From 13af07df9b7e49f1987cf36aa048dc6c49d0f93d Mon Sep 17 00:00:00 2001 From: Aristeu Rozanski Date: Thu, 23 Aug 2012 16:53:29 -0400 Subject: cgroup: revise how we re-populate root directory When remounting cgroupfs with some subsystems added to it and some removed, cgroup will remove all the files in root directory and then re-popluate it. What I'm doing here is, only remove files which belong to subsystems that are to be unbinded, and only create files for newly-added subsystems. The purpose is to have all other files untouched. This is a preparation for cgroup xattr support. v7: - checkpatch warnings fixed v6: - no changes v5: - no changes v4: - refactored cgroup_clear_directory() to not use cgroup_rm_file() - instead of going thru the list of files, get the file list using the subsystems - use 'subsys_mask' instead of {added,removed}_bits and made cgroup_populate_dir() to match the parameters with cgroup_clear_directory() v3: - refresh patches after recent refactoring Original-patch-by: Li Zefan Cc: Li Zefan Cc: Hugh Dickins Cc: Hillf Danton Cc: Lennart Poettering Signed-off-by: Li Zefan Signed-off-by: Aristeu Rozanski Signed-off-by: Tejun Heo --- kernel/cgroup.c | 61 +++++++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 48 insertions(+), 13 deletions(-) (limited to 'kernel/cgroup.c') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 79818507e444..875a7130647c 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -824,7 +824,8 @@ EXPORT_SYMBOL_GPL(cgroup_unlock); static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); static struct dentry *cgroup_lookup(struct inode *, struct dentry *, unsigned int); static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry); -static int cgroup_populate_dir(struct cgroup *cgrp); +static int cgroup_populate_dir(struct cgroup *cgrp, bool base_files, + unsigned long subsys_mask); static const struct inode_operations cgroup_dir_inode_operations; static const struct file_operations proc_cgroupstats_operations; @@ -963,12 +964,29 @@ static int cgroup_rm_file(struct cgroup *cgrp, const struct cftype *cft) return -ENOENT; } -static void cgroup_clear_directory(struct dentry *dir) +/** + * 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 + * @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) { struct cgroup *cgrp = __d_cgrp(dir); + struct cgroup_subsys *ss; - while (!list_empty(&cgrp->files)) - cgroup_rm_file(cgrp, NULL); + for_each_subsys(cgrp->root, ss) { + struct cftype_set *set; + if (!test_bit(ss->subsys_id, &subsys_mask)) + continue; + list_for_each_entry(set, &ss->cftsets, node) + cgroup_rm_file(cgrp, set->cfts); + } + if (base_files) { + while (!list_empty(&cgrp->files)) + cgroup_rm_file(cgrp, NULL); + } } /* @@ -977,8 +995,9 @@ static void cgroup_clear_directory(struct dentry *dir) 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); + cgroup_clear_directory(dentry, true, root->subsys_bits); parent = dentry->d_parent; spin_lock(&parent->d_lock); @@ -1339,6 +1358,7 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) struct cgroupfs_root *root = sb->s_fs_info; struct cgroup *cgrp = &root->top_cgroup; struct cgroup_sb_opts opts; + unsigned long added_bits, removed_bits; mutex_lock(&cgrp->dentry->d_inode->i_mutex); mutex_lock(&cgroup_mutex); @@ -1354,6 +1374,9 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) pr_warning("cgroup: option changes via remount are deprecated (pid=%d comm=%s)\n", task_tgid_nr(current), current->comm); + added_bits = opts.subsys_bits & ~root->subsys_bits; + removed_bits = root->subsys_bits & ~opts.subsys_bits; + /* Don't allow flags or name to change at remount */ if (opts.flags != root->flags || (opts.name && strcmp(opts.name, root->name))) { @@ -1369,8 +1392,9 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) } /* clear out any existing files and repopulate subsystem files */ - cgroup_clear_directory(cgrp->dentry); - cgroup_populate_dir(cgrp); + cgroup_clear_directory(cgrp->dentry, false, removed_bits); + /* re-populate subsystem files */ + cgroup_populate_dir(cgrp, false, added_bits); if (opts.release_agent) strcpy(root->release_agent_path, opts.release_agent); @@ -1669,7 +1693,7 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, BUG_ON(root->number_of_cgroups != 1); cred = override_creds(&init_cred); - cgroup_populate_dir(root_cgrp); + cgroup_populate_dir(root_cgrp, true, root->subsys_bits); revert_creds(cred); mutex_unlock(&cgroup_root_mutex); mutex_unlock(&cgroup_mutex); @@ -3843,18 +3867,29 @@ static struct cftype files[] = { { } /* terminate */ }; -static int cgroup_populate_dir(struct cgroup *cgrp) +/** + * cgroup_populate_dir - selectively creation of files in a 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 + */ +static int cgroup_populate_dir(struct cgroup *cgrp, bool base_files, + unsigned long subsys_mask) { int err; struct cgroup_subsys *ss; - err = cgroup_addrm_files(cgrp, NULL, files, true); - if (err < 0) - return err; + if (base_files) { + err = cgroup_addrm_files(cgrp, NULL, files, true); + if (err < 0) + return err; + } /* process cftsets of each subsystem */ for_each_subsys(cgrp->root, ss) { struct cftype_set *set; + if (!test_bit(ss->subsys_id, &subsys_mask)) + continue; list_for_each_entry(set, &ss->cftsets, node) cgroup_addrm_files(cgrp, ss, set->cfts, true); @@ -3988,7 +4023,7 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, list_add_tail(&cgrp->allcg_node, &root->allcg_list); - err = cgroup_populate_dir(cgrp); + err = cgroup_populate_dir(cgrp, true, root->subsys_bits); /* If err < 0, we have a half-filled directory - oh well ;) */ mutex_unlock(&cgroup_mutex); -- cgit v1.2.3 From 03b1cde6b22f625ae832b939bc7379ec1466aec5 Mon Sep 17 00:00:00 2001 From: Aristeu Rozanski Date: Thu, 23 Aug 2012 16:53:30 -0400 Subject: cgroup: add xattr support This is one of the items in the plumber's wish list. For use cases: >> What would the use case be for this? > > Attaching meta information to services, in an easily discoverable > way. For example, in systemd we create one cgroup for each service, and > could then store data like the main pid of the specific service as an > xattr on the cgroup itself. That way we'd have almost all service state > in the cgroupfs, which would make it possible to terminate systemd and > later restart it without losing any state information. But there's more: > for example, some very peculiar services cannot be terminated on > shutdown (i.e. fakeraid DM stuff) and it would be really nice if the > services in question could just mark that on their cgroup, by setting an > xattr. On the more desktopy side of things there are other > possibilities: for example there are plans defining what an application > is along the lines of a cgroup (i.e. an app being a collection of > processes). With xattrs one could then attach an icon or human readable > program name on the cgroup. > > The key idea is that this would allow attaching runtime meta information > to cgroups and everything they model (services, apps, vms), that doesn't > need any complex userspace infrastructure, has good access control > (i.e. because the file system enforces that anyway, and there's the > "trusted." xattr namespace), notifications (inotify), and can easily be > shared among applications. > > Lennart v7: - no changes v6: - remove user xattr namespace, only allow trusted and security v5: - check for capabilities before setting/removing xattrs v4: - no changes v3: - instead of config option, use mount option to enable xattr support Original-patch-by: Li Zefan Cc: Li Zefan Cc: Tejun Heo Cc: Hugh Dickins Cc: Hillf Danton Cc: Lennart Poettering Signed-off-by: Li Zefan Signed-off-by: Aristeu Rozanski Signed-off-by: Tejun Heo --- include/linux/cgroup.h | 13 +++++-- kernel/cgroup.c | 100 +++++++++++++++++++++++++++++++++++++++++++++---- 2 files changed, 103 insertions(+), 10 deletions(-) (limited to 'kernel/cgroup.c') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index c90eaa803440..145901f5ef99 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -17,6 +17,7 @@ #include #include #include +#include #ifdef CONFIG_CGROUPS @@ -216,6 +217,9 @@ struct cgroup { /* List of events which userspace want to receive */ struct list_head event_list; spinlock_t event_list_lock; + + /* directory xattrs */ + struct simple_xattrs xattrs; }; /* @@ -309,6 +313,9 @@ struct cftype { /* CFTYPE_* flags */ unsigned int flags; + /* file xattrs */ + struct simple_xattrs xattrs; + int (*open)(struct inode *inode, struct file *file); ssize_t (*read)(struct cgroup *cgrp, struct cftype *cft, struct file *file, @@ -394,7 +401,7 @@ struct cftype { */ struct cftype_set { struct list_head node; /* chained at subsys->cftsets */ - const struct cftype *cfts; + struct cftype *cfts; }; struct cgroup_scanner { @@ -406,8 +413,8 @@ struct cgroup_scanner { void *data; }; -int cgroup_add_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts); -int cgroup_rm_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts); +int cgroup_add_cftypes(struct cgroup_subsys *ss, struct cftype *cfts); +int cgroup_rm_cftypes(struct cgroup_subsys *ss, struct cftype *cfts); int cgroup_is_removed(const struct cgroup *cgrp); diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 875a7130647c..508b4a97ab19 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -276,7 +276,8 @@ inline int cgroup_is_removed(const struct cgroup *cgrp) /* bits in struct cgroupfs_root flags field */ enum { - ROOT_NOPREFIX, /* mounted subsystems have no named prefix */ + ROOT_NOPREFIX, /* mounted subsystems have no named prefix */ + ROOT_XATTR, /* supports extended attributes */ }; static int cgroup_is_releasable(const struct cgroup *cgrp) @@ -913,15 +914,19 @@ static void cgroup_diput(struct dentry *dentry, struct inode *inode) */ BUG_ON(!list_empty(&cgrp->pidlists)); + simple_xattrs_free(&cgrp->xattrs); + kfree_rcu(cgrp, rcu_head); } else { struct cfent *cfe = __d_cfe(dentry); struct cgroup *cgrp = dentry->d_parent->d_fsdata; + struct cftype *cft = cfe->type; WARN_ONCE(!list_empty(&cfe->node) && cgrp != &cgrp->root->top_cgroup, "cfe still linked for %s\n", cfe->type->name); kfree(cfe); + simple_xattrs_free(&cft->xattrs); } iput(inode); } @@ -1140,6 +1145,8 @@ static int cgroup_show_options(struct seq_file *seq, struct dentry *dentry) seq_printf(seq, ",%s", ss->name); if (test_bit(ROOT_NOPREFIX, &root->flags)) seq_puts(seq, ",noprefix"); + if (test_bit(ROOT_XATTR, &root->flags)) + seq_puts(seq, ",xattr"); if (strlen(root->release_agent_path)) seq_printf(seq, ",release_agent=%s", root->release_agent_path); if (clone_children(&root->top_cgroup)) @@ -1208,6 +1215,10 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) opts->clone_children = true; continue; } + if (!strcmp(token, "xattr")) { + set_bit(ROOT_XATTR, &opts->flags); + continue; + } if (!strncmp(token, "release_agent=", 14)) { /* Specifying two release agents is forbidden */ if (opts->release_agent) @@ -1425,6 +1436,7 @@ static void init_cgroup_housekeeping(struct cgroup *cgrp) mutex_init(&cgrp->pidlist_mutex); INIT_LIST_HEAD(&cgrp->event_list); spin_lock_init(&cgrp->event_list_lock); + simple_xattrs_init(&cgrp->xattrs); } static void init_cgroup_root(struct cgroupfs_root *root) @@ -1769,6 +1781,8 @@ static void cgroup_kill_sb(struct super_block *sb) { mutex_unlock(&cgroup_root_mutex); mutex_unlock(&cgroup_mutex); + simple_xattrs_free(&cgrp->xattrs); + kill_litter_super(sb); cgroup_drop_root(root); } @@ -2575,6 +2589,64 @@ static int cgroup_rename(struct inode *old_dir, struct dentry *old_dentry, return simple_rename(old_dir, old_dentry, new_dir, new_dentry); } +static struct simple_xattrs *__d_xattrs(struct dentry *dentry) +{ + if (S_ISDIR(dentry->d_inode->i_mode)) + return &__d_cgrp(dentry)->xattrs; + else + return &__d_cft(dentry)->xattrs; +} + +static inline int xattr_enabled(struct dentry *dentry) +{ + struct cgroupfs_root *root = dentry->d_sb->s_fs_info; + return test_bit(ROOT_XATTR, &root->flags); +} + +static bool is_valid_xattr(const char *name) +{ + if (!strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN) || + !strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN)) + return true; + return false; +} + +static int cgroup_setxattr(struct dentry *dentry, const char *name, + const void *val, size_t size, int flags) +{ + if (!xattr_enabled(dentry)) + return -EOPNOTSUPP; + if (!is_valid_xattr(name)) + return -EINVAL; + return simple_xattr_set(__d_xattrs(dentry), name, val, size, flags); +} + +static int cgroup_removexattr(struct dentry *dentry, const char *name) +{ + if (!xattr_enabled(dentry)) + return -EOPNOTSUPP; + if (!is_valid_xattr(name)) + return -EINVAL; + return simple_xattr_remove(__d_xattrs(dentry), name); +} + +static ssize_t cgroup_getxattr(struct dentry *dentry, const char *name, + void *buf, size_t size) +{ + if (!xattr_enabled(dentry)) + return -EOPNOTSUPP; + if (!is_valid_xattr(name)) + return -EINVAL; + return simple_xattr_get(__d_xattrs(dentry), name, buf, size); +} + +static ssize_t cgroup_listxattr(struct dentry *dentry, char *buf, size_t size) +{ + if (!xattr_enabled(dentry)) + return -EOPNOTSUPP; + return simple_xattr_list(__d_xattrs(dentry), buf, size); +} + static const struct file_operations cgroup_file_operations = { .read = cgroup_file_read, .write = cgroup_file_write, @@ -2583,11 +2655,22 @@ static const struct file_operations cgroup_file_operations = { .release = cgroup_file_release, }; +static const struct inode_operations cgroup_file_inode_operations = { + .setxattr = cgroup_setxattr, + .getxattr = cgroup_getxattr, + .listxattr = cgroup_listxattr, + .removexattr = cgroup_removexattr, +}; + static const struct inode_operations cgroup_dir_inode_operations = { .lookup = cgroup_lookup, .mkdir = cgroup_mkdir, .rmdir = cgroup_rmdir, .rename = cgroup_rename, + .setxattr = cgroup_setxattr, + .getxattr = cgroup_getxattr, + .listxattr = cgroup_listxattr, + .removexattr = cgroup_removexattr, }; static struct dentry *cgroup_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) @@ -2635,6 +2718,7 @@ static int cgroup_create_file(struct dentry *dentry, umode_t mode, } else if (S_ISREG(mode)) { inode->i_size = 0; inode->i_fop = &cgroup_file_operations; + inode->i_op = &cgroup_file_inode_operations; } d_instantiate(dentry, inode); dget(dentry); /* Extra count - pin the dentry in core */ @@ -2695,7 +2779,7 @@ static umode_t cgroup_file_mode(const struct cftype *cft) } static int cgroup_add_file(struct cgroup *cgrp, struct cgroup_subsys *subsys, - const struct cftype *cft) + struct cftype *cft) { struct dentry *dir = cgrp->dentry; struct cgroup *parent = __d_cgrp(dir); @@ -2705,6 +2789,8 @@ 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 }; + simple_xattrs_init(&cft->xattrs); + /* does @cft->flags tell us to skip creation on @cgrp? */ if ((cft->flags & CFTYPE_NOT_ON_ROOT) && !cgrp->parent) return 0; @@ -2745,9 +2831,9 @@ out: } static int cgroup_addrm_files(struct cgroup *cgrp, struct cgroup_subsys *subsys, - const struct cftype cfts[], bool is_add) + struct cftype cfts[], bool is_add) { - const struct cftype *cft; + struct cftype *cft; int err, ret = 0; for (cft = cfts; cft->name[0] != '\0'; cft++) { @@ -2781,7 +2867,7 @@ static void cgroup_cfts_prepare(void) } static void cgroup_cfts_commit(struct cgroup_subsys *ss, - const struct cftype *cfts, bool is_add) + struct cftype *cfts, bool is_add) __releases(&cgroup_mutex) __releases(&cgroup_cft_mutex) { LIST_HEAD(pending); @@ -2832,7 +2918,7 @@ static void cgroup_cfts_commit(struct cgroup_subsys *ss, * function currently returns 0 as long as @cfts registration is successful * even if some file creation attempts on existing cgroups fail. */ -int cgroup_add_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts) +int cgroup_add_cftypes(struct cgroup_subsys *ss, struct cftype *cfts) { struct cftype_set *set; @@ -2862,7 +2948,7 @@ EXPORT_SYMBOL_GPL(cgroup_add_cftypes); * Returns 0 on successful unregistration, -ENOENT if @cfts is not * registered with @ss. */ -int cgroup_rm_cftypes(struct cgroup_subsys *ss, const struct cftype *cfts) +int cgroup_rm_cftypes(struct cgroup_subsys *ss, struct cftype *cfts) { struct cftype_set *set; -- cgit v1.2.3 From a1a71b45a66fd3c3c453b55fbd180f8fccdd1daa Mon Sep 17 00:00:00 2001 From: Aristeu Rozanski Date: Thu, 23 Aug 2012 16:53:31 -0400 Subject: cgroup: rename subsys_bits to subsys_mask In a previous discussion, Tejun Heo suggested to rename references to subsys_bits (added_bits, removed_bits, etc) by something more meaningful. Cc: Li Zefan Cc: Tejun Heo Cc: Hugh Dickins Cc: Hillf Danton Cc: Lennart Poettering