendant(pos_cgrp); - continue; + cpuset_for_each_descendant_pre(cp, pos_css, root_cs) { + if (cp == root_cs) { + if (!update_root) + continue; + } else { + /* skip the whole subtree if @cp have some CPU */ + if (!cpumask_empty(cp->cpus_allowed)) { + pos_css = css_rightmost_descendant(pos_css); + continue; + } } if (!css_tryget(&cp->css)) continue; @@ -1055,20 +1048,24 @@ static void cpuset_change_task_nodemask(struct task_struct *tsk, task_unlock(tsk); } +struct cpuset_change_nodemask_arg { + struct cpuset *cs; + nodemask_t *newmems; +}; + /* * Update task's mems_allowed and rebind its mempolicy and vmas' mempolicy * of it to cpuset's new mems_allowed, and migrate pages to new nodes if * memory_migrate flag is set. Called with cpuset_mutex held. */ -static void cpuset_change_nodemask(struct task_struct *p, - struct cgroup_scanner *scan) +static void cpuset_change_nodemask(struct task_struct *p, void *data) { - struct cpuset *cs = cgroup_cs(scan->cg); + struct cpuset_change_nodemask_arg *arg = data; + struct cpuset *cs = arg->cs; struct mm_struct *mm; int migrate; - nodemask_t *newmems = scan->data; - cpuset_change_task_nodemask(p, newmems); + cpuset_change_task_nodemask(p, arg->newmems); mm = get_task_mm(p); if (!mm) @@ -1078,7 +1075,7 @@ static void cpuset_change_nodemask(struct task_struct *p, mpol_rebind_mm(mm, &cs->mems_allowed); if (migrate) - cpuset_migrate_mm(mm, &cs->old_mems_allowed, newmems); + cpuset_migrate_mm(mm, &cs->old_mems_allowed, arg->newmems); mmput(mm); } @@ -1087,28 +1084,22 @@ static void *cpuset_being_rebound; /** * update_tasks_nodemask - Update the nodemasks of tasks in the cpuset. * @cs: the cpuset in which each task's mems_allowed mask needs to be changed - * @heap: if NULL, defer allocating heap memory to cgroup_scan_tasks() + * @heap: if NULL, defer allocating heap memory to css_scan_tasks() * - * Called with cpuset_mutex held - * No return value. It's guaranteed that cgroup_scan_tasks() always returns 0 - * if @heap != NULL. + * Called with cpuset_mutex held. No return value. It's guaranteed that + * css_scan_tasks() always returns 0 if @heap != NULL. */ static void update_tasks_nodemask(struct cpuset *cs, struct ptr_heap *heap) { static nodemask_t newmems; /* protected by cpuset_mutex */ - struct cgroup_scanner scan; struct cpuset *mems_cs = effective_nodemask_cpuset(cs); + struct cpuset_change_nodemask_arg arg = { .cs = cs, + .newmems = &newmems }; cpuset_being_rebound = cs; /* causes mpol_dup() rebind */ guarantee_online_mems(mems_cs, &newmems); - scan.cg = cs->css.cgroup; - scan.test_task = NULL; - scan.process_task = cpuset_change_nodemask; - scan.heap = heap; - scan.data = &newmems; - /* * The mpol_rebind_mm() call takes mmap_sem, which we couldn't * take while holding tasklist_lock. Forks can happen - the @@ -1119,7 +1110,7 @@ static void update_tasks_nodemask(struct cpuset *cs, struct ptr_heap *heap) * It's ok if we rebind the same mm twice; mpol_rebind_mm() * is idempotent. Also migrate pages in each mm to new nodes. */ - cgroup_scan_tasks(&scan); + css_scan_tasks(&cs->css, NULL, cpuset_change_nodemask, &arg, heap); /* * All the tasks' nodemasks have been updated, update @@ -1135,7 +1126,7 @@ static void update_tasks_nodemask(struct cpuset *cs, struct ptr_heap *heap) * update_tasks_nodemask_hier - Update the nodemasks of tasks in the hierarchy. * @cs: the root cpuset of the hierarchy * @update_root: update the root cpuset or not? - * @heap: the heap used by cgroup_scan_tasks() + * @heap: the heap used by css_scan_tasks() * * This will update nodemasks of tasks in @root_cs and all other empty cpusets * which take on nodemask of @root_cs. @@ -1146,17 +1137,19 @@ static void update_tasks_nodemask_hier(struct cpuset *root_cs, bool update_root, struct ptr_heap *heap) { struct cpuset *cp; - struct cgroup *pos_cgrp; - - if (update_root) - update_tasks_nodemask(root_cs, heap); + struct cgroup_subsys_state *pos_css; rcu_read_lock(); - cpuset_for_each_descendant_pre(cp, pos_cgrp, root_cs) { - /* skip the whole subtree if @cp have some CPU */ - if (!nodes_empty(cp->mems_allowed)) { - pos_cgrp = cgroup_rightmost_descendant(pos_cgrp); - continue; + cpuset_for_each_descendant_pre(cp, pos_css, root_cs) { + if (cp == root_cs) { + if (!update_root) + continue; + } else { + /* skip the whole subtree if @cp have some CPU */ + if (!nodes_empty(cp->mems_allowed)) { + pos_css = css_rightmost_descendant(pos_css); + continue; + } } if (!css_tryget(&cp->css)) continue; @@ -1263,44 +1256,39 @@ static int update_relax_domain_level(struct cpuset *cs, s64 val) return 0; } -/* +/** * cpuset_change_flag - make a task's spread flags the same as its cpuset's * @tsk: task to be updated - * @scan: struct cgroup_scanner containing the cgroup of the task + * @data: cpuset to @tsk belongs to * - * Called by cgroup_scan_tasks() for each task in a cgroup. + * Called by css_scan_tasks() for each task in a cgroup. * * We don't need to re-check for the cgroup/cpuset membership, since we're * holding cpuset_mutex at this point. */ -static void cpuset_change_flag(struct task_struct *tsk, - struct cgroup_scanner *scan) +static void cpuset_change_flag(struct task_struct *tsk, void *data) { - cpuset_update_task_spread_flag(cgroup_cs(scan->cg), tsk); + struct cpuset *cs = data; + + cpuset_update_task_spread_flag(cs, tsk); } -/* +/** * update_tasks_flags - update the spread flags of tasks in the cpuset. * @cs: the cpuset in which each task's spread flags needs to be changed - * @heap: if NULL, defer allocating heap memory to cgroup_scan_tasks() + * @heap: if NULL, defer allocating heap memory to css_scan_tasks() * * Called with cpuset_mutex held * - * The cgroup_scan_tasks() function will scan all the tasks in a cgroup, + * The css_scan_tasks() function will scan all the tasks in a cgroup, * calling callback functions for each. * - * No return value. It's guaranteed that cgroup_scan_tasks() always returns 0 + * No return value. It's guaranteed that css_scan_tasks() always returns 0 * if @heap != NULL. */ static void update_tasks_flags(struct cpuset *cs, struct ptr_heap *heap) { - struct cgroup_scanner scan; - - scan.cg = cs->css.cgroup; - scan.test_task = NULL; - scan.process_task = cpuset_change_flag; - scan.heap = heap; - cgroup_scan_tasks(&scan); + css_scan_tasks(&cs->css, NULL, cpuset_change_flag, cs, heap); } /* @@ -1458,9 +1446,10 @@ static int fmeter_getrate(struct fmeter *fmp) } /* Called by cgroups to determine if a cpuset is usable; cpuset_mutex held */ -static int cpuset_can_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) +static int cpuset_can_attach(struct cgroup_subsys_state *css, + struct cgroup_taskset *tset) { - struct cpuset *cs = cgroup_cs(cgrp); + struct cpuset *cs = css_cs(css); struct task_struct *task; int ret; @@ -1471,11 +1460,11 @@ static int cpuset_can_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) * flag is set. */ ret = -ENOSPC; - if (!cgroup_sane_behavior(cgrp) && + if (!cgroup_sane_behavior(css->cgroup) && (cpumask_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed))) goto out_unlock; - cgroup_taskset_for_each(task, cgrp, tset) { + cgroup_taskset_for_each(task, css, tset) { /* * Kthreads which disallow setaffinity shouldn't be moved * to a new cpuset; we don't want to change their cpu @@ -1504,11 +1493,11 @@ out_unlock: return ret; } -static void cpuset_cancel_attach(struct cgroup *cgrp, +static void cpuset_cancel_attach(struct cgroup_subsys_state *css, struct cgroup_taskset *tset) { mutex_lock(&cpuset_mutex); - cgroup_cs(cgrp)->attach_in_progress--; + css_cs(css)->attach_in_progress--; mutex_unlock(&cpuset_mutex); } @@ -1519,16 +1508,18 @@ static void cpuset_cancel_attach(struct cgroup *cgrp, */ static cpumask_var_t cpus_attach; -static void cpuset_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) +static void cpuset_attach(struct cgroup_subsys_state *css, + struct cgroup_taskset *tset) { /* static buf protected by cpuset_mutex */ static nodemask_t cpuset_attach_nodemask_to; struct mm_struct *mm; struct task_struct *task; struct task_struct *leader = cgroup_taskset_first(tset); - struct cgroup *oldcgrp = cgroup_taskset_cur_cgroup(tset); - struct cpuset *cs = cgroup_cs(cgrp); - struct cpuset *oldcs = cgroup_cs(oldcgrp); + struct cgroup_subsys_state *oldcss = cgroup_taskset_cur_css(tset, + cpuset_subsys_id); + struct cpuset *cs = css_cs(css); + struct cpuset *oldcs = css_cs(oldcss); struct cpuset *cpus_cs = effective_cpumask_cpuset(cs); struct cpuset *mems_cs = effective_nodemask_cpuset(cs); @@ -1542,7 +1533,7 @@ static void cpuset_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) guarantee_online_mems(mems_cs, &cpuset_attach_nodemask_to); - cgroup_taskset_for_each(task, cgrp, tset) { + cgroup_taskset_for_each(task, css, tset) { /* * can_attach beforehand should guarantee that this doesn't * fail. TODO: have a better way to handle failure here @@ -1604,15 +1595,18 @@ typedef enum { FILE_SPREAD_SLAB, } cpuset_filetype_t; -static int cpuset_write_u64(struct cgroup *cgrp, struct cftype *cft, u64 val) +static int cpuset_write_u64(struct cgroup_subsys_state *css, struct cftype *cft, + u64 val) { - struct cpuset *cs = cgroup_cs(cgrp); + struct cpuset *cs = css_cs(css); cpuset_filetype_t type = cft->private; - int retval = -ENODEV; + int retval = 0; mutex_lock(&cpuset_mutex); - if (!is_cpuset_online(cs)) + if (!is_cpuset_online(cs)) { + retval = -ENODEV; goto out_unlock; + } switch (type) { case FILE_CPU_EXCLUSIVE: @@ -1651,9 +1645,10 @@ out_unlock: return retval; } -static int cpuset_write_s64(struct cgroup *cgrp, struct cftype *cft, s64 val) +static int cpuset_write_s64(struct cgroup_subsys_state *css, struct cftype *cft, + s64 val) { - struct cpuset *cs = cgroup_cs(cgrp); + struct cpuset *cs = css_cs(css); cpuset_filetype_t type = cft->private; int retval = -ENODEV; @@ -1677,10 +1672,10 @@ out_unlock: /* * Common handling for a write to a "cpus" or "mems" file. */ -static int cpuset_write_resmask(struct cgroup *cgrp, struct cftype *cft, - const char *buf) +static int cpuset_write_resmask(struct cgroup_subsys_state *css, + struct cftype *cft, const char *buf) { - struct cpuset *cs = cgroup_cs(cgrp); + struct cpuset *cs = css_cs(css); struct cpuset *trialcs; int retval = -ENODEV; @@ -1759,13 +1754,12 @@ static size_t cpuset_sprintf_memlist(char *page, struct cpuset *cs) return count; } -static ssize_t cpuset_common_file_read(struct cgroup *cgrp, - struct cftype *cft, - struct file *file, - char __user *buf, - size_t nbytes, loff_t *ppos) +static ssize_t cpuset_common_file_read(struct cgroup_subsys_state *css, + struct cftype *cft, struct file *file, + char __user *buf, size_t nbytes, + loff_t *ppos) { - struct cpuset *cs = cgroup_cs(cgrp); + struct cpuset *cs = css_cs(css); cpuset_filetype_t type = cft->private; char *page; ssize_t retval = 0; @@ -1795,9 +1789,9 @@ out: return retval; } -static u64 cpuset_read_u64(struct cgroup *cgrp, struct cftype *cft) +static u64 cpuset_read_u64(struct cgroup_subsys_state *css, struct cftype *cft) { - struct cpuset *cs = cgroup_cs(cgrp); + struct cpuset *cs = css_cs(css); cpuset_filetype_t type = cft->private; switch (type) { case FILE_CPU_EXCLUSIVE: @@ -1826,9 +1820,9 @@ static u64 cpuset_read_u64(struct cgroup *cgrp, struct cftype *cft) return 0; } -static s64 cpuset_read_s64(struct cgroup *cgrp, struct cftype *cft) +static s64 cpuset_read_s64(struct cgroup_subsys_state *css, struct cftype *cft) { - struct cpuset *cs = cgroup_cs(cgrp); + struct cpuset *cs = css_cs(css); cpuset_filetype_t type = cft->private; switch (type) { case FILE_SCHED_RELAX_DOMAIN_LEVEL: @@ -1943,11 +1937,12 @@ static struct cftype files[] = { * cgrp: control group that the new cpuset will be part of */ -static struct cgroup_subsys_state *cpuset_css_alloc(struct cgroup *cgrp) +static struct cgroup_subsys_state * +cpuset_css_alloc(struct cgroup_subsys_state *parent_css) { struct cpuset *cs; - if (!cgrp->parent) + if (!parent_css) return &top_cpuset.css; cs = kzalloc(sizeof(*cs), GFP_KERNEL); @@ -1967,12 +1962,12 @@ static struct cgroup_subsys_state *cpuset_css_alloc(struct cgroup *cgrp) return &cs->css; } -static int cpuset_css_online(struct cgroup *cgrp) +static int cpuset_css_online(struct cgroup_subsys_state *css) { - struct cpuset *cs = cgroup_cs(cgrp); + struct cpuset *cs = css_cs(css); struct cpuset *parent = parent_cs(cs); struct cpuset *tmp_cs; - struct cgroup *pos_cg; + struct cgroup_subsys_state *pos_css; if (!parent) return 0; @@ -1987,7 +1982,7 @@ static int cpuset_css_online(struct cgroup *cgrp) number_of_cpusets++; - if (!test_bit(CGRP_CPUSET_CLONE_CHILDREN, &cgrp->flags)) + if (!test_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags)) goto out_unlock; /* @@ -2004,7 +1999,7 @@ static int cpuset_css_online(struct cgroup *cgrp) * (and likewise for mems) to the new cgroup. */ rcu_read_lock(); - cpuset_for_each_child(tmp_cs, pos_cg, parent) { + cpuset_for_each_child(tmp_cs, pos_css, parent) { if (is_mem_exclusive(tmp_cs) || is_cpu_exclusive(tmp_cs)) { rcu_read_unlock(); goto out_unlock; @@ -2021,9 +2016,15 @@ out_unlock: return 0; } -static void cpuset_css_offline(struct cgroup *cgrp) +/* + * If the cpuset being removed has its flag 'sched_load_balance' + * enabled, then simulate turning sched_load_balance off, which + * will call rebuild_sched_domains_locked(). + */ + +static void cpuset_css_offline(struct cgroup_subsys_state *css) { - struct cpuset *cs = cgroup_cs(cgrp); + struct cpuset *cs = css_cs(css); mutex_lock(&cpuset_mutex); @@ -2036,15 +2037,9 @@ static void cpuset_css_offline(struct cgroup *cgrp) mutex_unlock(&cpuset_mutex); } -/* - * If the cpuset being removed has its flag 'sched_load_balance' - * enabled, then simulate turning sched_load_balance off, which - * will call rebuild_sched_domains_locked(). - */ - -static void cpuset_css_free(struct cgroup *cgrp) +static void cpuset_css_free(struct cgroup_subsys_state *css) { - struct cpuset *cs = cgroup_cs(cgrp); + struct cpuset *cs = css_cs(css); free_cpumask_var(cs->cpus_allowed); kfree(cs); @@ -2251,11 +2246,11 @@ static void cpuset_hotplug_workfn(struct work_struct *work) /* if cpus or mems changed, we need to propagate to descendants */ if (cpus_updated || mems_updated) { struct cpuset *cs; - struct cgroup *pos_cgrp; + struct cgroup_subsys_state *pos_css; rcu_read_lock(); - cpuset_for_each_descendant_pre(cs, pos_cgrp, &top_cpuset) { - if (!css_tryget(&cs->css)) + cpuset_for_each_descendant_pre(cs, pos_css, &top_cpuset) { + if (cs == &top_cpuset || !css_tryget(&cs->css)) continue; rcu_read_unlock(); @@ -2344,7 +2339,7 @@ void cpuset_cpus_allowed(struct task_struct *tsk, struct cpumask *pmask) void cpuset_cpus_allowed_fallback(struct task_struct *tsk) { - const struct cpuset *cpus_cs; + struct cpuset *cpus_cs; rcu_read_lock(); cpus_cs = effective_cpumask_cpuset(task_cs(tsk)); @@ -2417,7 +2412,7 @@ int cpuset_nodemask_valid_mems_allowed(nodemask_t *nodemask) * callback_mutex. If no ancestor is mem_exclusive or mem_hardwall * (an unusual configuration), then returns the root cpuset. */ -static const struct cpuset *nearest_hardwall_ancestor(const struct cpuset *cs) +static struct cpuset *nearest_hardwall_ancestor(struct cpuset *cs) { while (!(is_mem_exclusive(cs) || is_mem_hardwall(cs)) && parent_cs(cs)) cs = parent_cs(cs); @@ -2487,7 +2482,7 @@ static const struct cpuset *nearest_hardwall_ancestor(const struct cpuset *cs) */ int __cpuset_node_allowed_softwall(int node, gfp_t gfp_mask) { - const struct cpuset *cs; /* current cpuset ancestors */ + struct cpuset *cs; /* current cpuset ancestors */ int allowed; /* is allocation in zone z allowed? */ if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) @@ -2725,7 +2720,7 @@ int proc_cpuset_show(struct seq_file *m, void *unused_v) goto out_free; rcu_read_lock(); - css = task_subsys_state(tsk, cpuset_subsys_id); + css = task_css(tsk, cpuset_subsys_id); retval = cgroup_path(css->cgroup, buf, PAGE_SIZE); rcu_read_unlock(); if (retval < 0) diff --git a/kernel/events/core.c b/kernel/events/core.c index f86599e8c123..9300f5226077 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -340,8 +340,8 @@ struct perf_cgroup { static inline struct perf_cgroup * perf_cgroup_from_task(struct task_struct *task) { - return container_of(task_subsys_state(task, perf_subsys_id), - struct perf_cgroup, css); + return container_of(task_css(task, perf_subsys_id), + struct perf_cgroup, css); } static inline bool @@ -591,7 +591,9 @@ static inline int perf_cgroup_connect(int fd, struct perf_event *event, if (!f.file) return -EBADF; - css = cgroup_css_from_dir(f.file, perf_subsys_id); + rcu_read_lock(); + + css = css_from_dir(f.file->f_dentry, &perf_subsys); if (IS_ERR(css)) { ret = PTR_ERR(css); goto out; @@ -617,6 +619,7 @@ static inline int perf_cgroup_connect(int fd, struct perf_event *event, ret = -EINVAL; } out: + rcu_read_unlock(); fdput(f); return ret; } @@ -7798,7 +7801,8 @@ unlock: device_initcall(perf_event_sysfs_init); #ifdef CONFIG_CGROUP_PERF -static struct cgroup_subsys_state *perf_cgroup_css_alloc(struct cgroup *cont) +static struct cgroup_subsys_state * +perf_cgroup_css_alloc(struct cgroup_subsys_state *parent_css) { struct perf_cgroup *jc; @@ -7815,11 +7819,10 @@ static struct cgroup_subsys_state *perf_cgroup_css_alloc(struct cgroup *cont) return &jc->css; } -static void perf_cgroup_css_free(struct cgroup *cont) +static void perf_cgroup_css_free(struct cgroup_subsys_state *css) { - struct perf_cgroup *jc; - jc = container_of(cgroup_subsys_state(cont, perf_subsys_id), - struct perf_cgroup, css); + struct perf_cgroup *jc = container_of(css, struct perf_cgroup, css); + free_percpu(jc->info); kfree(jc); } @@ -7831,15 +7834,17 @@ static int __perf_cgroup_move(void *info) return 0; } -static void perf_cgroup_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) +static void perf_cgroup_attach(struct cgroup_subsys_state *css, + struct cgroup_taskset *tset) { struct task_struct *task; - cgroup_taskset_for_each(task, cgrp, tset) + cgroup_taskset_for_each(task, css, tset) task_function_call(task, __perf_cgroup_move, task); } -static void perf_cgroup_exit(struct cgroup *cgrp, struct cgroup *old_cgrp, +static void perf_cgroup_exit(struct cgroup_subsys_state *css, + struct cgroup_subsys_state *old_css, struct task_struct *task) { /* diff --git a/kernel/fork.c b/kernel/fork.c index 403d2bb8a968..bf46287c91a4 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1177,7 +1177,8 @@ static struct task_struct *copy_process(unsigned long clone_flags, * don't allow the creation of threads. */ if ((clone_flags & (CLONE_VM|CLONE_NEWPID)) && - (task_active_pid_ns(current) != current->nsproxy->pid_ns)) + (task_active_pid_ns(current) != + current->nsproxy->pid_ns_for_children)) return ERR_PTR(-EINVAL); retval = security_task_create(clone_flags); @@ -1351,7 +1352,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, if (pid != &init_struct_pid) { retval = -ENOMEM; - pid = alloc_pid(p->nsproxy->pid_ns); + pid = alloc_pid(p->nsproxy->pid_ns_for_children); if (!pid) goto bad_fork_cleanup_io; } @@ -1679,6 +1680,12 @@ SYSCALL_DEFINE5(clone, unsigned long, newsp, unsigned long, clone_flags, int __user *, parent_tidptr, int __user *, child_tidptr, int, tls_val) +#elif defined(CONFIG_CLONE_BACKWARDS3) +SYSCALL_DEFINE6(clone, unsigned long, clone_flags, unsigned long, newsp, + int, stack_size, + int __user *, parent_tidptr, + int __user *, child_tidptr, + int, tls_val) #else SYSCALL_DEFINE5(clone, unsigned long, clone_flags, unsigned long, newsp, int __user *, parent_tidptr, diff --git a/kernel/freezer.c b/kernel/freezer.c index 8b2afc1c9df0..b462fa197517 100644 --- a/kernel/freezer.c +++ b/kernel/freezer.c @@ -33,7 +33,7 @@ static DEFINE_SPINLOCK(freezer_lock); */ bool freezing_slow_path(struct task_struct *p) { - if (p->flags & PF_NOFREEZE) + if (p->flags & (PF_NOFREEZE | PF_SUSPEND_TASK)) return false; if (pm_nosig_freezing || cgroup_freezing(p)) diff --git a/kernel/mutex.c b/kernel/mutex.c index 98164a55a4dc..6d647aedffea 100644 --- a/kernel/mutex.c +++ b/kernel/mutex.c @@ -683,7 +683,7 @@ __ww_mutex_lock(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) might_sleep(); ret = __mutex_lock_common(&lock->base, TASK_UNINTERRUPTIBLE, 0, &ctx->dep_map, _RET_IP_, ctx); - if (!ret && ctx->acquired > 0) + if (!ret && ctx->acquired > 1) return ww_mutex_deadlock_injection(lock, ctx); return ret; @@ -699,7 +699,7 @@ __ww_mutex_lock_interruptible(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) ret = __mutex_lock_common(&lock->base, TASK_INTERRUPTIBLE, 0, &ctx->dep_map, _RET_IP_, ctx); - if (!ret && ctx->acquired > 0) + if (!ret && ctx->acquired > 1) return ww_mutex_deadlock_injection(lock, ctx); return ret; diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c index 364ceab15f0c..997cbb951a3b 100644 --- a/kernel/nsproxy.c +++ b/kernel/nsproxy.c @@ -29,15 +29,15 @@ static struct kmem_cache *nsproxy_cachep; struct nsproxy init_nsproxy = { - .count = ATOMIC_INIT(1), - .uts_ns = &init_uts_ns, + .count = ATOMIC_INIT(1), + .uts_ns = &init_uts_ns, #if defined(CONFIG_POSIX_MQUEUE) || defined(CONFIG_SYSVIPC) - .ipc_ns = &init_ipc_ns, + .ipc_ns = &init_ipc_ns, #endif - .mnt_ns = NULL, - .pid_ns = &init_pid_ns, + .mnt_ns = NULL, + .pid_ns_for_children = &init_pid_ns, #ifdef CONFIG_NET - .net_ns = &init_net, + .net_ns = &init_net, #endif }; @@ -85,9 +85,10 @@ static struct nsproxy *create_new_namespaces(unsigned long flags, goto out_ipc; } - new_nsp->pid_ns = copy_pid_ns(flags, user_ns, tsk->nsproxy->pid_ns); - if (IS_ERR(new_nsp->pid_ns)) { - err = PTR_ERR(new_nsp->pid_ns); + new_nsp->pid_ns_for_children = + copy_pid_ns(flags, user_ns, tsk->nsproxy->pid_ns_for_children); + if (IS_ERR(new_nsp->pid_ns_for_children)) { + err = PTR_ERR(new_nsp->pid_ns_for_children); goto out_pid; } @@ -100,8 +101,8 @@ static struct nsproxy *create_new_namespaces(unsigned long flags, return new_nsp; out_net: - if (new_nsp->pid_ns) - put_pid_ns(new_nsp->pid_ns); + if (new_nsp->pid_ns_for_children) + put_pid_ns(new_nsp->pid_ns_for_children); out_pid: if (new_nsp->ipc_ns) put_ipc_ns(new_nsp->ipc_ns); @@ -174,8 +175,8 @@ void free_nsproxy(struct nsproxy *ns) put_uts_ns(ns->uts_ns); if (ns->ipc_ns) put_ipc_ns(ns->ipc_ns); - if (ns->pid_ns) - put_pid_ns(ns->pid_ns); + if (ns->pid_ns_for_children) + put_pid_ns(ns->pid_ns_for_children); put_net(ns->net_ns); kmem_cache_free(nsproxy_cachep, ns); } diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index 6917e8edb48e..601bb361c235 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -349,8 +349,8 @@ static int pidns_install(struct nsproxy *nsproxy, void *ns) if (ancestor != active) return -EINVAL; - put_pid_ns(nsproxy->pid_ns); - nsproxy->pid_ns = get_pid_ns(new); + put_pid_ns(nsproxy->pid_ns_for_children); + nsproxy->pid_ns_for_children = get_pid_ns(new); return 0; } diff --git a/kernel/power/process.c b/kernel/power/process.c index fc0df8486449..06ec8869dbf1 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -109,6 +109,8 @@ static int try_to_freeze_tasks(bool user_only) /** * freeze_processes - Signal user space processes to enter the refrigerator. + * The current thread will not be frozen. The same process that calls + * freeze_processes must later call thaw_processes. * * On success, returns 0. On failure, -errno and system is fully thawed. */ @@ -120,6 +122,9 @@ int freeze_processes(void) if (error) return error; + /* Make sure this task doesn't get frozen */ + current->flags |= PF_SUSPEND_TASK; + if (!pm_freezing) atomic_inc(&system_freezing_cnt); @@ -168,6 +173,7 @@ int freeze_kernel_threads(void) void thaw_processes(void) { struct task_struct *g, *p; + struct task_struct *curr = current; if (pm_freezing) atomic_dec(&system_freezing_cnt); @@ -182,10 +188,15 @@ void thaw_processes(void) read_lock(&tasklist_lock); do_each_thread(g, p) { + /* No other threads should have PF_SUSPEND_TASK set */ + WARN_ON((p != curr) && (p->flags & PF_SUSPEND_TASK)); __thaw_task(p); } while_each_thread(g, p); read_unlock(&tasklist_lock); + WARN_ON(!(curr->flags & PF_SUSPEND_TASK)); + curr->flags &= ~PF_SUSPEND_TASK; + usermodehelper_enable(); schedule(); diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 06fe28589e9c..a394297f8b2f 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -296,6 +296,17 @@ int pm_qos_request_active(struct pm_qos_request *req) } EXPORT_SYMBOL_GPL(pm_qos_request_active); +static void __pm_qos_update_request(struct pm_qos_request *req, + s32 new_value) +{ + trace_pm_qos_update_request(req->pm_qos_class, new_value); + + if (new_value != req->node.prio) + pm_qos_update_target( + pm_qos_array[req->pm_qos_class]->constraints, + &req->node, PM_QOS_UPDATE_REQ, new_value); +} + /** * pm_qos_work_fn - the timeout handler of pm_qos_update_request_timeout * @work: work struct for the delayed work (timeout) @@ -308,7 +319,7 @@ static void pm_qos_work_fn(struct work_struct *work) struct pm_qos_request, work); - pm_qos_update_request(req, PM_QOS_DEFAULT_VALUE); + __pm_qos_update_request(req, PM_QOS_DEFAULT_VALUE); } /** @@ -364,12 +375,7 @@ void pm_qos_update_request(struct pm_qos_request *req, } cancel_delayed_work_sync(&req->work); - - trace_pm_qos_update_request(req->pm_qos_class, new_value); - if (new_value != req->node.prio) - pm_qos_update_target( - pm_qos_array[req->pm_qos_class]->constraints, - &req->node, PM_QOS_UPDATE_REQ, new_value); + __pm_qos_update_request(req, new_value); } EXPORT_SYMBOL_GPL(pm_qos_update_request); diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index ece04223bb1e..62ee437b5c7e 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -210,6 +210,7 @@ static int suspend_enter(suspend_state_t state, bool *wakeup) goto Platform_wake; } + ftrace_stop(); error = disable_nonboot_cpus(); if (error || suspend_test(TEST_CPUS)) goto Enable_cpus; @@ -232,6 +233,7 @@ static int suspend_enter(suspend_state_t state, bool *wakeup) Enable_cpus: enable_nonboot_cpus(); + ftrace_start(); Platform_wake: if (need_suspend_ops(state) && suspend_ops->wake) @@ -265,7 +267,6 @@ int suspend_devices_and_enter(suspend_state_t state) goto Close; } suspend_console(); - ftrace_stop(); suspend_test_start(); error = dpm_suspend_start(PMSG_SUSPEND); if (error) { @@ -285,7 +286,6 @@ int suspend_devices_and_enter(suspend_state_t state) suspend_test_start(); dpm_resume_end(PMSG_RESUME); suspend_test_finish("resume devices"); - ftrace_start(); resume_console(); Close: if (need_suspend_ops(state) && suspend_ops->end) diff --git a/kernel/printk/Makefile b/kernel/printk/Makefile new file mode 100644 index 000000000000..85405bdcf2b3 --- /dev/null +++ b/kernel/printk/Makefile @@ -0,0 +1,2 @@ +obj-y = printk.o +obj-$(CONFIG_A11Y_BRAILLE_CONSOLE) += braille.o diff --git a/kernel/printk/braille.c b/kernel/printk/braille.c new file mode 100644 index 000000000000..276762f3a460 --- /dev/null +++ b/kernel/printk/braille.c @@ -0,0 +1,49 @@ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/kernel.h> +#include <linux/console.h> +#include <linux/string.h> + +#include "console_cmdline.h" +#include "braille.h" + +char *_braille_console_setup(char **str, char **brl_options) +{ + if (!memcmp(*str, "brl,", 4)) { + *brl_options = ""; + *str += 4; + } else if (!memcmp(str, "brl=", 4)) { + *brl_options = *str + 4; + *str = strchr(*brl_options, ','); + if (!