diff options
Diffstat (limited to 'kernel/cpuset.c')
| -rw-r--r-- | kernel/cpuset.c | 3547 |
1 files changed, 1949 insertions, 1598 deletions
diff --git a/kernel/cpuset.c b/kernel/cpuset.c index fe2f71f92ae..116a4164720 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -4,22 +4,24 @@ * Processor and Memory placement constraints for sets of tasks. * * Copyright (C) 2003 BULL SA. - * Copyright (C) 2004 Silicon Graphics, Inc. + * Copyright (C) 2004-2007 Silicon Graphics, Inc. + * Copyright (C) 2006 Google, Inc * * Portions derived from Patrick Mochel's sysfs code. * sysfs is Copyright (c) 2001-3 Patrick Mochel - * Portions Copyright (c) 2004 Silicon Graphics, Inc. * - * 2003-10-10 Written by Simon Derr <simon.derr@bull.net> + * 2003-10-10 Written by Simon Derr. * 2003-10-22 Updates by Stephen Hemminger. - * 2004 May-July Rework by Paul Jackson <pj@sgi.com> + * 2004 May-July Rework by Paul Jackson. + * 2006 Rework by Paul Menage to use generic cgroups + * 2008 Rework of the scheduler domains and CPU hotplug handling + * by Max Krasnyansky * * This file is subject to the terms and conditions of the GNU General Public * License. See the file COPYING in the main directory of the Linux * distribution for more details. */ -#include <linux/config.h> #include <linux/cpu.h> #include <linux/cpumask.h> #include <linux/cpuset.h> @@ -34,7 +36,8 @@ #include <linux/list.h> #include <linux/mempolicy.h> #include <linux/mm.h> -#include <linux/module.h> +#include <linux/memory.h> +#include <linux/export.h> #include <linux/mount.h> #include <linux/namei.h> #include <linux/pagemap.h> @@ -42,8 +45,8 @@ #include <linux/rcupdate.h> #include <linux/sched.h> #include <linux/seq_file.h> +#include <linux/security.h> #include <linux/slab.h> -#include <linux/smp_lock.h> #include <linux/spinlock.h> #include <linux/stat.h> #include <linux/string.h> @@ -52,17 +55,13 @@ #include <linux/sort.h> #include <asm/uaccess.h> -#include <asm/atomic.h> -#include <asm/semaphore.h> +#include <linux/atomic.h> +#include <linux/mutex.h> +#include <linux/workqueue.h> +#include <linux/cgroup.h> +#include <linux/wait.h> -#define CPUSET_SUPER_MAGIC 0x27e0eb - -/* - * Tracks how many cpusets are currently defined in system. - * When there is only one cpuset (the root cpuset) we can - * short circuit some hooks. - */ -int number_of_cpusets __read_mostly; +struct static_key cpusets_enabled_key __read_mostly = STATIC_KEY_INIT_FALSE; /* See "Frequency meter" comments, below. */ @@ -74,935 +73,1211 @@ struct fmeter { }; struct cpuset { + struct cgroup_subsys_state css; + unsigned long flags; /* "unsigned long" so bitops work */ - cpumask_t cpus_allowed; /* CPUs allowed to tasks in cpuset */ + cpumask_var_t cpus_allowed; /* CPUs allowed to tasks in cpuset */ nodemask_t mems_allowed; /* Memory Nodes allowed to tasks */ /* - * Count is atomic so can incr (fork) or decr (exit) without a lock. - */ - atomic_t count; /* count tasks using this cpuset */ - - /* - * We link our 'sibling' struct into our parents 'children'. - * Our children link their 'sibling' into our 'children'. + * This is old Memory Nodes tasks took on. + * + * - top_cpuset.old_mems_allowed is initialized to mems_allowed. + * - A new cpuset's old_mems_allowed is initialized when some + * task is moved into it. + * - old_mems_allowed is used in cpuset_migrate_mm() when we change + * cpuset.mems_allowed and have tasks' nodemask updated, and + * then old_mems_allowed is updated to mems_allowed. */ - struct list_head sibling; /* my parents children */ - struct list_head children; /* my children */ + nodemask_t old_mems_allowed; - struct cpuset *parent; /* my parent */ - struct dentry *dentry; /* cpuset fs entry */ + struct fmeter fmeter; /* memory_pressure filter */ /* - * Copy of global cpuset_mems_generation as of the most - * recent time this cpuset changed its mems_allowed. + * Tasks are being attached to this cpuset. Used to prevent + * zeroing cpus/mems_allowed between ->can_attach() and ->attach(). */ - int mems_generation; + int attach_in_progress; - struct fmeter fmeter; /* memory_pressure filter */ + /* partition number for rebuild_sched_domains() */ + int pn; + + /* for custom sched domain */ + int relax_domain_level; }; +static inline struct cpuset *css_cs(struct cgroup_subsys_state *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 css_cs(task_css(task, cpuset_cgrp_id)); +} + +static inline struct cpuset *parent_cs(struct cpuset *cs) +{ + return css_cs(cs->css.parent); +} + +#ifdef CONFIG_NUMA +static inline bool task_has_mempolicy(struct task_struct *task) +{ + return task->mempolicy; +} +#else +static inline bool task_has_mempolicy(struct task_struct *task) +{ + return false; +} +#endif + + /* bits in struct cpuset flags field */ typedef enum { + CS_ONLINE, CS_CPU_EXCLUSIVE, CS_MEM_EXCLUSIVE, + CS_MEM_HARDWALL, CS_MEMORY_MIGRATE, - CS_REMOVED, - CS_NOTIFY_ON_RELEASE + CS_SCHED_LOAD_BALANCE, + CS_SPREAD_PAGE, + CS_SPREAD_SLAB, } cpuset_flagbits_t; /* convenient tests for these bits */ +static inline bool is_cpuset_online(const struct cpuset *cs) +{ + return test_bit(CS_ONLINE, &cs->flags); +} + static inline int is_cpu_exclusive(const struct cpuset *cs) { - return !!test_bit(CS_CPU_EXCLUSIVE, &cs->flags); + return test_bit(CS_CPU_EXCLUSIVE, &cs->flags); } static inline int is_mem_exclusive(const struct cpuset *cs) { - return !!test_bit(CS_MEM_EXCLUSIVE, &cs->flags); + return test_bit(CS_MEM_EXCLUSIVE, &cs->flags); } -static inline int is_removed(const struct cpuset *cs) +static inline int is_mem_hardwall(const struct cpuset *cs) { - return !!test_bit(CS_REMOVED, &cs->flags); + return test_bit(CS_MEM_HARDWALL, &cs->flags); } -static inline int notify_on_release(const struct cpuset *cs) +static inline int is_sched_load_balance(const struct cpuset *cs) { - return !!test_bit(CS_NOTIFY_ON_RELEASE, &cs->flags); + return test_bit(CS_SCHED_LOAD_BALANCE, &cs->flags); } static inline int is_memory_migrate(const struct cpuset *cs) { - return !!test_bit(CS_MEMORY_MIGRATE, &cs->flags); + return