Signed-off-by: Aristeu Rozanski Signed-off-by: Tejun Heo --- kernel/cgroup.c | 84 ++++++++++++++++++++++++++++----------------------------- 1 file changed, 42 insertions(+), 42 deletions(-) (limited to 'kernel/cgroup.c') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 508b4a97ab19..ced292d720b9 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -111,13 +111,13 @@ struct cgroupfs_root { * The bitmask of subsystems intended to be attached to this * hierarchy */ - unsigned long subsys_bits; + unsigned long subsys_mask; /* Unique id for this hierarchy. */ int hierarchy_id; /* The bitmask of subsystems currently attached to this hierarchy */ - unsigned long actual_subsys_bits; + unsigned long actual_subsys_mask; /* A list running through the attached subsystems */ struct list_head subsys_list; @@ -557,7 +557,7 @@ static struct css_set *find_existing_css_set( * won't change, so no need for locking. */ for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { - if (root->subsys_bits & (1UL << i)) { + if (root->subsys_mask & (1UL << i)) { /* Subsystem is in this hierarchy. So we want * the subsystem state from the new * cgroup */ @@ -1002,7 +1002,7 @@ 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_bits); + cgroup_clear_directory(dentry, true, root->subsys_mask); parent = dentry->d_parent; spin_lock(&parent->d_lock); @@ -1046,22 +1046,22 @@ void cgroup_release_and_wakeup_rmdir(struct cgroup_subsys_state *css) * returns an error, no reference counts are touched. */ static int rebind_subsystems(struct cgroupfs_root *root, - unsigned long final_bits) + unsigned long final_subsys_mask) { - unsigned long added_bits, removed_bits; + unsigned long added_mask, removed_mask; struct cgroup *cgrp = &root->top_cgroup; int i; BUG_ON(!mutex_is_locked(&cgroup_mutex)); BUG_ON(!mutex_is_locked(&cgroup_root_mutex)); - removed_bits = root->actual_subsys_bits & ~final_bits; - added_bits = final_bits & ~root->actual_subsys_bits; + removed_mask = root->actual_subsys_mask & ~final_subsys_mask; + added_mask = final_subsys_mask & ~root->actual_subsys_mask; /* Check that any added subsystems are currently free */ for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { unsigned long bit = 1UL << i; struct cgroup_subsys *ss = subsys[i]; - if (!(bit & added_bits)) + if (!(bit & added_mask)) continue; /* * Nobody should tell us to do a subsys that doesn't exist: @@ -1086,7 +1086,7 @@ static int rebind_subsystems(struct cgroupfs_root *root, for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; unsigned long bit = 1UL << i; - if (bit & added_bits) { + if (bit & added_mask) { /* We're binding this subsystem to this hierarchy */ BUG_ON(ss == NULL); BUG_ON(cgrp->subsys[i]); @@ -1099,7 +1099,7 @@ static int rebind_subsystems(struct cgroupfs_root *root, if (ss->bind) ss->bind(cgrp); /* refcount was already taken, and we're keeping it */ - } else if (bit & removed_bits) { + } else if (bit & removed_mask) { /* We're removing this subsystem */ BUG_ON(ss == NULL); BUG_ON(cgrp->subsys[i] != dummytop->subsys[i]); @@ -1112,7 +1112,7 @@ static int rebind_subsystems(struct cgroupfs_root *root, list_move(&ss->sibling, &rootnode.subsys_list); /* subsystem is now free - drop reference on module */ module_put(ss->module); - } else if (bit & final_bits) { + } else if (bit & final_subsys_mask) { /* Subsystem state should already exist */ BUG_ON(ss == NULL); BUG_ON(!cgrp->subsys[i]); @@ -1129,7 +1129,7 @@ static int rebind_subsystems(struct cgroupfs_root *root, BUG_ON(cgrp->subsys[i]); } } - root->subsys_bits = root->actual_subsys_bits = final_bits; + root->subsys_mask = root->actual_subsys_mask = final_subsys_mask; synchronize_rcu(); return 0; @@ -1158,7 +1158,7 @@ static int cgroup_show_options(struct seq_file *seq, struct dentry *dentry) } struct cgroup_sb_opts { - unsigned long subsys_bits; + unsigned long subsys_mask; unsigned long flags; char *release_agent; bool clone_children; @@ -1267,7 +1267,7 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) /* Mutually exclusive option 'all' + subsystem name */ if (all_ss) return -EINVAL; - set_bit(i, &opts->subsys_bits); + set_bit(i, &opts->subsys_mask); one_ss = true; break; @@ -1288,7 +1288,7 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) continue; if (ss->disabled) continue; - set_bit(i, &opts->subsys_bits); + set_bit(i, &opts->subsys_mask); } } @@ -1300,19 +1300,19 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) * the cpuset subsystem. */ if (test_bit(ROOT_NOPREFIX, &opts->flags) && - (opts->subsys_bits & mask)) + (opts->subsys_mask & mask)) return -EINVAL; /* Can't specify "none" and some subsystems */ - if (opts->subsys_bits && opts->none) + if (opts->subsys_mask && opts->none) return -EINVAL; /* * We either have to specify by name or by subsystems. (So all * empty hierarchies must have a name). */ - if (!opts->subsys_bits && !opts->name) + if (!opts->subsys_mask && !opts->name) return -EINVAL; /* @@ -1324,7 +1324,7 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) for (i = CGROUP_BUILTIN_SUBSYS_COUNT; i < CGROUP_SUBSYS_COUNT; i++) { unsigned long bit = 1UL << i; - if (!(bit & opts->subsys_bits)) + if (!(bit & opts->subsys_mask)) continue; if (!try_module_get(subsys[i]->module)) { module_pin_failed = true; @@ -1341,7 +1341,7 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) /* drop refcounts only on the ones we took */ unsigned long bit = 1UL << i; - if (!(bit & opts->subsys_bits)) + if (!(bit & opts->subsys_mask)) continue; module_put(subsys[i]->module); } @@ -1351,13 +1351,13 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) return 0; } -static void drop_parsed_module_refcounts(unsigned long subsys_bits) +static void drop_parsed_module_refcounts(unsigned long subsys_mask) { int i; for (i = CGROUP_BUILTIN_SUBSYS_COUNT; i < CGROUP_SUBSYS_COUNT; i++) { unsigned long bit = 1UL << i; - if (!(bit & subsys_bits)) + if (!(bit & subsys_mask)) continue; module_put(subsys[i]->module); } @@ -1369,7 +1369,7 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) struct cgroupfs_root *root = sb->s_fs_info; struct cgroup *cgrp = &root->top_cgroup; struct cgroup_sb_opts opts; - unsigned long added_bits, removed_bits; + unsigned long added_mask, removed_mask; mutex_lock(&cgrp->dentry->d_inode->i_mutex); mutex_lock(&cgroup_mutex); @@ -1381,31 +1381,31 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) goto out_unlock; /* See feature-removal-schedule.txt */ - if (opts.subsys_bits != root->actual_subsys_bits || opts.release_agent) + if (opts.subsys_mask != root->actual_subsys_mask || opts.release_agent) pr_warning("cgroup: option changes via remount are deprecated (pid=%d comm=%s)\n", task_tgid_nr(current), current->comm); - added_bits = opts.subsys_bits & ~root->subsys_bits; - removed_bits = root->subsys_bits & ~opts.subsys_bits; + added_mask = opts.subsys_mask & ~root->subsys_mask; + removed_mask = root->subsys_mask & ~opts.subsys_mask; /* Don't allow flags or name to change at remount */ if (opts.flags != root->flags || (opts.name && strcmp(opts.name, root->name))) { ret = -EINVAL; - drop_parsed_module_refcounts(opts.subsys_bits); + drop_parsed_module_refcounts(opts.subsys_mask); goto out_unlock; } - ret = rebind_subsystems(root, opts.subsys_bits); + ret = rebind_subsystems(root, opts.subsys_mask); if (ret) { - drop_parsed_module_refcounts(opts.subsys_bits); + drop_parsed_module_refcounts(opts.subsys_mask); goto out_unlock; } /* clear out any existing files and repopulate subsystem files */ - cgroup_clear_directory(cgrp->dentry, false, removed_bits); + cgroup_clear_directory(cgrp->dentry, false, removed_mask); /* re-populate subsystem files */ - cgroup_populate_dir(cgrp, false, added_bits); + cgroup_populate_dir(cgrp, false, added_mask); if (opts.release_agent) strcpy(root->release_agent_path, opts.release_agent); @@ -1491,8 +1491,8 @@ static int cgroup_test_super(struct super_block *sb, void *data) * If we asked for subsystems (or explicitly for no * subsystems) then they must match */ - if ((opts->subsys_bits || opts->none) - && (opts->subsys_bits != root->subsys_bits)) + if ((opts->subsys_mask || opts->none) + && (opts->subsys_mask != root->subsys_mask)) return 0; return 1; @@ -1502,7 +1502,7 @@ static struct cgroupfs_root *cgroup_root_from_opts(struct cgroup_sb_opts *opts) { struct cgroupfs_root *root; - if (!opts->subsys_bits && !opts->none) + if (!opts->subsys_mask && !opts->none) return NULL; root = kzalloc(sizeof(*root), GFP_KERNEL); @@ -1515,7 +1515,7 @@ static struct cgroupfs_root *cgroup_root_from_opts(struct cgroup_sb_opts *opts) } init_cgroup_root(root); - root->subsys_bits = opts->subsys_bits; + root->subsys_mask = opts->subsys_mask; root->flags = opts->flags; if (opts->release_agent) strcpy(root->release_agent_path, opts->release_agent); @@ -1547,7 +1547,7 @@ static int cgroup_set_super(struct super_block *sb, void *data) if (!opts->new_root) return -EINVAL; - BUG_ON(!opts->subsys_bits && !opts->none); + BUG_ON(!opts->subsys_mask && !opts->none); ret = set_anon_super(sb, NULL); if (ret) @@ -1665,7 +1665,7 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, if (ret) goto unlock_drop; - ret = rebind_subsystems(root, root->subsys_bits); + ret = rebind_subsystems(root, root->subsys_mask); if (ret == -EBUSY) { free_cg_links(&tmp_cg_links); goto unlock_drop; @@ -1705,7 +1705,7 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, BUG_ON(root->number_of_cgroups != 1); cred = override_creds(&init_cred); - cgroup_populate_dir(root_cgrp, true, root->subsys_bits); + cgroup_populate_dir(root_cgrp, true, root->subsys_mask); revert_creds(cred); mutex_unlock(&cgroup_root_mutex); mutex_unlock(&cgroup_mutex); @@ -1717,7 +1717,7 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, */ cgroup_drop_root(opts.new_root); /* no subsys rebinding, so refcounts don't change */ - drop_parsed_module_refcounts(opts.subsys_bits); + drop_parsed_module_refcounts(opts.subsys_mask); } kfree(opts.release_agent); @@ -1731,7 +1731,7 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, drop_new_super: deactivate_locked_super(sb); drop_modules: - drop_parsed_module_refcounts(opts.subsys_bits); + drop_parsed_module_refcounts(opts.subsys_mask); out_err: kfree(opts.release_agent); kfree(opts.name); @@ -4109,7 +4109,7 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, list_add_tail(&cgrp->allcg_node, &root->allcg_list); - err = cgroup_populate_dir(cgrp, true, root->subsys_bits); + err = cgroup_populate_dir(cgrp, true, root->subsys_mask); /* If err < 0, we have a half-filled directory - oh well ;) */ mutex_unlock(&cgroup_mutex); -- cgit v1.2.3 From be45c900fdc2c66baad5a7703fb8136991d88aeb Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Thu, 13 Sep 2012 09:50:55 +0200 Subject: cgroup: Remove CGROUP_BUILTIN_SUBSYS_COUNT MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit CGROUP_BUILTIN_SUBSYS_COUNT is used as start index or stop index when looping over the subsys array looking either at the builtin or the module subsystems. Since all the builtin subsystems have an id which is lower then CGROUP_BUILTIN_SUBSYS_COUNT we know that any module will have an id larger than CGROUP_BUILTIN_SUBSYS_COUNT. In short the ids are sorted. We are about to change id assignment to happen only at compile time later in this series. That means we can't rely on the above trick since all ids will always be defined at compile time. Furthermore, ordering the builtin subsystems and the module subsystems is not really necessary. So we need a different way to know which subsystem is a builtin or a module one. We can use the subsys[]->module pointer for this. Any place where we need to know if a subsys is module we just check for the pointer. If it is NULL then the subsystem is a builtin one. With this we are able to drop the CGROUP_BUILTIN_SUBSYS_COUNT enum. Though we need to introduce a temporary placeholder so that we don't get a compilation error when only CONFIG_CGROUP is selected and no single controller. An empty enum definition is not valid. Later in this series we are able to remove the placeholder again. And with this change we get a fix for this: kernel/cgroup.c: In function ‘cgroup_load_subsys’: kernel/cgroup.c:4326:38: warning: array subscript is below array bounds [-Warray-bounds] when CONFIG_CGROUP=y and no built in controller was enabled. Signed-off-by: Daniel Wagner Signed-off-by: Tejun Heo Acked-by: Li Zefan Acked-by: Neil Horman Cc: Gao feng Cc: Jamal Hadi Salim Cc: John Fastabend Cc: netdev@vger.kernel.org Cc: cgroups@vger.kernel.org --- include/linux/cgroup.h | 2 +- kernel/cgroup.c | 68 +++++++++++++++++++++++++++++--------------------- 2 files changed, 41 insertions(+), 29 deletions(-) (limited to 'kernel/cgroup.c') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 145901f5ef99..1916cdb071dc 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -48,7 +48,7 @@ extern const struct file_operations proc_cgroup_operations; #define SUBSYS(_x) _x ## _subsys_id, enum cgroup_subsys_id { #include - CGROUP_BUILTIN_SUBSYS_COUNT + __CGROUP_TEMPORARY_PLACEHOLDER }; #undef SUBSYS /* diff --git a/kernel/cgroup.c b/kernel/cgroup.c index ced292d720b9..1b18090269ad 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -88,7 +88,7 @@ static DEFINE_MUTEX(cgroup_root_mutex); /* * Generate an array of cgroup subsystem pointers. At boot time, this is - * populated up to CGROUP_BUILTIN_SUBSYS_COUNT, and modular subsystems are + * populated with the built in subsystems, and modular subsystems are * registered after that. The mutable section of this array is protected by * cgroup_mutex. */ @@ -1321,7 +1321,7 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) * take duplicate reference counts on a subsystem that's already used, * but rebind_subsystems handles this case. */ - for (i = CGROUP_BUILTIN_SUBSYS_COUNT; i < CGROUP_SUBSYS_COUNT; i++) { + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { unsigned long bit = 1UL << i; if (!(bit & opts->subsys_mask)) @@ -1337,7 +1337,7 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) * raced with a module_delete call, and to the user this is * essentially a "subsystem doesn't exist" case. */ - for (i--; i >= CGROUP_BUILTIN_SUBSYS_COUNT; i--) { + for (i--; i >= 0; i--) { /* drop refcounts only on the ones we took */ unsigned long bit = 1UL << i; @@ -1354,7 +1354,7 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) static void drop_parsed_module_refcounts(unsigned long subsys_mask) { int i; - for (i = CGROUP_BUILTIN_SUBSYS_COUNT; i < CGROUP_SUBSYS_COUNT; i++) { + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { unsigned long bit = 1UL << i; if (!(bit & subsys_mask)) @@ -4442,8 +4442,7 @@ int __init_or_module cgroup_load_subsys(struct cgroup_subsys *ss) * since cgroup_init_subsys will have already taken care of it. */ if (ss->module == NULL) { - /* a few sanity checks */ - BUG_ON(ss->subsys_id >= CGROUP_BUILTIN_SUBSYS_COUNT); + /* a sanity check */ BUG_ON(subsys[ss->subsys_id] != ss); return 0; } @@ -4457,7 +4456,7 @@ int __init_or_module cgroup_load_subsys(struct cgroup_subsys *ss) */ mutex_lock(&cgroup_mutex); /* find the first empty slot in the array */ - for (i = CGROUP_BUILTIN_SUBSYS_COUNT; i < CGROUP_SUBSYS_COUNT; i++) { + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { if (subsys[i] == NULL) break; } @@ -4560,7 +4559,6 @@ void cgroup_unload_subsys(struct cgroup_subsys *ss) mutex_lock(&cgroup_mutex); /* deassign the subsys_id */ - BUG_ON(ss->subsys_id < CGROUP_BUILTIN_SUBSYS_COUNT); subsys[ss->subsys_id] = NULL; /* remove subsystem from rootnode's list of subsystems */ @@ -4623,10 +4621,13 @@ int __init cgroup_init_early(void) for (i = 0; i < CSS_SET_TABLE_SIZE; i++) INIT_HLIST_HEAD(&css_set_table[i]); - /* at bootup time, we don't worry about modular subsystems */ - for (i = 0; i < CGROUP_BUILTIN_SUBSYS_COUNT; i++) { + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; + /* at bootup time, we don't worry about modular subsystems */ + if (!ss || ss->module) + continue; + BUG_ON(!ss->name); BUG_ON(strlen(ss->name) > MAX_CGROUP_TYPE_NAMELEN); BUG_ON(!ss->create); @@ -4659,9 +4660,12 @@ int __init cgroup_init(void) if (err) return err; - /* at bootup time, we don't worry about modular subsystems */ - for (i = 0; i < CGROUP_BUILTIN_SUBSYS_COUNT; i++) { + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; + + /* at bootup time, we don't worry about modular subsystems */ + if (!ss || ss->module) + continue; if (!ss->early_init) cgroup_init_subsys(ss); if (ss->use_id) @@ -4856,13 +4860,16 @@ void cgroup_fork_callbacks(struct task_struct *child) { if (need_forkexit_callback) { int i; - /* - * forkexit callbacks are only supported for builtin - * subsystems, and the builtin section of the subsys array is - * immutable, so we don't need to lock the subsys array here. - */ - for (i = 0; i < CGROUP_BUILTIN_SUBSYS_COUNT; i++) { + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; + + /* + * forkexit callbacks are only supported for + * builtin subsystems. + */ + if (!ss || ss->module) + continue; + if (ss->fork) ss->fork(child); } @@ -4967,12 +4974,13 @@ void cgroup_exit(struct task_struct *tsk, int run_callbacks) tsk->cgroups = &init_css_set; if (run_callbacks && need_forkexit_callback) { - /* - * modular subsystems can't use callbacks, so no need to lock - * the subsys array - */ - for (i = 0; i < CGROUP_BUILTIN_SUBSYS_COUNT; i++) { + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; + + /* modular subsystems can't use callbacks */ + if (!ss || ss->module) + continue; + if (ss->exit) { struct cgroup *old_cgrp = rcu_dereference_raw(cg->subsys[i])->cgroup; @@ -5158,13 +5166,17 @@ static int __init cgroup_disable(char *str) while ((token = strsep(&str, ",")) != NULL) { if (!