*str) + pr_err("need port name after brl=\n"); + else + *((*str)++) = 0; + } else + return NULL; + + return *str; +} + +int +_braille_register_console(struct console *console, struct console_cmdline *c) +{ + int rtn = 0; + + if (c->brl_options) { + console->flags |= CON_BRL; + rtn = braille_register_console(console, c->index, c->options, + c->brl_options); + } + + return rtn; +} + +int +_braille_unregister_console(struct console *console) +{ + if (console->flags & CON_BRL) + return braille_unregister_console(console); + + return 0; +} diff --git a/kernel/printk/braille.h b/kernel/printk/braille.h new file mode 100644 index 000000000000..769d771145c8 --- /dev/null +++ b/kernel/printk/braille.h @@ -0,0 +1,48 @@ +#ifndef _PRINTK_BRAILLE_H +#define _PRINTK_BRAILLE_H + +#ifdef CONFIG_A11Y_BRAILLE_CONSOLE + +static inline void +braille_set_options(struct console_cmdline *c, char *brl_options) +{ + c->brl_options = brl_options; +} + +char * +_braille_console_setup(char **str, char **brl_options); + +int +_braille_register_console(struct console *console, struct console_cmdline *c); + +int +_braille_unregister_console(struct console *console); + +#else + +static inline void +braille_set_options(struct console_cmdline *c, char *brl_options) +{ +} + +static inline char * +_braille_console_setup(char **str, char **brl_options) +{ + return NULL; +} + +static inline int +_braille_register_console(struct console *console, struct console_cmdline *c) +{ + return 0; +} + +static inline int +_braille_unregister_console(struct console *console) +{ + return 0; +} + +#endif + +#endif diff --git a/kernel/printk/console_cmdline.h b/kernel/printk/console_cmdline.h new file mode 100644 index 000000000000..cbd69d842341 --- /dev/null +++ b/kernel/printk/console_cmdline.h @@ -0,0 +1,14 @@ +#ifndef _CONSOLE_CMDLINE_H +#define _CONSOLE_CMDLINE_H + +struct console_cmdline +{ + char name[8]; /* Name of the driver */ + int index; /* Minor dev. to use */ + char *options; /* Options for the driver */ +#ifdef CONFIG_A11Y_BRAILLE_CONSOLE + char *brl_options; /* Options for braille driver */ +#endif +}; + +#endif diff --git a/kernel/printk.c b/kernel/printk/printk.c index 69b0890ed7e5..b4e8500afdb3 100644 --- a/kernel/printk.c +++ b/kernel/printk/printk.c @@ -51,6 +51,9 @@ #define CREATE_TRACE_POINTS #include <trace/events/printk.h> +#include "console_cmdline.h" +#include "braille.h" + /* printk's without a loglevel use this.. */ #define DEFAULT_MESSAGE_LOGLEVEL CONFIG_DEFAULT_MESSAGE_LOGLEVEL @@ -105,19 +108,11 @@ static struct console *exclusive_console; /* * Array of consoles built from command line options (console=) */ -struct console_cmdline -{ - char name[8]; /* Name of the driver */ - int index; /* Minor dev. to use */ - char *options; /* Options for the driver */ -#ifdef CONFIG_A11Y_BRAILLE_CONSOLE - char *brl_options; /* Options for braille driver */ -#endif -}; #define MAX_CMDLINECONSOLES 8 static struct console_cmdline console_cmdline[MAX_CMDLINECONSOLES]; + static int selected_console = -1; static int preferred_console = -1; int console_set_on_cmdline; @@ -178,7 +173,7 @@ static int console_may_schedule; * 67 "g" * 0032 00 00 00 padding to next message header * - * The 'struct log' buffer header must never be directly exported to + * The 'struct printk_log' buffer header must never be directly exported to * userspace, it is a kernel-private implementation detail that might * need to be changed in the future, when the requirements change. * @@ -200,7 +195,7 @@ enum log_flags { LOG_CONT = 8, /* text is a fragment of a continuation line */ }; -struct log { +struct printk_log { u64 ts_nsec; /* timestamp in nanoseconds */ u16 len; /* length of entire record */ u16 text_len; /* length of text buffer */ @@ -248,7 +243,7 @@ static u32 clear_idx; #if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) #define LOG_ALIGN 4 #else -#define LOG_ALIGN __alignof__(struct log) +#define LOG_ALIGN __alignof__(struct printk_log) #endif #define __LOG_BUF_LEN (1 << CONFIG_LOG_BUF_SHIFT) static char __log_buf[__LOG_BUF_LEN] __aligned(LOG_ALIGN); @@ -259,35 +254,35 @@ static u32 log_buf_len = __LOG_BUF_LEN; static volatile unsigned int logbuf_cpu = UINT_MAX; /* human readable text of the record */ -static char *log_text(const struct log *msg) +static char *log_text(const struct printk_log *msg) { - return (char *)msg + sizeof(struct log); + return (char *)msg + sizeof(struct printk_log); } /* optional key/value pair dictionary attached to the record */ -static char *log_dict(const struct log *msg) +static char *log_dict(const struct printk_log *msg) { - return (char *)msg + sizeof(struct log) + msg->text_len; + return (char *)msg + sizeof(struct printk_log) + msg->text_len; } /* get record by index; idx must point to valid msg */ -static struct log *log_from_idx(u32 idx) +static struct printk_log *log_from_idx(u32 idx) { - struct log *msg = (struct log *)(log_buf + idx); + struct printk_log *msg = (struct printk_log *)(log_buf + idx); /* * A length == 0 record is the end of buffer marker. Wrap around and * read the message at the start of the buffer. */ if (!msg->len) - return (struct log *)log_buf; + return (struct printk_log *)log_buf; return msg; } /* get next record; idx must point to valid msg */ static u32 log_next(u32 idx) { - struct log *msg = (struct log *)(log_buf + idx); + struct printk_log *msg = (struct printk_log *)(log_buf + idx); /* length == 0 indicates the end of the buffer; wrap */ /* @@ -296,7 +291,7 @@ static u32 log_next(u32 idx) * return the one after that. */ if (!msg->len) { - msg = (struct log *)log_buf; + msg = (struct printk_log *)log_buf; return msg->len; } return idx + msg->len; @@ -308,11 +303,11 @@ static void log_store(int facility, int level, const char *dict, u16 dict_len, const char *text, u16 text_len) { - struct log *msg; + struct printk_log *msg; u32 size, pad_len; /* number of '\0' padding bytes to next message */ - size = sizeof(struct log) + text_len + dict_len; + size = sizeof(struct printk_log) + text_len + dict_len; pad_len = (-size) & (LOG_ALIGN - 1); size += pad_len; @@ -324,7 +319,7 @@ static void log_store(int facility, int level, else free = log_first_idx - log_next_idx; - if (free > size + sizeof(struct log)) + if (free > size + sizeof(struct printk_log)) break; /* drop old messages until we have enough contiuous space */ @@ -332,18 +327,18 @@ static void log_store(int facility, int level, log_first_seq++; } - if (log_next_idx + size + sizeof(struct log) >= log_buf_len) { + if (log_next_idx + size + sizeof(struct printk_log) >= log_buf_len) { /* * This message + an additional empty header does not fit * at the end of the buffer. Add an empty header with len == 0 * to signify a wrap around. */ - memset(log_buf + log_next_idx, 0, sizeof(struct log)); + memset(log_buf + log_next_idx, 0, sizeof(struct printk_log)); log_next_idx = 0; } /* fill message */ - msg = (struct log *)(log_buf + log_next_idx); + msg = (struct printk_log *)(log_buf + log_next_idx); memcpy(log_text(msg), text, text_len); msg->text_len = text_len; memcpy(log_dict(msg), dict, dict_len); @@ -356,7 +351,7 @@ static void log_store(int facility, int level, else msg->ts_nsec = local_clock(); memset(log_dict(msg) + dict_len, 0, pad_len); - msg->len = sizeof(struct log) + text_len + dict_len + pad_len; + msg->len = sizeof(struct printk_log) + text_len + dict_len + pad_len; /* insert message */ log_next_idx += msg->len; @@ -479,7 +474,7 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { struct devkmsg_user *user = file->private_data; - struct log *msg; + struct printk_log *msg; u64 ts_usec; size_t i; char cont = '-'; @@ -724,14 +719,14 @@ void log_buf_kexec_setup(void) VMCOREINFO_SYMBOL(log_first_idx); VMCOREINFO_SYMBOL(log_next_idx); /* - * Export struct log size and field offsets. User space tools can + * Export struct printk_log size and field offsets. User space tools can * parse it and detect any changes to structure down the line. */ - VMCOREINFO_STRUCT_SIZE(log); - VMCOREINFO_OFFSET(log, ts_nsec); - VMCOREINFO_OFFSET(log, len); - VMCOREINFO_OFFSET(log, text_len); - VMCOREINFO_OFFSET(log, dict_len); + VMCOREINFO_STRUCT_SIZE(printk_log); + VMCOREINFO_OFFSET(printk_log, ts_nsec); + VMCOREINFO_OFFSET(printk_log, len); + VMCOREINFO_OFFSET(printk_log, text_len); + VMCOREINFO_OFFSET(printk_log, dict_len); } #endif @@ -884,7 +879,7 @@ static size_t print_time(u64 ts, char *buf) (unsigned long)ts, rem_nsec / 1000); } -static size_t print_prefix(const struct log *msg, bool syslog, char *buf) +static size_t print_prefix(const struct printk_log *msg, bool syslog, char *buf) { size_t len = 0; unsigned int prefix = (msg->facility << 3) | msg->level; @@ -907,7 +902,7 @@ static size_t print_prefix(const struct log *msg, bool syslog, char *buf) return len; } -static size_t msg_print_text(const struct log *msg, enum log_flags prev, +static size_t msg_print_text(const struct printk_log *msg, enum log_flags prev, bool syslog, char *buf, size_t size) { const char *text = log_text(msg); @@ -969,7 +964,7 @@ static size_t msg_print_text(const struct log *msg, enum log_flags prev, static int syslog_print(char __user *buf, int size) { char *text; - struct log *msg; + struct printk_log *msg; int len = 0; text = kmalloc(LOG_LINE_MAX + PREFIX_MAX, GFP_KERNEL); @@ -1060,7 +1055,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear) idx = clear_idx; prev = 0; while (seq < log_next_seq) { - struct log *msg = log_from_idx(idx); + struct printk_log *msg = log_from_idx(idx); len += msg_print_text(msg, prev, true, NULL, 0); prev = msg->flags; @@ -1073,7 +1068,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear) idx = clear_idx; prev = 0; while (len > size && seq < log_next_seq) { - struct log *msg = log_from_idx(idx); + struct printk_log *msg = log_from_idx(idx); len -= msg_print_text(msg, prev, true, NULL, 0); prev = msg->flags; @@ -1087,7 +1082,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear) len = 0; prev = 0; while (len >= 0 && seq < next_seq) { - struct log *msg = log_from_idx(idx); + struct printk_log *msg = log_from_idx(idx); int textlen; textlen = msg_print_text(msg, prev, true, text, @@ -1233,7 +1228,7 @@ int do_syslog(int type, char __user *buf, int len, bool from_file) error = 0; while (seq < log_next_seq) { - struct log *msg = log_from_idx(idx); + struct printk_log *msg = log_from_idx(idx); error += msg_print_text(msg, prev, true, NULL, 0); idx = log_next(idx); @@ -1719,10 +1714,10 @@ static struct cont { u8 level; bool flushed:1; } cont; -static struct log *log_from_idx(u32 idx) { return NULL; } +static struct printk_log *log_from_idx(u32 idx) { return NULL; } static u32 log_next(u32 idx) { return 0; } static void call_console_drivers(int level, const char *text, size_t len) {} -static size_t msg_print_text(const struct log *msg, enum log_flags prev, +static size_t msg_print_text(const struct printk_log *msg, enum log_flags prev, bool syslog, char *buf, size_t size) { return 0; } static size_t cont_print_text(char *text, size_t size) { return 0; } @@ -1761,23 +1756,23 @@ static int __add_preferred_console(char *name, int idx, char *options, * See if this tty is not yet registered, and * if we have a slot free. */ - for (i = 0; i < MAX_CMDLINECONSOLES && console_cmdline[i].name[0]; i++) - if (strcmp(console_cmdline[i].name, name) == 0 && - console_cmdline[i].index == idx) { - if (!brl_options) - selected_console = i; - return 0; + for (i = 0, c = console_cmdline; + i < MAX_CMDLINECONSOLES && c->name[0]; + i++, c++) { + if (strcmp(c->name, name) == 0 && c->index == idx) { + if (!brl_options) + selected_console = i; + return 0; } + } if (i == MAX_CMDLINECONSOLES) return -E2BIG; if (!brl_options) selected_console = i; - c = &console_cmdline[i]; strlcpy(c->name, name, sizeof(c->name)); c->options = options; -#ifdef CONFIG_A11Y_BRAILLE_CONSOLE - c->brl_options = brl_options; -#endif + braille_set_options(c, brl_options); + c->index = idx; return 0; } @@ -1790,20 +1785,8 @@ static int __init console_setup(char *str) char *s, *options, *brl_options = NULL; int idx; -#ifdef CONFIG_A11Y_BRAILLE_CONSOLE - if (!memcmp(str, "brl,", 4)) { - brl_options = ""; - str += 4; - } else if (!memcmp(str, "brl=", 4)) { - brl_options = str + 4; - str = strchr(brl_options, ','); - if (!str) { - printk(KERN_ERR "need port name after brl=\n"); - return 1; - } - *(str++) = 0; - } -#endif + if (_braille_console_setup(&str, &brl_options)) + return 1; /* * Decode str into name, index, options. @@ -1858,15 +1841,15 @@ int update_console_cmdline(char *name, int idx, char *name_new, int idx_new, cha struct console_cmdline *c; int i; - for (i = 0; i < MAX_CMDLINECONSOLES && console_cmdline[i].name[0]; i++) - if (strcmp(console_cmdline[i].name, name) == 0 && - console_cmdline[i].index == idx) { - c = &console_cmdline[i]; - strlcpy(c->name, name_new, sizeof(c->name)); - c->name[sizeof(c->name) - 1] = 0; - c->options = options; - c->index = idx_new; - return i; + for (i = 0, c = console_cmdline; + i < MAX_CMDLINECONSOLES && c->name[0]; + i++, c++) + if (strcmp(c->name, name) == 0 && c->index == idx) { + strlcpy(c->name, name_new, sizeof(c->name)); + c->name[sizeof(c->name) - 1] = 0; + c->options = options; + c->index = idx_new; + return i; } /* not found */ return -1; @@ -2046,7 +2029,7 @@ void console_unlock(void) console_cont_flush(text, sizeof(text)); again: for (;;) { - struct log *msg; + struct printk_log *msg; size_t len; int level; @@ -2241,6 +2224,14 @@ void register_console(struct console *newcon) int i; unsigned long flags; struct console *bcon = NULL; + struct console_cmdline *c; + + if (console_drivers) + for_each_console(bcon) + if (WARN(bcon == newcon, + "console '%s%d' already registered\n", + bcon->name, bcon->index)) + return; /* * before we register a new CON_BOOT console, make sure we don't @@ -2288,30 +2279,25 @@ void register_console(struct console *newcon) * See if this console matches one we selected on * the command line. */ - for (i = 0; i < MAX_CMDLINECONSOLES && console_cmdline[i].name[0]; - i++) { - if (strcmp(console_cmdline[i].name, newcon->name) != 0) + for (i = 0, c = console_cmdline; + i < MAX_CMDLINECONSOLES && c->name[0]; + i++, c++) { + if (strcmp(c->name, newcon->name) != 0) continue; if (newcon->index >= 0 && - newcon->index != console_cmdline[i].index) + newcon->index != c->index) continue; if (newcon->index < 0) - newcon->index = console_cmdline[i].index; -#ifdef CONFIG_A11Y_BRAILLE_CONSOLE - if (console_cmdline[i].brl_options) { - newcon->flags |= CON_BRL; - braille_register_console(newcon, - console_cmdline[i].index, - console_cmdline[i].options, - console_cmdline[i].brl_options); + newcon->index = c->index; + + if (_braille_register_console(newcon, c)) return; - } -#endif + if (newcon->setup && newcon->setup(newcon, console_cmdline[i].options) != 0) break; newcon->flags |= CON_ENABLED; - newcon->index = console_cmdline[i].index; + newcon->index = c->index; if (i == selected_console) { newcon->flags |= CON_CONSDEV; preferred_console = selected_console; @@ -2394,13 +2380,13 @@ EXPORT_SYMBOL(register_console); int unregister_console(struct console *console) { struct console *a, *b; - int res = 1; + int res; -#ifdef CONFIG_A11Y_BRAILLE_CONSOLE - if (console->flags & CON_BRL) - return braille_unregister_console(console); -#endif + res = _braille_unregister_console(console); + if (res) + return res; + res = 1; console_lock(); if (console_drivers == console) { console_drivers=console->next; @@ -2666,7 +2652,7 @@ void kmsg_dump(enum kmsg_dump_reason reason) bool kmsg_dump_get_line_nolock(struct kmsg_dumper *dumper, bool syslog, char *line, size_t size, size_t *len) { - struct log *msg; + struct printk_log *msg; size_t l = 0; bool ret = false; @@ -2778,7 +2764,7 @@ bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, idx = dumper->cur_idx; prev = 0; while (seq < dumper->next_seq) { - struct log *msg = log_from_idx(idx); + struct printk_log *msg = log_from_idx(idx); l += msg_print_text(msg, prev, true, NULL, 0); idx = log_next(idx); @@ -2791,7 +2777,7 @@ bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, idx = dumper->cur_idx; prev = 0; while (l > size && seq < dumper->next_seq) { - struct log *msg = log_from_idx(idx); + struct printk_log *msg = log_from_idx(idx); l -= msg_print_text(msg, prev, true, NULL, 0); idx = log_next(idx); @@ -2806,7 +2792,7 @@ bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, l = 0; prev = 0; while (seq < dumper->next_seq) { - struct log *msg = log_from_idx(idx); + struct printk_log *msg = log_from_idx(idx); l += msg_print_text(msg, prev, syslog, buf + l, size - l); idx = log_next(idx); diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 4041f5747e73..a146ee327f6a 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -469,7 +469,6 @@ static int ptrace_detach(struct task_struct *child, unsigned int data) /* Architecture-specific hardware disable .. */ ptrace_disable(child); clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - flush_ptrace_hw_breakpoint(child); write_lock_irq(&tasklist_lock); /* diff --git a/kernel/rcu.h b/kernel/rcu.h index 7f8e7590e3e5..77131966c4ad 100644 --- a/kernel/rcu.h +++ b/kernel/rcu.h @@ -67,12 +67,15 @@ extern struct debug_obj_descr rcuhead_debug_descr; -static inline void debug_rcu_head_queue(struct rcu_head *head) +static inline int debug_rcu_head_queue(struct rcu_head *head) { - debug_object_activate(head, &rcuhead_debug_descr); + int r1; + + r1 = debug_object_activate(head, &rcuhead_debug_descr); debug_object_active_state(head, &rcuhead_debug_descr, STATE_RCU_HEAD_READY, STATE_RCU_HEAD_QUEUED); + return r1; } static inline void debug_rcu_head_unqueue(struct rcu_head *head) @@ -83,8 +86,9 @@ static inline void debug_rcu_head_unqueue(struct rcu_head *head) debug_object_deactivate(head, &rcuhead_debug_descr); } #else /* !CONFIG_DEBUG_OBJECTS_RCU_HEAD */ -static inline void debug_rcu_head_queue(struct rcu_head *head) +static inline int debug_rcu_head_queue(struct rcu_head *head) { + return 0; } static inline void debug_rcu_head_unqueue(struct rcu_head *head) @@ -94,7 +98,7 @@ static inline void debug_rcu_head_unqueue(struct rcu_head *head) extern void kfree(const void *); -static inline bool __rcu_reclaim(char *rn, struct rcu_head *head) +static inline bool __rcu_reclaim(const char *rn, struct rcu_head *head) { unsigned long offset = (unsigned long)head->func; diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index cce6ba8bbace..33eb4620aa17 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -212,43 +212,6 @@ static inline void debug_rcu_head_free(struct rcu_head *head) } /* - * fixup_init is called when: - * - an active object is initialized - */ -static int rcuhead_fixup_init(void *addr, enum debug_obj_state state) -{ - struct rcu_head *head = addr; - - switch (state) { - case ODEBUG_STATE_ACTIVE: - /* - * Ensure that queued callbacks are all executed. - * If we detect that we are nested in a RCU read-side critical - * section, we should simply fail, otherwise we would deadlock. - * In !PREEMPT configurations, there is no way to tell if we are - * in a RCU read-side critical section or not, so we never - * attempt any fixup and just print a warning. - */ -#ifndef CONFIG_PREEMPT - WARN_ON_ONCE(1); - return 0; -#endif - if (rcu_preempt_depth() != 0 || preempt_count() != 0 || - irqs_disabled()) { - WARN_ON_ONCE(1); - return 0; - } - rcu_barrier(); - rcu_barrier_sched(); - rcu_barrier_bh(); - debug_object_init(head, &rcuhead_debug_descr); - return 1; - default: - return 0; - } -} - -/* * fixup_activate is called when: * - an active object is activated * - an unknown object is activated (might be a statically initialized object) @@ -268,69 +231,8 @@ static int rcuhead_fixup_activate(void *addr, enum debug_obj_state state) debug_object_init(head, &rcuhead_debug_descr); debug_object_activate(head, &rcuhead_debug_descr); return 0; - - case ODEBUG_STATE_ACTIVE: - /* - * Ensure that queued callbacks are all executed. - * If we detect that we are nested in a RCU read-side critical - * section, we should simply fail, otherwise we would deadlock. - * In !PREEMPT configurations, there is no way to tell if we are - * in a RCU read-side critical section or not, so we never - * attempt any fixup and just print a warning. - */ -#ifndef CONFIG_PREEMPT - WARN_ON_ONCE(1); - return 0; -#endif - if (rcu_preempt_depth() != 0 || preempt_count() != 0 || - irqs_disabled()) { - WARN_ON_ONCE(1); - return 0; - } - rcu_barrier(); - rcu_barrier_sched(); - rcu_barrier_bh(); - debug_object_activate(head, &rcuhead_debug_descr); - return 1; default: - return 0; - } -} - -/* - * fixup_free is called when: - * - an active object is freed - */ -static int rcuhead_fixup_free(void *addr, enum debug_obj_state state) -{ - struct rcu_head *head = addr; - - switch (state) { - case ODEBUG_STATE_ACTIVE: - /* - * Ensure that queued callbacks are all executed. - * If we detect that we are nested in a RCU read-side critical - * section, we should simply fail, otherwise we would deadlock. - * In !PREEMPT configurations, there is no way to tell if we are - * in a RCU read-side critical section or not, so we never - * attempt any fixup and just print a warning. - */ -#ifndef CONFIG_PREEMPT - WARN_ON_ONCE(1); - return 0; -#endif - if (rcu_preempt_depth() != 0 || preempt_count() != 0 || - irqs_disabled()) { - WARN_ON_ONCE(1); - return 0; - } - rcu_barrier(); - rcu_barrier_sched(); - rcu_barrier_bh(); - debug_object_free(head, &rcuhead_debug_descr); return 1; - default: - return 0; } } @@ -369,15 +271,13 @@ EXPORT_SYMBOL_GPL(destroy_rcu_head_on_stack); struct debug_obj_descr rcuhead_debug_descr = { .name = "rcu_head", - .fixup_init = rcuhead_fixup_init, .fixup_activate = rcuhead_fixup_activate, - .fixup_free = rcuhead_fixup_free, }; EXPORT_SYMBOL_GPL(rcuhead_debug_descr); #endif /* #ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD */ #if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU) || defined(CONFIG_RCU_TRACE) -void do_trace_rcu_torture_read(char *rcutorturename, struct rcu_head *rhp, +void do_trace_rcu_torture_read(const char *rcutorturename, struct rcu_head *rhp, unsigned long secs, unsigned long c_old, unsigned long c) { diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index aa344111de3e..9ed6075dc562 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -264,7 +264,7 @@ void rcu_check_callbacks(int cpu, int user) */ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) { - char *rn = NULL; + const char *rn = NULL; struct rcu_head *next, *list; unsigned long flags; RCU_TRACE(int cb_count = 0); diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h index 0cd385acccfa..280d06cae352 100644 --- a/kernel/rcutiny_plugin.h +++ b/kernel/rcutiny_plugin.h @@ -36,7 +36,7 @@ struct rcu_ctrlblk { RCU_TRACE(unsigned long gp_start); /* Start time for stalls. */ RCU_TRACE(unsigned long ticks_this_gp); /* Statistic for stalls. */ RCU_TRACE(unsigned long jiffies_stall); /* Jiffies at next stall. */ - RCU_TRACE(char *name); /* Name of RCU type. */ + RCU_TRACE(const char *name); /* Name of RCU type. */ }; /* Definition for rcupdate control block. */ diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index f4871e52c546..be63101c6175 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -52,72 +52,78 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Paul E. McKenney <paulmck@us.ibm.com> and Josh Triplett <josh@freedesktop.org>"); -static int nreaders = -1; /* # reader threads, defaults to 2*ncpus */ -static int nfakewriters = 4; /* # fake writer threads */ -static int stat_interval = 60; /* Interval between stats, in seconds. */ - /* Zero means "only at end of test". */ -static bool verbose; /* Print more debug info. */ -static bool test_no_idle_hz = true; - /* Test RCU support for tickless idle CPUs. */ -static int shuffle_interval = 3; /* Interval between shuffles (in sec)*/ -static int stutter = 5; /* Start/stop testing interval (in sec) */ -static int irqreader = 1; /* RCU readers from irq (timers). */ -static int fqs_duration; /* Duration of bursts (us), 0 to disable. */ -static int fqs_holdoff; /* Hold time within burst (us). */ -static int fqs_stutter = 3; /* Wait time between bursts (s). */ -static int n_barrier_cbs; /* Number of callbacks to test RCU barriers. */ -static int onoff_interval; /* Wait time between CPU hotplugs, 0=disable. */ -static int onoff_holdoff; /* Seconds after boot before CPU hotplugs. */ -static int shutdown_secs; /* Shutdown time (s). <=0 for no shutdown. */ -static int stall_cpu; /* CPU-stall duration (s). 