test_bit(CS_MEMORY_MIGRATE, &cs->flags); } -/* - * Increment this atomic integer everytime any cpuset changes its - * mems_allowed value. Users of cpusets can track this generation - * number, and avoid having to lock and reload mems_allowed unless - * the cpuset they're using changes generation. - * - * A single, global generation is needed because attach_task() could - * reattach a task to a different cpuset, which must not have its - * generation numbers aliased with those of that tasks previous cpuset. - * - * Generations are needed for mems_allowed because one task cannot - * modify anothers memory placement. So we must enable every task, - * on every visit to __alloc_pages(), to efficiently check whether - * its current->cpuset->mems_allowed has changed, requiring an update - * of its current->mems_allowed. - */ -static atomic_t cpuset_mems_generation = ATOMIC_INIT(1); +static inline int is_spread_page(const struct cpuset *cs) +{ + return test_bit(CS_SPREAD_PAGE, &cs->flags); +} + +static inline int is_spread_slab(const struct cpuset *cs) +{ + return test_bit(CS_SPREAD_SLAB, &cs->flags); +} static struct cpuset top_cpuset = { - .flags = ((1 << CS_CPU_EXCLUSIVE) | (1 << CS_MEM_EXCLUSIVE)), - .cpus_allowed = CPU_MASK_ALL, - .mems_allowed = NODE_MASK_ALL, - .count = ATOMIC_INIT(0), - .sibling = LIST_HEAD_INIT(top_cpuset.sibling), - .children = LIST_HEAD_INIT(top_cpuset.children), + .flags = ((1 << CS_ONLINE) | (1 << CS_CPU_EXCLUSIVE) | + (1 << CS_MEM_EXCLUSIVE)), }; -static struct vfsmount *cpuset_mount; -static struct super_block *cpuset_sb; +/** + * cpuset_for_each_child - traverse online children of a cpuset + * @child_cs: loop cursor pointing to the current child + * @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_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_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_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_css, root_cs) \ + css_for_each_descendant_pre((pos_css), &(root_cs)->css) \ + if (is_cpuset_online(((des_cs) = css_cs((pos_css))))) /* - * We have two global cpuset semaphores below. They can nest. - * It is ok to first take manage_sem, then nest callback_sem. We also - * require taking task_lock() when dereferencing a tasks cpuset pointer. + * There are two global mutexes guarding cpuset structures - cpuset_mutex + * and callback_mutex. The latter may nest inside the former. We also + * require taking task_lock() when dereferencing a task's cpuset pointer. * See "The task_lock() exception", at the end of this comment. * - * A task must hold both semaphores to modify cpusets. If a task - * holds manage_sem, then it blocks others wanting that semaphore, - * ensuring that it is the only task able to also acquire callback_sem - * and be able to modify cpusets. It can perform various checks on - * the cpuset structure first, knowing nothing will change. It can - * also allocate memory while just holding manage_sem. While it is - * performing these checks, various callback routines can briefly - * acquire callback_sem to query cpusets. Once it is ready to make - * the changes, it takes callback_sem, blocking everyone else. + * A task must hold both mutexes to modify cpusets. If a task holds + * cpuset_mutex, then it blocks others wanting that mutex, ensuring that it + * is the only task able to also acquire callback_mutex and be able to + * modify cpusets. It can perform various checks on the cpuset structure + * first, knowing nothing will change. It can also allocate memory while + * just holding cpuset_mutex. While it is performing these checks, various + * callback routines can briefly acquire callback_mutex to query cpusets. + * Once it is ready to make the changes, it takes callback_mutex, blocking + * everyone else. * * Calls to the kernel memory allocator can not be made while holding - * callback_sem, as that would risk double tripping on callback_sem + * callback_mutex, as that would risk double tripping on callback_mutex * from one of the callbacks into the cpuset code from within * __alloc_pages(). * - * If a task is only holding callback_sem, then it has read-only + * If a task is only holding callback_mutex, then it has read-only * access to cpusets. * - * The task_struct fields mems_allowed and mems_generation may only - * be accessed in the context of that task, so require no locks. - * - * Any task can increment and decrement the count field without lock. - * So in general, code holding manage_sem or callback_sem can't rely - * on the count field not changing. However, if the count goes to - * zero, then only attach_task(), which holds both semaphores, can - * increment it again. Because a count of zero means that no tasks - * are currently attached, therefore there is no way a task attached - * to that cpuset can fork (the other way to increment the count). - * So code holding manage_sem or callback_sem can safely assume that - * if the count is zero, it will stay zero. Similarly, if a task - * holds manage_sem or callback_sem on a cpuset with zero count, it - * knows that the cpuset won't be removed, as cpuset_rmdir() needs - * both of those semaphores. - * - * A possible optimization to improve parallelism would be to make - * callback_sem a R/W semaphore (rwsem), allowing the callback routines - * to proceed in parallel, with read access, until the holder of - * manage_sem needed to take this rwsem for exclusive write access - * and modify some cpusets. - * - * The cpuset_common_file_write handler for operations that modify - * the cpuset hierarchy holds manage_sem across the entire operation, - * single threading all such cpuset modifications across the system. - * - * The cpuset_common_file_read() handlers only hold callback_sem across + * Now, the task_struct fields mems_allowed and mempolicy may be changed + * by other task, we use alloc_lock in the task_struct fields to protect + * them. + * + * The cpuset_common_file_read() handlers only hold callback_mutex across * small pieces of code, such as when reading out possibly multi-word * cpumasks and nodemasks. * - * The fork and exit callbacks cpuset_fork() and cpuset_exit(), don't - * (usually) take either semaphore. These are the two most performance - * critical pieces of code here. The exception occurs on cpuset_exit(), - * when a task in a notify_on_release cpuset exits. Then manage_sem - * is taken, and if the cpuset count is zero, a usermode call made - * to /sbin/cpuset_release_agent with the name of the cpuset (path - * relative to the root of cpuset file system) as the argument. - * - * A cpuset can only be deleted if both its 'count' of using tasks - * is zero, and its list of 'children' cpusets is empty. Since all - * tasks in the system use _some_ cpuset, and since there is always at - * least one task in the system (init, pid == 1), therefore, top_cpuset - * always has either children cpusets and/or using tasks. So we don't - * need a special hack to ensure that top_cpuset cannot be deleted. - * - * The above "Tale of Two Semaphores" would be complete, but for: - * - * The task_lock() exception - * - * The need for this exception arises from the action of attach_task(), - * which overwrites one tasks cpuset pointer with another. It does - * so using both semaphores, however there are several performance - * critical places that need to reference task->cpuset without the - * expense of grabbing a system global semaphore. Therefore except as - * noted below, when dereferencing or, as in attach_task(), modifying - * a tasks cpuset pointer we use task_lock(), which acts on a spinlock - * (task->alloc_lock) already in the task_struct routinely used for - * such matters. - * - * P.S. One more locking exception. RCU is used to guard the - * update of a tasks cpuset pointer by attach_task() and the - * access of task->cpuset->mems_generation via that pointer in - * the routine cpuset_update_task_memory_state(). + * Accessing a task's cpuset should be done in accordance with the + * guidelines for accessing subsystem state in kernel/cgroup.c */ -static DECLARE_MUTEX(manage_sem); -static DECLARE_MUTEX(callback_sem); +static DEFINE_MUTEX(cpuset_mutex); +static DEFINE_MUTEX(callback_mutex); /* - * A couple of forward declarations required, due to cyclic reference loop: - * cpuset_mkdir -> cpuset_create -> cpuset_populate_dir -> cpuset_add_file - * -> cpuset_create_file -> cpuset_dir_inode_operations -> cpuset_mkdir. + * CPU / memory hotplug is handled asynchronously. */ +static void cpuset_hotplug_workfn(struct work_struct *work); +static DECLARE_WORK(cpuset_hotplug_work, cpuset_hotplug_workfn); -static int cpuset_mkdir(struct inode *dir, struct dentry *dentry, int mode); -static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry); - -static struct backing_dev_info cpuset_backing_dev_info = { - .ra_pages = 0, /* No readahead */ - .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, -}; +static DECLARE_WAIT_QUEUE_HEAD(cpuset_attach_wq); -static struct inode *cpuset_new_inode(mode_t mode) -{ - struct inode *inode = new_inode(cpuset_sb); - - if (inode) { - inode->i_mode = mode; - inode->i_uid = current->fsuid; - inode->i_gid = current->fsgid; - inode->i_blksize = PAGE_CACHE_SIZE; - inode->i_blocks = 0; - inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; - inode->i_mapping->backing_dev_info = &cpuset_backing_dev_info; - } - return inode; -} - -static void cpuset_diput(struct dentry *dentry, struct inode *inode) -{ - /* is dentry a directory ? if so, kfree() associated cpuset */ - if (S_ISDIR(inode->i_mode)) { - struct cpuset *cs = dentry->d_fsdata; - BUG_ON(!(is_removed(cs))); - kfree(cs); +/* + * This is ugly, but preserves the userspace API for existing cpuset + * users. If someone tries to mount the "cpuset" filesystem, we + * silently switch it to mount "cgroup" instead + */ +static struct dentry *cpuset_mount(struct file_system_type *fs_type, + int flags, const char *unused_dev_name, void *data) +{ + struct file_system_type *cgroup_fs = get_fs_type("cgroup"); + struct dentry *ret = ERR_PTR(-ENODEV); + if (cgroup_fs) { + char mountopts[] = + "cpuset,noprefix," + "release_agent=/sbin/cpuset_release_agent"; + ret = cgroup_fs->mount(cgroup_fs, flags, + unused_dev_name, mountopts); + put_filesystem(cgroup_fs); } - iput(inode); + return ret; } -static struct dentry_operations cpuset_dops = { - .d_iput = cpuset_diput, +static struct file_system_type cpuset_fs_type = { + .name = "cpuset", + .mount = cpuset_mount, }; -static struct dentry *cpuset_get_dentry(struct dentry *parent, const char *name) +/* + * Return in pmask the portion of a cpusets's cpus_allowed that + * are online. If none are online, walk up the cpuset hierarchy + * until we find one that does have some online cpus. The top + * cpuset always has some cpus online. + * + * One way or another, we guarantee to return some non-empty subset + * of cpu_online_mask. + * + * Call with callback_mutex held. + */ +static void guarantee_online_cpus(struct cpuset *cs, struct cpumask *pmask) { - struct dentry *d = lookup_one_len(name, parent, strlen(name)); - if (!IS_ERR(d)) - d->d_op = &cpuset_dops; - return d; + while (!cpumask_intersects(cs->cpus_allowed, cpu_online_mask)) + cs = parent_cs(cs); + cpumask_and(pmask, cs->cpus_allowed, cpu_online_mask); } -static void remove_dir(struct dentry *d) +/* + * Return in *pmask the portion of a cpusets's mems_allowed that + * are online, with memory. If none are online with memory, walk + * up the cpuset hierarchy until we find one that does have some + * online mems. The top cpuset always has some mems online. + * + * One way or another, we guarantee to return some non-empty subset + * of node_states[N_MEMORY]. + * + * Call with callback_mutex held. + */ +static void guarantee_online_mems(struct cpuset *cs, nodemask_t *pmask) { - struct dentry *parent = dget(d->d_parent); - - d_delete(d); - simple_rmdir(parent->d_inode, d); - dput(parent); + while (!nodes_intersects(cs->mems_allowed, node_states[N_MEMORY])) + cs = parent_cs(cs); + nodes_and(*pmask, cs->mems_allowed, node_states[N_MEMORY]); } /* - * NOTE : the dentry must have been dget()'ed + * update task's spread flag if cpuset's page/slab spread flag is set + * + * Called with callback_mutex/cpuset_mutex held */ -static void cpuset_d_remove_dir(struct dentry *dentry) -{ - struct list_head *node; - - spin_lock(&dcache_lock); - node = dentry->d_subdirs.next; - while (node != &dentry->d_subdirs) { - struct dentry *d = list_entry(node, struct dentry, d_u.d_child); - list_del_init(node); - if (d->d_inode) { - d = dget_locked(d); - spin_unlock(&dcache_lock); - d_delete(d); - simple_unlink(dentry->d_inode, d); - dput(d); - spin_lock(&dcache_lock); - } - node = dentry->d_subdirs.next; - } - list_del_init(&dentry->d_u.d_child); - spin_unlock(&dcache_lock); - remove_dir(dentry); +static void cpuset_update_task_spread_flag(struct cpuset *cs, + struct task_struct *tsk) +{ + if (is_spread_page(cs)) + tsk->flags |= PF_SPREAD_PAGE; + else + tsk->flags &= ~PF_SPREAD_PAGE; + if (is_spread_slab(cs)) + tsk->flags |= PF_SPREAD_SLAB; + else + tsk->flags &= ~PF_SPREAD_SLAB; } -static struct super_operations cpuset_ops = { - .statfs = simple_statfs, - .drop_inode = generic_delete_inode, -}; +/* + * is_cpuset_subset(p, q) - Is cpuset p a subset of cpuset q? + * + * One cpuset is a subset of another if all its allowed CPUs and + * Memory Nodes are a subset of the other, and its exclusive flags + * are only set if the other's are set. Call holding cpuset_mutex. + */ -static int cpuset_fill_super(struct super_block *sb, void *unused_data, - int unused_silent) +static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q) { - struct inode *inode; - struct dentry *root; + return cpumask_subset(p->cpus_allowed, q->cpus_allowed) && + nodes_subset(p->mems_allowed, q->mems_allowed) && + is_cpu_exclusive(p) <= is_cpu_exclusive(q) && + is_mem_exclusive(p) <= is_mem_exclusive(q); +} - sb->s_blocksize = PAGE_CACHE_SIZE; - sb->s_blocksize_bits = PAGE_CACHE_SHIFT; - sb->s_magic = CPUSET_SUPER_MAGIC; - sb->s_op = &cpuset_ops; - cpuset_sb = sb; +/** + * alloc_trial_cpuset - allocate a trial cpuset + * @cs: the cpuset that the trial cpuset duplicates + */ +static struct cpuset *alloc_trial_cpuset(struct cpuset *cs) +{ + struct cpuset *trial; - inode = cpuset_new_inode(S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR); - if (inode) { - inode->i_op = &simple_dir_inode_operations; - inode->i_fop = &simple_dir_operations; - /* directories start off with i_nlink == 2 (for "." entry) */ - inode->i_nlink++; - } else { - return -ENOMEM; - } + trial = kmemdup(cs, sizeof(*cs), GFP_KERNEL); + if (!trial) + return NULL; - root = d_alloc_root(inode); - if (!root) { - iput(inode); - return -ENOMEM; + if (!alloc_cpumask_var(&trial->cpus_allowed, GFP_KERNEL)) { + kfree(trial); + return NULL; } - sb->s_root = root; - return 0; + cpumask_copy(trial->cpus_allowed, cs->cpus_allowed); + + return trial; } -static struct super_block *cpuset_get_sb(struct file_system_type *fs_type, - int flags, const char *unused_dev_name, - void *data) +/** + * free_trial_cpuset - free the trial cpuset + * @trial: the trial cpuset to be freed + */ +static void free_trial_cpuset(struct cpuset *trial) { - return get_sb_single(fs_type, flags, data, cpuset_fill_super); + free_cpumask_var(trial->cpus_allowed); + kfree(trial); } -static struct file_system_type cpuset_fs_type = { - .name = "cpuset", - .get_sb = cpuset_get_sb, - .kill_sb = kill_litter_super, -}; - -/* struct cftype: +/* + * validate_change() - Used to validate that any proposed cpuset change + * follows the structural rules for cpusets. + * + * If we replaced the flag and mask values of the current cpuset + * (cur) with those values in the trial cpuset (trial), would + * our various subset and exclusive rules still be valid? Presumes + * cpuset_mutex held. * - * The files in the cpuset filesystem mostly have a very simple read/write - * handling, some common function will take care of it. Nevertheless some cases - * (read tasks) are special and therefore I define this structure for every - * kind of file. + * 'cur' is the address of an actual, in-use cpuset. Operations + * such as list traversal that depend on the actual address of the + * cpuset in the list must use cur below, not trial. * + * 'trial' is the address of bulk structure copy of cur, with + * perhaps one or more of the fields cpus_allowed, mems_allowed, + * or flags changed to new, trial values. * - * When reading/writing to a file: - * - the cpuset to use in file->f_dentry->d_parent->d_fsdata - * - the 'cftype' of the file is file->f_dentry->d_fsdata + * Return 0 if valid, -errno if not. */ -struct cftype { - char *name; - int private; - int (*open) (struct inode *inode, struct file *file); - ssize_t (*read) (struct file *file, char __user *buf, size_t nbytes, - loff_t *ppos); - int (*write) (struct file *file, const char __user *buf, size_t nbytes, - loff_t *ppos); - int (*release) (struct inode *inode, struct file *file); -}; - -static inline struct cpuset *__d_cs(struct dentry *dentry) +static int validate_change(struct cpuset *cur, struct cpuset *trial) { - return dentry->d_fsdata; -} + struct cgroup_subsys_state *css; + struct cpuset *c, *par; + int ret; -static inline struct cftype *__d_cft(struct dentry *dentry) -{ - return dentry->d_fsdata; + rcu_read_lock(); + + /* Each of our child cpusets must be a subset of us */ + ret = -EBUSY; + cpuset_for_each_child(c, css, cur) + if (!is_cpuset_subset(c, trial)) + goto out; + + /* Remaining checks don't apply to root cpuset */ + ret = 0; + if (cur == &top_cpuset) + goto out; + + par = parent_cs(cur); + + /* We must be a subset of our parent cpuset */ + ret = -EACCES; + if (!is_cpuset_subset(trial, par)) + goto out; + + /* + * If either I or some sibling (!