*token) continue; - /* - * cgroup_disable, being at boot time, can't know about module - * subsystems, so we don't worry about them. - */ - for (i = 0; i < CGROUP_BUILTIN_SUBSYS_COUNT; i++) { + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; + /* + * cgroup_disable, being at boot time, can't + * know about module subsystems, so we don't + * worry about them. + */ + if (!ss || ss->module) + continue; + if (!strcmp(token, ss->name)) { ss->disabled = 1; printk(KERN_INFO "Disabling %s control group" -- cgit v1.2.3 From 5fc0b02544b3b9bd3db5a8156b5f3e7350f8e797 Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Wed, 12 Sep 2012 16:12:05 +0200 Subject: cgroup: Wrap subsystem selection macro Before we are able to define all subsystem ids at compile time we need a more fine grained control what gets defined when we include cgroup_subsys.h. For example we define the enums for the subsystems or to declare for struct cgroup_subsys (builtin subsystem) by including cgroup_subsys.h and defining SUBSYS accordingly. Currently, the decision if a subsys is used is defined inside the header by testing if CONFIG_*=y is true. By moving this test outside of cgroup_subsys.h we are able to control it on the include level. This is done by introducing IS_SUBSYS_ENABLED which then is defined according the task, e.g. is CONFIG_*=y or CONFIG_*=m. Signed-off-by: Daniel Wagner Signed-off-by: Tejun Heo Acked-by: Li Zefan Acked-by: Neil Horman Cc: Gao feng Cc: Jamal Hadi Salim Cc: John Fastabend Cc: netdev@vger.kernel.org Cc: cgroups@vger.kernel.org --- include/linux/cgroup.h | 4 ++++ include/linux/cgroup_subsys.h | 24 ++++++++++++------------ kernel/cgroup.c | 1 + 3 files changed, 17 insertions(+), 12 deletions(-) (limited to 'kernel/cgroup.c') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 1916cdb071dc..a5ab5651441b 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -46,10 +46,12 @@ extern const struct file_operations proc_cgroup_operations; /* Define the enumeration of all builtin cgroup subsystems */ #define SUBSYS(_x) _x ## _subsys_id, +#define IS_SUBSYS_ENABLED(option) IS_BUILTIN(option) enum cgroup_subsys_id { #include __CGROUP_TEMPORARY_PLACEHOLDER }; +#undef IS_SUBSYS_ENABLED #undef SUBSYS /* * This define indicates the maximum number of subsystems that can be loaded @@ -528,7 +530,9 @@ struct cgroup_subsys { }; #define SUBSYS(_x) extern struct cgroup_subsys _x ## _subsys; +#define IS_SUBSYS_ENABLED(option) IS_BUILTIN(option) #include +#undef IS_SUBSYS_ENABLED #undef SUBSYS static inline struct cgroup_subsys_state *cgroup_subsys_state( diff --git a/include/linux/cgroup_subsys.h b/include/linux/cgroup_subsys.h index dfae957398c3..f204a7a9cf38 100644 --- a/include/linux/cgroup_subsys.h +++ b/include/linux/cgroup_subsys.h @@ -7,73 +7,73 @@ /* */ -#ifdef CONFIG_CPUSETS +#if IS_SUBSYS_ENABLED(CONFIG_CPUSETS) SUBSYS(cpuset) #endif /* */ -#ifdef CONFIG_CGROUP_DEBUG +#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_DEBUG) SUBSYS(debug) #endif /* */ -#ifdef CONFIG_CGROUP_SCHED +#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_SCHED) SUBSYS(cpu_cgroup) #endif /* */ -#ifdef CONFIG_CGROUP_CPUACCT +#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_CPUACCT) SUBSYS(cpuacct) #endif /* */ -#ifdef CONFIG_MEMCG +#if IS_SUBSYS_ENABLED(CONFIG_MEMCG) SUBSYS(mem_cgroup) #endif /* */ -#ifdef CONFIG_CGROUP_DEVICE +#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_DEVICE) SUBSYS(devices) #endif /* */ -#ifdef CONFIG_CGROUP_FREEZER +#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_FREEZER) SUBSYS(freezer) #endif /* */ -#ifdef CONFIG_NET_CLS_CGROUP +#if IS_SUBSYS_ENABLED(CONFIG_NET_CLS_CGROUP) SUBSYS(net_cls) #endif /* */ -#ifdef CONFIG_BLK_CGROUP +#if IS_SUBSYS_ENABLED(CONFIG_BLK_CGROUP) SUBSYS(blkio) #endif /* */ -#ifdef CONFIG_CGROUP_PERF +#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_PERF) SUBSYS(perf) #endif /* */ -#ifdef CONFIG_NETPRIO_CGROUP +#if IS_SUBSYS_ENABLED(CONFIG_NETPRIO_CGROUP) SUBSYS(net_prio) #endif /* */ -#ifdef CONFIG_CGROUP_HUGETLB +#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_HUGETLB) SUBSYS(hugetlb) #endif diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 1b18090269ad..95e9f3fdb729 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -93,6 +93,7 @@ static DEFINE_MUTEX(cgroup_root_mutex); * cgroup_mutex. */ #define SUBSYS(_x) &_x ## _subsys, +#define IS_SUBSYS_ENABLED(option) IS_BUILTIN(option) static struct cgroup_subsys *subsys[CGROUP_SUBSYS_COUNT] = { #include }; -- cgit v1.2.3 From 80f4c87774721e864d5a5a1f7aca3e95fd90e194 Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Wed, 12 Sep 2012 16:12:06 +0200 Subject: cgroup: Do not depend on a given order when populating the subsys array The *_subsys_id will be used as index to access the subsys. Therefore we need to care we populate the subsystem at the correct position by using designated initialization. With this change we are able to interleave builtin and modules in the subsys array. Signed-off-by: Daniel Wagner Signed-off-by: Tejun Heo Acked-by: Li Zefan Acked-by: Neil Horman Cc: Gao feng Cc: Jamal Hadi Salim Cc: John Fastabend Cc: netdev@vger.kernel.org Cc: cgroups@vger.kernel.org --- kernel/cgroup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/cgroup.c') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 95e9f3fdb729..2bfc78f531b6 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -92,7 +92,7 @@ static DEFINE_MUTEX(cgroup_root_mutex); * registered after that. The mutable section of this array is protected by * cgroup_mutex. */ -#define SUBSYS(_x) &_x ## _subsys, +#define SUBSYS(_x) [_x ## _subsys_id] = &_x ## _subsys, #define IS_SUBSYS_ENABLED(option) IS_BUILTIN(option) static struct cgroup_subsys *subsys[CGROUP_SUBSYS_COUNT] = { #include -- cgit v1.2.3 From 8a8e04df4747661daaee77e98e102d99c9e09b98 Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Wed, 12 Sep 2012 16:12:07 +0200 Subject: cgroup: Assign subsystem IDs during compile time WARNING: With this change it is impossible to load external built controllers anymore. In case where CONFIG_NETPRIO_CGROUP=m and CONFIG_NET_CLS_CGROUP=m is set, corresponding subsys_id should also be a constant. Up to now, net_prio_subsys_id and net_cls_subsys_id would be of the type int and the value would be assigned during runtime. By switching the macro definition IS_SUBSYS_ENABLED from IS_BUILTIN to IS_ENABLED, all *_subsys_id will have constant value. That means we need to remove all the code which assumes a value can be assigned to net_prio_subsys_id and net_cls_subsys_id. A close look is necessary on the RCU part which was introduces by following patch: commit f845172531fb7410c7fb7780b1a6e51ee6df7d52 Author: Herbert Xu Mon May 24 09:12:34 2010 Committer: David S. Miller Mon May 24 09:12:34 2010 cls_cgroup: Store classid in struct sock Tis code was added to init_cgroup_cls() /* We can't use rcu_assign_pointer because this is an int. */ smp_wmb(); net_cls_subsys_id = net_cls_subsys.subsys_id; respectively to exit_cgroup_cls() net_cls_subsys_id = -1; synchronize_rcu(); and in module version of task_cls_classid() rcu_read_lock(); id = rcu_dereference(net_cls_subsys_id); if (id >= 0) classid = container_of(task_subsys_state(p, id), struct cgroup_cls_state, css)->classid; rcu_read_unlock(); Without an explicit explaination why the RCU part is needed. (The rcu_deference was fixed by exchanging it to rcu_derefence_index_check() in a later commit, but that is a minor detail.) So here is my pondering why it was introduced and why it safe to remove it now. Note that this code was copied over to net_prio the reasoning holds for that subsystem too. The idea behind the RCU use for net_cls_subsys_id is to make sure we get a valid pointer back from task_subsys_state(). task_subsys_state() is just blindly accessing the subsys array and returning the pointer. Obviously, passing in -1 as id into task_subsys_state() returns an invalid value (out of lower bound). So this code makes sure that only after module is loaded and the subsystem registered, the id is assigned. Before unregistering the module all old readers must have left the critical section. This is done by assigning -1 to the id and issuing a synchronized_rcu(). Any new readers wont call task_subsys_state() anymore and therefore it is safe to unregister the subsystem. The new code relies on the same trick, but it looks at the subsys pointer return by task_subsys_state() (remember the id is constant and therefore we allways have a valid index into the subsys array). No precautions need to be taken during module loading module. Eventually, all CPUs will get a valid pointer back from task_subsys_state() because rebind_subsystem() which is called after the module init() function will assigned subsys[net_cls_subsys_id] the newly loaded module subsystem pointer. When the subsystem is about to be removed, rebind_subsystem() will called before the module exit() function. In this case, rebind_subsys() will assign subsys[net_cls_subsys_id] a NULL pointer and then it calls synchronize_rcu(). All old readers have left by then the critical section. Any new reader wont access the subsystem anymore. At this point we are safe to unregister the subsystem. No synchronize_rcu() call is needed. Signed-off-by: Daniel Wagner Signed-off-by: Tejun Heo Acked-by: Li Zefan Acked-by: Neil Horman Cc: "David S. Miller" Cc: "Paul E. McKenney" Cc: Andrew Morton Cc: Eric Dumazet Cc: Gao feng Cc: Glauber Costa Cc: Herbert Xu Cc: Jamal Hadi Salim Cc: John Fastabend Cc: Kamezawa Hiroyuki Cc: netdev@vger.kernel.org Cc: cgroups@vger.kernel.org --- include/linux/cgroup.h | 2 +- include/net/cls_cgroup.h | 12 ++++-------- include/net/netprio_cgroup.h | 18 +++++------------- kernel/cgroup.c | 22 +++------------------- net/core/netprio_cgroup.c | 11 ----------- net/core/sock.c | 11 ----------- net/sched/cls_cgroup.c | 13 ------------- 7 files changed, 13 insertions(+), 76 deletions(-) (limited to 'kernel/cgroup.c') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index a5ab5651441b..018f819405c8 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -46,7 +46,7 @@ extern const struct file_operations proc_cgroup_operations; /* Define the enumeration of all builtin cgroup subsystems */ #define SUBSYS(_x) _x ## _subsys_id, -#define IS_SUBSYS_ENABLED(option) IS_BUILTIN(option) +#define IS_SUBSYS_ENABLED(option) IS_ENABLED(option) enum cgroup_subsys_id { #include __CGROUP_TEMPORARY_PLACEHOLDER diff --git a/include/net/cls_cgroup.h b/include/net/cls_cgroup.h index 9bd5db9e10ba..b6a6eeb3905f 100644 --- a/include/net/cls_cgroup.h +++ b/include/net/cls_cgroup.h @@ -42,22 +42,18 @@ static inline u32 task_cls_classid(struct task_struct *p) return classid; } #elif IS_MODULE(CONFIG_NET_CLS_CGROUP) - -extern int net_cls_subsys_id; - static inline u32 task_cls_classid(struct task_struct *p) { - int id; + struct cgroup_subsys_state *css; u32 classid = 0; if (in_interrupt()) return 0; rcu_read_lock(); - id = rcu_dereference_index_check(net_cls_subsys_id, - rcu_read_lock_held()); - if (id >= 0) - classid = container_of(task_subsys_state(p, id), + css = task_subsys_state(p, net_cls_subsys_id); + if (css) + classid = container_of(css, struct cgroup_cls_state, css)->classid; rcu_read_unlock(); diff --git a/include/net/netprio_cgroup.h b/include/net/netprio_cgroup.h index b202de882489..2760f4f4ae9b 100644 --- a/include/net/netprio_cgroup.h +++ b/include/net/netprio_cgroup.h @@ -30,10 +30,6 @@ struct cgroup_netprio_state { u32 prioidx; }; -#ifndef CONFIG_NETPRIO_CGROUP -extern int net_prio_subsys_id; -#endif - extern void sock_update_netprioidx(struct sock *sk, struct task_struct *task); #if IS_BUILTIN(CONFIG_NETPRIO_CGROUP) @@ -55,18 +51,14 @@ static inline u32 task_netprioidx(struct task_struct *p) static inline u32 task_netprioidx(struct task_struct *p) { - struct cgroup_netprio_state *state; - int subsys_id; + struct cgroup_subsys_state *css; u32 idx = 0; rcu_read_lock(); - subsys_id = rcu_dereference_index_check(net_prio_subsys_id, - rcu_read_lock_held()); - if (subsys_id >= 0) { - state = container_of(task_subsys_state(p, subsys_id), - struct cgroup_netprio_state, css); - idx = state->prioidx; - } + css = task_subsys_state(p, net_prio_subsys_id); + if (css) + idx = container_of(css, + struct cgroup_netprio_state, css)->prioidx; rcu_read_unlock(); return idx; } diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 2bfc78f531b6..485cc1487ea2 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -4451,24 +4451,8 @@ int __init_or_module cgroup_load_subsys(struct cgroup_subsys *ss) /* init base cftset */ cgroup_init_cftsets(ss); - /* - * need to register a subsys id before anything else - for example, - * init_cgroup_css needs it. - */ mutex_lock(&cgroup_mutex); - /* find the first empty slot in the array */ - for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { - if (subsys[i] == NULL) - break; - } - if (i == CGROUP_SUBSYS_COUNT) { - /* maximum number of subsystems already registered! */ - mutex_unlock(&cgroup_mutex); - return -EBUSY; - } - /* assign ourselves the subsys_id */ - ss->subsys_id = i; - subsys[i] = ss; + subsys[ss->subsys_id] = ss; /* * no ss->create seems to need anything important in the ss struct, so @@ -4477,7 +4461,7 @@ int __init_or_module cgroup_load_subsys(struct cgroup_subsys *ss) css = ss->create(dummytop); if (IS_ERR(css)) { /* failure case - need to deassign the subsys[] slot. */ - subsys[i] = NULL; + subsys[ss->subsys_id] = NULL; mutex_unlock(&cgroup_mutex); return PTR_ERR(css); } @@ -4493,7 +4477,7 @@ int __init_or_module cgroup_load_subsys(struct cgroup_subsys *ss) if (ret) { dummytop->subsys[ss->subsys_id] = NULL; ss->destroy(dummytop); - subsys[i] = NULL; + subsys[ss->subsys_id] = NULL; mutex_unlock(&cgroup_mutex); return ret; } diff --git a/net/core/netprio_cgroup.c b/net/core/netprio_cgroup.c index c75e3f9d060f..6bc460c38e4f 100644 --- a/net/core/netprio_cgroup.c +++ b/net/core/netprio_cgroup.c @@ -326,9 +326,7 @@ struct cgroup_subsys net_prio_subsys = { .create = cgrp_create, .destroy = cgrp_destroy, .attach = net_prio_attach, -#ifdef CONFIG_NETPRIO_CGROUP .subsys_id = net_prio_subsys_id, -#endif .base_cftypes = ss_files, .module = THIS_MODULE }; @@ -366,10 +364,6 @@ static int __init init_cgroup_netprio(void) ret = cgroup_load_subsys(&net_prio_subsys); if (ret) goto out; -#ifndef CONFIG_NETPRIO_CGROUP - smp_wmb(); - net_prio_subsys_id = net_prio_subsys.subsys_id; -#endif register_netdevice_notifier(&netprio_device_notifier); @@ -386,11 +380,6 @@ static void __exit exit_cgroup_netprio(void) cgroup_unload_subsys(&net_prio_subsys); -#ifndef CONFIG_NETPRIO_CGROUP - net_prio_subsys_id = -1; - synchronize_rcu(); -#endif - rtnl_lock(); for_each_netdev(&init_net, dev) { old = rtnl_dereference(dev->priomap); diff --git a/net/core/sock.c b/net/core/sock.c index ca3eaee66056..47b4ac048e88 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -326,17 +326,6 @@ int __sk_backlog_rcv(struct sock *sk, struct sk_buff *skb) } EXPORT_SYMBOL(__sk_backlog_rcv); -#if defined(CONFIG_CGROUPS) -#if !defined(CONFIG_NET_CLS_CGROUP) -int net_cls_subsys_id = -1; -EXPORT_SYMBOL_GPL(net_cls_subsys_id); -#endif -#if !defined(CONFIG_NETPRIO_CGROUP) -int net_prio_subsys_id = -1; -EXPORT_SYMBOL_GPL(net_prio_subsys_id); -#endif -#endif - static int sock_set_timeout(long *timeo_p, char __user *optval, int optlen) { struct timeval tv; diff --git a/net/sched/cls_cgroup.c