0 for no stall. */ -static int stall_cpu_holdoff = 10; /* Time to wait until stall (s). */ -static int test_boost = 1; /* Test RCU prio boost: 0=no, 1=maybe, 2=yes. */ -static int test_boost_interval = 7; /* Interval between boost tests, seconds. */ -static int test_boost_duration = 4; /* Duration of each boost test, seconds. */ -static char *torture_type = "rcu"; /* What RCU implementation to torture. */ - -module_param(nreaders, int, 0444); -MODULE_PARM_DESC(nreaders, "Number of RCU reader threads"); -module_param(nfakewriters, int, 0444); -MODULE_PARM_DESC(nfakewriters, "Number of RCU fake writer threads"); -module_param(stat_interval, int, 0644); -MODULE_PARM_DESC(stat_interval, "Number of seconds between stats printk()s"); -module_param(verbose, bool, 0444); -MODULE_PARM_DESC(verbose, "Enable verbose debugging printk()s"); -module_param(test_no_idle_hz, bool, 0444); -MODULE_PARM_DESC(test_no_idle_hz, "Test support for tickless idle CPUs"); -module_param(shuffle_interval, int, 0444); -MODULE_PARM_DESC(shuffle_interval, "Number of seconds between shuffles"); -module_param(stutter, int, 0444); -MODULE_PARM_DESC(stutter, "Number of seconds to run/halt test"); -module_param(irqreader, int, 0444); -MODULE_PARM_DESC(irqreader, "Allow RCU readers from irq handlers"); +static int fqs_duration; module_param(fqs_duration, int, 0444); -MODULE_PARM_DESC(fqs_duration, "Duration of fqs bursts (us)"); +MODULE_PARM_DESC(fqs_duration, "Duration of fqs bursts (us), 0 to disable"); +static int fqs_holdoff; module_param(fqs_holdoff, int, 0444); MODULE_PARM_DESC(fqs_holdoff, "Holdoff time within fqs bursts (us)"); +static int fqs_stutter = 3; module_param(fqs_stutter, int, 0444); MODULE_PARM_DESC(fqs_stutter, "Wait time between fqs bursts (s)"); +static bool gp_exp; +module_param(gp_exp, bool, 0444); +MODULE_PARM_DESC(gp_exp, "Use expedited GP wait primitives"); +static bool gp_normal; +module_param(gp_normal, bool, 0444); +MODULE_PARM_DESC(gp_normal, "Use normal (non-expedited) GP wait primitives"); +static int irqreader = 1; +module_param(irqreader, int, 0444); +MODULE_PARM_DESC(irqreader, "Allow RCU readers from irq handlers"); +static int n_barrier_cbs; module_param(n_barrier_cbs, int, 0444); MODULE_PARM_DESC(n_barrier_cbs, "# of callbacks/kthreads for barrier testing"); -module_param(onoff_interval, int, 0444); -MODULE_PARM_DESC(onoff_interval, "Time between CPU hotplugs (s), 0=disable"); +static int nfakewriters = 4; +module_param(nfakewriters, int, 0444); +MODULE_PARM_DESC(nfakewriters, "Number of RCU fake writer threads"); +static int nreaders = -1; +module_param(nreaders, int, 0444); +MODULE_PARM_DESC(nreaders, "Number of RCU reader threads"); +static int object_debug; +module_param(object_debug, int, 0444); +MODULE_PARM_DESC(object_debug, "Enable debug-object double call_rcu() testing"); +static int onoff_holdoff; module_param(onoff_holdoff, int, 0444); MODULE_PARM_DESC(onoff_holdoff, "Time after boot before CPU hotplugs (s)"); +static int onoff_interval; +module_param(onoff_interval, int, 0444); +MODULE_PARM_DESC(onoff_interval, "Time between CPU hotplugs (s), 0=disable"); +static int shuffle_interval = 3; +module_param(shuffle_interval, int, 0444); +MODULE_PARM_DESC(shuffle_interval, "Number of seconds between shuffles"); +static int shutdown_secs; module_param(shutdown_secs, int, 0444); -MODULE_PARM_DESC(shutdown_secs, "Shutdown time (s), zero to disable."); +MODULE_PARM_DESC(shutdown_secs, "Shutdown time (s), <= zero to disable."); +static int stall_cpu; module_param(stall_cpu, int, 0444); MODULE_PARM_DESC(stall_cpu, "Stall duration (s), zero to disable."); +static int stall_cpu_holdoff = 10; module_param(stall_cpu_holdoff, int, 0444); MODULE_PARM_DESC(stall_cpu_holdoff, "Time to wait before starting stall (s)."); +static int stat_interval = 60; +module_param(stat_interval, int, 0644); +MODULE_PARM_DESC(stat_interval, "Number of seconds between stats printk()s"); +static int stutter = 5; +module_param(stutter, int, 0444); +MODULE_PARM_DESC(stutter, "Number of seconds to run/halt test"); +static int test_boost = 1; module_param(test_boost, int, 0444); MODULE_PARM_DESC(test_boost, "Test RCU prio boost: 0=no, 1=maybe, 2=yes."); -module_param(test_boost_interval, int, 0444); -MODULE_PARM_DESC(test_boost_interval, "Interval between boost tests, seconds."); +static int test_boost_duration = 4; module_param(test_boost_duration, int, 0444); MODULE_PARM_DESC(test_boost_duration, "Duration of each boost test, seconds."); +static int test_boost_interval = 7; +module_param(test_boost_interval, int, 0444); +MODULE_PARM_DESC(test_boost_interval, "Interval between boost tests, seconds."); +static bool test_no_idle_hz = true; +module_param(test_no_idle_hz, bool, 0444); +MODULE_PARM_DESC(test_no_idle_hz, "Test support for tickless idle CPUs"); +static char *torture_type = "rcu"; module_param(torture_type, charp, 0444); -MODULE_PARM_DESC(torture_type, "Type of RCU to torture (rcu, rcu_bh, srcu)"); +MODULE_PARM_DESC(torture_type, "Type of RCU to torture (rcu, rcu_bh, ...)"); +static bool verbose; +module_param(verbose, bool, 0444); +MODULE_PARM_DESC(verbose, "Enable verbose debugging printk()s"); #define TORTURE_FLAG "-torture:" #define PRINTK_STRING(s) \ @@ -267,7 +273,7 @@ rcutorture_shutdown_notify(struct notifier_block *unused1, * Absorb kthreads into a kernel function that won't return, so that * they won't ever access module text or data again. */ -static void rcutorture_shutdown_absorb(char *title) +static void rcutorture_shutdown_absorb(const char *title) { if (ACCESS_ONCE(fullstop) == FULLSTOP_SHUTDOWN) { pr_notice( @@ -337,7 +343,7 @@ rcu_random(struct rcu_random_state *rrsp) } static void -rcu_stutter_wait(char *title) +rcu_stutter_wait(const char *title) { while (stutter_pause_test || !rcutorture_runnable) { if (rcutorture_runnable) @@ -360,13 +366,14 @@ struct rcu_torture_ops { int (*completed)(void); void (*deferred_free)(struct rcu_torture *p); void (*sync)(void); + void (*exp_sync)(void); void (*call)(struct rcu_head *head, void (*func)(struct rcu_head *rcu)); void (*cb_barrier)(void); void (*fqs)(void); int (*stats)(char *page); int irq_capable; int can_boost; - char *name; + const char *name; }; static struct rcu_torture_ops *cur_ops; @@ -443,81 +450,27 @@ static void rcu_torture_deferred_free(struct rcu_torture *p) call_rcu(&p->rtort_rcu, rcu_torture_cb); } -static struct rcu_torture_ops rcu_ops = { - .init = NULL, - .readlock = rcu_torture_read_lock, - .read_delay = rcu_read_delay, - .readunlock = rcu_torture_read_unlock, - .completed = rcu_torture_completed, - .deferred_free = rcu_torture_deferred_free, - .sync = synchronize_rcu, - .call = call_rcu, - .cb_barrier = rcu_barrier, - .fqs = rcu_force_quiescent_state, - .stats = NULL, - .irq_capable = 1, - .can_boost = rcu_can_boost(), - .name = "rcu" -}; - -static void rcu_sync_torture_deferred_free(struct rcu_torture *p) -{ - int i; - struct rcu_torture *rp; - struct rcu_torture *rp1; - - cur_ops->sync(); - list_add(&p->rtort_free, &rcu_torture_removed); - list_for_each_entry_safe(rp, rp1, &rcu_torture_removed, rtort_free) { - i = rp->rtort_pipe_count; - if (i > RCU_TORTURE_PIPE_LEN) - i = RCU_TORTURE_PIPE_LEN; - atomic_inc(&rcu_torture_wcount[i]); - if (++rp->rtort_pipe_count >= RCU_TORTURE_PIPE_LEN) { - rp->rtort_mbtest = 0; - list_del(&rp->rtort_free); - rcu_torture_free(rp); - } - } -} - static void rcu_sync_torture_init(void) { INIT_LIST_HEAD(&rcu_torture_removed); } -static struct rcu_torture_ops rcu_sync_ops = { +static struct rcu_torture_ops rcu_ops = { .init = rcu_sync_torture_init, .readlock = rcu_torture_read_lock, .read_delay = rcu_read_delay, .readunlock = rcu_torture_read_unlock, .completed = rcu_torture_completed, - .deferred_free = rcu_sync_torture_deferred_free, + .deferred_free = rcu_torture_deferred_free, .sync = synchronize_rcu, - .call = NULL, - .cb_barrier = NULL, - .fqs = rcu_force_quiescent_state, - .stats = NULL, - .irq_capable = 1, - .can_boost = rcu_can_boost(), - .name = "rcu_sync" -}; - -static struct rcu_torture_ops rcu_expedited_ops = { - .init = rcu_sync_torture_init, - .readlock = rcu_torture_read_lock, - .read_delay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = rcu_torture_read_unlock, - .completed = rcu_no_completed, - .deferred_free = rcu_sync_torture_deferred_free, - .sync = synchronize_rcu_expedited, - .call = NULL, - .cb_barrier = NULL, + .exp_sync = synchronize_rcu_expedited, + .call = call_rcu, + .cb_barrier = rcu_barrier, .fqs = rcu_force_quiescent_state, .stats = NULL, .irq_capable = 1, .can_boost = rcu_can_boost(), - .name = "rcu_expedited" + .name = "rcu" }; /* @@ -546,13 +499,14 @@ static void rcu_bh_torture_deferred_free(struct rcu_torture *p) } static struct rcu_torture_ops rcu_bh_ops = { - .init = NULL, + .init = rcu_sync_torture_init, .readlock = rcu_bh_torture_read_lock, .read_delay = rcu_read_delay, /* just reuse rcu's version. */ .readunlock = rcu_bh_torture_read_unlock, .completed = rcu_bh_torture_completed, .deferred_free = rcu_bh_torture_deferred_free, .sync = synchronize_rcu_bh, + .exp_sync = synchronize_rcu_bh_expedited, .call = call_rcu_bh, .cb_barrier = rcu_barrier_bh, .fqs = rcu_bh_force_quiescent_state, @@ -561,38 +515,6 @@ static struct rcu_torture_ops rcu_bh_ops = { .name = "rcu_bh" }; -static struct rcu_torture_ops rcu_bh_sync_ops = { - .init = rcu_sync_torture_init, - .readlock = rcu_bh_torture_read_lock, - .read_delay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = rcu_bh_torture_read_unlock, - .completed = rcu_bh_torture_completed, - .deferred_free = rcu_sync_torture_deferred_free, - .sync = synchronize_rcu_bh, - .call = NULL, - .cb_barrier = NULL, - .fqs = rcu_bh_force_quiescent_state, - .stats = NULL, - .irq_capable = 1, - .name = "rcu_bh_sync" -}; - -static struct rcu_torture_ops rcu_bh_expedited_ops = { - .init = rcu_sync_torture_init, - .readlock = rcu_bh_torture_read_lock, - .read_delay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = rcu_bh_torture_read_unlock, - .completed = rcu_bh_torture_completed, - .deferred_free = rcu_sync_torture_deferred_free, - .sync = synchronize_rcu_bh_expedited, - .call = NULL, - .cb_barrier = NULL, - .fqs = rcu_bh_force_quiescent_state, - .stats = NULL, - .irq_capable = 1, - .name = "rcu_bh_expedited" -}; - /* * Definitions for srcu torture testing. */ @@ -667,6 +589,11 @@ static int srcu_torture_stats(char *page) return cnt; } +static void srcu_torture_synchronize_expedited(void) +{ + synchronize_srcu_expedited(&srcu_ctl); +} + static struct rcu_torture_ops srcu_ops = { .init = rcu_sync_torture_init, .readlock = srcu_torture_read_lock, @@ -675,45 +602,13 @@ static struct rcu_torture_ops srcu_ops = { .completed = srcu_torture_completed, .deferred_free = srcu_torture_deferred_free, .sync = srcu_torture_synchronize, + .exp_sync = srcu_torture_synchronize_expedited, .call = srcu_torture_call, .cb_barrier = srcu_torture_barrier, .stats = srcu_torture_stats, .name = "srcu" }; -static struct rcu_torture_ops srcu_sync_ops = { - .init = rcu_sync_torture_init, - .readlock = srcu_torture_read_lock, - .read_delay = srcu_read_delay, - .readunlock = srcu_torture_read_unlock, - .completed = srcu_torture_completed, - .deferred_free = rcu_sync_torture_deferred_free, - .sync = srcu_torture_synchronize, - .call = NULL, - .cb_barrier = NULL, - .stats = srcu_torture_stats, - .name = "srcu_sync" -}; - -static void srcu_torture_synchronize_expedited(void) -{ - synchronize_srcu_expedited(&srcu_ctl); -} - -static struct rcu_torture_ops srcu_expedited_ops = { - .init = rcu_sync_torture_init, - .readlock = srcu_torture_read_lock, - .read_delay = srcu_read_delay, - .readunlock = srcu_torture_read_unlock, - .completed = srcu_torture_completed, - .deferred_free = rcu_sync_torture_deferred_free, - .sync = srcu_torture_synchronize_expedited, - .call = NULL, - .cb_barrier = NULL, - .stats = srcu_torture_stats, - .name = "srcu_expedited" -}; - /* * Definitions for sched torture testing. */ @@ -742,6 +637,8 @@ static struct rcu_torture_ops sched_ops = { .completed = rcu_no_completed, .deferred_free = rcu_sched_torture_deferred_free, .sync = synchronize_sched, + .exp_sync = synchronize_sched_expedited, + .call = call_rcu_sched, .cb_barrier = rcu_barrier_sched, .fqs = rcu_sched_force_quiescent_state, .stats = NULL, @@ -749,35 +646,6 @@ static struct rcu_torture_ops sched_ops = { .name = "sched" }; -static struct rcu_torture_ops sched_sync_ops = { - .init = rcu_sync_torture_init, - .readlock = sched_torture_read_lock, - .read_delay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = sched_torture_read_unlock, - .completed = rcu_no_completed, - .deferred_free = rcu_sync_torture_deferred_free, - .sync = synchronize_sched, - .cb_barrier = NULL, - .fqs = rcu_sched_force_quiescent_state, - .stats = NULL, - .name = "sched_sync" -}; - -static struct rcu_torture_ops sched_expedited_ops = { - .init = rcu_sync_torture_init, - .readlock = sched_torture_read_lock, - .read_delay = rcu_read_delay, /* just reuse rcu's version. */ - .readunlock = sched_torture_read_unlock, - .completed = rcu_no_completed, - .deferred_free = rcu_sync_torture_deferred_free, - .sync = synchronize_sched_expedited, - .cb_barrier = NULL, - .fqs = rcu_sched_force_quiescent_state, - .stats = NULL, - .irq_capable = 1, - .name = "sched_expedited" -}; - /* * RCU torture priority-boost testing. Runs one real-time thread per * CPU for moderate bursts, repeatedly registering RCU callbacks and @@ -927,9 +795,10 @@ rcu_torture_fqs(void *arg) static int rcu_torture_writer(void *arg) { + bool exp; int i; - long oldbatch = rcu_batches_completed(); struct rcu_torture *rp; + struct rcu_torture *rp1; struct rcu_torture *old_rp; static DEFINE_RCU_RANDOM(rand); @@ -954,10 +823,33 @@ rcu_torture_writer(void *arg) i = RCU_TORTURE_PIPE_LEN; atomic_inc(&rcu_torture_wcount[i]); old_rp->rtort_pipe_count++; - cur_ops->deferred_free(old_rp); + if (gp_normal == gp_exp) + exp = !!(rcu_random(&rand) & 0x80); + else + exp = gp_exp; + if (!exp) { + cur_ops->deferred_free(old_rp); + } else { + cur_ops->exp_sync(); + list_add(&old_rp->rtort_free, + &rcu_torture_removed); + list_for_each_entry_safe(rp, rp1, + &rcu_torture_removed, + rtort_free) { + i = rp->rtort_pipe_count; + if (i > RCU_TORTURE_PIPE_LEN) + i = RCU_TORTURE_PIPE_LEN; + atomic_inc(&rcu_torture_wcount[i]); + if (++rp->rtort_pipe_count >= + RCU_TORTURE_PIPE_LEN) { + rp->rtort_mbtest = 0; + list_del(&rp->rtort_free); + rcu_torture_free(rp); + } + } + } } rcutorture_record_progress(++rcu_torture_current_version); - oldbatch = cur_ops->completed(); rcu_stutter_wait("rcu_torture_writer"); } while (!kthread_should_stop() && fullstop == FULLSTOP_DONTSTOP); VERBOSE_PRINTK_STRING("rcu_torture_writer task stopping"); @@ -983,10 +875,18 @@ rcu_torture_fakewriter(void *arg) schedule_timeout_uninterruptible(1 + rcu_random(&rand)%10); udelay(rcu_random(&rand) & 0x3ff); if (cur_ops->cb_barrier != NULL && - rcu_random(&rand) % (nfakewriters * 8) == 0) + rcu_random(&rand) % (nfakewriters * 8) == 0) { cur_ops->cb_barrier(); - else + } else if (gp_normal == gp_exp) { + if (rcu_random(&rand) & 0x80) + cur_ops->sync(); + else + cur_ops->exp_sync(); + } else if (gp_normal) { cur_ops->sync(); + } else { + cur_ops->exp_sync(); + } rcu_stutter_wait("rcu_torture_fakewriter"); } while (!kthread_should_stop() && fullstop == FULLSTOP_DONTSTOP); @@ -1364,7 +1264,7 @@ rcu_torture_stutter(void *arg) } static inline void -rcu_torture_print_module_parms(struct rcu_torture_ops *cur_ops, char *tag) +rcu_torture_print_module_parms(struct rcu_torture_ops *cur_ops, const char *tag) { pr_alert("%s" TORTURE_FLAG "--- %s: nreaders=%d nfakewriters=%d " @@ -1534,7 +1434,13 @@ rcu_torture_onoff(void *arg) torture_type, cpu); starttime = jiffies; n_online_attempts++; - if (cpu_up(cpu) == 0) { + ret = cpu_up(cpu); + if (ret) { + if (verbose) + pr_alert("%s" TORTURE_FLAG + "rcu_torture_onoff task: online %d failed: errno %d\n", + torture_type, cpu, ret); + } else { if (verbose) pr_alert("%s" TORTURE_FLAG "rcu_torture_onoff task: onlined %d\n", @@ -1934,6 +1840,62 @@ rcu_torture_cleanup(void) rcu_torture_print_module_parms(cur_ops, "End of test: SUCCESS"); } +#ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD +static void rcu_torture_leak_cb(struct rcu_head *rhp) +{ +} + +static void rcu_torture_err_cb(struct rcu_head *rhp) +{ + /* + * This -might- happen due to race conditions, but is unlikely. + * The scenario that leads to this happening is that the + * first of the pair of duplicate callbacks is queued, + * someone else starts a grace period that includes that + * callback, then the second of the pair must wait for the + * next grace period. Unlikely, but can happen. If it + * does happen, the debug-objects subsystem won't have splatted. + */ + pr_alert("rcutorture: duplicated callback was invoked.\n"); +} +#endif /* #ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD */ + +/* + * Verify that double-free causes debug-objects to complain, but only + * if CONFIG_DEBUG_OBJECTS_RCU_HEAD=y. Otherwise, say that the test + * cannot be carried out. + */ +static void rcu_test_debug_objects(void) +{ +#ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD + struct rcu_head rh1; + struct rcu_head rh2; + + init_rcu_head_on_stack(&rh1); + init_rcu_head_on_stack(&rh2); + pr_alert("rcutorture: WARN: Duplicate call_rcu() test starting.\n"); + + /* Try to queue the rh2 pair of callbacks for the same grace period. */ + preempt_disable(); /* Prevent preemption from interrupting test. */ + rcu_read_lock(); /* Make it impossible to finish a grace period. */ + call_rcu(&rh1, rcu_torture_leak_cb); /* Start grace period. */ + local_irq_disable(); /* Make it harder to start a new grace period. */ + call_rcu(&rh2, rcu_torture_leak_cb); + call_rcu(&rh2, rcu_torture_err_cb); /* Duplicate callback. */ + local_irq_enable(); + rcu_read_unlock(); + preempt_enable(); + + /* Wait for them all to get done so we can safely return. */ + rcu_barrier(); + pr_alert("rcutorture: WARN: Duplicate call_rcu() test complete.\n"); + destroy_rcu_head_on_stack(&rh1); + destroy_rcu_head_on_stack(&rh2); +#else /* #ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD */ + pr_alert("rcutorture: !CONFIG_DEBUG_OBJECTS_RCU_HEAD, not testing duplicate call_rcu()\n"); +#endif /* #else #ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD */ +} + static int __init rcu_torture_init(void) { @@ -1941,11 +1903,9 @@ rcu_torture_init(void) int cpu; int firsterr = 0; int retval; - static struct rcu_torture_ops *torture_ops[] = - { &rcu_ops, &rcu_sync_ops, &rcu_expedited_ops, - &rcu_bh_ops, &rcu_bh_sync_ops, &rcu_bh_expedited_ops, - &srcu_ops, &srcu_sync_ops, &srcu_expedited_ops, - &sched_ops, &sched_sync_ops, &sched_expedited_ops, }; + static struct rcu_torture_ops *torture_ops[] = { + &rcu_ops, &rcu_bh_ops, &srcu_ops, &sched_ops, + }; mutex_lock(&fullstop_mutex); @@ -2163,6 +2123,8 @@ rcu_torture_init(void) firsterr = retval; goto unwind; } + if (object_debug) + rcu_test_debug_objects(); rcutorture_record_test_transition(); mutex_unlock(&fullstop_mutex); return 0; diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 068de3a93606..32618b3fe4e6 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -53,18 +53,38 @@ #include <linux/delay.h> #include <linux/stop_machine.h> #include <linux/random.h> +#include <linux/ftrace_event.h> +#include <linux/suspend.h> #include "rcutree.h" #include <trace/events/rcu.h> #include "rcu.h" +/* + * Strings used in tracepoints need to be exported via the + * tracing system such that tools like perf and trace-cmd can + * translate the string address pointers to actual text. + */ +#define TPS(x) tracepoint_string(x) + /* Data structures. */ static struct lock_class_key rcu_node_class[RCU_NUM_LVLS]; static struct lock_class_key rcu_fqs_class[RCU_NUM_LVLS]; -#define RCU_STATE_INITIALIZER(sname, sabbr, cr) { \ +/* + * In order to export the rcu_state name to the tracing tools, it + * needs to be added in the __tracepoint_string section. + * This requires defining a separate variable tp_<sname>_varname + * that points to the string being used, and this will allow + * the tracing userspace tools to be able to decipher the string + * address to the matching string. + */ +#define RCU_STATE_INITIALIZER(sname, sabbr, cr) \ +static char sname##_varname[] = #sname; \ +static const char *tp_##sname##_varname __used __tracepoint_string = sname##_varname; \ +struct rcu_state sname##_state = { \ .level = { &sname##_state.node[0] }, \ .call = cr, \ .fqs_state = RCU_GP_IDLE, \ @@ -75,16 +95,13 @@ static struct lock_class_key rcu_fqs_class[RCU_NUM_LVLS]; .orphan_donetail = &sname##_state.orphan_donelist, \ .barrier_mutex = __MUTEX_INITIALIZER(sname##_state.barrier_mutex), \ .onoff_mutex = __MUTEX_INITIALIZER(sname##_state.onoff_mutex), \ - .name = #sname, \ + .name = sname##_varname, \ .abbr = sabbr, \ -} - -struct rcu_state rcu_sched_state = - RCU_STATE_INITIALIZER(rcu_sched, 's', call_rcu_sched); -DEFINE_PER_CPU(struct rcu_data, rcu_sched_data); +}; \ +DEFINE_PER_CPU(struct rcu_data, sname##_data) -struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh, 'b', call_rcu_bh); -DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); +RCU_STATE_INITIALIZER(rcu_sched, 's', call_rcu_sched); +RCU_STATE_INITIALIZER(rcu_bh, 'b', call_rcu_bh); static struct rcu_state *rcu_state; LIST_HEAD(rcu_struct_flavors); @@ -178,7 +195,7 @@ void rcu_sched_qs(int cpu) struct rcu_data *rdp = &per_cpu(rcu_sched_data, cpu); if (rdp->passed_quiesce == 0) - trace_rcu_grace_period("rcu_sched", rdp->gpnum, "cpuqs"); + trace_rcu_grace_period(TPS("rcu_sched"), rdp->gpnum, TPS("cpuqs")); rdp->passed_quiesce = 1; } @@ -187,7 +204,7 @@ void rcu_bh_qs(int cpu) struct rcu_data *rdp = &per_cpu(rcu_bh_data, cpu); if (rdp->passed_quiesce == 0) - trace_rcu_grace_period("rcu_bh", rdp->gpnum, "cpuqs"); + trace_rcu_grace_period(TPS("rcu_bh"), rdp->gpnum, TPS("cpuqs")); rdp->passed_quiesce = 1; } @@ -198,16 +215,20 @@ void rcu_bh_qs(int cpu) */ void rcu_note_context_switch(int cpu) { - trace_rcu_utilization("Start context switch"); + trace_rcu_utilization(TPS("Start context switch")); rcu_sched_qs(cpu); rcu_preempt_note_context_switch(cpu); - trace_rcu_utilization("End context switch"); + trace_rcu_utilization(TPS("End context switch")); } EXPORT_SYMBOL_GPL(rcu_note_context_switch); DEFINE_PER_CPU(struct rcu_dynticks, rcu_dynticks) = { .dynticks_nesting = DYNTICK_TASK_EXIT_IDLE, .dynticks = ATOMIC_INIT(1), +#ifdef CONFIG_NO_HZ_FULL_SYSIDLE + .dynticks_idle_nesting = DYNTICK_TASK_NEST_VALUE, + .dynticks_idle = ATOMIC_INIT(1), +#endif /* #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ }; static long blimit = 10; /* Maximum callbacks per rcu_do_batch. */ @@ -226,7 +247,10 @@ module_param(jiffies_till_next_fqs, ulong, 0644); static void rcu_start_gp_advanced(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp); -static void force_qs_rnp(struct rcu_state *rsp, int (*f)(struct rcu_data *)); +static void force_qs_rnp(struct rcu_state *rsp, + int (*f)(struct rcu_data *rsp, bool *isidle, + unsigned long *maxj), + bool *isidle, unsigned long *maxj); static void force_quiescent_state(struct rcu_state *rsp); static int rcu_pending(int cpu); @@ -345,11 +369,11 @@ static struct rcu_node *rcu_get_root(struct rcu_state *rsp) static void rcu_eqs_enter_common(struct rcu_dynticks *rdtp, long long oldval, bool user) { - trace_rcu_dyntick("Start", oldval, rdtp->dynticks_nesting); + trace_rcu_dyntick(TPS("Start"), oldval, rdtp->dynticks_nesting); if (!user && !is_idle_task(current)) { struct task_struct *idle = idle_task(smp_processor_id()); - trace_rcu_dyntick("Error on entry: not idle task", oldval, 0); + trace_rcu_dyntick(TPS("Error on entry: not idle task"), oldval, 0); ftrace_dump(DUMP_ORIG); WARN_ONCE(1, "Current pid: %d comm: %s / Idle pid: %d comm: %s", current->pid, current->comm, @@ -411,6 +435,7 @@ void rcu_idle_enter(void) local_irq_save(flags); rcu_eqs_enter(false); + rcu_sysidle_enter(&__get_cpu_var(rcu_dynticks), 0); local_irq_restore(flags); } EXPORT_SYMBOL_GPL(rcu_idle_enter); @@ -428,27 +453,6 @@ void rcu_user_enter(void) { rcu_eqs_enter(1); } - -/** - * rcu_user_enter_after_irq - inform RCU that we are going to resume userspace - * after the current irq returns. - * - * This is similar to rcu_user_enter() but in the context of a non-nesting - * irq. After this call, RCU enters into idle mode when the interrupt - * returns. - */ -void rcu_user_enter_after_irq(void) -{ - unsigned long flags; - struct rcu_dynticks *rdtp; - - local_irq_save(flags); - rdtp = &__get_cpu_var(rcu_dynticks); - /* Ensure this irq is interrupting a non-idle RCU state. */ - WARN_ON_ONCE(!(rdtp->dynticks_nesting & DYNTICK_TASK_MASK)); - rdtp->dynticks_nesting = 1; - local_irq_restore(flags); -} #endif /* CONFIG_RCU_USER_QS */ /** @@ -479,9 +483,10 @@ void rcu_irq_exit(void) rdtp->dynticks_nesting--; WARN_ON_ONCE(rdtp->dynticks_nesting < 0); if (rdtp->dynticks_nesting) - trace_rcu_dyntick("--=", oldval, rdtp->dynticks_nesting); + trace_rcu_dyntick(TPS("--="), oldval, rdtp->dynticks_nesting); else rcu_eqs_enter_common(rdtp, oldval, true); + rcu_sysidle_enter(rdtp, 1); local_irq_restore(flags); } @@ -501,11 +506,11 @@ static void rcu_eqs_exit_common(struct rcu_dynticks *rdtp, long long oldval, smp_mb__after_atomic_inc(); /* See above. */ WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1)); rcu_cleanup_after_idle(smp_processor_id()); - trace_rcu_dyntick("End", oldval, rdtp->dynticks_nesting); + trace_rcu_dyntick(TPS("End"), oldval, rdtp->dynticks_nesting); if (!user && !is_idle_task(current)) { struct task_struct *idle = idle_task(smp_processor_id()); - trace_rcu_dyntick("Error on exit: not idle task", + trace_rcu_dyntick(TPS("Error on exit: not idle task"), oldval, rdtp->dynticks_nesting); ftrace_dump(DUMP_ORIG); WARN_ONCE(1, "Current pid: %d comm: %s / Idle pid: %d comm: %s", @@ -550,6 +555,7 @@ void rcu_idle_exit(void) local_irq_save(flags); rcu_eqs_exit(false); + rcu_sysidle_exit(&__get_cpu_var(rcu_dynticks), 0); local_irq_restore(flags); } EXPORT_SYMBOL_GPL(rcu_idle_exit); @@ -565,28 +571,6 @@ void rcu_user_exit(void) { rcu_eqs_exit(1); } - -/** - * rcu_user_exit_after_irq - inform RCU that we won't resume to userspace - * idle mode after the current non-nesting irq returns. - * - * This is similar to rcu_user_exit() but in the context of an irq. - * This is called when the irq has interrupted a userspace RCU idle mode - * context. When the current non-nesting interrupt returns after this call, - * the CPU won't restore the RCU idle mode. - */ -void rcu_user_exit_after_irq(void) -{ - unsigned long flags; - struct rcu_dynticks *rdtp; - - local_irq_save(flags); - rdtp = &__get_cpu_var(rcu_dynticks); - /* Ensure we are interrupting an RCU idle mode. */ - WARN_ON_ONCE(rdtp->dynticks_nesting & DYNTICK_TASK_NEST_MASK); - rdtp->dynticks_nesting += DYNTICK_TASK_EXIT_IDLE; - local_irq_restore(flags); -} #endif /* CONFIG_RCU_USER_QS */ /** @@ -620,9 +604,10 @@ void rcu_irq_enter(void) rdtp->dynticks_nesting++; WARN_ON_ONCE(rdtp->dynticks_nesting == 0); if (oldval) - trace_rcu_dyntick("++=", oldval, rdtp->dynticks_nesting); + trace_rcu_dyntick(TPS("++="), oldval, rdtp->dynticks_nesting); else rcu_eqs_exit_common(rdtp, oldval, true); + rcu_sysidle_exit(rdtp, 1); local_irq_restore(flags); } @@ -746,9 +731,11 @@ static int rcu_is_cpu_rrupt_from_idle(void) * credit them with an implicit quiescent state. Return 1 if this CPU * is in dynticks idle mode, which is an extended quiescent state. */ -static int dyntick_save_progress_counter(struct rcu_data *rdp) +static int dyntick_save_progress_counter(struct rcu_data *rdp, + bool *isidle, unsigned long *maxj) { rdp->dynticks_snap = atomic_add_return(0, &rdp->dynticks->dynticks); + rcu_sysidle_check_cpu(rdp, isidle, maxj); return (rdp->dynticks_snap & 0x1) == 0; } @@ -758,7 +745,8 @@ static int dyntick_save_progress_counter(struct rcu_data *rdp) * idle state since the last call to dyntick_save_progress_counter() * for this same CPU, or by virtue of having been offline. */ -static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) +static int rcu_implicit_dynticks_qs(struct rcu_data *rdp, + bool *isidle, unsigned long *maxj) { unsigned int curr; unsigned int snap; @@ -775,7 +763,7 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) * of the current RCU grace period. */ if ((curr & 0x1) == 0 || UINT_CMP_GE(curr, snap + 2)) { - trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, "dti"); + trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("dti")); rdp->dynticks_fqs++; return 1; } @@ -795,7 +783,7 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) return 0; /* Grace period is not old enough. */ barrier(); if (cpu_is_offline(rdp->cpu)) { - trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, "ofl"); + trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("ofl")); rdp->offline_fqs++; return 1; } @@ -1032,7 +1020,7 @@ static unsigned long rcu_cbs_completed(struct rcu_state *rsp, * rcu_nocb_wait_gp(). */ static void trace_rcu_future_gp(struct rcu_node *rnp, struct rcu_data *rdp, - unsigned long c, char *s) + unsigned long c, const char *s) { trace_rcu_future_grace_period(rdp->rsp->name, rnp->gpnum, rnp->completed, c, rnp->level, @@ -1058,9 +1046,9 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp) * grace period is already marked as needed, return to the caller. */ c = rcu_cbs_completed(rdp->rsp, rnp); - trace_rcu_future_gp(rnp, rdp, c, "Startleaf"); + trace_rcu_future_gp(rnp, rdp, c, TPS("Startleaf")); if (rnp->need_future_gp[c & 0x1]) { - trace_rcu_future_gp(rnp, rdp, c, "Prestartleaf"); + trace_rcu_future_gp(rnp, rdp, c, TPS("Prestartleaf")); return c; } @@ -1074,7 +1062,7 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp) if (rnp->gpnum != rnp->completed || ACCESS_ONCE(rnp->gpnum) != ACCESS_ONCE(rnp->completed)) { rnp->need_future_gp[c & 0x1]++; - trace_rcu_future_gp(rnp, rdp, c, "Startedleaf"); + trace_rcu_future_gp(rnp, rdp, c, TPS("Startedleaf")); return c; } @@ -1102,7 +1090,7 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp) * recorded, trace and leave. */ if (rnp_root->need_future_gp[c & 0x1]) { - trace_rcu_future_gp(rnp, rdp, c, "Prestartedroot"); + trace_rcu_future_gp(rnp, rdp, c, TPS("Prestartedroot")); goto unlock_out; } @@ -1111,9 +1099,9 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp) /* If a grace period is not already in progress, start one. */ if (rnp_root->gpnum != rnp_root->completed) { - trace_rcu_future_gp(rnp, rdp, c, "Startedleafroot"); + trace_rcu_future_gp(rnp, rdp, c, TPS("Startedleafroot")); } else { - trace_rcu_future_gp(rnp, rdp, c, "Startedroot"); + trace_rcu_future_gp(rnp, rdp, c, TPS("Startedroot")); rcu_start_gp_advanced(rdp->rsp, rnp_root, rdp); } unlock_out: @@ -1137,7 +1125,8 @@ static int rcu_future_gp_cleanup(struct rcu_state *rsp, struct rcu_node *rnp) rcu_nocb_gp_cleanup(rsp, rnp); rnp->need_future_gp[c & 0x1] = 0; needmore = rnp->need_future_gp[(c + 1) & 0x1]; - trace_rcu_future_gp(rnp, rdp, c, needmore ? "CleanupMore" : "Cleanup"); + trace_rcu_future_gp(rnp, rdp, c, + needmore ? TPS("CleanupMore") : TPS("Cleanup")); return needmore; } @@ -1205,9 +1194,9 @@ static void rcu_accelerate_cbs(struct rcu_state *rsp, struct rcu_node *rnp, /* Trace depending on how much we were able to accelerate. */ if (!*rdp->nxttail[RCU_WAIT_TAIL]) - trace_rcu_grace_period(rsp->name, rdp->gpnum, "AccWaitCB"); + trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("AccWaitCB")); else - trace_rcu_grace_period(rsp->name, rdp->gpnum, "AccReadyCB"); + trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("AccReadyCB")); } /* @@ -1273,7 +1262,7 @@ static void __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, struc /* Remember that we saw this grace-period completion. */ rdp->completed = rnp->completed; - trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpuend"); + trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuend")); } if (rdp->gpnum != rnp->gpnum) { @@ -1283,7 +1272,7 @@ static void __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp, struc * go looking for one. */ rdp->gpnum = rnp->gpnum; - trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpustart"); + trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpustart")); rdp->passed_quiesce = 0; rdp->qs_pending = !!(rnp->qsmask & rdp->grpmask); zero_cpu_stall_ticks(rdp); @@ -1315,6 +1304,7 @@ static int rcu_gp_init(struct rcu_state *rsp) struct rcu_data *rdp; struct rcu_node *rnp = rcu_get_root(rsp); + rcu_bind_gp_kthread(); raw_spin_lock_irq(&rnp->lock); rsp->gp_flags = 0; /* Clear all flags: New grace period. */ @@ -1326,7 +1316,7 @@ static int rcu_gp_init(struct rcu_state *rsp) /* Advance to a new grace period and initialize state. */ rsp->gpnum++; - trace_rcu_grace_period(rsp->name, rsp->gpnum, "start"); + trace_rcu_grace_period(rsp->name, rsp->gpnum, TPS("start")); record_gp_stall_check_time(rsp); raw_spin_unlock_irq(&rnp->lock); @@ -1379,16 +1369,25 @@ static int rcu_gp_init(struct rcu_state *rsp) int rcu_gp_fqs(struct rcu_state *rsp, int fqs_state_in) { int fqs_state = fqs_state_in; + bool isidle = false; + unsigned long maxj; struct rcu_node *rnp = rcu_get_root(rsp); rsp->n_force_qs++; if (fqs_state == RCU_SAVE_DYNTICK) { /* Collect dyntick-idle snapshots. */ - force_qs_rnp(rsp, dyntick_save_progress_counter); + if (is_sysidle_rcu_state(rsp)) { + isidle = 1; + maxj = jiffies - ULONG_MAX / 4; + } + force_qs_rnp(rsp, dyntick_save_progress_counter, + &isidle, &maxj); + rcu_sysidle_report_gp(rsp, isidle, maxj); fqs_state = RCU_FORCE_QS; } else { /* Handle dyntick-idle and offline CPUs. */ - force_qs_rnp(rsp, rcu_implicit_dynticks_qs); + isidle = 0; + force_qs_rnp(rsp, rcu_implicit_dynticks_qs, &isidle, &maxj); } /* Clear flag to prevent immediate re-entry. */ if (ACCESS_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS) { @@ -1448,7 +1447,7 @@ static void rcu_gp_cleanup(struct rcu_state *rsp) rcu_nocb_gp_set(rnp, nocb); rsp->completed = rsp->gpnum; /* Declare grace period done. */ - trace_rcu_grace_period(rsp->name, rsp->completed, "end"); + trace_rcu_grace_period(rsp->name, rsp->completed, TPS("end")); rsp->fqs_state = RCU_GP_IDLE; rdp = this_cpu_ptr(rsp->rda); rcu_advance_cbs(rsp, rnp, rdp); /* Reduce false positives below. */ @@ -1558,10 +1557,12 @@ rcu_start_gp_advanced(struct rcu_state *rsp, struct rcu_node *rnp, /* * We can't do wakeups while holding the rnp->lock, as that - * could cause possible deadlocks with the rq->lock. Deter - * the wakeup to interrupt context. + * could cause possible deadlocks with the rq->lock. Defer + * the wakeup to interrupt context. And don't bother waking + * up the running kthread. */ - irq_work_queue(&rsp->wakeup_work); + if (current != rsp->gp_kthread) + irq_work_queue(&rsp->wakeup_work); } /* @@ -1857,7 +1858,7 @@ static void rcu_cleanup_dying_cpu(struct rcu_state *rsp) RCU_TRACE(mask = rdp->grpmask); trace_rcu_grace_period(rsp->name, rnp->gpnum + 1 - !!(rnp->qsmask & mask), - "cpuofl"); + TPS("cpuofl")); } /* @@ -2044,7 +2045,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) */ void rcu_check_callbacks(int cpu, int user) { - trace_rcu_utilization("Start scheduler-tick"); + trace_rcu_utilization(TPS("Start scheduler-tick")); increment_cpu_stall_ticks(); if (user || rcu_is_cpu_rrupt_from_idle()) { @@ -2077,7 +2078,7 @@ void rcu_check_callbacks(int cpu, int user) rcu_preempt_check_callbacks(cpu); if (rcu_pending(cpu)) invoke_rcu_core(); - trace_rcu_utilization("End scheduler-tick"); + trace_rcu_utilization(TPS("End scheduler-tick")); } /* @@ -2087,7 +2088,10 @@ void rcu_check_callbacks(int cpu, int user) * * The caller must have suppressed start of new grace periods. */ -static void force_qs_rnp(struct rcu_state *rsp, int (*f)(struct rcu_data *)) +static void force_qs_rnp(struct rcu_state *rsp, + int (*f)(struct rcu_data *rsp, bool *isidle, + unsigned long *maxj), + bool *isidle, unsigned long *maxj) { unsigned long bit; int cpu; @@ -2110,9 +2114,12 @@ static void force_qs_rnp(struct rcu_state *rsp, int (*f)(struct rcu_data *)) cpu = rnp->grplo; bit = 1; for (; cpu <= rnp->grphi; cpu++, bit <<= 1) { - if ((rnp->qsmask & bit) != 0 && - f(per_cpu_ptr(rsp->rda, cpu))) - mask |= bit; + if ((rnp->qsmask & bit) != 0) { + if ((rnp->qsmaskinit & bit) != 0) + *isidle = 0; + if (f(per_cpu_ptr(rsp->rda, cpu), isidle, maxj)) + mask |= bit; + } } if (mask != 0) { @@ -2208,10 +2215,10 @@ static void rcu_process_callbacks(struct softirq_action *unused) if (cpu_is_offline(smp_processor_id())) return; - trace_rcu_utilization("Start RCU core"); + trace_rcu_utilization(TPS("Start RCU core")); for_each_rcu_flavor(rsp) __rcu_process_callbacks(rsp); - trace_rcu_utilization("End RCU core"); + trace_rcu_utilization(TPS("End RCU core")); } /* @@ -2287,6 +2294,13 @@ static void __call_rcu_core(struct rcu_state *rsp, struct rcu_data *rdp, } /* + * RCU callback function to leak a callback. + */ +static void rcu_leak_callback(struct rcu_head *rhp) +{ +} + +/* * Helper function for call_rcu() and friends. The cpu argument will * normally be -1, indicating "currently running CPU". It may specify * a CPU only if that CPU is a no-CBs CPU. Currently, only _rcu_barrier() @@ -2300,7 +2314,12 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), struct rcu_data *rdp; WARN_ON_ONCE((unsigned long)head & 0x3); /* Misaligned rcu_head! */ - debug_rcu_head_queue(head); + if (debug_rcu_head_queue(head)) { + /* Probable double call_rcu(), so leak the callback. */ + ACCESS_ONCE(head->func) = rcu_leak_callback; + WARN_ONCE(1, "__call_rcu(): Leaked duplicate callback\n"); + return; + } head->func = func; head->next = NULL; @@ -2720,7 +2739,7 @@ static int rcu_cpu_has_callbacks(int cpu, bool *all_lazy) * Helper function for _rcu_barrier() tracing. If tracing is disabled, * the compiler is expected to optimize this away. */ -static void _rcu_barrier_trace(struct rcu_state *rsp, char *s, +static void _rcu_barrier_trace(struct rcu_state *rsp, const char *s, int cpu, unsigned long done) { trace_rcu_barrier(rsp->name, s, cpu, @@ -2785,9 +2804,20 @@ static void _rcu_barrier(struct rcu_state *rsp) * transition. The "if" expression below therefore rounds the old * value up to the next even number and adds two before comparing. */ - snap_done = ACCESS_ONCE(rsp->n_barrier_done); + snap_done = rsp->n_barrier_done; _rcu_barrier_trace(rsp, "Check", -1, snap_done); - if (ULONG_CMP_GE(snap_done, ((snap + 1) & ~0x1) + 2)) { + + /* + * If the value in snap is odd, we needed to wait for the current + * rcu_barrier() to complete, then wait for the next one, in other + * words, we need the value of snap_done to be three larger than + * the value of snap. On the other hand, if the value in snap is + * even, we only had to wait for the next rcu_barrier() to complete, + * in other words, we need the value of snap_done to be only two + * greater than the value of snap. The "(snap + 3) & ~0x1" computes + * this for us (thank you, Linus!). + */ + if (ULONG_CMP_GE(snap_done, (snap + 3) & ~0x1)) { _rcu_barrier_trace(rsp, "EarlyExit", -1, snap_done); smp_mb(); /* caller's subsequent code after above check. */ mutex_unlock(&rsp->barrier_mutex); @@ -2930,6 +2960,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptible) rdp->blimit = blimit; init_callback_list(rdp); /* Re-enable callbacks on this CPU. */ rdp->dynticks->dynticks_nesting = DYNTICK_TASK_EXIT_IDLE; + rcu_sysidle_init_percpu_data(rdp->dynticks); atomic_set(&rdp->dynticks->dynticks, (atomic_read(&rdp->dynticks->dynticks) & ~0x1) + 1); raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ @@ -2952,7 +2983,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptible) rdp->completed = rnp->completed; rdp->passed_quiesce = 0; rdp->qs_pending = 0; - trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpuonl"); + trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuonl")); } raw_spin_unlock(&rnp->lock); /* irqs already disabled. */ rnp = rnp->parent; @@ -2982,7 +3013,7 @@ static int rcu_cpu_notify(struct notifier_block *self, struct rcu_node *rnp = rdp->mynode; struct rcu_state *rsp; - trace_rcu_utilization("Start CPU hotplug"); + trace_rcu_utilization(TPS("Start CPU hotplug")); switch (action) { case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: @@ -3011,7 +3042,26 @@ static int rcu_cpu_notify(struct notifier_block *self, default: break; } - trace_rcu_utilization("End CPU hotplug"); + trace_rcu_utilization(TPS("End CPU hotplug")); + return NOTIFY_OK; +} + +static int rcu_pm_notify(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + switch (action) { + case PM_HIBERNATION_PREPARE: + case PM_SUSPEND_PREPARE: + if (nr_cpu_ids <= 256) /* Expediting bad for large systems. */ + rcu_expedited = 1; + break; + case PM_POST_HIBERNATION: + case PM_POST_SUSPEND: + rcu_expedited = 0; + break; + default: + break; + } return NOTIFY_OK; } @@ -3256,6 +3306,7 @@ void __init rcu_init(void) * or the scheduler are operational. */ cpu_notifier(rcu_cpu_notify, 0); + pm_notifier(rcu_pm_notify, 0); for_each_online_cpu(cpu) rcu_cpu_notify(NULL, CPU_UP_PREPARE, (void *)(long)cpu); } diff --git a/kernel/rcutree.h b/kernel/rcutree.h index b3832581043c..5f97eab602cd 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -88,6 +88,14 @@ struct rcu_dynticks { /* Process level is worth LLONG_MAX/2. */ int dynticks_nmi_nesting; /* Track NMI nesting level. */ atomic_t dynticks; /* Even value for idle, else odd. */ +#ifdef CONFIG_NO_HZ_FULL_SYSIDLE + long long dynticks_idle_nesting; + /* irq/process nesting level from idle. */ + atomic_t dynticks_idle; /* Even value for idle, else odd. */ + /* "Idle" excludes userspace execution. */ + unsigned long dynticks_idle_jiffies; + /* End of last non-NMI non-idle period. */ +#endif /* #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ #ifdef CONFIG_RCU_FAST_NO_HZ bool all_lazy; /* Are all CPU's CBs lazy? */ unsigned long nonlazy_posted; @@ -445,7 +453,7 @@ struct rcu_state { /* for CPU stalls. */ unsigned long gp_max; /* Maximum GP duration in */ /* jiffies. */ - char *name; /* Name of structure. */ + const char *name; /* Name of structure. */ char abbr; /* Abbreviated name. */ struct list_head flavors; /* List of RCU flavors. */ struct irq_work wakeup_work; /* Postponed wakeups */ @@ -545,6 +553,15 @@ static void rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp); static void rcu_spawn_nocb_kthreads(struct rcu_state *rsp); static void rcu_kick_nohz_cpu(int cpu); static bool init_nocb_callback_list(struct rcu_data *rdp); +static void rcu_sysidle_enter(struct rcu_dynticks *rdtp, int irq); +static void rcu_sysidle_exit(struct rcu_dynticks *rdtp, int irq); +static void rcu_sysidle_check_cpu(struct rcu_data *rdp, bool *isidle, + unsigned long *maxj); +static bool is_sysidle_rcu_state(struct rcu_state *rsp); +static void rcu_sysidle_report_gp(struct rcu_state *rsp, int isidle, + unsigned long maxj); +static void rcu_bind_gp_kthread(void); +static void rcu_sysidle_init_percpu_data(struct rcu_dynticks *rdtp); #endif /* #ifndef RCU_TREE_NONCORE */ diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 769e12e3151b..130c97b027f2 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -28,7 +28,7 @@ #include <linux/gfp.h> #include <linux/oom.h> #include <linux/smpboot.h> -#include <linux/tick.h> +#include "time/tick-internal.h" #define RCU_KTHREAD_PRIO 1 @@ -110,9 +110,7 @@ static void __init rcu_bootup_announce_oddness(void) #ifdef CONFIG_TREE_PREEMPT_RCU -struct rcu_state rcu_preempt_state = - RCU_STATE_INITIALIZER(rcu_preempt, 'p', call_rcu); -DEFINE_PER_CPU(struct rcu_data, rcu_preempt_data); +RCU_STATE_INITIALIZER(rcu_preempt, 'p', call_rcu); static struct rcu_state *rcu_state = &rcu_preempt_state; static int rcu_preempted_readers_exp(struct rcu_node *rnp); @@ -169,7 +167,7 @@ static void rcu_preempt_qs(int cpu) struct rcu_data *rdp = &per_cpu(rcu_preempt_data, cpu); if (rdp->passed_quiesce == 0) - trace_rcu_grace_period("rcu_preempt", rdp->gpnum, "cpuqs"); + trace_rcu_grace_period(TPS("rcu_preempt"), rdp->gpnum, TPS("cpuqs")); rdp->passed_quiesce = 1; current->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS; } @@ -388,7 +386,7 @@ void rcu_read_unlock_special(struct task_struct *t) np = rcu_next_node_entry(t, rnp); list_del_init(&t->rcu_node_entry); t->rcu_blocked_node = NULL; - trace_rcu_unlock_preempted_task("rcu_preempt", + trace_rcu_unlock_preempted_task(TPS("rcu_preempt"), rnp->gpnum, t->pid); if (&t->rcu_node_entry == rnp->gp_tasks) rnp->gp_tasks = np; @@ -412,7 +410,7 @@ void rcu_read_unlock_special(struct task_struct *t) */ empty_exp_now = !rcu_preempted_readers_exp(rnp); if (!empty && !rcu_preempt_blocked_readers_cgp(rnp)) { - trace_rcu_quiescent_state_report("preempt_rcu", + trace_rcu_quiescent_state_report(TPS("preempt_rcu"), rnp->gpnum, 0, rnp->qsmask, rnp->level, @@ -1250,12 +1248,12 @@ static int rcu_boost_kthread(void *arg) int spincnt = 0; int more2boost; - trace_rcu_utilization("Start boost kthread@init"); + trace_rcu_utilization(TPS("Start boost kthread@init")); for (;;) { rnp->boost_kthread_status = RCU_KTHREAD_WAITING; - trace_rcu_utilization("End boost kthread@rcu_wait"); + trace_rcu_utilization(TPS("End boost kthread@rcu_wait")); rcu_wait(rnp->boost_tasks || rnp->exp_tasks); - trace_rcu_utilization("Start boost kthread@rcu_wait"); + trace_rcu_utilization(TPS("Start boost kthread@rcu_wait")); rnp->boost_kthread_status = RCU_KTHREAD_RUNNING; more2boost = rcu_boost(rnp); if (more2boost) @@ -1264,14 +1262,14 @@ static int rcu_boost_kthread(void *arg) spincnt = 0; if (spincnt > 10) { rnp->boost_kthread_status = RCU_KTHREAD_YIELDING; - trace_rcu_utilization("End boost kthread@rcu_yield"); + trace_rcu_utilization(TPS("End boost kthread@rcu_yield")); schedule_timeout_interruptible(2); - trace_rcu_utilization("Start boost kthread@rcu_yield"); + trace_rcu_utilization(TPS("Start boost kthread@rcu_yield")); spincnt = 0; } } /* NOTREACHED */ - trace_rcu_utilization("End boost kthread@notreached"); + trace_rcu_utilization(TPS("End boost kthread@notreached")); return 0; } @@ -1419,7 +1417,7 @@ static void rcu_cpu_kthread(unsigned int cpu) int spincnt; for (spincnt = 0; spincnt < 10; spincnt++) { - trace_rcu_utilization("Start CPU kthread@rcu_wait"); + trace_rcu_utilization(TPS("Start CPU kthread@rcu_wait")); local_bh_disable(); *statusp = RCU_KTHREAD_RUNNING; this_cpu_inc(rcu_cpu_kthread_loops); @@ -1431,15 +1429,15 @@ static void rcu_cpu_kthread(unsigned int cpu) rcu_kthread_do_work(); local_bh_enable(); if (*workp == 0) { - trace_rcu_utilization("End CPU kthread@rcu_wait"); + trace_rcu_utilization(TPS("End CPU kthread@rcu_wait")); *statusp = RCU_KTHREAD_WAITING; return; } } *statusp = RCU_KTHREAD_YIELDING; - trace_rcu_utilization("Start CPU kthread@rcu_yield"); + trace_rcu_utilization(TPS("Start CPU kthread@rcu_yield")); schedule_timeout_interruptible(2); - trace_rcu_utilization("End CPU kthread@rcu_yield"); + trace_rcu_utilization(TPS("End CPU kthread@rcu_yield")); *statusp = RCU_KTHREAD_WAITING; } @@ -2202,7 +2200,7 @@ static void rcu_nocb_wait_gp(struct rcu_data *rdp) * Wait for the grace period. Do so interruptibly to avoid messing * up the load average. */ - trace_rcu_future_gp(rnp, rdp, c, "StartWait"); + trace_rcu_future_gp(rnp, rdp, c, TPS("StartWait")); for (;;) { wait_event_interruptible( rnp->nocb_gp_wq[c & 0x1], @@ -2210,9 +2208,9 @@ static void rcu_nocb_wait_gp(struct rcu_data *rdp) if (likely(d)) break; flush_signals(current); - trace_rcu_future_gp(rnp, rdp, c, "ResumeWait"); + trace_rcu_future_gp(rnp, rdp, c, TPS("ResumeWait")); } - trace_rcu_future_gp(rnp, rdp, c, "EndWait"); + trace_rcu_future_gp(rnp, rdp, c, TPS("EndWait")); smp_mb(); /* Ensure that CB invocation happens after GP end. */ } @@ -2375,3 +2373,425 @@ static void rcu_kick_nohz_cpu(int cpu) smp_send_reschedule(cpu); #endif /* #ifdef CONFIG_NO_HZ_FULL */ } + + +#ifdef CONFIG_NO_HZ_FULL_SYSIDLE + +/* + * Define RCU flavor that holds sysidle state. This needs to be the + * most active flavor of RCU. + */ +#ifdef CONFIG_PREEMPT_RCU +static struct rcu_state *rcu_sysidle_state = &rcu_preempt_state; +#else /* #ifdef CONFIG_PREEMPT_RCU */ +static struct rcu_state *rcu_sysidle_state = &rcu_sched_state; +#endif /* #else #ifdef CONFIG_PREEMPT_RCU */ + +static int full_sysidle_state; /* Current system-idle state. */ +#define RCU_SYSIDLE_NOT 0 /* Some CPU is not idle. */ +#define RCU_SYSIDLE_SHORT 1 /* All CPUs idle for brief period. */ +#define RCU_SYSIDLE_LONG 2 /* All CPUs idle for long enough. */ +#define RCU_SYSIDLE_FULL 3 /* All CPUs idle, ready for sysidle. */ +#define RCU_SYSIDLE_FULL_NOTED 4 /* Actually entered sysidle state. */ + +/* + * Invoked to note exit from irq or task transition to idle. Note that + * usermode execution does -not- count as idle here! After all, we want + * to detect full-system idle states, not RCU quiescent states and grace + * periods. The caller must have disabled interrupts. + */ +static void rcu_sysidle_enter(struct rcu_dynticks *rdtp, int irq) +{ + unsigned long j; + + /* Adjust nesting, check for fully idle. */ + if (irq) { + rdtp->dynticks_idle_nesting--; + WARN_ON_ONCE(rdtp->dynticks_idle_nesting < 0); + if (rdtp->dynticks_idle_nesting != 0) + return; /* Still not fully idle. */ + } else { + if ((rdtp->dynticks_idle_nesting & DYNTICK_TASK_NEST_MASK) == + DYNTICK_TASK_NEST_VALUE) { + rdtp->dynticks_idle_nesting = 0; + } else { + rdtp->dynticks_idle_nesting -= DYNTICK_TASK_NEST_VALUE; + WARN_ON_ONCE(rdtp->dynticks_idle_nesting < 0); + return; /* Still not fully idle. */ + } + } + + /* Record start of fully idle period. */ + j = jiffies; + ACCESS_ONCE(rdtp->dynticks_idle_jiffies) = j; + smp_mb__before_atomic_inc(); + atomic_inc(&rdtp->dynticks_idle); + smp_mb__after_atomic_inc(); + WARN_ON_ONCE(atomic_read(&rdtp->dynticks_idle) & 0x1); +} + +/* + * Unconditionally force exit from full system-idle state. This is + * invoked when a normal CPU exits idle, but must be called separately + * for the timekeeping CPU (tick_do_timer_cpu). The reason for this + * is that the timekeeping CPU is permitted to take scheduling-clock + * interrupts while the system is in system-idle state, and of course + * rcu_sysidle_exit() has no way of distinguishing a scheduling-clock + * interrupt from any other type of interrupt. + */ +void rcu_sysidle_force_exit(void) +{ + int oldstate = ACCESS_ONCE(full_sysidle_state); + int newoldstate; + + /* + * Each pass through the following loop attempts to exit full + * system-idle state. If contention proves to be a problem, + * a trylock-based contention tree could be used here. + */ + while (oldstate > RCU_SYSIDLE_SHORT) { + newoldstate = cmpxchg(&full_sysidle_state, + oldstate, RCU_SYSIDLE_NOT); + if (oldstate == newoldstate && + oldstate == RCU_SYSIDLE_FULL_NOTED) { + rcu_kick_nohz_cpu(tick_do_timer_cpu); + return; /* We cleared it, done! */ + } + oldstate = newoldstate; + } + smp_mb(); /* Order initial oldstate fetch vs. later non-idle work. */ +} + +/* + * Invoked to note entry to irq or task transition from idle. Note that + * usermode execution does -not- count as idle here! The caller must + * have disabled interrupts. + */ +static void rcu_sysidle_exit(struct rcu_dynticks *rdtp, int irq) +{ + /* Adjust nesting, check for already non-idle. */ + if (irq) { + rdtp->dynticks_idle_nesting++; + WARN_ON_ONCE(rdtp->dynticks_idle_nesting <= 0); + if (rdtp->dynticks_idle_nesting != 1) + return; /* Already non-idle. */ + } else { + /* + * Allow for irq misnesting. Yes, it really is possible + * to enter an irq handler then never leave it, and maybe + * also vice versa. Handle both possibilities. + */ + if (rdtp->dynticks_idle_nesting & DYNTICK_TASK_NEST_MASK) { + rdtp->dynticks_idle_nesting += DYNTICK_TASK_NEST_VALUE; + WARN_ON_ONCE(rdtp->dynticks_idle_nesting <= 0); + return; /* Already non-idle. */ + } else { + rdtp->dynticks_idle_nesting = DYNTICK_TASK_EXIT_IDLE; + } + } + + /* Record end of idle period. */ + smp_mb__before_atomic_inc(); + atomic_inc(&rdtp->dynticks_idle); + smp_mb__after_atomic_inc(); + WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks_idle) & 0x1)); + + /* + * If we are the timekeeping CPU, we are permitted to be non-idle + * during a system-idle state. This must be the case, because + * the timekeeping CPU has to take scheduling-clock interrupts + * during the time that the system is transitioning to full + * system-idle state. This means that the timekeeping CPU must + * invoke rcu_sysidle_force_exit() directly if it does anything + * more than take a scheduling-clock interrupt. + */ + if (smp_processor_id() == tick_do_timer_cpu) + return; + + /* Update system-idle state: We are clearly no longer fully idle! */ + rcu_sysidle_force_exit(); +} + +/* + * Check to see if the current CPU is idle. Note that usermode execution + * does not count as idle. The caller must have disabled interrupts. + */ +static void rcu_sysidle_check_cpu(struct rcu_data *rdp, bool *isidle, + unsigned long *maxj) +{ + int cur; + unsigned long j; + struct rcu_dynticks *rdtp = rdp->dynticks; + + /* + * If some other CPU has already reported non-idle, if this is + * not the flavor of RCU that tracks sysidle state, or if this + * is an offline or the timekeeping CPU, nothing to do. + */ + if (!*isidle || rdp->rsp != rcu_sysidle_state || + cpu_is_offline(rdp->cpu) || rdp->cpu == tick_do_timer_cpu) + return; + if (rcu_gp_in_progress(rdp->rsp)) + WARN_ON_ONCE(smp_processor_id() != tick_do_timer_cpu); + + /* Pick up current idle and NMI-nesting counter and check. */ + cur = atomic_read(&rdtp->dynticks_idle); + if (cur & 0x1) { + *isidle = false; /* We are not idle! */ + return; + } + smp_mb(); /* Read counters before timestamps. */ + + /* Pick up timestamps. */ + j = ACCESS_ONCE(rdtp->dynticks_idle_jiffies); + /* If this CPU entered idle more recently, update maxj timestamp. */ + if (ULONG_CMP_LT(*maxj, j)) + *maxj = j; +} + +/* + * Is this the flavor of RCU that is handling full-system idle? + */ +static bool is_sysidle_rcu_state(struct rcu_state *rsp) +{ + return rsp == rcu_sysidle_state; +} + +/* + * Bind the grace-period kthread for the sysidle flavor of RCU to the + * timekeeping CPU. + */ +static void rcu_bind_gp_kthread(void) +{ + int cpu = ACCESS_ONCE(tick_do_timer_cpu); + + if (cpu < 0 || cpu >= nr_cpu_ids) + return; + if (raw_smp_processor_id() != cpu) + set_cpus_allowed_ptr(current, cpumask_of(cpu)); +} + +/* + * Return a delay in jiffies based on the number of CPUs, rcu_node + * leaf fanout, and jiffies tick rate. The idea is to allow larger + * systems more time to transition to full-idle state in order to + * avoid the cache thrashing that otherwise occur on the state variable. + * Really small systems (less than a couple of tens of CPUs) should + * instead use a single global atomically incremented counter, and later + * versions of this will automatically reconfigure themselves accordingly. + */ +static unsigned long rcu_sysidle_delay(void) +{ + if (nr_cpu_ids <= CONFIG_NO_HZ_FULL_SYSIDLE_SMALL) + return 0; + return DIV_ROUND_UP(nr_cpu_ids * HZ, rcu_fanout_leaf * 1000); +} + +/* + * Advance the full-system-idle state. This is invoked when all of + * the non-timekeeping CPUs are idle. + */ +static void rcu_sysidle(unsigned long j) +{ + /* Check the current state. */ + switch (ACCESS_ONCE(full_sysidle_state)) { + case RCU_SYSIDLE_NOT: + + /* First time all are idle, so note a short idle period. */ + ACCESS_ONCE(full_sysidle_state) = RCU_SYSIDLE_SHORT; + break; + + case RCU_SYSIDLE_SHORT: + + /* + * Idle for a bit, time to advance to next state? + * cmpxchg failure means race with non-idle, let them win. + */ + if (ULONG_CMP_GE(jiffies, j + rcu_sysidle_delay())) + (void)cmpxchg(&full_sysidle_state, + RCU_SYSIDLE_SHORT, RCU_SYSIDLE_LONG); + break; + + case RCU_SYSIDLE_LONG: + + /* + * Do an additional check pass before advancing to full. + * cmpxchg failure means race with non-idle, let them win. + */ + if (ULONG_CMP_GE(jiffies, j + rcu_sysidle_delay())) + (void)cmpxchg(&full_sysidle_state, + RCU_SYSIDLE_LONG, RCU_SYSIDLE_FULL); + break; + + default: + break; + } +} + +/* + * Found a non-idle non-timekeeping CPU, so kick the system-idle state + * back to the beginning. + */ +static void rcu_sysidle_cancel(void) +{ + smp_mb(); + ACCESS_ONCE(full_sysidle_state) = RCU_SYSIDLE_NOT; +} + +/* + * Update the sysidle state based on the results of a force-quiescent-state + * scan of the CPUs' dyntick-idle state. + */ +static void rcu_sysidle_report(struct rcu_state *rsp, int isidle, + unsigned long maxj, bool gpkt) +{ + if (rsp != rcu_sysidle_state) + return; /* Wrong flavor, ignore. */ + if (gpkt && nr_cpu_ids <= CONFIG_NO_HZ_FULL_SYSIDLE_SMALL) + return; /* Running state machine from timekeeping CPU. */ + if (isidle) + rcu_sysidle(maxj); /* More idle! */ + else + rcu_sysidle_cancel(); /* Idle is over. */ +} + +/* + * Wrapper for rcu_sysidle_report() when called from the grace-period + * kthread's context. + */ +static void rcu_sysidle_report_gp(struct rcu_state *rsp, int isidle, + unsigned long maxj) +{ + rcu_sysidle_report(rsp, isidle, maxj, true); +} + +/* Callback and function for forcing an RCU grace period. */ +struct rcu_sysidle_head { + struct rcu_head rh; + int inuse; +}; + +static void rcu_sysidle_cb(struct rcu_head *rhp) +{ + struct rcu_sysidle_head *rshp; + + /* + * The following memory barrier is needed to replace the + * memory barriers that would normally be in the memory + * allocator. + */ + smp_mb(); /* grace period precedes setting inuse. */ + + rshp = container_of(rhp, struct rcu_sysidle_head, rh); + ACCESS_ONCE(rshp->inuse) = 0; +} + +/* + * Check to see if the system is fully idle, other than the timekeeping CPU. + * The caller must have disabled interrupts. + */ +bool rcu_sys_is_idle(void) +{ + static struct rcu_sysidle_head rsh; + int rss = ACCESS_ONCE(full_sysidle_state); + + if (WARN_ON_ONCE(smp_processor_id() != tick_do_timer_cpu)) + return false; + + /* Handle small-system case by doing a full scan of CPUs. */ + if (nr_cpu_ids <= CONFIG_NO_HZ_FULL_SYSIDLE_SMALL) { + int oldrss = rss - 1; + + /* + * One pass to advance to each state up to _FULL. + * Give up if any pass fails to advance the state. + */ + while (rss < RCU_SYSIDLE_FULL && oldrss < rss) { + int cpu; + bool isidle = true; + unsigned long maxj = jiffies - ULONG_MAX / 4; + struct rcu_data *rdp; + + /* Scan all the CPUs looking for nonidle CPUs. */ + for_each_possible_cpu(cpu) { + rdp = per_cpu_ptr(rcu_sysidle_state->rda, cpu); + rcu_sysidle_check_cpu(rdp, &isidle, &maxj); + if (!isidle) + break; + } + rcu_sysidle_report(rcu_sysidle_state, + isidle, maxj, false); + oldrss = rss; + rss = ACCESS_ONCE(full_sysidle_state); + } + } + + /* If this is the first observation of an idle period, record it. */ + if (rss == RCU_SYSIDLE_FULL) { + rss = cmpxchg(&full_sysidle_state, + RCU_SYSIDLE_FULL, RCU_SYSIDLE_FULL_NOTED); + return rss == RCU_SYSIDLE_FULL; + } + + smp_mb(); /* ensure rss load happens before later caller actions. */ + + /* If already fully idle, tell the caller (in case of races). */ + if (rss == RCU_SYSIDLE_FULL_NOTED) + return true; + + /* + * If we aren't there yet, and a grace period is not in flight, + * initiate a grace period. Either way, tell the caller that + * we are not there yet. We use an xchg() rather than an assignment + * to make up for the memory barriers that would otherwise be + * provided by the memory allocator. + */ + if (nr_cpu_ids > CONFIG_NO_HZ_FULL_SYSIDLE_SMALL && + !rcu_gp_in_progress(rcu_sysidle_state) && + !rsh.inuse && xchg(&rsh.inuse, 1) == 0) + call_rcu(&rsh.rh, rcu_sysidle_cb); + return false; +} + +/* + * Initialize dynticks sysidle state for CPUs coming online. + */ +static void rcu_sysidle_init_percpu_data(struct rcu_dynticks *rdtp) +{ + rdtp->dynticks_idle_nesting = DYNTICK_TASK_NEST_VALUE; +} + +#else /* #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ + +static void rcu_sysidle_enter(struct rcu_dynticks *rdtp, int irq) +{ +} + +static void rcu_sysidle_exit(struct rcu_dynticks *rdtp, int irq) +{ +} + +static void rcu_sysidle_check_cpu(struct rcu_data *rdp, bool *isidle, + unsigned long *maxj) +{ +} + +static bool is_sysidle_rcu_state(struct rcu_state *rsp) +{ + return false; +} + +static void rcu_bind_gp_kthread(void) +{ +} + +static void rcu_sysidle_report_gp(struct rcu_state *rsp, int isidle, + unsigned long maxj) +{ +} + +static void rcu_sysidle_init_percpu_data(struct rcu_dynticks *rdtp) +{ +} + +#endif /* #else #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ diff --git a/kernel/sched/core.c b/kernel/sched/core.c index b7c32cb7bfeb..e53bda3ff2f1 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -933,6 +933,8 @@ static int effective_prio(struct task_struct *p) /** * task_curr - is this task currently executing on a CPU? * @p: the task in question. + * + * Return: 1 if the task is currently executing. 