b/net/sched/cls_cgroup.c index 7743ea8d1d38..67cf90d962f4 100644 --- a/net/sched/cls_cgroup.c +++ b/net/sched/cls_cgroup.c @@ -77,9 +77,7 @@ struct cgroup_subsys net_cls_subsys = { .name = "net_cls", .create = cgrp_create, .destroy = cgrp_destroy, -#ifdef CONFIG_NET_CLS_CGROUP .subsys_id = net_cls_subsys_id, -#endif .base_cftypes = ss_files, .module = THIS_MODULE, }; @@ -283,12 +281,6 @@ static int __init init_cgroup_cls(void) if (ret) goto out; -#ifndef CONFIG_NET_CLS_CGROUP - /* We can't use rcu_assign_pointer because this is an int. */ - smp_wmb(); - net_cls_subsys_id = net_cls_subsys.subsys_id; -#endif - ret = register_tcf_proto_ops(&cls_cgroup_ops); if (ret) cgroup_unload_subsys(&net_cls_subsys); @@ -301,11 +293,6 @@ static void __exit exit_cgroup_cls(void) { unregister_tcf_proto_ops(&cls_cgroup_ops); -#ifndef CONFIG_NET_CLS_CGROUP - net_cls_subsys_id = -1; - synchronize_rcu(); -#endif - cgroup_unload_subsys(&net_cls_subsys); } -- cgit v1.2.3 From 8c7f6edbda01f1b1a2e60ad61f14fe38023e433b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 13 Sep 2012 12:20:58 -0700 Subject: cgroup: mark subsystems with broken hierarchy support and whine if cgroups are nested for them Currently, cgroup hierarchy support is a mess. cpu related subsystems behave correctly - configuration, accounting and control on a parent properly cover its children. blkio and freezer completely ignore hierarchy and treat all cgroups as if they're directly under the root cgroup. Others show yet different behaviors. These differing interpretations of cgroup hierarchy make using cgroup confusing and it impossible to co-mount controllers into the same hierarchy and obtain sane behavior. Eventually, we want full hierarchy support from all subsystems and probably a unified hierarchy. Users using separate hierarchies expecting completely different behaviors depending on the mounted subsystem is deterimental to making any progress on this front. This patch adds cgroup_subsys.broken_hierarchy and sets it to %true for controllers which are lacking in hierarchy support. The goal of this patch is two-fold. * Move users away from using hierarchy on currently non-hierarchical subsystems, so that implementing proper hierarchy support on those doesn't surprise them. * Keep track of which controllers are broken how and nudge the subsystems to implement proper hierarchy support. For now, start with a single warning message. We can whine louder later on. v2: Fixed a typo spotted by Michal. Warning message updated. v3: Updated memcg part so that it doesn't generate warning in the cases where .use_hierarchy=false doesn't make the behavior different from root.use_hierarchy=true. Fixed a typo spotted by Glauber. v4: Check ->broken_hierarchy after cgroup creation is complete so that ->create() can affect the result per Michal. Dropped unnecessary memcg root handling per Michal. Signed-off-by: Tejun Heo Acked-by: Michal Hocko Acked-by: Li Zefan Acked-by: Serge E. Hallyn Cc: Glauber Costa Cc: Peter Zijlstra Cc: Paul Turner Cc: Johannes Weiner Cc: Thomas Graf Cc: Vivek Goyal Cc: Paul Mackerras Cc: Ingo Molnar Cc: Arnaldo Carvalho de Melo Cc: Neil Horman Cc: Aneesh Kumar K.V --- block/blk-cgroup.c | 8 ++++++++ include/linux/cgroup.h | 15 +++++++++++++++ kernel/cgroup.c | 12 +++++++++++- kernel/cgroup_freezer.c | 8 ++++++++ kernel/events/core.c | 7 +++++++ mm/memcontrol.c | 7 +++++++ net/core/netprio_cgroup.c | 12 +++++++++++- net/sched/cls_cgroup.c | 9 +++++++++ security/device_cgroup.c | 9 +++++++++ 9 files changed, 85 insertions(+), 2 deletions(-) (limited to 'kernel/cgroup.c') diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index f3b44a65fc7a..cafcd7431189 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -737,6 +737,14 @@ struct cgroup_subsys blkio_subsys = { .subsys_id = blkio_subsys_id, .base_cftypes = blkcg_files, .module = THIS_MODULE, + + /* + * blkio subsystem is utterly broken in terms of hierarchy support. + * It treats all cgroups equally regardless of where they're + * located in the hierarchy - all cgroups are treated as if they're + * right below the root. Fix it and remove the following. + */ + .broken_hierarchy = true, }; EXPORT_SYMBOL_GPL(blkio_subsys); diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index c90eaa803440..68e8df70487e 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -496,6 +496,21 @@ struct cgroup_subsys { */ bool __DEPRECATED_clear_css_refs; + /* + * If %false, this subsystem is properly hierarchical - + * configuration, resource accounting and restriction on a parent + * cgroup cover those of its children. If %true, hierarchy support + * is broken in some ways - some subsystems ignore hierarchy + * completely while others are only implemented half-way. + * + * It's now disallowed to create nested cgroups if the subsystem is + * broken and cgroup core will emit a warning message on such + * cases. Eventually, all subsystems will be made properly + * hierarchical and this will go away. + */ + bool broken_hierarchy; + bool warned_broken_hierarchy; + #define MAX_CGROUP_TYPE_NAMELEN 32 const char *name; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 79818507e444..b7d9606b17d7 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -3954,8 +3954,9 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, set_bit(CGRP_CLONE_CHILDREN, &cgrp->flags); for_each_subsys(root, ss) { - struct cgroup_subsys_state *css = ss->create(cgrp); + struct cgroup_subsys_state *css; + css = ss->create(cgrp); if (IS_ERR(css)) { err = PTR_ERR(css); goto err_destroy; @@ -3969,6 +3970,15 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, /* At error, ->destroy() callback has to free assigned ID. */ if (clone_children(parent) && ss->post_clone) ss->post_clone(cgrp); + + if (ss->broken_hierarchy && !ss->warned_broken_hierarchy && + parent->parent) { + pr_warning("cgroup: %s (%d) created nested cgroup for controller \"%s\" which has incomplete hierarchy support. Nested cgroups may change behavior in the future.\n", + current->comm, current->pid, ss->name); + if (!strcmp(ss->name, "memory")) + pr_warning("cgroup: \"memory\" requires setting use_hierarchy to 1 on the root.\n"); + ss->warned_broken_hierarchy = true; + } } list_add(&cgrp->sibling, &cgrp->parent->children); diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup_freezer.c index 3649fc6b3eaa..b1724ce98981 100644 --- a/kernel/cgroup_freezer.c +++ b/kernel/cgroup_freezer.c @@ -373,4 +373,12 @@ struct cgroup_subsys freezer_subsys = { .can_attach = freezer_can_attach, .fork = freezer_fork, .base_cftypes = files, + + /* + * freezer subsys doesn't handle hierarchy at all. Frozen state + * should be inherited through the hierarchy - if a parent is + * frozen, all its children should be frozen. Fix it and remove + * the following. + */ + .broken_hierarchy = true, }; diff --git a/kernel/events/core.c b/kernel/events/core.c index b7935fcec7d9..f18a0a56e5aa 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -7285,5 +7285,12 @@ struct cgroup_subsys perf_subsys = { .destroy = perf_cgroup_destroy, .exit = perf_cgroup_exit, .attach = perf_cgroup_attach, + + /* + * perf_event cgroup doesn't handle nesting correctly. + * ctx->nr_cgroups adjustments should be propagated through the + * cgroup hierarchy. Fix it and remove the following. + */ + .broken_hierarchy = true, }; #endif /* CONFIG_CGROUP_PERF */ diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 795e525afaba..a72f2ffdc3d0 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -4973,6 +4973,13 @@ mem_cgroup_create(struct cgroup *cont) } else { res_counter_init(&memcg->res, NULL); res_counter_init(&memcg->memsw, NULL); + /* + * Deeper hierachy with use_hierarchy == false doesn't make + * much sense so let cgroup subsystem know about this + * unfortunate state in our controller. + */ + if (parent && parent != root_mem_cgroup) + mem_cgroup_subsys.broken_hierarchy = true; } memcg->last_scanned_node = MAX_NUMNODES; INIT_LIST_HEAD(&memcg->oom_notify); diff --git a/net/core/netprio_cgroup.c b/net/core/netprio_cgroup.c index c75e3f9d060f..34f3615b30ca 100644 --- a/net/core/netprio_cgroup.c +++ b/net/core/netprio_cgroup.c @@ -330,7 +330,17 @@ struct cgroup_subsys net_prio_subsys = { .subsys_id = net_prio_subsys_id, #endif .base_cftypes = ss_files, - .module = THIS_MODULE + .module = THIS_MODULE, + + /* + * net_prio has artificial limit on the number of cgroups and + * disallows nesting making it impossible to co-mount it with other + * hierarchical subsystems. Remove the artificially low PRIOIDX_SZ + * limit and properly nest configuration such that children follow + * their parents' configurations by default and are allowed to + * override and remove the following. + */ + .broken_hierarchy = true, }; static int netprio_device_event(struct notifier_block *unused, diff --git a/net/sched/cls_cgroup.c b/net/sched/cls_cgroup.c index 7743ea8d1d38..907daf99ab2e 100644 --- a/net/sched/cls_cgroup.c +++ b/net/sched/cls_cgroup.c @@ -82,6 +82,15 @@ struct cgroup_subsys net_cls_subsys = { #endif .base_cftypes = ss_files, .module = THIS_MODULE, + + /* + * While net_cls cgroup has the rudimentary hierarchy support of + * inheriting the parent's classid on cgroup creation, it doesn't + * properly propagates config changes in ancestors to their + * descendents. A child should follow the parent's configuration + * but be allowed to override it. Fix it and remove the following. + */ + .broken_hierarchy = true, }; struct cls_cgroup_head { diff --git a/security/device_cgroup.c b/security/device_cgroup.c index 442204cc22d9..4b877a92a7ea 100644 --- a/security/device_cgroup.c +++ b/security/device_cgroup.c @@ -457,6 +457,15 @@ struct cgroup_subsys devices_subsys = { .destroy = devcgroup_destroy, .subsys_id = devices_subsys_id, .base_cftypes = dev_cgroup_files, + + /* + * While devices cgroup has the rudimentary hierarchy support which + * checks the parent's restriction, it doesn't properly propagates + * config changes in ancestors to their descendents. A child + * should only be allowed to add more restrictions to the parent's + * configuration. Fix it and remove the following. + */ + .broken_hierarchy = true, }; int __devcgroup_inode_permission(struct inode *inode, int mask) -- cgit v1.2.3 From 1f5320d5972aa50d3e8d2b227b636b370e608359 Mon Sep 17 00:00:00 2001 From: Daisuke Nishimura Date: Thu, 4 Oct 2012 16:37:16 +0900 Subject: cgroup: notify_on_release may not be triggered in some cases notify_on_release must be triggered when the last process in a cgroup is move to another. But if the first(and only) process in a cgroup is moved to another, notify_on_release is not triggered. # mkdir /cgroup/cpu/SRC # mkdir /cgroup/cpu/DST # # echo 1 >/cgroup/cpu/SRC/notify_on_release # echo 1 >/cgroup/cpu/DST/notify_on_release # # sleep 300 & [1] 8629 # # echo 8629 >/cgroup/cpu/SRC/tasks # echo 8629 >/cgroup/cpu/DST/tasks -> notify_on_release for /SRC must be triggered at this point, but it isn't. This is because put_css_set() is called before setting CGRP_RELEASABLE in cgroup_task_migrate(), and is a regression introduce by the commit:74a1166d(cgroups: make procs file writable), which was merged into v3.0. Cc: Ben Blum Cc: # v3.0.x and later Acked-by: Li Zefan Signed-off-by: Daisuke Nishimura Signed-off-by: Tejun Heo --- kernel/cgroup.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel/cgroup.c') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 13774b3b39aa..d1739fc7eb94 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -1962,9 +1962,8 @@ static void cgroup_task_migrate(struct cgroup *cgrp, struct cgroup *oldcgrp, * trading it for newcg is protected by cgroup_mutex, we're safe to drop * it here; it will be freed under RCU. */ - put_css_set(oldcg); - set_bit(CGRP_RELEASABLE, &oldcgrp->flags); + put_css_set(oldcg); } /** -- cgit v1.2.3 From 9bb71308b8133d643648776243e4d5599b1c193d Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 18 Oct 2012 17:52:07 -0700 Subject: Revert "cgroup: Drop task_lock(parent) on cgroup_fork()" This reverts commit 7e381b0eb1e1a9805c37335562e8dc02e7d7848c. The commit incorrectly assumed that fork path always performed threadgroup_change_begin/end() and depended on that for synchronization against task exit and cgroup migration paths instead of explicitly grabbing task_lock(). threadgroup_change is not locked when forking a new process (as opposed to a new thread in the same process) and even if it were it wouldn't be effective as different processes use different threadgroup locks. Revert the incorrect optimization. Signed-off-by: Tejun Heo LKML-Reference: <20121008020000.GB2575@localhost> Acked-by: Li Zefan Bitterly-Acked-by: Frederic Weisbecker Cc: stable@vger.kernel.org --- kernel/cgroup.c | 23 ++++++----------------- 1 file changed, 6 insertions(+), 17 deletions(-) (limited to 'kernel/cgroup.c') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index d1739fc7eb94..75aec12c78a0 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -4814,31 +4814,20 @@ static const struct file_operations proc_cgroupstats_operations = { * * A pointer to the shared css_set was automatically copied in * fork.c by dup_task_struct(). However, we ignore that copy, since - * it was not made under the protection of RCU, cgroup_mutex or - * threadgroup_change_begin(), so it might no longer be a valid - * cgroup pointer. cgroup_attach_task() might have already changed - * current->cgroups, allowing the previously referenced cgroup - * group to be removed and freed. - * - * Outside the pointer validity we also need to process the css_set - * inheritance between threadgoup_change_begin() and - * threadgoup_change_end(), this way there is no leak in any process - * wide migration performed by cgroup_attach_proc() that could otherwise - * miss a thread because it is too early or too late in the fork stage. + * it was not made under the protection of RCU or cgroup_mutex, so + * might no longer be a valid cgroup pointer. cgroup_attach_task() might + * have already changed current->cgroups, allowing the previously + * referenced cgroup group to be removed and freed. * * At the point that cgroup_fork() is called, 'current' is the parent * task, and the passed argument 'child' points to the child task. */ void cgroup_fork(struct task_struct *child) { - /* - * We don't need to task_lock() current because current->cgroups - * can't be changed concurrently here. The parent obviously hasn't - * exited and called cgroup_exit(), and we are synchronized against - * cgroup migration through threadgroup_change_begin(). - */ + task_lock(current); child->cgroups = current->cgroups; get_css_set(child->cgroups); + task_unlock(current); INIT_LIST_HEAD(&child->cg_list); } -- cgit v1.2.3 From d87838321124061f6c935069d97f37010fa417e6 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 18 Oct 2012 17:40:30 -0700 Subject: Revert "cgroup: Remove task_lock() from cgroup_post_fork()" This reverts commit 7e3aa30ac8c904a706518b725c451bb486daaae9. The commit incorrectly assumed that fork path always performed threadgroup_change_begin/end() and depended on that for synchronization against task exit and cgroup migration paths instead of explicitly grabbing task_lock(). threadgroup_change is not locked when forking a new process (as opposed to a new thread in the same process) and even if it were it wouldn't be effective as different processes use different threadgroup locks. Revert the incorrect optimization. Signed-off-by: Tejun Heo LKML-Reference: <20121008020000.GB2575@localhost> Acked-by: Li Zefan Cc: Frederic Weisbecker Cc: stable@vger.kernel.org --- kernel/cgroup.c | 15 +++------------ 1 file changed, 3 insertions(+), 12 deletions(-) (limited to 'kernel/cgroup.c') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 75aec12c78a0..f24f724620dd 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -4883,19 +4883,10 @@ void cgroup_post_fork(struct task_struct *child) */ if (use_task_css_set_links) { write_lock(&css_set_lock); - if (list_empty(&child->cg_list)) { - /* - * It's safe to use child->cgroups without task_lock() - * here because we are protected through - * threadgroup_change_begin() against concurrent - * css_set change in cgroup_task_migrate(). Also - * the task can't exit at that point until - * wake_up_new_task() is called, so we are protected - * against cgroup_exit() setting child->cgroup to - * init_css_set. - */ + task_lock(child); + if (list_empty(&child->cg_list)) list_add(&child->cg_list, &child->cgroups->tasks); - } + task_unlock(child); write_unlock(&css_set_lock); } } -- cgit v1.2.3