0 otherwise. */ inline int task_curr(const struct task_struct *p) { @@ -1482,7 +1484,7 @@ static void ttwu_queue(struct task_struct *p, int cpu) * the simpler "current->state = TASK_RUNNING" to mark yourself * runnable without the overhead of this. * - * Returns %true if @p was woken up, %false if it was already running + * Return: %true if @p was woken up, %false if it was already running. * or @state didn't match @p's state. */ static int @@ -1491,7 +1493,13 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) unsigned long flags; int cpu, success = 0; - smp_wmb(); + /* + * If we are going to wake up a thread waiting for CONDITION we + * need to ensure that CONDITION=1 done by the caller can not be + * reordered with p->state check below. This pairs with mb() in + * set_current_state() the waiting thread does. + */ + smp_mb__before_spinlock(); raw_spin_lock_irqsave(&p->pi_lock, flags); if (!(p->state & state)) goto out; @@ -1577,8 +1585,9 @@ out: * @p: The process to be woken up. * * Attempt to wake up the nominated process and move it to the set of runnable - * processes. Returns 1 if the process was woken up, 0 if it was already - * running. + * processes. + * + * Return: 1 if the process was woken up, 0 if it was already running. * * It may be assumed that this function implies a write memory barrier before * changing the task state if and only if any tasks are woken up. @@ -2191,6 +2200,8 @@ void scheduler_tick(void) * This makes sure that uptime, CFS vruntime, load * balancing, etc... continue to move forward, even * with a very low granularity. + * + * Return: Maximum deferment in nanoseconds. */ u64 scheduler_tick_max_deferment(void) { @@ -2394,6 +2405,12 @@ need_resched: if (sched_feat(HRTICK)) hrtick_clear(rq); + /* + * Make sure that signal_pending_state()->signal_pending() below + * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE) + * done by the caller to avoid the race with signal_wake_up(). + */ + smp_mb__before_spinlock(); raw_spin_lock_irq(&rq->lock); switch_count = &prev->nivcsw; @@ -2796,8 +2813,8 @@ EXPORT_SYMBOL(wait_for_completion); * specified timeout to expire. The timeout is in jiffies. It is not * interruptible. * - * The return value is 0 if timed out, and positive (at least 1, or number of - * jiffies left till timeout) if completed. + * Return: 0 if timed out, and positive (at least 1, or number of jiffies left + * till timeout) if completed. */ unsigned long __sched wait_for_completion_timeout(struct completion *x, unsigned long timeout) @@ -2829,8 +2846,8 @@ EXPORT_SYMBOL(wait_for_completion_io); * specified timeout to expire. The timeout is in jiffies. It is not * interruptible. The caller is accounted as waiting for IO. * - * The return value is 0 if timed out, and positive (at least 1, or number of - * jiffies left till timeout) if completed. + * Return: 0 if timed out, and positive (at least 1, or number of jiffies left + * till timeout) if completed. */ unsigned long __sched wait_for_completion_io_timeout(struct completion *x, unsigned long timeout) @@ -2846,7 +2863,7 @@ EXPORT_SYMBOL(wait_for_completion_io_timeout); * This waits for completion of a specific task to be signaled. It is * interruptible. * - * The return value is -ERESTARTSYS if interrupted, 0 if completed. + * Return: -ERESTARTSYS if interrupted, 0 if completed. */ int __sched wait_for_completion_interruptible(struct completion *x) { @@ -2865,8 +2882,8 @@ EXPORT_SYMBOL(wait_for_completion_interruptible); * This waits for either a completion of a specific task to be signaled or for a * specified timeout to expire. It is interruptible. The timeout is in jiffies. * - * The return value is -ERESTARTSYS if interrupted, 0 if timed out, - * positive (at least 1, or number of jiffies left till timeout) if completed. + * Return: -ERESTARTSYS if interrupted, 0 if timed out, positive (at least 1, + * or number of jiffies left till timeout) if completed. */ long __sched wait_for_completion_interruptible_timeout(struct completion *x, @@ -2883,7 +2900,7 @@ EXPORT_SYMBOL(wait_for_completion_interruptible_timeout); * This waits to be signaled for completion of a specific task. It can be * interrupted by a kill signal. * - * The return value is -ERESTARTSYS if interrupted, 0 if completed. + * Return: -ERESTARTSYS if interrupted, 0 if completed. */ int __sched wait_for_completion_killable(struct completion *x) { @@ -2903,8 +2920,8 @@ EXPORT_SYMBOL(wait_for_completion_killable); * signaled or for a specified timeout to expire. It can be * interrupted by a kill signal. The timeout is in jiffies. * - * The return value is -ERESTARTSYS if interrupted, 0 if timed out, - * positive (at least 1, or number of jiffies left till timeout) if completed. + * Return: -ERESTARTSYS if interrupted, 0 if timed out, positive (at least 1, + * or number of jiffies left till timeout) if completed. */ long __sched wait_for_completion_killable_timeout(struct completion *x, @@ -2918,7 +2935,7 @@ EXPORT_SYMBOL(wait_for_completion_killable_timeout); * try_wait_for_completion - try to decrement a completion without blocking * @x: completion structure * - * Returns: 0 if a decrement cannot be done without blocking + * Return: 0 if a decrement cannot be done without blocking * 1 if a decrement succeeded. * * If a completion is being used as a counting completion, @@ -2945,7 +2962,7 @@ EXPORT_SYMBOL(try_wait_for_completion); * completion_done - Test to see if a completion has any waiters * @x: completion structure * - * Returns: 0 if there are waiters (wait_for_completion() in progress) + * Return: 0 if there are waiters (wait_for_completion() in progress) * 1 if there are no waiters. * */ @@ -3182,7 +3199,7 @@ SYSCALL_DEFINE1(nice, int, increment) * task_prio - return the priority value of a given task. * @p: the task in question. * - * This is the priority value as seen by users in /proc. + * Return: The priority value as seen by users in /proc. * RT tasks are offset by -200. Normal tasks are centered * around 0, value goes from -16 to +15. */ @@ -3194,6 +3211,8 @@ int task_prio(const struct task_struct *p) /** * task_nice - return the nice value of a given task. * @p: the task in question. + * + * Return: The nice value [ -20 ... 0 ... 19 ]. */ int task_nice(const struct task_struct *p) { @@ -3204,6 +3223,8 @@ EXPORT_SYMBOL(task_nice); /** * idle_cpu - is a given cpu idle currently? * @cpu: the processor in question. + * + * Return: 1 if the CPU is currently idle. 0 otherwise. */ int idle_cpu(int cpu) { @@ -3226,6 +3247,8 @@ int idle_cpu(int cpu) /** * idle_task - return the idle task for a given cpu. * @cpu: the processor in question. + * + * Return: The idle task for the cpu @cpu. */ struct task_struct *idle_task(int cpu) { @@ -3235,6 +3258,8 @@ struct task_struct *idle_task(int cpu) /** * find_process_by_pid - find a process with a matching PID value. * @pid: the pid in question. + * + * The task of @pid, if found. %NULL otherwise. */ static struct task_struct *find_process_by_pid(pid_t pid) { @@ -3432,6 +3457,8 @@ recheck: * @policy: new policy. * @param: structure containing the new RT priority. * + * Return: 0 on success. An error code otherwise. + * * NOTE that the task may be already dead. */ int sched_setscheduler(struct task_struct *p, int policy, @@ -3451,6 +3478,8 @@ EXPORT_SYMBOL_GPL(sched_setscheduler); * current context has permission. For example, this is needed in * stop_machine(): we create temporary high priority worker threads, * but our caller might not have that capability. + * + * Return: 0 on success. An error code otherwise. */ int sched_setscheduler_nocheck(struct task_struct *p, int policy, const struct sched_param *param) @@ -3485,6 +3514,8 @@ do_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) * @pid: the pid in question. * @policy: new policy. * @param: structure containing the new RT priority. + * + * Return: 0 on success. An error code otherwise. */ SYSCALL_DEFINE3(sched_setscheduler, pid_t, pid, int, policy, struct sched_param __user *, param) @@ -3500,6 +3531,8 @@ SYSCALL_DEFINE3(sched_setscheduler, pid_t, pid, int, policy, * sys_sched_setparam - set/change the RT priority of a thread * @pid: the pid in question. * @param: structure containing the new RT priority. + * + * Return: 0 on success. An error code otherwise. */ SYSCALL_DEFINE2(sched_setparam, pid_t, pid, struct sched_param __user *, param) { @@ -3509,6 +3542,9 @@ SYSCALL_DEFINE2(sched_setparam, pid_t, pid, struct sched_param __user *, param) /** * sys_sched_getscheduler - get the policy (scheduling class) of a thread * @pid: the pid in question. + * + * Return: On success, the policy of the thread. Otherwise, a negative error + * code. */ SYSCALL_DEFINE1(sched_getscheduler, pid_t, pid) { @@ -3535,6 +3571,9 @@ SYSCALL_DEFINE1(sched_getscheduler, pid_t, pid) * sys_sched_getparam - get the RT priority of a thread * @pid: the pid in question. * @param: structure containing the RT priority. + * + * Return: On success, 0 and the RT priority is in @param. Otherwise, an error + * code. */ SYSCALL_DEFINE2(sched_getparam, pid_t, pid, struct sched_param __user *, param) { @@ -3659,6 +3698,8 @@ static int get_user_cpu_mask(unsigned long __user *user_mask_ptr, unsigned len, * @pid: pid of the process * @len: length in bytes of the bitmask pointed to by user_mask_ptr * @user_mask_ptr: user-space pointer to the new cpu mask + * + * Return: 0 on success. An error code otherwise. */ SYSCALL_DEFINE3(sched_setaffinity, pid_t, pid, unsigned int, len, unsigned long __user *, user_mask_ptr) @@ -3710,6 +3751,8 @@ out_unlock: * @pid: pid of the process * @len: length in bytes of the bitmask pointed to by user_mask_ptr * @user_mask_ptr: user-space pointer to hold the current cpu mask + * + * Return: 0 on success. An error code otherwise. */ SYSCALL_DEFINE3(sched_getaffinity, pid_t, pid, unsigned int, len, unsigned long __user *, user_mask_ptr) @@ -3744,6 +3787,8 @@ SYSCALL_DEFINE3(sched_getaffinity, pid_t, pid, unsigned int, len, * * This function yields the current CPU to other tasks. If there are no * other threads running on this CPU then this function will return. + * + * Return: 0. */ SYSCALL_DEFINE0(sched_yield) { @@ -3869,7 +3914,7 @@ EXPORT_SYMBOL(yield); * It's the caller's job to ensure that the target task struct * can't go away on us before we can do any checks. * - * Returns: + * Return: * true (>0) if we indeed boosted the target task. * false (0) if we failed to boost the target. * -ESRCH if there's no task to yield to. @@ -3972,8 +4017,9 @@ long __sched io_schedule_timeout(long timeout) * sys_sched_get_priority_max - return maximum RT priority. * @policy: scheduling class. * - * this syscall returns the maximum rt_priority that can be used - * by a given scheduling class. + * Return: On success, this syscall returns the maximum + * rt_priority that can be used by a given scheduling class. + * On failure, a negative error code is returned. */ SYSCALL_DEFINE1(sched_get_priority_max, int, policy) { @@ -3997,8 +4043,9 @@ SYSCALL_DEFINE1(sched_get_priority_max, int, policy) * sys_sched_get_priority_min - return minimum RT priority. * @policy: scheduling class. * - * this syscall returns the minimum rt_priority that can be used - * by a given scheduling class. + * Return: On success, this syscall returns the minimum + * rt_priority that can be used by a given scheduling class. + * On failure, a negative error code is returned. */ SYSCALL_DEFINE1(sched_get_priority_min, int, policy) { @@ -4024,6 +4071,9 @@ SYSCALL_DEFINE1(sched_get_priority_min, int, policy) * * this syscall writes the default timeslice value of a given process * into the user-space timespec buffer. A value of '0' means infinity. + * + * Return: On success, 0 and the timeslice is in @interval. Otherwise, + * an error code. */ SYSCALL_DEFINE2(sched_rr_get_interval, pid_t, pid, struct timespec __user *, interval) @@ -6632,6 +6682,8 @@ void normalize_rt_tasks(void) * @cpu: the processor in question. * * ONLY VALID WHEN THE WHOLE SYSTEM IS STOPPED! + * + * Return: The current task for @cpu. */ struct task_struct *curr_task(int cpu) { @@ -6763,7 +6815,7 @@ void sched_move_task(struct task_struct *tsk) if (unlikely(running)) tsk->sched_class->put_prev_task(rq, tsk); - tg = container_of(task_subsys_state_check(tsk, cpu_cgroup_subsys_id, + tg = container_of(task_css_check(tsk, cpu_cgroup_subsys_id, lockdep_is_held(&tsk->sighand->siglock)), struct task_group, css); tg = autogroup_task_group(tsk, tg); @@ -7085,23 +7137,22 @@ int sched_rt_handler(struct ctl_table *table, int write, #ifdef CONFIG_CGROUP_SCHED -/* return corresponding task_group object of a cgroup */ -static inline struct task_group *cgroup_tg(struct cgroup *cgrp) +static inline struct task_group *css_tg(struct cgroup_subsys_state *css) { - return container_of(cgroup_subsys_state(cgrp, cpu_cgroup_subsys_id), - struct task_group, css); + return css ? container_of(css, struct task_group, css) : NULL; } -static struct cgroup_subsys_state *cpu_cgroup_css_alloc(struct cgroup *cgrp) +static struct cgroup_subsys_state * +cpu_cgroup_css_alloc(struct cgroup_subsys_state *parent_css) { - struct task_group *tg, *parent; + struct task_group *parent = css_tg(parent_css); + struct task_group *tg; - if (!cgrp->parent) { + if (!parent) { /* This is early initialization for the top cgroup */ return &root_task_group.css; } - parent = cgroup_tg(cgrp->parent); tg = sched_create_group(parent); if (IS_ERR(tg)) return ERR_PTR(-ENOMEM); @@ -7109,41 +7160,38 @@ static struct cgroup_subsys_state *cpu_cgroup_css_alloc(struct cgroup *cgrp) return &tg->css; } -static int cpu_cgroup_css_online(struct cgroup *cgrp) +static int cpu_cgroup_css_online(struct cgroup_subsys_state *css) { - struct task_group *tg = cgroup_tg(cgrp); - struct task_group *parent; - - if (!cgrp->parent) - return 0; + struct task_group *tg = css_tg(css); + struct task_group *parent = css_tg(css_parent(css)); - parent = cgroup_tg(cgrp->parent); - sched_online_group(tg, parent); + if (parent) + sched_online_group(tg, parent); return 0; } -static void cpu_cgroup_css_free(struct cgroup *cgrp) +static void cpu_cgroup_css_free(struct cgroup_subsys_state *css) { - struct task_group *tg = cgroup_tg(cgrp); + struct task_group *tg = css_tg(css); sched_destroy_group(tg); } -static void cpu_cgroup_css_offline(struct cgroup *cgrp) +static void cpu_cgroup_css_offline(struct cgroup_subsys_state *css) { - struct task_group *tg = cgroup_tg(cgrp); + struct task_group *tg = css_tg(css); sched_offline_group(tg); } -static int cpu_cgroup_can_attach(struct cgroup *cgrp, +static int cpu_cgroup_can_attach(struct cgroup_subsys_state *css, struct cgroup_taskset *tset) { struct task_struct *task; - cgroup_taskset_for_each(task, cgrp, tset) { + cgroup_taskset_for_each(task, css, tset) { #ifdef CONFIG_RT_GROUP_SCHED - if (!sched_rt_can_attach(cgroup_tg(cgrp), task)) + if (!sched_rt_can_attach(css_tg(css), task)) return -EINVAL; #else /* We don't support RT-tasks being in separate groups */ @@ -7154,18 +7202,18 @@ static int cpu_cgroup_can_attach(struct cgroup *cgrp, return 0; } -static void cpu_cgroup_attach(struct cgroup *cgrp, +static void cpu_cgroup_attach(struct cgroup_subsys_state *css, struct cgroup_taskset *tset) { struct task_struct *task; - cgroup_taskset_for_each(task, cgrp, tset) + cgroup_taskset_for_each(task, css, tset) sched_move_task(task); } -static void -cpu_cgroup_exit(struct cgroup *cgrp, struct cgroup *old_cgrp, - struct task_struct *task) +static void cpu_cgroup_exit(struct cgroup_subsys_state *css, + struct cgroup_subsys_state *old_css, + struct task_struct *task) { /* * cgroup_exit() is called in the copy_process() failure path. @@ -7179,15 +7227,16 @@ cpu_cgroup_exit(struct cgroup *cgrp, struct cgroup *old_cgrp, } #ifdef CONFIG_FAIR_GROUP_SCHED -static int cpu_shares_write_u64(struct cgroup *cgrp, struct cftype *cftype, - u64 shareval) +static int cpu_shares_write_u64(struct cgroup_subsys_state *css, + struct cftype *cftype, u64 shareval) { - return sched_group_set_shares(cgroup_tg(cgrp), scale_load(shareval)); + return sched_group_set_shares(css_tg(css), scale_load(shareval)); } -static u64 cpu_shares_read_u64(struct cgroup *cgrp, struct cftype *cft) +static u64 cpu_shares_read_u64(struct cgroup_subsys_state *css, + struct cftype *cft) { - struct task_group *tg = cgroup_tg(cgrp); + struct task_group *tg = css_tg(css); return (u64) scale_load_down(tg->shares); } @@ -7309,26 +7358,28 @@ long tg_get_cfs_period(struct task_group *tg) return cfs_period_us; } -static s64 cpu_cfs_quota_read_s64(struct cgroup *cgrp, struct cftype *cft) +static s64 cpu_cfs_quota_read_s64(struct cgroup_subsys_state *css, + struct cftype *cft) { - return tg_get_cfs_quota(cgroup_tg(cgrp)); + return tg_get_cfs_quota(css_tg(css)); } -static int cpu_cfs_quota_write_s64(struct cgroup *cgrp, struct cftype *cftype, - s64 cfs_quota_us) +static int cpu_cfs_quota_write_s64(struct cgroup_subsys_state *css, + struct cftype *cftype, s64 cfs_quota_us) { - return tg_set_cfs_quota(cgroup_tg(cgrp), cfs_quota_us); + return tg_set_cfs_quota(css_tg(css), cfs_quota_us); } -static u64 cpu_cfs_period_read_u64(struct cgroup *cgrp, struct cftype *cft) +static u64 cpu_cfs_period_read_u64(struct cgroup_subsys_state *css, + struct cftype *cft) { - return tg_get_cfs_period(cgroup_tg(cgrp)); + return tg_get_cfs_period(css_tg(css)); } -static int cpu_cfs_period_write_u64(struct cgroup *cgrp, struct cftype *cftype, - u64 cfs_period_us) +static int cpu_cfs_period_write_u64(struct cgroup_subsys_state *css, + struct cftype *cftype, u64 cfs_period_us) { - return tg_set_cfs_period(cgroup_tg(cgrp), cfs_period_us); + return tg_set_cfs_period(css_tg(css), cfs_period_us); } struct cfs_schedulable_data { @@ -7409,10 +7460,10 @@ static int __cfs_schedulable(struct task_group *tg, u64 period, u64 quota) return ret; } -static int cpu_stats_show(struct cgroup *cgrp, struct cftype *cft, +static int cpu_stats_show(struct cgroup_subsys_state *css, struct cftype *cft, struct cgroup_map_cb *cb) { - struct task_group *tg = cgroup_tg(cgrp); + struct task_group *tg = css_tg(css); struct cfs_bandwidth *cfs_b = &tg->cfs_bandwidth; cb->fill(cb, "nr_periods", cfs_b->nr_periods); @@ -7425,26 +7476,28 @@ static int cpu_stats_show(struct cgroup *cgrp, struct cftype *cft, #endif /* CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_RT_GROUP_SCHED -static int cpu_rt_runtime_write(struct cgroup *cgrp, struct cftype *cft, - s64 val) +static int cpu_rt_runtime_write(struct cgroup_subsys_state *css, + struct cftype *cft, s64 val) { - return sched_group_set_rt_runtime(cgroup_tg(cgrp), val); + return sched_group_set_rt_runtime(css_tg(css), val); } -static s64 cpu_rt_runtime_read(struct cgroup *cgrp, struct cftype *cft) +static s64 cpu_rt_runtime_read(struct cgroup_subsys_state *css, + struct cftype *cft) { - return sched_group_rt_runtime(cgroup_tg(cgrp)); + return sched_group_rt_runtime(css_tg(css)); } -static int cpu_rt_period_write_uint(struct cgroup *cgrp, struct cftype *cftype, - u64 rt_period_us) +static int cpu_rt_period_write_uint(struct cgroup_subsys_state *css, + struct cftype *cftype, u64 rt_period_us) { - return sched_group_set_rt_period(cgroup_tg(cgrp), rt_period_us); + return sched_group_set_rt_period(css_tg(css), rt_period_us); } -static u64 cpu_rt_period_read_uint(struct cgroup *cgrp, struct cftype *cft) +static u64 cpu_rt_period_read_uint(struct cgroup_subsys_state *css, + struct cftype *cft) { - return sched_group_rt_period(cgroup_tg(cgrp)); + return sched_group_rt_period(css_tg(css)); } #endif /* CONFIG_RT_GROUP_SCHED */ diff --git a/kernel/sched/cpuacct.c b/kernel/sched/cpuacct.c index dbb7e2cd95eb..f64722ff0299 100644 --- a/kernel/sched/cpuacct.c +++ b/kernel/sched/cpuacct.c @@ -33,30 +33,20 @@ struct cpuacct { struct kernel_cpustat __percpu *cpustat; }; -/* return cpu accounting group corresponding to this container */ -static inline struct cpuacct *cgroup_ca(struct cgroup *cgrp) +static inline struct cpuacct *css_ca(struct cgroup_subsys_state *css) { - return container_of(cgroup_subsys_state(cgrp, cpuacct_subsys_id), - struct cpuacct, css); + return css ? container_of(css, struct cpuacct, css) : NULL; } /* return cpu accounting group to which this task belongs */ static inline struct cpuacct *task_ca(struct task_struct *tsk) { - return container_of(task_subsys_state(tsk, cpuacct_subsys_id), - struct cpuacct, css); -} - -static inline struct cpuacct *__parent_ca(struct cpuacct *ca) -{ - return cgroup_ca(ca->css.cgroup->parent); + return css_ca(task_css(tsk, cpuacct_subsys_id)); } static inline struct cpuacct *parent_ca(struct cpuacct *ca) { - if (!ca->css.cgroup->parent) - return NULL; - return cgroup_ca(ca->css.cgroup->parent); + return css_ca(css_parent(&ca->css)); } static DEFINE_PER_CPU(u64, root_cpuacct_cpuusage); @@ -66,11 +56,12 @@ static struct cpuacct root_cpuacct = { }; /* create a new cpu accounting group */ -static struct cgroup_subsys_state *cpuacct_css_alloc(struct cgroup *cgrp) +static struct cgroup_subsys_state * +cpuacct_css_alloc(struct cgroup_subsys_state *parent_css) { struct cpuacct *ca; - if (!cgrp->parent) + if (!parent_css) return &root_cpuacct.css; ca = kzalloc(sizeof(*ca), GFP_KERNEL); @@ -96,9 +87,9 @@ out: } /* destroy an existing cpu accounting group */ -static void cpuacct_css_free(struct cgroup *cgrp) +static void cpuacct_css_free(struct cgroup_subsys_state *css) { - struct cpuacct *ca = cgroup_ca(cgrp); + struct cpuacct *ca = css_ca(css); free_percpu(ca->cpustat); free_percpu(ca->cpuusage); @@ -141,9 +132,9 @@ static void cpuacct_cpuusage_write(struct cpuacct *ca, int cpu, u64 val) } /* return total cpu usage (in nanoseconds) of a group */ -static u64 cpuusage_read(struct cgroup *cgrp, struct cftype *cft) +static u64 cpuusage_read(struct cgroup_subsys_state *css, struct cftype *cft) { - struct cpuacct *ca = cgroup_ca(cgrp); + struct cpuacct *ca = css_ca(css); u64 totalcpuusage = 0; int i; @@ -153,10 +144,10 @@ static u64 cpuusage_read(struct cgroup *cgrp, struct cftype *cft) return totalcpuusage; } -static int cpuusage_write(struct cgroup *cgrp, struct cftype *cftype, - u64 reset) +static int cpuusage_write(struct cgroup_subsys_state *css, struct cftype *cft, + u64 reset) { - struct cpuacct *ca = cgroup_ca(cgrp); + struct cpuacct *ca = css_ca(css); int err = 0; int i; @@ -172,10 +163,10 @@ out: return err; } -static int cpuacct_percpu_seq_read(struct cgroup *cgroup, struct cftype *cft, - struct seq_file *m) +static int cpuacct_percpu_seq_read(struct cgroup_subsys_state *css, + struct cftype *cft, struct seq_file *m) { - struct cpuacct *ca = cgroup_ca(cgroup); + struct cpuacct *ca = css_ca(css); u64 percpu; int i; @@ -192,10 +183,10 @@ static const char * const cpuacct_stat_desc[] = { [CPUACCT_STAT_SYSTEM] = "system", }; -static int cpuacct_stats_show(struct cgroup *cgrp, struct cftype *cft, - struct cgroup_map_cb *cb) +static int cpuacct_stats_show(struct cgroup_subsys_state *css, + struct cftype *cft, struct cgroup_map_cb *cb) { - struct cpuacct *ca = cgroup_ca(cgrp); + struct cpuacct *ca = css_ca(css); int cpu; s64 val = 0; @@ -281,7 +272,7 @@ void cpuacct_account_field(struct task_struct *p, int index, u64 val) while (ca != &root_cpuacct) { kcpustat = this_cpu_ptr(ca->cpustat); kcpustat->cpustat[index] += val; - ca = __parent_ca(ca); + ca = parent_ca(ca); } rcu_read_unlock(); } diff --git a/kernel/sched/cpupri.c b/kernel/sched/cpupri.c index 1095e878a46f..8b836b376d91 100644 --- a/kernel/sched/cpupri.c +++ b/kernel/sched/cpupri.c @@ -62,7 +62,7 @@ static int convert_prio(int prio) * any discrepancies created by racing against the uncertainty of the current * priority configuration. * - * Returns: (int)bool - CPUs were found + * Return: (int)bool - CPUs were found */ int cpupri_find(struct cpupri *cp, struct task_struct *p, struct cpumask *lowest_mask) @@ -203,7 +203,7 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) * cpupri_init - initialize the cpupri structure * @cp: The cpupri context * - * Returns: -ENOMEM if memory fails. + * Return: -ENOMEM on memory allocation failure. */ int cpupri_init(struct cpupri *cp) { diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index bb456f44b7b1..68f1609ca149 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -851,7 +851,7 @@ void task_numa_fault(int node, int pages, bool migrated) { struct task_struct *p = current; - if (!sched_feat_numa(NUMA)) + if (!numabalancing_enabled) return; /* FIXME: Allocate task-specific structure for placement policy here */ @@ -2032,6 +2032,7 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) */ update_entity_load_avg(curr, 1); update_cfs_rq_blocked_load(cfs_rq, 1); + update_cfs_shares(cfs_rq); #ifdef CONFIG_SCHED_HRTICK /* @@ -4280,6 +4281,8 @@ struct sg_lb_stats { * get_sd_load_idx - Obtain the load index for a given sched domain. * @sd: The sched_domain whose load_idx is to be obtained. * @idle: The Idle status of the CPU for whose sd load_icx is obtained. + * + * Return: The load index. */ static inline int get_sd_load_idx(struct sched_domain *sd, enum cpu_idle_type idle) @@ -4574,6 +4577,9 @@ static inline void update_sg_lb_stats(struct lb_env *env, * * Determine if @sg is a busier group than the previously selected * busiest group. + * + * Return: %true if @sg is a busier group than the previously selected + * busiest group. %false otherwise. */ static bool update_sd_pick_busiest(struct lb_env *env, struct sd_lb_stats *sds, @@ -4691,7 +4697,7 @@ static inline void update_sd_lb_stats(struct lb_env *env, * assuming lower CPU number will be equivalent to lower a SMT thread * number. * - * Returns 1 when packing is required and a task should be moved to + * Return: 1 when packing is required and a task should be moved to * this CPU. The amount of the imbalance is returned in *imbalance. * * @env: The load balancing environment. @@ -4869,7 +4875,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s * @balance: Pointer to a variable indicating if this_cpu * is the appropriate cpu to perform load balancing at this_level. * - * Returns: - the busiest group if imbalance exists. + * Return: - The busiest group if imbalance exists. * - If no imbalance and user has opted for power-savings balance, * return the least loaded group whose CPUs can be * put to idle by rebalancing its tasks onto our group. @@ -5786,7 +5792,7 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued) entity_tick(cfs_rq, se, queued); } - if (sched_feat_numa(NUMA)) + if (numabalancing_enabled) task_tick_numa(rq, curr); update_rq_runnable_avg(rq, 1); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index ef0a7b2439dd..471a56db05ea 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -665,9 +665,9 @@ extern int group_balance_cpu(struct sched_group *sg); /* * Return the group to which this tasks belongs. * - * We cannot use task_subsys_state() and friends because the cgroup - * subsystem changes that value before the cgroup_subsys::attach() method - * is called, therefore we cannot pin it and might observe the wrong value. + * We cannot use task_css() and friends because the cgroup subsystem + * changes that value before the cgroup_subsys::attach() method is called, + * therefore we cannot pin it and might observe the wrong value. * * The same is true for autogroup's p->signal->autogroup->tg, the autogroup * core changes this before calling sched_move_task(). diff --git a/kernel/sysctl.c b/kernel/sysctl.c index ac09d98490aa..07f6fc468e17 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2346,7 +2346,11 @@ static int do_proc_dointvec_ms_jiffies_conv(bool *negp, unsigned long *lvalp, int write, void *data) { if (write) { - *valp = msecs_to_jiffies(*negp ? -*lvalp : *lvalp); + unsigned long jif = msecs_to_jiffies(*negp ? -*lvalp : *lvalp); + + if (jif > INT_MAX) + return 1; + *valp = (int)jif; } else { int val = *valp; unsigned long lval; diff --git a/kernel/time/Kconfig b/kernel/time/Kconfig index 70f27e89012b..3381f098070f 100644 --- a/kernel/time/Kconfig +++ b/kernel/time/Kconfig @@ -134,6 +134,56 @@ config NO_HZ_FULL_ALL Note the boot CPU will still be kept outside the range to handle the timekeeping duty. +config NO_HZ_FULL_SYSIDLE + bool "Detect full-system idle state for full dynticks system" + depends on NO_HZ_FULL + default n + help + At least one CPU must keep the scheduling-clock tick running for + timekeeping purposes whenever there is a non-idle CPU, where + "non-idle" also includes dynticks CPUs as long as they are + running non-idle tasks. Because the underlying adaptive-tick + support cannot distinguish between all CPUs being idle and + all CPUs each running a single task in dynticks mode, the + underlying support simply ensures that there is always a CPU + handling the scheduling-clock tick, whether or not all CPUs + are idle. This Kconfig option enables scalable detection of + the all-CPUs-idle state, thus allowing the scheduling-clock + tick to be disabled when all CPUs are idle. Note that scalable + detection of the all-CPUs-idle state means that larger systems + will be slower to declare the all-CPUs-idle state. + + Say Y if you would like to help debug all-CPUs-idle detection. + + Say N if you are unsure. + +config NO_HZ_FULL_SYSIDLE_SMALL + int "Number of CPUs above which large-system approach is used" + depends on NO_HZ_FULL_SYSIDLE + range 1 NR_CPUS + default 8 + help + The full-system idle detection mechanism takes a lazy approach + on large systems, as is required to attain decent scalability. + However, on smaller systems, scalability is not anywhere near as + large a concern as is energy efficiency. The sysidle subsystem + therefore uses a fast but non-scalable algorithm for small + systems and a lazier but scalable algorithm for large systems. + This Kconfig parameter defines the number of CPUs in the largest + system that will be considered to be "small". + + The default value will be fine in most cases. Battery-powered + systems that (1) enable NO_HZ_FULL_SYSIDLE, (2) have larger + numbers of CPUs, and (3) are suffering from battery-lifetime + problems due to long sysidle latencies might wish to experiment + with larger values for this Kconfig parameter. On the other + hand, they might be even better served by disabling NO_HZ_FULL + entirely, given that NO_HZ_FULL is intended for HPC and + real-time workloads that at present do not tend to be run on + battery-powered systems. + + Take the default if you are unsure. + config NO_HZ bool "Old Idle dynticks config" depends on !ARCH_USES_GETTIMEOFFSET && GENERIC_CLOCKEVENTS diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c index a326f27d7f09..0b479a6a22bb 100644 --- a/kernel/time/sched_clock.c +++ b/kernel/time/sched_clock.c @@ -121,7 +121,7 @@ void __init setup_sched_clock(u32 (*read)(void), int bits, unsigned long rate) BUG_ON(bits > 32); WARN_ON(!irqs_disabled()); read_sched_clock = read; - sched_clock_mask = (1 << bits) - 1; + sched_clock_mask = (1ULL << bits) - 1; cd.rate = rate; /* calculate the mult/shift to convert counter ticks to ns. */ diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index e80183f4a6c4..e8a1516cc0a3 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -182,7 +182,8 @@ static bool can_stop_full_tick(void) * Don't allow the user to think they can get * full NO_HZ with this machine. */ - WARN_ONCE(1, "NO_HZ FULL will not work with unstable sched clock"); + WARN_ONCE(have_nohz_full_mask, + "NO_HZ FULL will not work with unstable sched clock"); return false; } #endif @@ -343,8 +344,6 @@ static int tick_nohz_init_all(void) void __init tick_nohz_init(void) { - int cpu; - if (!have_nohz_full_mask) { if (tick_nohz_init_all() < 0) return; @@ -827,13 +826,10 @@ void tick_nohz_irq_exit(void) { struct tick_sched *ts = &__get_cpu_var(tick_cpu_sched); - if (ts->inidle) { - /* Cancel the timer because CPU already waken up from the C-states*/ - menu_hrtimer_cancel(); + if (ts->inidle) __tick_nohz_idle_enter(ts); - } else { + else tick_nohz_full_stop_tick(ts); - } } /** @@ -931,8 +927,6 @@ void tick_nohz_idle_exit(void) ts->inidle = 0; - /* Cancel the timer because CPU already waken up from the C-states*/ - menu_hrtimer_cancel(); if (ts->idle_active || ts->tick_stopped) now = ktime_get(); diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index 3bdf28323012..61ed862cdd37 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -265,10 +265,9 @@ static inline void timer_list_header(struct seq_file *m, u64 now) static int timer_list_show(struct seq_file *m, void *v) { struct timer_list_iter *iter = v; - u64 now = ktime_to_ns(ktime_get()); if (iter->cpu == -1 && !iter->second_pass) - timer_list_header(m, now); + timer_list_header(m, iter->now); else if (!iter->second_pass) print_cpu(m, iter->cpu, iter->now); #ifdef CONFIG_GENERIC_CLOCKEVENTS @@ -298,33 +297,41 @@ void sysrq_timer_list_show(void) return; } -static void *timer_list_start(struct seq_file *file, loff_t *offset) +static void *move_iter(struct timer_list_iter *iter, loff_t offset) { - struct timer_list_iter *iter = file->private; - - if (!*offset) { - iter->cpu = -1; - iter->now = ktime_to_ns(ktime_get()); - } else if (iter->cpu >= nr_cpu_ids) { + for (; offset; offset--) { + iter->cpu = cpumask_next(iter->cpu, cpu_online_mask); + if (iter->cpu >= nr_cpu_ids) { #ifdef CONFIG_GENERIC_CLOCKEVENTS - if (!iter->second_pass) { - iter->cpu = -1; - iter->second_pass = true; - } else - return NULL; + if (!iter->second_pass) { + iter->cpu = -1; + iter->second_pass = true; + } else + return NULL; #else - return NULL; + return NULL; #endif + } } return iter; } +static void *timer_list_start(struct seq_file *file, loff_t *offset) +{ + struct timer_list_iter *iter = file->private; + + if (!*offset) + iter->now = ktime_to_ns(ktime_get()); + iter->cpu = -1; + iter->second_pass = false; + return move_iter(iter, *offset); +} + static void *timer_list_next(struct seq_file *file, void *v, loff_t *offset) { struct timer_list_iter *iter = file->private; - iter->cpu = cpumask_next(iter->cpu, cpu_online_mask); ++*offset; - return timer_list_start(file, offset); + return move_iter(iter, 1); } static void timer_list_stop(struct seq_file *seq, void *v) diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 67708f46baae..a6d098c6df3f 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1441,12 +1441,22 @@ ftrace_hash_move(struct ftrace_ops *ops, int enable, * the hashes are freed with call_rcu_sched(). */ static int -ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip) +ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs) { struct ftrace_hash *filter_hash; struct ftrace_hash *notrace_hash; int ret; +#ifdef CONFIG_DYNAMIC_FTRACE_WITH_REGS + /* + * There's a small race when adding ops that the ftrace handler + * that wants regs, may be called without them. We can not + * allow that handler to be called if regs is NULL. + */ + if (regs == NULL && (ops->flags & FTRACE_OPS_FL_SAVE_REGS)) + return 0; +#endif + filter_hash = rcu_dereference_raw_notrace(ops->filter_hash); notrace_hash = rcu_dereference_raw_notrace(ops->notrace_hash); @@ -2159,12 +2169,57 @@ static cycle_t ftrace_update_time; static unsigned long ftrace_update_cnt; unsigned long ftrace_update_tot_cnt; -static int ops_traces_mod(struct ftrace_ops *ops) +static inline int ops_traces_mod(struct ftrace_ops *ops) { - struct ftrace_hash *hash; + /* + * Filter_hash being empty will default to trace module. + * But notrace hash requires a test of individual module functions. + */ + return ftrace_hash_empty(ops->filter_hash) && + ftrace_hash_empty(ops->notrace_hash); +} + +/* + * Check if the current ops references the record. + * + * If the ops traces all functions, then it was already accounted for. + * If the ops does not trace the current record function, skip it. + * If the ops ignores the function via notrace filter, skip it. + */ +static inline bool +ops_references_rec(struct ftrace_ops *ops, struct dyn_ftrace *rec) +{ + /* If ops isn't enabled, ignore it */ + if (!(ops->flags & FTRACE_OPS_FL_ENABLED)) + return 0; + + /* If ops traces all mods, we already accounted for it */ + if (ops_traces_mod(ops)) + return 0; + + /* The function must be in the filter */ + if (!ftrace_hash_empty(ops->filter_hash) && + !ftrace_lookup_ip(ops->filter_hash, rec->ip)) + return 0; + + /* If in notrace hash, we ignore it too */ + if (ftrace_lookup_ip(ops->notrace_hash, rec->ip)) + return 0; - hash = ops->filter_hash; - return ftrace_hash_empty(hash); + return 1; +} + +static int referenced_filters(struct dyn_ftrace *rec) +{ + struct ftrace_ops *ops; + int cnt = 0; + + for (ops = ftrace_ops_list; ops != &ftrace_list_end; ops = ops->next) { + if (ops_references_rec(ops, rec)) + cnt++; + } + + return cnt; } static int ftrace_update_code(struct module *mod) @@ -2173,6 +2228,7 @@ static int ftrace_update_code(struct module *mod) struct dyn_ftrace *p; cycle_t start, stop; unsigned long ref = 0; + bool test = false; int i; /* @@ -2186,9 +2242,12 @@ static int ftrace_update_code(struct module *mod) for (ops = ftrace_ops_list; ops != &ftrace_list_end; ops = ops->next) { - if (ops->flags & FTRACE_OPS_FL_ENABLED && - ops_traces_mod(ops)) - ref++; + if (ops->flags & FTRACE_OPS_FL_ENABLED) { + if (ops_traces_mod(ops)) + ref++; + else + test = true; + } } } @@ -2198,12 +2257,16 @@ static int ftrace_update_code(struct module *mod) for (pg = ftrace_new_pgs; pg; pg = pg->next) { for (i = 0; i < pg->index; i++) { + int cnt = ref; + /* If something went wrong, bail without enabling anything */ if (unlikely(ftrace_disabled)) return -1; p = &pg->records[i]; - p->flags = ref; + if (test) + cnt += referenced_filters(p); + p->flags = cnt; /* * Do the initial record conversion from mcount jump @@ -2223,7 +2286,7 @@ static int ftrace_update_code(struct module *mod) * conversion puts the module to the correct state, thus * passing the ftrace_make_call check. */ - if (ftrace_start_up && ref) { + if (ftrace_start_up && cnt) { int failed = __ftrace_replace_code(p, 1); if (failed) ftrace_bug(failed, p->ip); @@ -3374,6 +3437,12 @@ ftrace_match_addr(struct ftrace_hash *hash, unsigned long ip, int remove) return add_hash_entry(hash, ip); } +static void ftrace_ops_update_code(struct ftrace_ops *ops) +{ + if (ops->flags & FTRACE_OPS_FL_ENABLED && ftrace_enabled) + ftrace_run_update_code(FTRACE_UPDATE_CALLS); +} + static int ftrace_set_hash(struct ftrace_ops *ops, unsigned char *buf, int len, unsigned long ip, int remove, int reset, int enable) @@ -3416,9 +3485,8 @@ ftrace_set_hash(struct ftrace_ops *ops, unsigned char *buf, int len, mutex_lock(&ftrace_lock); ret = ftrace_hash_move(ops, enable, orig_hash, hash); - if (!ret && ops->flags & FTRACE_OPS_FL_ENABLED - && ftrace_enabled) - ftrace_run_update_code(FTRACE_UPDATE_CALLS); + if (!ret) + ftrace_ops_update_code(ops); mutex_unlock(&ftrace_lock); @@ -3645,9 +3713,8 @@ int ftrace_regex_release(struct inode *inode, struct file *file) mutex_lock(&ftrace_lock); ret = ftrace_hash_move(iter->ops, filter_hash, orig_hash, iter->hash); - if (!ret && (iter->ops->flags & FTRACE_OPS_FL_ENABLED) - && ftrace_enabled) - ftrace_run_update_code(FTRACE_UPDATE_CALLS); + if (!ret) + ftrace_ops_update_code(iter->ops); mutex_unlock(&ftrace_lock); } @@ -4218,7 +4285,7 @@ static inline void ftrace_startup_enable(int command) { } # define ftrace_shutdown_sysctl() do { } while (0) static inline int -ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip) +ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs) { return 1; } @@ -4241,7 +4308,7 @@ ftrace_ops_control_func(unsigned long ip, unsigned long parent_ip, do_for_each_ftrace_op(op, ftrace_control_list) { if (!(op->flags & FTRACE_OPS_FL_STUB) && !ftrace_function_local_disabled(op) && - ftrace_ops_test(op, ip)) + ftrace_ops_test(op, ip, regs)) op->func(ip, parent_ip, op, regs); } while_for_each_ftrace_op(op); trace_recursion_clear(TRACE_CONTROL_BIT); @@ -4274,7 +4341,7 @@ __ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip, */ preempt_disable_notrace(); do_for_each_ftrace_op(op, ftrace_ops_list) { - if (ftrace_ops_test(op, ip)) + if (ftrace_ops_test(op, ip, regs)) op->func(ip, parent_ip, op, regs); } while_for_each_ftrace_op(op); preempt_enable_notrace(); diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index e444ff88f0a4..cc2f66f68dc5 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -36,11 +36,11 @@ int ring_buffer_print_entry_header(struct trace_seq *s) { int ret; - ret = trace_seq_printf(s, "# compressed entry header\n"); - ret = trace_seq_printf(s, "\ttype_len : 5 bits\n"); - ret = trace_seq_printf(s, "\ttime_delta : 27 bits\n"); - ret = trace_seq_printf(s, "\tarray : 32 bits\n"); - ret = trace_seq_printf(s, "\n"); + ret = trace_seq_puts(s, "# compressed entry header\n"); + ret = trace_seq_puts(s, "\ttype_len : 5 bits\n"); + ret = trace_seq_puts(s, "\ttime_delta : 27 bits\n"); + ret = trace_seq_puts(s, "\tarray : 32 bits\n"); + ret = trace_seq_putc(s, '\n'); ret = trace_seq_printf(s, "\tpadding : type == %d\n", RINGBUF_TYPE_PADDING); ret = trace_seq_printf(s, "\ttime_extend : type == %d\n", @@ -1066,7 +1066,7 @@ static int rb_check_list(struct ring_buffer_per_cpu *cpu_buffer, } /** - * check_pages - integrity check of buffer pages + * rb_check_pages - integrity check of buffer pages * @cpu_buffer: CPU buffer with pages to test * * As a safety measure we check to make sure the data pages have not @@ -1258,7 +1258,7 @@ static int rb_cpu_notify(struct notifier_block *self, #endif /** - * ring_buffer_alloc - allocate a new ring_buffer + * __ring_buffer_alloc - allocate a new ring_buffer * @size: the size in bytes per cpu that is needed. * @flags: attributes to set for the ring buffer. * @@ -1607,6 +1607,7 @@ static void update_pages_handler(struct work_struct *work) * ring_buffer_resize - resize the ring buffer * @buffer: the buffer to resize. * @size: the new size. + * @cpu_id: the cpu buffer to resize * * Minimum size is 2 * BUF_PAGE_SIZE. * @@ -3956,11 +3957,11 @@ EXPORT_SYMBOL_GPL(ring_buffer_consume); * expected. * * After a sequence of ring_buffer_read_prepare calls, the user is - * expected to make at least one call to ring_buffer_prepare_sync. + * expected to make at least one call to ring_buffer_read_prepare_sync. * Afterwards, ring_buffer_read_start is invoked to get things going * for real. * - * This overall must be paired with ring_buffer_finish. + * This overall must be paired with ring_buffer_read_finish. */ struct ring_buffer_iter * ring_buffer_read_prepare(struct ring_buffer *buffer, int cpu) @@ -4009,7 +4010,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_read_prepare_sync); * an intervening ring_buffer_read_prepare_sync must have been * performed. * - * Must be paired with ring_buffer_finish. + * Must be paired with ring_buffer_read_finish. */ void ring_buffer_read_start(struct ring_buffer_iter *iter) @@ -4031,7 +4032,7 @@ ring_buffer_read_start(struct ring_buffer_iter *iter) EXPORT_SYMBOL_GPL(ring_buffer_read_start); /** - * ring_buffer_finish - finish reading the iterator of the buffer + * ring_buffer_read_finish - finish reading the iterator of the buffer * @iter: The iterator retrieved by ring_buffer_start * * This re-enables the recording to the buffer, and frees the @@ -4346,6 +4347,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_swap_cpu); /** * ring_buffer_alloc_read_page - allocate a page to read from buffer * @buffer: the buffer to allocate for. + * @cpu: the cpu buffer to allocate. * * This function is used in conjunction with ring_buffer_read_page. * When reading a full page from the ring buffer, these functions @@ -4403,7 +4405,7 @@ EXPORT_SYMBOL_GPL(ring_buffer_free_read_page); * to swap with a page in the ring buffer. * * for example: - * rpage = ring_buffer_alloc_read_page(buffer); + * rpage = ring_buffer_alloc_read_page(buffer, cpu); * if (!rpage) * return error; * ret = ring_buffer_read_page(buffer, &rpage, len, cpu, 0); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 0cd500bffd9b..496f94d57698 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -243,20 +243,25 @@ int filter_current_check_discard(struct ring_buffer *buffer, } EXPORT_SYMBOL_GPL(filter_current_check_discard); -cycle_t ftrace_now(int cpu) +cycle_t buffer_ftrace_now(struct trace_buffer *buf, int cpu) { u64 ts; /* Early boot up does not have a buffer yet */ - if (!global_trace.trace_buffer.buffer) + if (!buf->buffer) return trace_clock_local(); - ts = ring_buffer_time_stamp(global_trace.trace_buffer.buffer, cpu); - ring_buffer_normalize_time_stamp(global_trace.trace_buffer.buffer, cpu, &ts); + ts = ring_buffer_time_stamp(buf->buffer, cpu); + ring_buffer_normalize_time_stamp(buf->buffer, cpu, &ts); return ts; } +cycle_t ftrace_now(int cpu) +{ + return buffer_ftrace_now(&global_trace.trace_buffer, cpu); +} + /** * tracing_is_enabled - Show if global_trace has been disabled * @@ -1211,7 +1216,7 @@ void tracing_reset_online_cpus(struct trace_buffer *buf) /* Make sure all commits have finished */ synchronize_sched(); - buf->time_start = ftrace_now(buf->cpu); + buf->time_start = buffer_ftrace_now(buf, buf->cpu); for_each_online_cpu(cpu) ring_buffer_reset_cpu(buffer, cpu); @@ -1219,23 +1224,17 @@ void tracing_reset_online_cpus(struct trace_buffer *buf) ring_buffer_record_enable(buffer); } -void tracing_reset_current(int cpu) -{ - tracing_reset(&global_trace.trace_buffer, cpu); -} - +/* Must have trace_types_lock held */ void tracing_reset_all_online_cpus(void) { struct trace_array *tr; - mutex_lock(&trace_types_lock); list_for_each_entry(tr, &ftrace_trace_arrays, list) { tracing_reset_online_cpus(&tr->trace_buffer); #ifdef CONFIG_TRACER_MAX_TRACE tracing_reset_online_cpus(&tr->max_buffer); #endif } - mutex_unlock(&trace_types_lock); } #define SAVED_CMDLINES 128 @@ -2843,6 +2842,17 @@ static int s_show(struct seq_file *m, void *v) return 0; } +/* + * Should be used after trace_array_get(), trace_types_lock + * ensures that i_cdev was already initialized. + */ +static inline int tracing_get_cpu(struct inode *inode) +{ + if (inode->i_cdev) /* See trace_create_cpu_file() */ + return (long)inode->i_cdev - 1; + return RING_BUFFER_ALL_CPUS; +} + static const struct seq_operations tracer_seq_ops = { .start = s_start, .next = s_next, @@ -2851,9 +2861,9 @@ static const struct seq_operations tracer_seq_ops = { }; static struct trace_iterator * -__tracing_open(struct trace_array *tr, struct trace_cpu *tc, - struct inode *inode, struct file *file, bool snapshot) +__tracing_open(struct inode *inode, struct file *file, bool snapshot) { + struct trace_array *tr = inode->i_private; struct trace_iterator *iter; int cpu; @@ -2894,8 +2904,8 @@ __tracing_open(struct trace_array *tr, struct trace_cpu *tc, iter->trace_buffer = &tr->trace_buffer; iter->snapshot = snapshot; iter->pos = -1; + iter->cpu_file = tracing_get_cpu(inode); mutex_init(&iter->mutex); - iter->cpu_file = tc->cpu; /* Notify the tracer early; before we stop tracing. */ if (iter->trace && iter->trace->open) @@ -2971,45 +2981,22 @@ static int tracing_open_generic_tr(struct inode *inode, struct file *filp) filp->private_data = inode->i_private; return 0; - -} - -static int tracing_open_generic_tc(struct inode *inode, struct file *filp) -{ - struct trace_cpu *tc = inode->i_private; - struct trace_array *tr = tc->tr; - - if (tracing_disabled) - return -ENODEV; - - if (trace_array_get(tr) < 0) - return -ENODEV; - - filp->private_data = inode->i_private; - - return 0; - } static int tracing_release(struct inode *inode, struct file *file) { + struct trace_array *tr = inode->i_private; struct seq_file *m = file->private_data; struct trace_iterator *iter; - struct trace_array *tr; int cpu; - /* Writes do not use seq_file, need to grab tr from inode */ if (!(file->f_mode & FMODE_READ)) { - struct trace_cpu *tc = inode->i_private; - - trace_array_put(tc->tr); + trace_array_put(tr); return 0; } + /* Writes do not use seq_file */ iter = m->private; - tr = iter->tr; - trace_array_put(tr); - mutex_lock(&trace_types_lock); for_each_tracing_cpu(cpu) { @@ -3023,6 +3010,9 @@ static int tracing_release(struct inode *inode, struct file *file) if (!iter->snapshot) /* reenable tracing if it was previously enabled */ tracing_start_tr(tr); + + __trace_array_put(tr); + mutex_unlock(&trace_types_lock); mutex_destroy(&iter->mutex); @@ -3042,15 +3032,6 @@ static int tracing_release_generic_tr(struct inode *inode, struct file *file) return 0; } -static int tracing_release_generic_tc(struct inode *inode, struct file *file) -{ - struct trace_cpu *tc = inode->i_private; - struct trace_array *tr = tc->tr; - - trace_array_put(tr); - return 0; -} - static int tracing_single_release_tr(struct inode *inode, struct file *file) { struct trace_array *tr = inode->i_private; @@ -3062,8 +3043,7 @@ static int tracing_single_release_tr(struct inode *inode, struct file *file) static int tracing_open(struct inode *inode, struct file *file) { - struct trace_cpu *tc = inode->i_private; - struct trace_array *tr = tc->tr; + struct trace_array *tr = inode->i_private; struct trace_iterator *iter; int ret = 0; @@ -3071,16 +3051,17 @@ static int tracing_open(struct inode *inode, struct file *file) return -ENODEV; /* If this file was open for write, then erase contents */ - if ((file->f_mode & FMODE_WRITE) && - (file->f_flags & O_TRUNC)) { - if (tc->cpu == RING_BUFFER_ALL_CPUS) + if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) { + int cpu = tracing_get_cpu(inode); + + if (cpu == RING_BUFFER_ALL_CPUS) tracing_reset_online_cpus(&tr->trace_buffer); else - tracing_reset(&tr->trace_buffer, tc->cpu); + tracing_reset(&tr->trace_buffer, cpu); } if (file->f_mode & FMODE_READ) { - iter = __tracing_open(tr, tc, inode, file, false); + iter = __tracing_open(inode, file, false); if (IS_ERR(iter)) ret = PTR_ERR(iter); else if (trace_flags & TRACE_ITER_LATENCY_FMT) @@ -3447,6 +3428,7 @@ tracing_trace_options_write(struct file *filp, const char __user *ubuf, static int tracing_trace_options_open(struct inode *inode, struct file *file) { struct trace_array *tr = inode->i_private; + int ret; if (tracing_disabled) return -ENODEV; @@ -3454,7 +3436,11 @@ static int tracing_trace_options_open(struct inode *inode, struct file *file) if (trace_array_get(tr) < 0) return -ENODEV; - return single_open(file, tracing_trace_options_show, inode->i_private); + ret = single_open(file, tracing_trace_options_show, inode->i_private); + if (ret < 0) + trace_array_put(tr); + + return ret; } static const struct file_operations tracing_iter_fops = { @@ -3537,14 +3523,14 @@ static const char readme_msg[] = "\n snapshot\t\t- Like 'trace' but shows the content of the static snapshot buffer\n" "\t\t\t Read the contents for more information\n" #endif -#ifdef CONFIG_STACKTRACE +#ifdef CONFIG_STACK_TRACER " stack_trace\t\t- Shows the max stack trace when active\n" " stack_max_size\t- Shows current max stack size that was traced\n" "\t\t\t Write into this file to reset the max size (trigger a new trace)\n" #ifdef CONFIG_DYNAMIC_FTRACE " stack_trace_filter\t- Like set_ftrace_filter but limits what stack_trace traces\n" #endif -#endif /* CONFIG_STACKTRACE */ +#endif /* CONFIG_STACK_TRACER */ ; static ssize_t @@ -3941,8 +3927,7 @@ tracing_max_lat_write(struct file *filp, const char __user *ubuf, static int tracing_open_pipe(struct inode *inode, struct file *filp) { - struct trace_cpu *tc = inode->i_private; - struct trace_array *tr = tc->tr; + struct trace_array *tr = inode->i_private; struct trace_iterator *iter; int ret = 0; @@ -3958,6 +3943,7 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) iter = kzalloc(sizeof(*iter), GFP_KERNEL); if (!iter) { ret = -ENOMEM; + __trace_array_put(tr); goto out; } @@ -3987,9 +3973,9 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) if (trace_clocks[tr->clock_id].in_ns) iter->iter_flags |= TRACE_FILE_TIME_IN_NS; - iter->cpu_file = tc->cpu; - iter->tr = tc->tr; - iter->trace_buffer = &tc->tr->trace_buffer; + iter->tr = tr; + iter->trace_buffer = &tr->trace_buffer; + iter->cpu_file = tracing_get_cpu(inode); mutex_init(&iter->mutex); filp->private_data = iter; @@ -4012,8 +3998,7 @@ fail: static int tracing_release_pipe(struct inode *inode, struct file *file) { struct trace_iterator *iter = file->private_data; - struct trace_cpu *tc = inode->i_private; - struct trace_array *tr = tc->tr; + struct trace_array *tr = inode->i_private; mutex_lock(&trace_types_lock); @@ -4166,6 +4151,7 @@ waitagain: memset(&iter->seq, 0, sizeof(struct trace_iterator) - offsetof(struct trace_iterator, seq)); + cpumask_clear(iter->started); iter->pos = -1; trace_event_read_lock(); @@ -4366,15 +4352,16 @@ static ssize_t tracing_entries_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - struct trace_cpu *tc = filp->private_data; - struct trace_array *tr = tc->tr; + struct inode *inode = file_inode(filp); + struct trace_array *tr = inode->i_private; + int cpu = tracing_get_cpu(inode); char buf[64]; int r = 0; ssize_t ret; mutex_lock(&trace_types_lock); - if (tc->cpu == RING_BUFFER_ALL_CPUS) { + if (cpu == RING_BUFFER_ALL_CPUS) { int cpu, buf_size_same; unsigned long size; @@ -4401,7 +4388,7 @@ tracing_entries_read(struct file *filp, char __user *ubuf, } else r = sprintf(buf, "X\n"); } else - r = sprintf(buf, "%lu\n", per_cpu_ptr(tr->trace_buffer.data, tc->cpu)->entries >> 10); + r = sprintf(buf, "%lu\n", per_cpu_ptr(tr->trace_buffer.data, cpu)->entries >> 10); mutex_unlock(&trace_types_lock); @@ -4413,7 +4400,8 @@ static ssize_t tracing_entries_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - struct trace_cpu *tc = filp->private_data; + struct inode *inode = file_inode(filp); + struct trace_array *tr = inode->i_private; unsigned long val; int ret; @@ -4427,8 +4415,7 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, /* value is in KB */ val <<= 10; - - ret = tracing_resize_ring_buffer(tc->tr, val, tc->cpu); + ret = tracing_resize_ring_buffer(tr, val, tracing_get_cpu(inode)); if (ret < 0) return ret; @@ -4482,7 +4469,7 @@ tracing_free_buffer_release(struct inode *inode, struct file *filp) /* disable tracing ? */ if (trace_flags & TRACE_ITER_STOP_ON_FREE) - tracing_off(); + tracer_tracing_off(tr); /* resize the ring buffer to 0 */ tracing_resize_ring_buffer(tr, 0, RING_BUFFER_ALL_CPUS); @@ -4647,12 +4634,12 @@ static ssize_t tracing_clock_write(struct file *filp, const char __user *ubuf, * New clock may not be consistent with the previous clock. * Reset the buffer so that it doesn't have incomparable timestamps. */ - tracing_reset_online_cpus(&global_trace.trace_buffer); + tracing_reset_online_cpus(&tr->trace_buffer); #ifdef CONFIG_TRACER_MAX_TRACE if (tr->flags & TRACE_ARRAY_FL_GLOBAL && tr->max_buffer.buffer) ring_buffer_set_clock(tr->max_buffer.buffer, trace_clocks[i].func); - tracing_reset_online_cpus(&global_trace.max_buffer); + tracing_reset_online_cpus(&tr->max_buffer); #endif mutex_unlock(&trace_types_lock); @@ -4689,8 +4676,7 @@ struct ftrace_buffer_info { #ifdef CONFIG_TRACER_SNAPSHOT static int tracing_snapshot_open(struct inode *inode, struct file *file) { - struct trace_cpu *tc = inode->i_private; - struct trace_array *tr = tc->tr; + struct trace_array *tr = inode->i_private; struct trace_iterator *iter; struct seq_file *m; int ret = 0; @@ -4699,26 +4685,29 @@ static int tracing_snapshot_open(struct inode *inode, struct file *file) return -ENODEV; if (file->f_mode & FMODE_READ) { - iter = __tracing_open(tr, tc, inode, file, true); + iter = __tracing_open(inode, file, true); if (IS_ERR(iter)) ret = PTR_ERR(iter); } else { /* Writes still need the seq_file to hold the private data */ + ret = -ENOMEM; m = kzalloc(sizeof(*m), GFP_KERNEL); if (!m) - return -ENOMEM; + goto out; iter = kzalloc(sizeof(*iter), GFP_KERNEL); if (!iter) { kfree(m); - return -ENOMEM; + goto out; } + ret = 0; + iter->tr = tr; - iter->trace_buffer = &tc->tr->max_buffer; - iter->cpu_file = tc->cpu; + iter->trace_buffer = &tr->max_buffer; + iter->cpu_file = tracing_get_cpu(inode); m->private = iter; file->private_data = m; } - +out: if (ret < 0) trace_array_put(tr); @@ -4873,11 +4862,11 @@ static const struct file_operations tracing_pipe_fops = { }; static const struct file_operations tracing_entries_fops = { - .open = tracing_open_generic_tc, + .open = tracing_open_generic_tr, .read = tracing_entries_read, .write = tracing_entries_write, .llseek = generic_file_llseek, - .release = tracing_release_generic_tc, + .release = tracing_release_generic_tr, }; static const struct file_operations tracing_total_entries_fops = { @@ -4929,8 +4918,7 @@ static const struct file_operations snapshot_raw_fops = { static int tracing_buffers_open(struct inode *inode, struct file *filp) { - struct trace_cpu *tc = inode->i_private; - struct trace_array *tr = tc->tr; + struct trace_array *tr = inode->i_private; struct ftrace_buffer_info *info; int ret; @@ -4948,10 +4936,8 @@ static int tracing_buffers_open(struct inode *inode, struct file *filp) mutex_lock(&trace_types_lock); - tr->ref++; - info->iter.tr = tr; - info->iter.cpu_file = tc->cpu; + info->iter.cpu_file = tracing_get_cpu(inode); info->iter.trace = tr->current_trace; info->iter.trace_buffer = &tr->trace_buffer; info->spare = NULL; @@ -5268,14 +5254,14 @@ static ssize_t tracing_stats_read(struct file *filp, char __user *ubuf, size_t count, loff_t *ppos) { - struct trace_cpu *tc = filp->private_data; - struct trace_array *tr = tc->tr; + struct inode *inode = file_inode(filp); + struct trace_array *tr = inode->i_private; struct trace_buffer *trace_buf = &tr->trace_buffer; + int cpu = tracing_get_cpu(inode); struct trace_seq *s; unsigned long cnt; unsigned long long t; unsigned long usec_rem; - int cpu = tc->cpu; s = kmalloc(sizeof(*s), GFP_KERNEL); if (!s) @@ -5328,9 +5314,10 @@ tracing_stats_read(struct file *filp, char __user *ubuf, } static const struct file_operations tracing_stats_fops = { - .open = tracing_open_generic, + .open = tracing_open_generic_tr, .read = tracing_stats_read, .llseek = generic_file_llseek, + .release = tracing_release_generic_tr, }; #ifdef CONFIG_DYNAMIC_FTRACE @@ -5519,10 +5506,20 @@ static struct dentry *tracing_dentry_percpu(struct trace_array *tr, int cpu) return tr->percpu_dir; } +static struct dentry * +trace_create_cpu_file(const char *name, umode_t mode, struct dentry *parent, + void *data, long cpu, const struct file_operations *fops) +{ + struct dentry *ret = trace_create_file(name, mode, parent, data, fops); + + if (ret) /* See tracing_get_cpu() */ + ret->d_inode->i_cdev = (void *)(cpu + 1); + return ret; +} + static void tracing_init_debugfs_percpu(struct trace_array *tr, long cpu) { - struct trace_array_cpu *data = per_cpu_ptr(tr->trace_buffer.data, cpu); struct dentry *d_percpu = tracing_dentry_percpu(tr, cpu); struct dentry *d_cpu; char cpu_dir[30]; /* 30 characters should be more than enough */ @@ -5538,28 +5535,28 @@ tracing_init_debugfs_percpu(struct trace_array *tr, long cpu) } /* per cpu trace_pipe */ - trace_create_file("trace_pipe", 0444, d_cpu, - (void *)&data->trace_cpu, &tracing_pipe_fops); + trace_create_cpu_file("trace_pipe", 0444, d_cpu, + tr, cpu, &tracing_pipe_fops); /* per cpu trace */ - trace_create_file("trace", 0644, d_cpu, - (void *)&data->trace_cpu, &tracing_fops); + trace_create_cpu_file("trace", 0644, d_cpu, + tr, cpu, &tracing_fops); - trace_create_file("trace_pipe_raw", 0444, d_cpu, - (void *)&data->trace_cpu, &tracing_buffers_fops); + trace_create_cpu_file("trace_pipe_raw", 0444, d_cpu, + tr, cpu, &tracing_buffers_fops); - trace_create_file("stats", 0444, d_cpu, - (void *)&data->trace_cpu, &tracing_stats_fops); + trace_create_cpu_file("stats", 0444, d_cpu, + tr, cpu, &tracing_stats_fops); - trace_create_file("buffer_size_kb", 0444, d_cpu, - (void *)&data->trace_cpu, &tracing_entries_fops); + trace_create_cpu_file("buffer_size_kb", 0444, d_cpu, + tr, cpu, &tracing_entries_fops); #ifdef CONFIG_TRACER_SNAPSHOT - trace_create_file("snapshot", 0644, d_cpu, - (void *)&data->trace_cpu, &snapshot_fops); + trace_create_cpu_file("snapshot", 0644, d_cpu, + tr, cpu, &snapshot_fops); - trace_create_file("snapshot_raw", 0444, d_cpu, - (void *)&data->trace_cpu, &snapshot_raw_fops); + trace_create_cpu_file("snapshot_raw", 0444, d_cpu, + tr, cpu, &snapshot_raw_fops); #endif } @@ -5868,17 +5865,6 @@ struct dentry *trace_instance_dir; static void init_tracer_debugfs(struct trace_array *tr, struct dentry *d_tracer); -static void init_trace_buffers(struct trace_array *tr, struct trace_buffer *buf) -{ - int cpu; - - for_each_tracing_cpu(cpu) { - memset(per_cpu_ptr(buf->data, cpu), 0, sizeof(struct trace_array_cpu)); - per_cpu_ptr(buf->data, cpu)->trace_cpu.cpu = cpu; - per_cpu_ptr(buf->data, cpu)->trace_cpu.tr = tr; - } -} - static int allocate_trace_buffer(struct trace_array *tr, struct trace_buffer *buf, int size) { @@ -5896,8 +5882,6 @@ allocate_trace_buffer(struct trace_array *tr, struct trace_buffer *buf, int size return -ENOMEM; } - init_trace_buffers(tr, buf); - /* Allocate the first page for all buffers */ set_buffer_entries(&tr->trace_buffer, ring_buffer_size(tr->trace_buffer.buffer, 0)); @@ -5964,17 +5948,15 @@ static int new_instance_create(const char *name) if (allocate_trace_buffers(tr, trace_buf_size) < 0) goto out_free_tr; - /* Holder for file callbacks */ - tr->trace_cpu.cpu = RING_BUFFER_ALL_CPUS; - tr->trace_cpu.tr = tr; - tr->dir = debugfs_create_dir(name, trace_instance_dir); if (!tr->dir) goto out_free_tr; ret = event_trace_add_tracer(tr->dir, tr); - if (ret) + if (ret) { + debugfs_remove_recursive(tr->dir); goto out_free_tr; + } init_tracer_debugfs(tr, tr->dir); @@ -6120,13 +6102,13 @@ init_tracer_debugfs(struct trace_array *tr, struct dentry *d_tracer) tr, &tracing_iter_fops); trace_create_file("trace", 0644, d_tracer, - (void *)&tr->trace_cpu, &tracing_fops); + tr, &tracing_fops); trace_create_file("trace_pipe", 0444, d_tracer, - (void *)&tr->trace_cpu, &tracing_pipe_fops); + tr, &tracing_pipe_fops); trace_create_file("buffer_size_kb", 0644, d_tracer, - (void *)&tr->trace_cpu, &tracing_entries_fops); + tr, &tracing_entries_fops); trace_create_file("buffer_total_size_kb", 0444, d_tracer, tr, &tracing_total_entries_fops); @@ -6141,11 +6123,11 @@ init_tracer_debugfs(struct trace_array *tr, struct dentry *d_tracer) &trace_clock_fops); trace_create_file("tracing_on", 0644, d_tracer, - tr, &rb_simple_fops); + tr, &rb_simple_fops); #ifdef CONFIG_TRACER_SNAPSHOT trace_create_file("snapshot", 0644, d_tracer, - (void *)&tr->trace_cpu, &snapshot_fops); + tr, &snapshot_fops); #endif for_each_tracing_cpu(cpu) @@ -6439,10 +6421,6 @@ __init static int tracer_alloc_buffers(void) global_trace.flags = TRACE_ARRAY_FL_GLOBAL; - /* Holder for file callbacks */ - global_trace.trace_cpu.cpu = RING_BUFFER_ALL_CPUS; - global_trace.trace_cpu.tr = &global_trace; - INIT_LIST_HEAD(&global_trace.systems); INIT_LIST_HEAD(&global_trace.events); list_add(&global_trace.list, &ftrace_trace_arrays); diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 4a4f6e1828b6..fe39acd4c1aa 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -130,19 +130,12 @@ enum trace_flag_type { struct trace_array; -struct trace_cpu { - struct trace_array *tr; - struct dentry *dir; - int cpu; -}; - /* * The CPU trace array - it consists of thousands of trace entries * plus some other descriptor data: (for example which task started * the trace, etc.) */ struct trace_array_cpu { - struct trace_cpu trace_cpu; atomic_t disabled; void *buffer_page; /* ring buffer spare */ @@ -196,7 +189,6 @@ struct trace_array { bool allocated_snapshot; #endif int buffer_disabled; - struct trace_cpu trace_cpu; /* place holder */ #ifdef CONFIG_FTRACE_SYSCALLS int sys_refcount_enter; int sys_refcount_exit; @@ -214,7 +206,6 @@ struct trace_array { struct dentry *event_dir; struct list_head systems; struct list_head events; - struct task_struct *waiter; int ref; }; @@ -680,6 +671,15 @@ extern int trace_selftest_startup_sched_switch(struct tracer *trace, struct trace_array *tr); extern int trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr); +/* + * Tracer data references selftest functions that only occur + * on boot up. These can be __init functions. Thus, when selftests + * are enabled, then the tracers need to reference __init functions. + */ +#define __tracer_data __refdata +#else +/* Tracers are seldom changed. Optimize when selftests are disabled. */ +#define __tracer_data __read_mostly #endif /* CONFIG_FTRACE_STARTUP_TEST */ extern void *head_page(struct trace_array_cpu *data); @@ -1022,6 +1022,9 @@ extern struct list_head ftrace_events; extern const char *__start___trace_bprintk_fmt[]; extern const char *__stop___trace_bprintk_fmt[]; +extern const char *__start___tracepoint_str[]; +extern const char *__stop___tracepoint_str[]; + void trace_printk_init_buffers(void); void trace_printk_start_comm(void); int trace_keep_overwrite(struct tracer *tracer, u32 mask, int set); diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index 84b1e045faba..80c36bcf66e8 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -236,6 +236,10 @@ __kprobes void *perf_trace_buf_prepare(int size, unsigned short type, BUILD_BUG_ON(PERF_MAX_TRACE_SIZE % sizeof(unsigned long)); + if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, + "perf buffer not large enough")) + return NULL; + pc = preempt_count(); *rctxp = perf_swevent_get_recursion_context(); @@ -266,6 +270,10 @@ perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip, struct pt_regs regs; int rctx; + head = this_cpu_ptr(event_function.perf_events); + if (hlist_empty(head)) + return; + #define ENTRY_SIZE (ALIGN(sizeof(struct ftrace_entry) + sizeof(u32), \ sizeof(u64)) - sizeof(u32)) @@ -279,8 +287,6 @@ perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip, entry->ip = ip; entry->parent_ip = parent_ip; - - head = this_cpu_ptr(event_function.perf_events); perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, 0, 1, ®s, head, NULL); diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 7d854290bf81..29a7ebcfb426 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -409,33 +409,42 @@ static void put_system(struct ftrace_subsystem_dir *dir) mutex_unlock(&event_mutex); } -/* - * Open and update trace_array ref count. - * Must have the current trace_array passed to it. - */ -static int tracing_open_generic_file(struct inode *inode, struct file *filp) +static void remove_subsystem(struct ftrace_subsystem_dir *dir) { - struct ftrace_event_file *file = inode->i_private; - struct trace_array *tr = file->tr; - int ret; + if (!dir) + return; - if (trace_array_get(tr) < 0) - return -ENODEV; + if (!--dir->nr_events) { + debugfs_remove_recursive(dir->entry); + list_del(&dir->list); + __put_system_dir(dir); + } +} - ret = tracing_open_generic(inode, filp); - if (ret < 0) - trace_array_put(tr); - return ret; +static void *event_file_data(struct file *filp) +{ + return ACCESS_ONCE(file_inode(filp)->i_private); } -static int tracing_release_generic_file(struct inode *inode, struct file *filp) +static void remove_event_file_dir(struct ftrace_event_file *file) { - struct ftrace_event_file *file = inode->i_private; - struct trace_array *tr = file->tr; + struct dentry *dir = file->dir; + struct dentry *child; - trace_array_put(tr); + if (dir) { + spin_lock(&dir->d_lock); /* probably unneeded */ + list_for_each_entry(child, &dir->d_subdirs, d_u.d_child) { + if (child->d_inode) /* probably unneeded */ + child->d_inode->i_private = NULL; + } + spin_unlock(&dir->d_lock); - return 0; + debugfs_remove_recursive(dir); + } + + list_del(&file->list); + remove_subsystem(file->system); + kmem_cache_free(file_cachep, file); } /* @@ -679,15 +688,25 @@ static ssize_t event_enable_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - struct ftrace_event_file *file = filp->private_data; + struct ftrace_event_file *file; + unsigned long flags; char buf[4] = "0"; - if (file->flags & FTRACE_EVENT_FL_ENABLED && - !(file->flags & FTRACE_EVENT_FL_SOFT_DISABLED)) + mutex_lock(&event_mutex); + file = event_file_data(filp); + if (likely(file)) + flags = file->flags; + mutex_unlock(&event_mutex); + + if (!file) + return -ENODEV; + + if (flags & FTRACE_EVENT_FL_ENABLED && + !(flags & FTRACE_EVENT_FL_SOFT_DISABLED)) strcpy(buf, "1"); - if (file->flags & FTRACE_EVENT_FL_SOFT_DISABLED || - file->flags & FTRACE_EVENT_FL_SOFT_MODE) + if (flags & FTRACE_EVENT_FL_SOFT_DISABLED || + flags & FTRACE_EVENT_FL_SOFT_MODE) strcat(buf, "*"); strcat(buf, "\n"); @@ -699,13 +718,10 @@ static ssize_t event_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - struct ftrace_event_file *file = filp->private_data; + struct ftrace_event_file *file; unsigned long val; int ret; - if (!file) - return -EINVAL; - ret = kstrtoul_from_user(ubuf, cnt, 10, &val); if (ret) return ret; @@ -717,8 +733,11 @@ event_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, switch (val) { case 0: case 1: + ret = -ENODEV; mutex_lock(&event_mutex); - ret = ftrace_event_enable_disable(file, val); + file = event_file_data(filp); + if (likely(file)) + ret = ftrace_event_enable_disable(file, val); mutex_unlock(&event_mutex); break; @@ -825,65 +844,39 @@ enum { static void *f_next(struct seq_file *m, void *v, loff_t *pos) { - struct ftrace_event_call *call = m->private; - struct ftrace_event_field *field; + struct ftrace_event_call *call = event_file_data(m->private); struct list_head *common_head = &ftrace_common_fields; struct list_head *head = trace_get_fields(call); + struct list_head *node = v; (*pos)++; switch ((unsigned long)v) { case FORMAT_HEADER: - if (unlikely(list_empty(common_head))) - return NULL; - - field = list_entry(common_head->prev, - struct ftrace_event_field, link); - return field; + node = common_head; + break; case FORMAT_FIELD_SEPERATOR: - if (unlikely(list_empty(head))) - return NULL; - - field = list_entry(head->prev, struct ftrace_event_field, link); - return field; + node = head; + break; case FORMAT_PRINTFMT: /* all done */ return NULL; } - field = v; - if (field->link.prev == common_head) + node = node->prev; + if (node == common_head) return (void *)FORMAT_FIELD_SEPERATOR; - else if (field->link.prev == head) + else if (node == head) return (void *)FORMAT_PRINTFMT; - - field = list_entry(field->link.prev, struct ftrace_event_field, link); - - return field; -} - -static void *f_start(struct seq_file *m, loff_t *pos) -{ - loff_t l = 0; - void *p; - - /* Start by showing the header */ - if (!*pos) - return (void *)FORMAT_HEADER; - - p = (void *)FORMAT_HEADER; - do { - p = f_next(m, p, &l); - } while (p && l < *pos); - - return p; + else + return node; } static int f_show(struct seq_file *m, void *v) { - struct ftrace_event_call *call = m->private; + struct ftrace_event_call *call = event_file_data(m->private); struct ftrace_event_field *field; const char *array_descriptor; @@ -904,8 +897,7 @@ static int f_show(struct seq_file *m, void *v) return 0; } - field = v; - + field = list_entry(v, struct ftrace_event_field, link); /* * Smartly shows the array type(except dynamic array). * Normal: @@ -932,8 +924,25 @@ static int f_show(struct seq_file *m, void *v) return 0; } +static void *f_start(struct seq_file *m, loff_t *pos) +{ + void *p = (void *)FORMAT_HEADER; + loff_t l = 0; + + /* ->stop() is called even if ->start() fails */ + mutex_lock(&event_mutex); + if (!event_file_data(m->private)) + return ERR_PTR(-ENODEV); + + while (l < *pos && p) + p = f_next(m, p, &l); + + return p; +} + static void f_stop(struct seq_file *m, void *p) { + mutex_unlock(&event_mutex); } static const struct seq_operations trace_format_seq_ops = { @@ -945,7 +954,6 @@ static const struct seq_operations trace_format_seq_ops = { static int trace_format_open(struct inode *inode, struct file *file) { - struct ftrace_event_call *call = inode->i_private; struct seq_file *m; int ret; @@ -954,7 +962,7 @@ static int trace_format_open(struct inode *inode, struct file *file) return ret; m = file->private_data; - m->private = call; + m->private = file; return 0; } @@ -962,45 +970,47 @@ static int trace_format_open(struct inode *inode, struct file *file) static ssize_t event_id_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - struct ftrace_event_call *call = filp->private_data; - struct trace_seq *s; - int r; + int id = (long)event_file_data(filp); + char buf[32]; + int len; if (*ppos) return 0; - s = kmalloc(sizeof(*s), GFP_KERNEL); - if (!s) - return -ENOMEM; + if (unlikely(!id)) + return -ENODEV; - trace_seq_init(s); - trace_seq_printf(s, "%d\n", call->event.type); + len = sprintf(buf, "%d\n", id); - r = simple_read_from_buffer(ubuf, cnt, ppos, - s->buffer, s->len); - kfree(s); - return r; + return simple_read_from_buffer(ubuf, cnt, ppos, buf, len); } static ssize_t event_filter_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - struct ftrace_event_call *call = filp->private_data; + struct ftrace_event_call *call; struct trace_seq *s; - int r; + int r = -ENODEV; if (*ppos) return 0; s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) return -ENOMEM; trace_seq_init(s); - print_event_filter(call, s); - r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len); + mutex_lock(&event_mutex); + call = event_file_data(filp); + if (call) + print_event_filter(call, s); + mutex_unlock(&event_mutex); + + if (call) + r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len); kfree(s); @@ -1011,9 +1021,9 @@ static ssize_t event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - struct ftrace_event_call *call = filp->private_data; + struct ftrace_event_call *call; char *buf; - int err; + int err = -ENODEV; if (cnt >= PAGE_SIZE) return -EINVAL; @@ -1028,7 +1038,12 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, } buf[cnt] = '\0'; - err = apply_event_filter(call, buf); + mutex_lock(&event_mutex); + call = event_file_data(filp); + if (call) + err = apply_event_filter(call, buf); + mutex_unlock(&event_mutex); + free_page((unsigned long) buf); if (err < 0) return err; @@ -1218,6 +1233,7 @@ show_header(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) static int ftrace_event_avail_open(struct inode *inode, struct file *file); static int ftrace_event_set_open(struct inode *inode, struct file *file); +static int ftrace_event_release(struct inode *inode, struct file *file); static const struct seq_operations show_event_seq_ops = { .start = t_start, @@ -1245,14 +1261,13 @@ static const struct file_operations ftrace_set_event_fops = { .read = seq_read, .write = ftrace_event_write, .llseek = seq_lseek, - .release = seq_release, + .release = ftrace_event_release, }; static const struct file_operations ftrace_enable_fops = { - .open = tracing_open_generic_file, + .open = tracing_open_generic, .read = event_enable_read, .write = event_enable_write, - .release = tracing_release_generic_file, .llseek = default_llseek, }; @@ -1264,7 +1279,6 @@ static const struct file_operations ftrace_event_format_fops = { }; static const struct file_operations ftrace_event_id_fops = { - .open = tracing_open_generic, .read = event_id_read, .llseek = default_llseek, }; @@ -1323,6 +1337,15 @@ ftrace_event_open(struct inode *inode, struct file *file, return ret; } +static int ftrace_event_release(struct inode *inode, struct file *file) +{ + struct trace_array *tr = inode->i_private; + + trace_array_put(tr); + + return seq_release(inode, file); +} + static int ftrace_event_avail_open(struct inode *inode, struct file *file) { @@ -1336,12 +1359,19 @@ ftrace_event_set_open(struct inode *inode, struct file *file) { const struct seq_operations *seq_ops = &show_set_event_seq_ops; struct trace_array *tr = inode->i_private; + int ret; + + if (trace_array_get(tr) < 0) + return -ENODEV; if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) ftrace_clear_events(tr); - return ftrace_event_open(inode, file, seq_ops); + ret = ftrace_event_open(inode, file, seq_ops); + if (ret < 0) + trace_array_put(tr); + return ret; } static struct event_subsystem * @@ -1496,8 +1526,8 @@ event_create_dir(struct dentry *parent, #ifdef CONFIG_PERF_EVENTS if (call->event.type && call->class->reg) - trace_create_file("id", 0444, file->dir, call, - id); + trace_create_file("id", 0444, file->dir, + (void *)(long)call->event.type, id); #endif /* @@ -1522,33 +1552,16 @@ event_create_dir(struct dentry *parent, return 0; } -static void remove_subsystem(struct ftrace_subsystem_dir *dir) -{ - if (!dir) - return; - - if (!--dir->nr_events) { - debugfs_remove_recursive(dir->entry); - list_del(&dir->list); - __put_system_dir(dir); - } -} - static void remove_event_from_tracers(struct ftrace_event_call *call) { struct ftrace_event_file *file; struct trace_array *tr; do_for_each_event_file_safe(tr, file) { - if (file->event_call != call) continue; - list_del(&file->list); - debugfs_remove_recursive(file->dir); - remove_subsystem(file->system); - kmem_cache_free(file_cachep, file); - + remove_event_file_dir(file); /* * The do_for_each_event_file_safe() is * a double loop. After finding the call for this @@ -1700,16 +1713,53 @@ static void __trace_remove_event_call(struct ftrace_event_call *call) destroy_preds(call); } +static int probe_remove_event_call(struct ftrace_event_call *call) +{ + struct trace_array *tr; + struct ftrace_event_file *file; + +#ifdef CONFIG_PERF_EVENTS + if (call->perf_refcount) + return -EBUSY; +#endif + do_for_each_event_file(tr, file) { + if (file->event_call != call) + continue; + /* + * We can't rely on ftrace_event_enable_disable(enable => 0) + * we are going to do, FTRACE_EVENT_FL_SOFT_MODE can suppress + * TRACE_REG_UNREGISTER. + */ + if (file->flags & FTRACE_EVENT_FL_ENABLED) + return -EBUSY; + /* + * The do_for_each_event_file_safe() is + * a double loop. After finding the call for this + * trace_array, we use break to jump to the next + * trace_array. + */ + break; + } while_for_each_event_file(); + + __trace_remove_event_call(call); + + return 0; +} + /* Remove an event_call */ -void trace_remove_event_call(struct ftrace_event_call *call) +int trace_remove_event_call(struct ftrace_event_call *call) { + int ret; + mutex_lock(&trace_types_lock); mutex_lock(&event_mutex); down_write(&trace_event_sem); - __trace_remove_event_call(call); + ret = probe_remove_event_call(call); up_write(&trace_event_sem); mutex_unlock(&event_mutex); mutex_unlock(&trace_types_lock); + + return ret; } #define for_each_event(event, start, end) \ @@ -2278,12 +2328,8 @@ __trace_remove_event_dirs(struct trace_array *tr) { struct ftrace_event_file *file, *next; - list_for_each_entry_safe(file, next, &tr->events, list) { - list_del(&file->list); - debugfs_remove_recursive(file->dir); - remove_subsystem(file->system); - kmem_cache_free(file_cachep, file); - } + list_for_each_entry_safe(file, next, &tr->events, list) + remove_event_file_dir(file); } static void diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 0d883dc057d6..97daa8cf958d 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -637,17 +637,15 @@ static void append_filter_err(struct filter_parse_state *ps, free_page((unsigned long) buf); } +/* caller must hold event_mutex */ void print_event_filter(struct ftrace_event_call *call, struct trace_seq *s) { - struct event_filter *filter; + struct event_filter *filter = call->filter; - mutex_lock(&event_mutex); - filter = call->filter; if (filter && filter->filter_string) trace_seq_printf(s, "%s\n", filter->filter_string); else - trace_seq_printf(s, "none\n"); - mutex_unlock(&event_mutex); + trace_seq_puts(s, "none\n"); } void print_subsystem_event_filter(struct event_subsystem *system, @@ -660,7 +658,7 @@ void print_subsystem_event_filter(struct event_subsystem *system, if (filter && filter->filter_string) trace_seq_printf(s, "%s\n", filter->filter_string); else - trace_seq_printf(s, DEFAULT_SYS_FILTER_MESSAGE "\n"); + trace_seq_puts(s, DEFAULT_SYS_FILTER_MESSAGE "\n"); mutex_unlock(&event_mutex); } @@ -1841,23 +1839,22 @@ static int create_system_filter(struct event_subsystem *system, return err; } +/* caller must hold event_mutex */ int apply_event_filter(struct ftrace_event_call *call, char *filter_string) { struct event_filter *filter; - int err = 0; - - mutex_lock(&event_mutex); + int err; if (!strcmp(strstrip(filter_string), "0")) { filter_disable(call); filter = call->filter; if (!filter) - goto out_unlock; + return 0; RCU_INIT_POINTER(call->filter, NULL); /* Make sure the filter is not being used */ synchronize_sched(); __free_filter(filter); - goto out_unlock; + return 0; } err = create_filter(call, filter_string, true, &filter); @@ -1884,8 +1881,6 @@ int apply_event_filter(struct ftrace_event_call *call, char *filter_string) __free_filter(tmp); } } -out_unlock: - mutex_unlock(&event_mutex); return err; } diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index b863f93b30f3..38fe1483c508 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -199,7 +199,7 @@ static int func_set_flag(u32 old_flags, u32 bit, int set) return 0; } -static struct tracer function_trace __read_mostly = +static struct tracer function_trace __tracer_data = { .name = "function", .init = function_trace_init, diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 8388bc99f2ee..b5c09242683d 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -446,7 +446,7 @@ print_graph_proc(struct trace_seq *s, pid_t pid) /* First spaces to align center */ for (i = 0; i < spaces / 2; i++) { - ret = trace_seq_printf(s, " "); + ret = trace_seq_putc(s, ' '); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } @@ -457,7 +457,7 @@ print_graph_proc(struct trace_seq *s, pid_t pid) /* Last spaces to align center */ for (i = 0; i < spaces - (spaces / 2); i++) { - ret = trace_seq_printf(s, " "); + ret = trace_seq_putc(s, ' '); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } @@ -503,7 +503,7 @@ verif_pid(struct trace_seq *s, pid_t pid, int cpu, struct fgraph_data *data) ------------------------------------------ */ - ret = trace_seq_printf(s, + ret = trace_seq_puts(s, " ------------------------------------------\n"); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -516,7 +516,7 @@ verif_pid(struct trace_seq *s, pid_t pid, int cpu, struct fgraph_data *data) if (ret == TRACE_TYPE_PARTIAL_LINE) return TRACE_TYPE_PARTIAL_LINE; - ret = trace_seq_printf(s, " => "); + ret = trace_seq_puts(s, " => "); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -524,7 +524,7 @@ verif_pid(struct trace_seq *s, pid_t pid, int cpu, struct fgraph_data *data) if (ret == TRACE_TYPE_PARTIAL_LINE) return TRACE_TYPE_PARTIAL_LINE; - ret = trace_seq_printf(s, + ret = trace_seq_puts(s, "\n ------------------------------------------\n\n"); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -645,7 +645,7 @@ print_graph_irq(struct trace_iterator *iter, unsigned long addr, ret = print_graph_proc(s, pid); if (ret == TRACE_TYPE_PARTIAL_LINE) return TRACE_TYPE_PARTIAL_LINE; - ret = trace_seq_printf(s, " | "); + ret = trace_seq_puts(s, " | "); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } @@ -657,9 +657,9 @@ print_graph_irq(struct trace_iterator *iter, unsigned long addr, return ret; if (type == TRACE_GRAPH_ENT) - ret = trace_seq_printf(s, "==========>"); + ret = trace_seq_puts(s, "==========>"); else - ret = trace_seq_printf(s, "<=========="); + ret = trace_seq_puts(s, "<=========="); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -668,7 +668,7 @@ print_graph_irq(struct trace_iterator *iter, unsigned long addr, if (ret != TRACE_TYPE_HANDLED) return ret; - ret = trace_seq_printf(s, "\n"); + ret = trace_seq_putc(s, '\n'); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -705,13 +705,13 @@ trace_print_graph_duration(unsigned long long duration, struct trace_seq *s) len += strlen(nsecs_str); } - ret = trace_seq_printf(s, " us "); + ret = trace_seq_puts(s, " us "); if (!ret) return TRACE_TYPE_PARTIAL_LINE; /* Print remaining spaces to fit the row's width */ for (i = len; i < 7; i++) { - ret = trace_seq_printf(s, " "); + ret = trace_seq_putc(s, ' '); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } @@ -731,13 +731,13 @@ print_graph_duration(unsigned long long duration, struct trace_seq *s, /* No real adata, just filling the column with spaces */ switch (duration) { case DURATION_FILL_FULL: - ret = trace_seq_printf(s, " | "); + ret = trace_seq_puts(s, " | "); return ret ? TRACE_TYPE_HANDLED : TRACE_TYPE_PARTIAL_LINE; case DURATION_FILL_START: - ret = trace_seq_printf(s, " "); + ret = trace_seq_puts(s, " "); return ret ? TRACE_TYPE_HANDLED : TRACE_TYPE_PARTIAL_LINE; case DURATION_FILL_END: - ret = trace_seq_printf(s, " |"); + ret = trace_seq_puts(s, " |"); return ret ? TRACE_TYPE_HANDLED : TRACE_TYPE_PARTIAL_LINE; } @@ -745,10 +745,10 @@ print_graph_duration(unsigned long long duration, struct trace_seq *s, if (flags & TRACE_GRAPH_PRINT_OVERHEAD) { /* Duration exceeded 100 msecs */ if (duration > 100000ULL) - ret = trace_seq_printf(s, "! "); + ret = trace_seq_puts(s, "! "); /* Duration exceeded 10 msecs */ else if (duration > 10000ULL) - ret = trace_seq_printf(s, "+ "); + ret = trace_seq_puts(s, "+ "); } /* @@ -757,7 +757,7 @@ print_graph_duration(unsigned long long duration, struct trace_seq *s, * to fill out the space. */ if (ret == -1) - ret = trace_seq_printf(s, " "); + ret = trace_seq_puts(s, " "); /* Catching here any failure happenned above */ if (!ret) @@ -767,7 +767,7 @@ print_graph_duration(unsigned long long duration, struct trace_seq *s, if (ret != TRACE_TYPE_HANDLED) return ret; - ret = trace_seq_printf(s, "| "); + ret = trace_seq_puts(s, "| "); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -817,7 +817,7 @@ print_graph_entry_leaf(struct trace_iterator *iter, /* Function */ for (i = 0; i < call->depth * TRACE_GRAPH_INDENT; i++) { - ret = trace_seq_printf(s, " "); + ret = trace_seq_putc(s, ' '); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } @@ -858,7 +858,7 @@ print_graph_entry_nested(struct trace_iterator *iter, /* Function */ for (i = 0; i < call->depth * TRACE_GRAPH_INDENT; i++) { - ret = trace_seq_printf(s, " "); + ret = trace_seq_putc(s, ' '); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } @@ -917,7 +917,7 @@ print_graph_prologue(struct trace_iterator *iter, struct trace_seq *s, if (ret == TRACE_TYPE_PARTIAL_LINE) return TRACE_TYPE_PARTIAL_LINE; - ret = trace_seq_printf(s, " | "); + ret = trace_seq_puts(s, " | "); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } @@ -1117,7 +1117,7 @@ print_graph_return(struct ftrace_graph_ret *trace, struct trace_seq *s, /* Closing brace */ for (i = 0; i < trace->depth * TRACE_GRAPH_INDENT; i++) { - ret = trace_seq_printf(s, " "); + ret = trace_seq_putc(s, ' '); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } @@ -1129,7 +1129,7 @@ print_graph_return(struct ftrace_graph_ret *trace, struct trace_seq *s, * belongs to, write out the function name. */ if (func_match) { - ret = trace_seq_printf(s, "}\n"); + ret = trace_seq_puts(s, "}\n"); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } else { @@ -1179,13 +1179,13 @@ print_graph_comment(struct trace_seq *s, struct trace_entry *ent, /* Indentation */ if (depth > 0) for (i = 0; i < (depth + 1) * TRACE_GRAPH_INDENT; i++) { - ret = trace_seq_printf(s, " "); + ret = trace_seq_putc(s, ' '); if (!ret) return TRACE_TYPE_PARTIAL_LINE; } /* The comment */ - ret = trace_seq_printf(s, "/* "); + ret = trace_seq_puts(s, "/* "); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -1216,7 +1216,7 @@ print_graph_comment(struct trace_seq *s, struct trace_entry *ent, s->len--; } - ret = trace_seq_printf(s, " */\n"); + ret = trace_seq_puts(s, " */\n"); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -1448,7 +1448,7 @@ static struct trace_event graph_trace_ret_event = { .funcs = &graph_functions }; -static struct tracer graph_trace __read_mostly = { +static struct tracer graph_trace __tracer_data = { .name = "function_graph", .open = graph_trace_open, .pipe_open = graph_trace_open, diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 7ed6976493c8..243f6834d026 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -95,7 +95,7 @@ static __kprobes bool trace_probe_is_on_module(struct trace_probe *tp) } static int register_probe_event(struct trace_probe *tp); -static void unregister_probe_event(struct trace_probe *tp); +static int unregister_probe_event(struct trace_probe *tp); static DEFINE_MUTEX(probe_lock); static LIST_HEAD(probe_list); @@ -243,11 +243,11 @@ find_event_file_link(struct trace_probe *tp, struct ftrace_event_file *file) static int disable_trace_probe(struct trace_probe *tp, struct ftrace_event_file *file) { + struct event_file_link *link = NULL; + int wait = 0; int ret = 0; if (file) { - struct event_file_link *link; - link = find_event_file_link(tp, file); if (!link) { ret = -EINVAL; @@ -255,10 +255,7 @@ disable_trace_probe(struct trace_probe *tp, struct ftrace_event_file *file) } list_del_rcu(&link->list); - /* synchronize with kprobe_trace_func/kretprobe_trace_func */ - synchronize_sched(); - kfree(link); - + wait = 1; if (!list_empty(&tp->files)) goto out; @@ -271,8 +268,22 @@ disable_trace_probe(struct trace_probe *tp, struct ftrace_event_file *file) disable_kretprobe(&tp->rp); else disable_kprobe(&tp->rp.kp); + wait = 1; } out: + if (wait) { + /* + * Synchronize with kprobe_trace_func/kretprobe_trace_func + * to ensure disabled (all running handlers are finished). + * This is not only for kfree(), but also the caller, + * trace_remove_event_call() supposes it for releasing + * event_call related objects, which will be accessed in + * the kprobe_trace_func/kretprobe_trace_func. + */ + synchronize_sched(); + kfree(link); /* Ignored if link == NULL */ + } + return ret; } @@ -340,9 +351,12 @@ static int unregister_trace_probe(struct trace_probe *tp) if (trace_probe_is_enabled(tp)) return -EBUSY; + /* Will fail if probe is being used by ftrace or perf */ + if (unregister_probe_event(tp)) + return -EBUSY; + __unregister_trace_probe(tp); list_del(&tp->list); - unregister_probe_event(tp); return 0; } @@ -621,7 +635,9 @@ static int release_all_trace_probes(void) /* TODO: Use batch unregistration */ while (!list_empty(&probe_list)) { tp = list_entry(probe_list.next, struct trace_probe, list); - unregister_trace_probe(tp); + ret = unregister_trace_probe(tp); + if (ret) + goto end; free_trace_probe(tp); } @@ -1087,9 +1103,6 @@ kprobe_perf_func(struct trace_probe *tp, struct pt_regs *regs) __size = sizeof(*entry) + tp->size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); - if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, - "profile buffer not large enough")) - return; entry = perf_trace_buf_prepare(size, call->event.type, regs, &rctx); if (!entry) @@ -1120,9 +1133,6 @@ kretprobe_perf_func(struct trace_probe *tp, struct kretprobe_instance *ri, __size = sizeof(*entry) + tp->size + dsize; size = ALIGN(__size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); - if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, - "profile buffer not large enough")) - return; entry = perf_trace_buf_prepare(size, call->event.type, regs, &rctx); if (!entry) @@ -1242,11 +1252,15 @@ static int register_probe_event(struct trace_probe *tp) return ret; } -static void unregister_probe_event(struct trace_probe *tp) +static int unregister_probe_event(struct trace_probe *tp) { + int ret; + /* tp->event is unregistered in trace_remove_event_call() */ - trace_remove_event_call(&tp->call); - kfree(tp->call.print_fmt); + ret = trace_remove_event_call(&tp->call); + if (!ret) + kfree(tp->call.print_fmt); + return ret; } /* Make a debugfs interface for controlling probe points */ diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c index a5e8f4878bfa..b3dcfb2f0fef 100644 --- a/kernel/trace/trace_mmiotrace.c +++ b/kernel/trace/trace_mmiotrace.c @@ -90,7 +90,7 @@ static int mmio_print_pcidev(struct trace_seq *s, const struct pci_dev *dev) if (drv) ret += trace_seq_printf(s, " %s\n", drv->name); else - ret += trace_seq_printf(s, " \n"); + ret += trace_seq_puts(s, " \n"); return ret; } @@ -107,7 +107,7 @@ static void mmio_pipe_open(struct trace_iterator *iter) struct header_iter *hiter; struct trace_seq *s = &iter->seq; - trace_seq_printf(s, "VERSION 20070824\n"); + trace_seq_puts(s, "VERSION 20070824\n"); hiter = kzalloc(sizeof(*hiter), GFP_KERNEL); if (!hiter) @@ -209,7 +209,7 @@ static enum print_line_t mmio_print_rw(struct trace_iterator *iter) (rw->value >> 0) & 0xff, rw->pc, 0); break; default: - ret = trace_seq_printf(s, "rw what?\n"); + ret = trace_seq_puts(s, "rw what?\n"); break; } if (ret) @@ -245,7 +245,7 @@ static enum print_line_t mmio_print_map(struct trace_iterator *iter) secs, usec_rem, m->map_id, 0UL, 0); break; default: - ret = trace_seq_printf(s, "map what?\n"); + ret = trace_seq_puts(s, "map what?\n"); break; } if (ret) diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index bb922d9ee51b..34e7cbac0c9c 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -78,7 +78,7 @@ enum print_line_t trace_print_printk_msg_only(struct trace_iterator *iter) trace_assign_type(field, entry); - ret = trace_seq_printf(s, "%s", field->buf); + ret = trace_seq_puts(s, field->buf); if (!ret) return TRACE_TYPE_PARTIAL_LINE; @@ -558,14 +558,14 @@ seq_print_userip_objs(const struct userstack_entry *entry, struct trace_seq *s, if (ret) ret = trace_seq_puts(s, "??"); if (ret) - ret = trace_seq_puts(s, "\n"); + ret = trace_seq_putc(s, '\n'); continue; } if (!ret) break; if (ret) ret = seq_print_user_ip(s, mm, ip, sym_flags); - ret = trace_seq_puts(s, "\n"); + ret = trace_seq_putc(s, '\n'); } if (mm) @@ -579,7 +579,7 @@ seq_print_ip_sym(struct trace_seq *s, unsigned long ip, unsigned long sym_flags) int ret; if (!ip) - return trace_seq_printf(s, "0"); + return trace_seq_putc(s, '0'); if (sym_flags & TRACE_ITER_SYM_OFFSET) ret = seq_print_sym_offset(s, "%s", ip); @@ -964,14 +964,14 @@ static enum print_line_t trace_fn_trace(struct trace_iterator *iter, int flags, goto partial; if ((flags & TRACE_ITER_PRINT_PARENT) && field->parent_ip) { - if (!trace_seq_printf(s, " <-")) + if (!trace_seq_puts(s, " <-")) goto partial; if (!seq_print_ip_sym(s, field->parent_ip, flags)) goto partial; } - if (!trace_seq_printf(s, "\n")) + if (!trace_seq_putc(s, '\n')) goto partial; return TRACE_TYPE_HANDLED; @@ -1210,7 +1210,7 @@ static enum print_line_t trace_stack_print(struct trace_iterator *iter, if (!seq_print_ip_sym(s, *p, flags)) goto partial; - if (!trace_seq_puts(s, "\n")) + if (!trace_seq_putc(s, '\n')) goto partial; } diff --git a/kernel/trace/trace_printk.c b/kernel/trace/trace_printk.c index a9077c1b4ad3..2900817ba65c 100644 --- a/kernel/trace/trace_printk.c +++ b/kernel/trace/trace_printk.c @@ -244,12 +244,31 @@ static const char **find_next(void *v, loff_t *pos) { const char **fmt = v; int start_index; + int last_index; start_index = __stop___trace_bprintk_fmt - __start___trace_bprintk_fmt; if (*pos < start_index) return __start___trace_bprintk_fmt + *pos; + /* + * The __tracepoint_str section is treated the same as the + * __trace_printk_fmt section. The difference is that the + * __trace_printk_fmt section should only be used by trace_printk() + * in a debugging environment, as if anything exists in that section + * the trace_prink() helper buffers are allocated, which would just + * waste space in a production environment. + * + * The __tracepoint_str sections on the other hand are used by + * tracepoints which need to map pointers to their strings to + * the ASCII text for userspace. + */ + last_index = start_index; + start_index = __stop___tracepoint_str - __start___tracepoint_str; + + if (*pos < last_index + start_index) + return __start___tracepoint_str + (*pos - last_index); + return find_next_mod_format(start_index, v, fmt, pos); } diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 322e16461072..8fd03657bc7d 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -175,7 +175,7 @@ print_syscall_exit(struct trace_iterator *iter, int flags, entry = syscall_nr_to_meta(syscall); if (!entry) { - trace_seq_printf(s, "\n"); + trace_seq_putc(s, '\n'); return TRACE_TYPE_HANDLED; } @@ -566,15 +566,15 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) if (!sys_data) return; + head = this_cpu_ptr(sys_data->enter_event->perf_events); + if (hlist_empty(head)) + return; + /* get the size after alignment with the u32 buffer size field */ size = sizeof(unsigned long) * sys_data->nb_args + sizeof(*rec); size = ALIGN(size + sizeof(u32), sizeof(u64)); size -= sizeof(u32); - if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, - "perf buffer not large enough")) - return; - rec = (struct syscall_trace_enter *)perf_trace_buf_prepare(size, sys_data->enter_event->event.type, regs, &rctx); if (!rec) @@ -583,8 +583,6 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) rec->nr = syscall_nr; syscall_get_arguments(current, regs, 0, sys_data->nb_args, (unsigned long *)&rec->args); - - head = this_cpu_ptr(sys_data->enter_event->perf_events); perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head, NULL); } @@ -642,18 +640,14 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) if (!sys_data) return; + head = this_cpu_ptr(sys_data->exit_event->perf_events); + if (hlist_empty(head)) + return; + /* We can probably do that at build time */ size = ALIGN(sizeof(*rec) + sizeof(u32), sizeof(u64)); size -= sizeof(u32); - /* - * Impossible, but be paranoid with the future - * How to put this check outside runtime? - */ - if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, - "exit event has grown above perf buffer size")) - return; - rec = (struct syscall_trace_exit *)perf_trace_buf_prepare(size, sys_data->exit_event->event.type, regs, &rctx); if (!rec) @@ -661,8 +655,6 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) rec->nr = syscall_nr; rec->ret = syscall_get_return_value(current, regs); - - head = this_cpu_ptr(sys_data->exit_event->perf_events); perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head, NULL); } diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index d5d0cd368a56..272261b5f94f 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -70,7 +70,7 @@ struct trace_uprobe { (sizeof(struct probe_arg) * (n))) static int register_uprobe_event(struct trace_uprobe *tu); -static void unregister_uprobe_event(struct trace_uprobe *tu); +static int unregister_uprobe_event(struct trace_uprobe *tu); static DEFINE_MUTEX(uprobe_lock); static LIST_HEAD(uprobe_list); @@ -164,11 +164,17 @@ static struct trace_uprobe *find_probe_event(const char *event, const char *grou } /* Unregister a trace_uprobe and probe_event: call with locking uprobe_lock */ -static void unregister_trace_uprobe(struct trace_uprobe *tu) +static int unregister_trace_uprobe(struct trace_uprobe *tu) { + int ret; + + ret = unregister_uprobe_event(tu); + if (ret) + return ret; + list_del(&tu->list); - unregister_uprobe_event(tu); free_trace_uprobe(tu); + return 0; } /* Register a trace_uprobe and probe_event */ @@ -181,9 +187,12 @@ static int register_trace_uprobe(struct trace_uprobe *tu) /* register as an event */ old_tp = find_probe_event(tu->call.name, tu->call.class->system); - if (old_tp) + if (old_tp) { /* delete old event */ - unregister_trace_uprobe(old_tp); + ret = unregister_trace_uprobe(old_tp); + if (ret) + goto end; + } ret = register_uprobe_event(tu); if (ret) { @@ -256,6 +265,8 @@ static int create_trace_uprobe(int argc, char **argv) group = UPROBE_EVENT_SYSTEM; if (is_delete) { + int ret; + if (!event) { pr_info("Delete command needs an event name.\n"); return -EINVAL; @@ -269,9 +280,9 @@ static int create_trace_uprobe(int argc, char **argv) return -ENOENT; } /* delete an event */ - unregister_trace_uprobe(tu); + ret = unregister_trace_uprobe(tu); mutex_unlock(&uprobe_lock); - return 0; + return ret; } if (argc < 2) { @@ -408,16 +419,20 @@ fail_address_parse: return ret; } -static void cleanup_all_probes(void) +static int cleanup_all_probes(void) { struct trace_uprobe *tu; + int ret = 0; mutex_lock(&uprobe_lock); while (!list_empty(&uprobe_list)) { tu = list_entry(uprobe_list.next, struct trace_uprobe, list); - unregister_trace_uprobe(tu); + ret = unregister_trace_uprobe(tu); + if (ret) + break; } mutex_unlock(&uprobe_lock); + return ret; } /* Probes listing interfaces */ @@ -462,8 +477,13 @@ static const struct seq_operations probes_seq_op = { static int probes_open(struct inode *inode, struct file *file) { - if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) - cleanup_all_probes(); + int ret; + + if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) { + ret = cleanup_all_probes(); + if (ret) + return ret; + } return seq_open(file, &probes_seq_op); } @@ -818,8 +838,6 @@ static void uprobe_perf_print(struct trace_uprobe *tu, size = SIZEOF_TRACE_ENTRY(is_ret_probe(tu)); size = ALIGN(size + tu->size + sizeof(u32), sizeof(u64)) - sizeof(u32); - if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE, "profile buffer not large enough")) - return; preempt_disable(); head = this_cpu_ptr(call->perf_events); @@ -970,12 +988,17 @@ static int register_uprobe_event(struct trace_uprobe *tu) return ret; } -static void unregister_uprobe_event(struct trace_uprobe *tu) +static int unregister_uprobe_event(struct trace_uprobe *tu) { + int ret; + /* tu->event is unregistered in trace_remove_event_call() */ - trace_remove_event_call(&tu->call); + ret = trace_remove_event_call(&tu->call); + if (ret) + return ret; kfree(tu->call.print_fmt); tu->call.print_fmt = NULL; + return 0; } /* Make a trace interface for controling probe points */ diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c index d8c30db06c5b..9064b919a406 100644 --- a/kernel/user_namespace.c +++ b/kernel/user_namespace.c @@ -62,6 +62,9 @@ int create_user_ns(struct cred *new) kgid_t group = new->egid; int ret; + if (parent_ns->level > 32) + return -EUSERS; + /* * Verify that we can not violate the policy of which files * may be accessed that is specified by the root directory, @@ -92,6 +95,7 @@ int create_user_ns(struct cred *new) atomic_set(&ns->count, 1); /* Leave the new->user_ns reference with the new user namespace. */ ns->parent = parent_ns; + ns->level = parent_ns->level + 1; ns->owner = owner; ns->group = group; @@ -105,16 +109,21 @@ int create_user_ns(struct cred *new) int unshare_userns(unsigned long unshare_flags, struct cred **new_cred) { struct cred *cred; + int err = -ENOMEM; if (!(unshare_flags & CLONE_NEWUSER)) return 0; cred = prepare_creds(); - if (!cred) - return -ENOMEM; + if (cred) { + err = create_user_ns(cred); + if (err) + put_cred(cred); + else + *new_cred = cred; + } - *new_cred = cred; - return create_user_ns(cred); + return err; } void free_user_ns(struct user_namespace *ns) diff --git a/kernel/wait.c b/kernel/wait.c index ce0daa320a26..d550920e040c 100644 --- a/kernel/wait.c +++ b/kernel/wait.c @@ -333,7 +333,8 @@ int __wait_on_atomic_t(wait_queue_head_t *wq, struct wait_bit_queue *q, prepare_to_wait(wq, &q->wait, mode); val = q->key.flags; if (atomic_read(val) == 0) - ret = (*action)(val); + break; + ret = (*action)(val); } while (!ret && atomic_read(val) != 0); finish_wait(wq, &q->wait); return ret; @@ -362,8 +363,7 @@ EXPORT_SYMBOL(out_of_line_wait_on_atomic_t); /** * wake_up_atomic_t - Wake up a waiter on a atomic_t - * @word: The word being waited on, a kernel virtual address - * @bit: The bit of the word being waited on + * @p: The atomic_t being waited on, a kernel virtual address * * Wake up anyone waiting for the atomic_t to go to zero. * diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 0b72e816b8d0..29b79852a845 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -16,9 +16,10 @@ * * This is the generic async execution mechanism. Work items as are * executed in process context. The worker pool is shared and - * automatically managed. There is one worker pool for each CPU and - * one extra for works which are better served by workers which are - * not bound to any specific CPU. + * automatically managed. There are two worker pools for each CPU (one for + * normal work items and the other for high priority ones) and some extra + * pools for workqueues which are not bound to any specific CPU - the + * number of these backing pools is dynamic. * * Please read Documentation/workqueue.txt for details. */ @@ -2033,8 +2034,11 @@ static bool maybe_destroy_workers(struct worker_pool *pool) * multiple times. Does GFP_KERNEL allocations. * * RETURNS: - * spin_lock_irq(pool->lock) which may be released and regrabbed - * multiple times. Does GFP_KERNEL allocations. + * %false if the pool don't need management and the caller can safely start + * processing works, %true indicates that the function released pool->lock + * and reacquired it to perform some management function and that the + * conditions that the caller verified while holding the lock before + * calling the function might no longer be true. */ static bool manage_workers(struct worker *worker) { @@ -2201,6 +2205,15 @@ __acquires(&pool->lock) dump_stack(); } + /* + * The following prevents a kworker from hogging CPU on !PREEMPT + * kernels, where a requeueing work item waiting for something to + * happen could deadlock with stop_machine as such work item could + * indefinitely requeue itself while all other CPUs are trapped in + * stop_machine. + */ + cond_resched(); + spin_lock_irq(&pool->lock); /* clear cpu intensive status */ @@ -2817,6 +2830,19 @@ already_gone: return false; } +static bool __flush_work(struct work_struct *work) +{ + struct wq_barrier barr; + + if (start_flush_work(work, &barr)) { + wait_for_completion(&barr.done); + destroy_work_on_stack(&barr.work); + return true; + } else { + return false; + } +} + /** * flush_work - wait for a work to finish executing the last queueing instance * @work: the work to flush @@ -2830,18 +2856,10 @@ already_gone: */ bool flush_work(struct work_struct *work) { - struct wq_barrier barr; - lock_map_acquire(&work->lockdep_map); lock_map_release(&work->lockdep_map); - if (start_flush_work(work, &barr)) { - wait_for_completion(&barr.done); - destroy_work_on_stack(&barr.work); - return true; - } else { - return false; - } + return __flush_work(work); } EXPORT_SYMBOL_GPL(flush_work); @@ -3081,25 +3099,26 @@ static struct workqueue_struct *dev_to_wq(struct device *dev) return wq_dev->wq; } -static ssize_t wq_per_cpu_show(struct device *dev, - struct device_attribute *attr, char *buf) +static ssize_t per_cpu_show(struct device *dev, struct device_attribute *attr, + char *buf) { struct workqueue_struct *wq = dev_to_wq(dev); return scnprintf(buf, PAGE_SIZE, "%d\n", (bool)!(wq->flags & WQ_UNBOUND)); } +static DEVICE_ATTR_RO(per_cpu); -static ssize_t wq_max_active_show(struct device *dev, - struct device_attribute *attr, char *buf) +static ssize_t max_active_show(struct device *dev, + struct device_attribute *attr, char *buf) { struct workqueue_struct *wq = dev_to_wq(dev); return scnprintf(buf, PAGE_SIZE, "%d\n", wq->saved_max_active); } -static ssize_t wq_max_active_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) +static ssize_t max_active_store(struct device *dev, + struct device_attribute *attr, const char *buf, + size_t count) { struct workqueue_struct *wq = dev_to_wq(dev); int val; @@ -3110,12 +3129,14 @@ static ssize_t wq_max_active_store(struct device *dev, workqueue_set_max_active(wq, val); return count; } +static DEVICE_ATTR_RW(max_active); -static struct device_attribute wq_sysfs_attrs[] = { - __ATTR(per_cpu, 0444, wq_per_cpu_show, NULL), - __ATTR(max_active, 0644, wq_max_active_show, wq_max_active_store), - __ATTR_NULL, +static struct attribute *wq_sysfs_attrs[] = { + &dev_attr_per_cpu.attr, + &dev_attr_max_active.attr, + NULL, }; +ATTRIBUTE_GROUPS(wq_sysfs); static ssize_t wq_pool_ids_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -3265,7 +3286,7 @@ static struct device_attribute wq_sysfs_unbound_attrs[] = { static struct bus_type wq_subsys = { .name = "workqueue", - .dev_attrs = wq_sysfs_attrs, + .dev_groups = wq_sysfs_groups, }; static int __init wq_sysfs_init(void) @@ -3411,6 +3432,12 @@ static void copy_workqueue_attrs(struct workqueue_attrs *to, { to->nice = from->nice; cpumask_copy(to->cpumask, from->cpumask); + /* + * Unlike hash and equality test, this function doesn't ignore + * ->no_numa as it is used for both pool and wq attrs. Instead, + * get_unbound_pool() explicitly clears ->no_numa after copying. + */ + to->no_numa = from->no_numa; } /* hash value of the content of @attr */ @@ -3578,6 +3605,12 @@ static struct worker_pool *get_unbound_pool(const struct workqueue_attrs *attrs) lockdep_set_subclass(&pool->lock, 1); /* see put_pwq() */ copy_workqueue_attrs(pool->attrs, attrs); + /* + * no_numa isn't a worker_pool attribute, always clear it. See + * 'struct workqueue_attrs' comments for detail. + */ + pool->attrs->no_numa = false; + /* if cpumask is contained inside a NUMA node, we belong to that node */ if (wq_numa_enabled) { for_each_node(node) { @@ -4756,7 +4789,14 @@ long work_on_cpu(int cpu, long (*fn)(void *), void *arg) INIT_WORK_ONSTACK(&wfc.work, work_for_cpu_fn); schedule_work_on(cpu, &wfc.work); - flush_work(&wfc.work); + + /* + * The work item is on-stack and can't lead to deadlock through + * flushing. Use __flush_work() to avoid spurious lockdep warnings + * when work_on_cpu()s are nested. + */ + __flush_work(&wfc.work); + return wfc.ret; } EXPORT_SYMBOL_GPL(work_on_cpu); |