= me) is exclusive, we can't + * overlap + */ + ret = -EINVAL; + 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)) + goto out; + if ((is_mem_exclusive(trial) || is_mem_exclusive(c)) && + c != cur && + nodes_intersects(trial->mems_allowed, c->mems_allowed)) + goto out; + } + + /* + * Cpusets with tasks - existing or newly being attached - can't + * be changed to have empty cpus_allowed or mems_allowed. + */ + ret = -ENOSPC; + if ((cgroup_has_tasks(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: + rcu_read_unlock(); + return ret; } +#ifdef CONFIG_SMP /* - * Call with manage_sem held. Writes path of cpuset into buf. - * Returns 0 on success, -errno on error. + * Helper routine for generate_sched_domains(). + * Do cpusets a, b have overlapping cpus_allowed masks? */ +static int cpusets_overlap(struct cpuset *a, struct cpuset *b) +{ + return cpumask_intersects(a->cpus_allowed, b->cpus_allowed); +} -static int cpuset_path(const struct cpuset *cs, char *buf, int buflen) +static void +update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c) { - char *start; + if (dattr->relax_domain_level < c->relax_domain_level) + dattr->relax_domain_level = c->relax_domain_level; + return; +} - start = buf + buflen; +static void update_domain_attr_tree(struct sched_domain_attr *dattr, + struct cpuset *root_cs) +{ + struct cpuset *cp; + struct cgroup_subsys_state *pos_css; + + rcu_read_lock(); + cpuset_for_each_descendant_pre(cp, pos_css, root_cs) { + if (cp == root_cs) + continue; - *--start = '\0'; - for (;;) { - int len = cs->dentry->d_name.len; - if ((start -= len) < buf) - return -ENAMETOOLONG; - memcpy(start, cs->dentry->d_name.name, len); - cs = cs->parent; - if (!cs) - break; - if (!cs->parent) + /* skip the whole subtree if @cp doesn't have any CPU */ + if (cpumask_empty(cp->cpus_allowed)) { + pos_css = css_rightmost_descendant(pos_css); continue; - if (--start < buf) - return -ENAMETOOLONG; - *start = '/'; + } + + if (is_sched_load_balance(cp)) + update_domain_attr(dattr, cp); } - memmove(buf, start, buf + buflen - start); - return 0; + rcu_read_unlock(); } /* - * Notify userspace when a cpuset is released, by running - * /sbin/cpuset_release_agent with the name of the cpuset (path - * relative to the root of cpuset file system) as the argument. - * - * Most likely, this user command will try to rmdir this cpuset. - * - * This races with the possibility that some other task will be - * attached to this cpuset before it is removed, or that some other - * user task will 'mkdir' a child cpuset of this cpuset. That's ok. - * The presumed 'rmdir' will fail quietly if this cpuset is no longer - * unused, and this cpuset will be reprieved from its death sentence, - * to continue to serve a useful existence. Next time it's released, - * we will get notified again, if it still has 'notify_on_release' set. - * - * The final arg to call_usermodehelper() is 0, which means don't - * wait. The separate /sbin/cpuset_release_agent task is forked by - * call_usermodehelper(), then control in this thread returns here, - * without waiting for the release agent task. We don't bother to - * wait because the caller of this routine has no use for the exit - * status of the /sbin/cpuset_release_agent task, so no sense holding - * our caller up for that. - * - * When we had only one cpuset semaphore, we had to call this - * without holding it, to avoid deadlock when call_usermodehelper() - * allocated memory. With two locks, we could now call this while - * holding manage_sem, but we still don't, so as to minimize - * the time manage_sem is held. + * generate_sched_domains() + * + * This function builds a partial partition of the systems CPUs + * A 'partial partition' is a set of non-overlapping subsets whose + * union is a subset of that set. + * The output of this function needs to be passed to kernel/sched/core.c + * partition_sched_domains() routine, which will rebuild the scheduler's + * load balancing domains (sched domains) as specified by that partial + * partition. + * + * See "What is sched_load_balance" in Documentation/cgroups/cpusets.txt + * for a background explanation of this. + * + * Does not return errors, on the theory that the callers of this + * routine would rather not worry about failures to rebuild sched + * domains when operating in the severe memory shortage situations + * that could cause allocation failures below. + * + * Must be called with cpuset_mutex held. + * + * The three key local variables below are: + * q - a linked-list queue of cpuset pointers, used to implement a + * top-down scan of all cpusets. This scan loads a pointer + * to each cpuset marked is_sched_load_balance into the + * array 'csa'. For our purposes, rebuilding the schedulers + * sched domains, we can ignore !is_sched_load_balance cpusets. + * csa - (for CpuSet Array) Array of pointers to all the cpusets + * that need to be load balanced, for convenient iterative + * access by the subsequent code that finds the best partition, + * i.e the set of domains (subsets) of CPUs such that the + * cpus_allowed of every cpuset marked is_sched_load_balance + * is a subset of one of these domains, while there are as + * many such domains as possible, each as small as possible. + * doms - Conversion of 'csa' to an array of cpumasks, for passing to + * the kernel/sched/core.c routine partition_sched_domains() in a + * convenient format, that can be easily compared to the prior + * value to determine what partition elements (sched domains) + * were changed (added or removed.) + * + * Finding the best partition (set of domains): + * The triple nested loops below over i, j, k scan over the + * load balanced cpusets (using the array of cpuset pointers in + * csa[]) looking for pairs of cpusets that have overlapping + * cpus_allowed, but which don't have the same 'pn' partition + * number and gives them in the same partition number. It keeps + * looping on the 'restart' label until it can no longer find + * any such pairs. + * + * The union of the cpus_allowed masks from the set of + * all cpusets having the same 'pn' value then form the one + * element of the partition (one sched domain) to be passed to + * partition_sched_domains(). */ +static int generate_sched_domains(cpumask_var_t **domains, + struct sched_domain_attr **attributes) +{ + struct cpuset *cp; /* scans q */ + struct cpuset **csa; /* array of all cpuset ptrs */ + int csn; /* how many cpuset ptrs in csa so far */ + int i, j, k; /* indices for partition finding loops */ + cpumask_var_t *doms; /* resulting partition; i.e. sched 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_subsys_state *pos_css; + + doms = NULL; + dattr = NULL; + csa = NULL; + + /* Special case for the 99% of systems with one, full, sched domain */ + if (is_sched_load_balance(&top_cpuset)) { + ndoms = 1; + doms = alloc_sched_domains(ndoms); + if (!doms) + goto done; -static void cpuset_release_agent(const char *pathbuf) -{ - char *argv[3], *envp[3]; - int i; + dattr = kmalloc(sizeof(struct sched_domain_attr), GFP_KERNEL); + if (dattr) { + *dattr = SD_ATTR_INIT; + update_domain_attr_tree(dattr, &top_cpuset); + } + cpumask_copy(doms[0], top_cpuset.cpus_allowed); - if (!pathbuf) - return; + goto done; + } + + csa = kmalloc(nr_cpusets() * sizeof(cp), GFP_KERNEL); + if (!csa) + goto done; + csn = 0; + + rcu_read_lock(); + 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 + * latter: All child cpusets contain a subset of the + * parent's cpus, so just skip them, and then we call + * update_domain_attr_tree() to calc relax_domain_level of + * the corresponding sched domain. + */ + if (!cpumask_empty(cp->cpus_allowed) && + !is_sched_load_balance(cp)) + continue; + + if (is_sched_load_balance(cp)) + csa[csn++] = cp; + + /* skip @cp's subtree */ + pos_css = css_rightmost_descendant(pos_css); + } + rcu_read_unlock(); + + for (i = 0; i < csn; i++) + csa[i]->pn = i; + ndoms = csn; + +restart: + /* Find the best partition (set of sched domains) */ + for (i = 0; i < csn; i++) { + struct cpuset *a = csa[i]; + int apn = a->pn; + + for (j = 0; j < csn; j++) { + struct cpuset *b = csa[j]; + int bpn = b->pn; + + if (apn != bpn && cpusets_overlap(a, b)) { + for (k = 0; k < csn; k++) { + struct cpuset *c = csa[k]; + + if (c->pn == bpn) + c->pn = apn; + } + ndoms--; /* one less element */ + goto restart; + } + } + } + + /* + * Now we know how many domains to create. + * Convert <csn, csa> to <ndoms, doms> and populate cpu masks. + */ + doms = alloc_sched_domains(ndoms); + if (!doms) + goto done; + + /* + * The rest of the code, including the scheduler, can deal with + * dattr==NULL case. No need to abort if alloc fails. + */ + dattr = kmalloc(ndoms * sizeof(struct sched_domain_attr), GFP_KERNEL); + + for (nslot = 0, i = 0; i < csn; i++) { + struct cpuset *a = csa[i]; + struct cpumask *dp; + int apn = a->pn; + + if (apn < 0) { + /* Skip completed partitions */ + continue; + } - i = 0; - argv[i++] = "/sbin/cpuset_release_agent"; - argv[i++] = (char *)pathbuf; - argv[i] = NULL; + dp = doms[nslot]; - i = 0; - /* minimal command environment */ - envp[i++] = "HOME=/"; - envp[i++] = "PATH=/sbin:/bin:/usr/sbin:/usr/bin"; - envp[i] = NULL; + if (nslot == ndoms) { + static int warnings = 10; + if (warnings) { + pr_warn("rebuild_sched_domains confused: nslot %d, ndoms %d, csn %d, i %d, apn %d\n", + nslot, ndoms, csn, i, apn); + warnings--; + } + continue; + } + + cpumask_clear(dp); + if (dattr) + *(dattr + nslot) = SD_ATTR_INIT; + for (j = i; j < csn; j++) { + struct cpuset *b = csa[j]; + + if (apn == b->pn) { + cpumask_or(dp, dp, b->cpus_allowed); + if (dattr) + update_domain_attr_tree(dattr + nslot, b); + + /* Done with this partition */ + b->pn = -1; + } + } + nslot++; + } + BUG_ON(nslot != ndoms); - call_usermodehelper(argv[0], argv, envp, 0); - kfree(pathbuf); +done: + kfree(csa); + + /* + * Fallback to the default domain if kmalloc() failed. + * See comments in partition_sched_domains(). + */ + if (doms == NULL) + ndoms = 1; + + *domains = doms; + *attributes = dattr; + return ndoms; } /* - * Either cs->count of using tasks transitioned to zero, or the - * cs->children list of child cpusets just became empty. If this - * cs is notify_on_release() and now both the user count is zero and - * the list of children is empty, prepare cpuset path in a kmalloc'd - * buffer, to be returned via ppathbuf, so that the caller can invoke - * cpuset_release_agent() with it later on, once manage_sem is dropped. - * Call here with manage_sem held. - * - * This check_for_release() routine is responsible for kmalloc'ing - * pathbuf. The above cpuset_release_agent() is responsible for - * kfree'ing pathbuf. The caller of these routines is responsible - * for providing a pathbuf pointer, initialized to NULL, then - * calling check_for_release() with manage_sem held and the address - * of the pathbuf pointer, then dropping manage_sem, then calling - * cpuset_release_agent() with pathbuf, as set by check_for_release(). + * Rebuild scheduler domains. + * + * If the flag 'sched_load_balance' of any cpuset with non-empty + * 'cpus' changes, or if the 'cpus' allowed changes in any cpuset + * which has that flag enabled, or if any cpuset with a non-empty + * 'cpus' is removed, then call this routine to rebuild the + * scheduler's dynamic sched domains. + * + * Call with cpuset_mutex held. Takes get_online_cpus(). */ +static void rebuild_sched_domains_locked(void) +{ + struct sched_domain_attr *attr; + cpumask_var_t *doms; + int ndoms; + + lockdep_assert_held(&cpuset_mutex); + get_online_cpus(); + + /* + * We have raced with CPU hotplug. Don't do anything to avoid + * passing doms with offlined cpu to partition_sched_domains(). + * Anyways, hotplug work item will rebuild sched domains. + */ + if (!cpumask_equal(top_cpuset.cpus_allowed, cpu_active_mask)) + goto out; -static void check_for_release(struct cpuset *cs, char **ppathbuf) + /* Generate domain masks and attrs */ + ndoms = generate_sched_domains(&doms, &attr); + + /* Have scheduler rebuild the domains */ + partition_sched_domains(ndoms, doms, attr); +out: + put_online_cpus(); +} +#else /* !CONFIG_SMP */ +static void rebuild_sched_domains_locked(void) { - if (notify_on_release(cs) && atomic_read(&cs->count) == 0 && - list_empty(&cs->children)) { - char *buf; +} +#endif /* CONFIG_SMP */ - buf = kmalloc(PAGE_SIZE, GFP_KERNEL); - if (!buf) - return; - if (cpuset_path(cs, buf, PAGE_SIZE) < 0) - kfree(buf); - else - *ppathbuf = buf; - } +void rebuild_sched_domains(void) +{ + mutex_lock(&cpuset_mutex); + rebuild_sched_domains_locked(); + mutex_unlock(&cpuset_mutex); } /* - * Return in *pmask the portion of a cpusets's cpus_allowed that - * are online. If none are online, walk up the cpuset hierarchy - * until we find one that does have some online cpus. If we get - * all the way to the top and still haven't found any online cpus, - * return cpu_online_map. Or if passed a NULL cs from an exit'ing - * task, return cpu_online_map. + * effective_cpumask_cpuset - return nearest ancestor with non-empty cpus + * @cs: the cpuset in interest * - * One way or another, we guarantee to return some non-empty subset - * of cpu_online_map. + * A cpuset's effective cpumask is the cpumask of the nearest ancestor + * with non-empty cpus. We use effective cpumask whenever: + * - we update tasks' cpus_allowed. (they take on the ancestor's cpumask + * if the cpuset they reside in has no cpus) + * - we want to retrieve task_cs(tsk)'s cpus_allowed. * - * Call with callback_sem held. + * Called with cpuset_mutex held. cpuset_cpus_allowed_fallback() is an + * exception. See comments there. */ - -static void guarantee_online_cpus(const struct cpuset *cs, cpumask_t *pmask) +static struct cpuset *effective_cpumask_cpuset(struct cpuset *cs) { - while (cs && !cpus_intersects(cs->cpus_allowed, cpu_online_map)) - cs = cs->parent; - if (cs) - cpus_and(*pmask, cs->cpus_allowed, cpu_online_map); - else - *pmask = cpu_online_map; - BUG_ON(!cpus_intersects(*pmask, cpu_online_map)); + while (cpumask_empty(cs->cpus_allowed)) + cs = parent_cs(cs); + return cs; } /* - * Return in *pmask the portion of a cpusets's mems_allowed that - * are online. If none are online, walk up the cpuset hierarchy - * until we find one that does have some online mems. If we get - * all the way to the top and still haven't found any online mems, - * return node_online_map. + * effective_nodemask_cpuset - return nearest ancestor with non-empty mems + * @cs: the cpuset in interest * - * One way or another, we guarantee to return some non-empty subset - * of node_online_map. + * A cpuset's effective nodemask is the nodemask of the nearest ancestor + * with non-empty memss. We use effective nodemask whenever: + * - we update tasks' mems_allowed. (they take on the ancestor's nodemask + * if the cpuset they reside in has no mems) + * - we want to retrieve task_cs(tsk)'s mems_allowed. * - * Call with callback_sem held. + * Called with cpuset_mutex held. */ - -static void guarantee_online_mems(const struct cpuset *cs, nodemask_t *pmask) +static struct cpuset *effective_nodemask_cpuset(struct cpuset *cs) { - while (cs && !nodes_intersects(cs->mems_allowed, node_online_map)) - cs = cs->parent; - if (cs) - nodes_and(*pmask, cs->mems_allowed, node_online_map); - else - *pmask = node_online_map; - BUG_ON(!nodes_intersects(*pmask, node_online_map)); + while (nodes_empty(cs->mems_allowed)) + cs = parent_cs(cs); + return cs; } /** - * cpuset_update_task_memory_state - update task memory placement - * - * If the current tasks cpusets mems_allowed changed behind our - * backs, update current->mems_allowed, mems_generation and task NUMA - * mempolicy to the new value. - * - * Task mempolicy is updated by rebinding it relative to the - * current->cpuset if a task has its memory placement changed. - * Do not call this routine if in_interrupt(). - * - * Call without callback_sem or task_lock() held. May be called - * with or without manage_sem held. Doesn't need task_lock to guard - * against another task changing a non-NULL cpuset pointer to NULL, - * as that is only done by a task on itself, and if the current task - * is here, it is not simultaneously in the exit code NULL'ing its - * cpuset pointer. This routine also might acquire callback_sem and - * current->mm->mmap_sem during call. - * - * Reading current->cpuset->mems_generation doesn't need task_lock - * to guard the current->cpuset derefence, because it is guarded - * from concurrent freeing of current->cpuset by attach_task(), - * using RCU. - * - * The rcu_dereference() is technically probably not needed, - * as I don't actually mind if I see a new cpuset pointer but - * an old value of mems_generation. However this really only - * matters on alpha systems using cpusets heavily. If I dropped - * that rcu_dereference(), it would save them a memory barrier. - * For all other arch's, rcu_dereference is a no-op anyway, and for - * alpha systems not using cpusets, another planned optimization, - * avoiding the rcu critical section for tasks in the root cpuset - * which is statically allocated, so can't vanish, will make this - * irrelevant. Better to use RCU as intended, than to engage in - * some cute trick to save a memory barrier that is impossible to - * test, for alpha systems using cpusets heavily, which might not - * even exist. - * - * This routine is needed to update the per-task mems_allowed data, - * within the tasks context, when it is trying to allocate memory - * (in various mm/mempolicy.c routines) and notices that some other - * task has been modifying its cpuset. + * 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 + * + * Iterate through each task of @cs updating its cpus_allowed to the + * effective cpuset's. As this function is called with cpuset_mutex held, + * cpuset membership stays stable. */ - -void cpuset_update_task_memory_state() +static void update_tasks_cpumask(struct cpuset *cs) { - int my_cpusets_mem_gen; - struct task_struct *tsk = current; - struct cpuset *cs; + struct cpuset *cpus_cs = effective_cpumask_cpuset(cs); + struct css_task_iter it; + struct task_struct *task; - if (tsk->cpuset == &top_cpuset) { - /* Don't need rcu for top_cpuset. It's never freed. */ - my_cpusets_mem_gen = top_cpuset.mems_generation; - } else { - rcu_read_lock(); - cs = rcu_dereference(tsk->cpuset); - my_cpusets_mem_gen = cs->mems_generation; + css_task_iter_start(&cs->css, &it); + while ((task = css_task_iter_next(&it))) + set_cpus_allowed_ptr(task, cpus_cs->cpus_allowed); + css_task_iter_end(&it); +} + +/* + * 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? + * + * This will update cpumasks of tasks in @root_cs and all other empty cpusets + * which take on cpumask of @root_cs. + * + * Called with cpuset_mutex held + */ +static void update_tasks_cpumask_hier(struct cpuset *root_cs, bool update_root) +{ + struct cpuset *cp; + struct cgroup_subsys_state *pos_css; + + rcu_read_lock(); + 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_online(&cp->css)) + continue; rcu_read_unlock(); + + update_tasks_cpumask(cp); + + rcu_read_lock(); + css_put(&cp->css); } + rcu_read_unlock(); +} + +/** + * update_cpumask - update the cpus_allowed mask of a cpuset and all tasks in it + * @cs: the cpuset to consider + * @trialcs: trial cpuset + * @buf: buffer of cpu numbers written to this cpuset + */ +static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs, + const char *buf) +{ + int retval; + int is_load_balanced; + + /* top_cpuset.cpus_allowed tracks cpu_online_mask; it's read-only */ + if (cs == &top_cpuset) + return -EACCES; + + /* + * An empty cpus_allowed is ok only if the cpuset has no tasks. + * Since cpulist_parse() fails on an empty mask, we special case + * that parsing. The validate_change() call ensures that cpusets + * with tasks have cpus. + */ + if (!*buf) { + cpumask_clear(trialcs->cpus_allowed); + } else { + retval = cpulist_parse(buf, trialcs->cpus_allowed); + if (retval < 0) + return retval; - if (my_cpusets_mem_gen != tsk->cpuset_mems_generation) { - down(&callback_sem); - task_lock(tsk); - cs = tsk->cpuset; /* Maybe changed when task not locked */ - guarantee_online_mems(cs, &tsk->mems_allowed); - tsk->cpuset_mems_generation = cs->mems_generation; - task_unlock(tsk); - up(&callback_sem); - mpol_rebind_task(tsk, &tsk->mems_allowed); + if (!cpumask_subset(trialcs->cpus_allowed, cpu_active_mask)) + return -EINVAL; } + + /* Nothing to do if the cpus didn't change */ + if (cpumask_equal(cs->cpus_allowed, trialcs->cpus_allowed)) + return 0; + + retval = validate_change(cs, trialcs); + if (retval < 0) + return retval; + + is_load_balanced = is_sched_load_balance(trialcs); + + mutex_lock(&callback_mutex); + cpumask_copy(cs->cpus_allowed, trialcs->cpus_allowed); + mutex_unlock(&callback_mutex); + + update_tasks_cpumask_hier(cs, true); + + if (is_load_balanced) + rebuild_sched_domains_locked(); + return 0; } /* - * is_cpuset_subset(p, q) - Is cpuset p a subset of cpuset q? + * cpuset_migrate_mm * - * One cpuset is a subset of another if all its allowed CPUs and - * Memory Nodes are a subset of the other, and its exclusive flags - * are only set if the other's are set. Call holding manage_sem. + * Migrate memory region from one set of nodes to another. + * + * Temporarilly set tasks mems_allowed to target nodes of migration, + * so that the migration code can allocate pages on these nodes. + * + * While the mm_struct we are migrating is typically from some + * other task, the task_struct mems_allowed that we are hacking + * is for our current task, which must allocate new pages for that + * migrating memory region. */ -static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q) +static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from, + const nodemask_t *to) { - return cpus_subset(p->cpus_allowed, q->cpus_allowed) && - nodes_subset(p->mems_allowed, q->mems_allowed) && - is_cpu_exclusive(p) <= is_cpu_exclusive(q) && - is_mem_exclusive(p) <= is_mem_exclusive(q); + struct task_struct *tsk = current; + struct cpuset *mems_cs; + + tsk->mems_allowed = *to; + + do_migrate_pages(mm, from, to, MPOL_MF_MOVE_ALL); + + rcu_read_lock(); + mems_cs = effective_nodemask_cpuset(task_cs(tsk)); + guarantee_online_mems(mems_cs, &tsk->mems_allowed); + rcu_read_unlock(); } /* - * validate_change() - Used to validate that any proposed cpuset change - * follows the structural rules for cpusets. + * cpuset_change_task_nodemask - change task's mems_allowed and mempolicy + * @tsk: the task to change + * @newmems: new nodes that the task will be set * - * If we replaced the flag and mask values of the current cpuset - * (cur) with those values in the trial cpuset (trial), would - * our various subset and exclusive rules still be valid? Presumes - * manage_sem held. - * - * 'cur' is the address of an actual, in-use cpuset. Operations - * such as list traversal that depend on the actual address of the - * cpuset in the list must use cur below, not trial. - * - * 'trial' is the address of bulk structure copy of cur, with - * perhaps one or more of the fields cpus_allowed, mems_allowed, - * or flags changed to new, trial values. - * - * Return 0 if valid, -errno if not. + * In order to avoid seeing no nodes if the old and new nodes are disjoint, + * we structure updates as setting all new allowed nodes, then clearing newly + * disallowed ones. */ - -static int validate_change(const struct cpuset *cur, const struct cpuset *trial) +static void cpuset_change_task_nodemask(struct task_struct *tsk, + nodemask_t *newmems) { - struct cpuset *c, *par; + bool need_loop; - /* Each of our child cpusets must be a subset of us */ - list_for_each_entry(c, &cur->children, sibling) { - if (!is_cpuset_subset(c, trial)) - return -EBUSY; + /* + * Allow tasks that have access to memory reserves because they have + * been OOM killed to get memory anywhere. + */ + if (unlikely(test_thread_flag(TIF_MEMDIE))) + return; + if (current->flags & PF_EXITING) /* Let dying task have memory */ + return; + + task_lock(tsk); + /* + * Determine if a loop is necessary if another thread is doing + * read_mems_allowed_begin(). If at least one node remains unchanged and + * tsk does not have a mempolicy, then an empty nodemask will not be + * possible when mems_allowed is larger than a word. + */ + need_loop = task_has_mempolicy(tsk) || + !nodes_intersects(*newmems, tsk->mems_allowed); + + if (need_loop) { + local_irq_disable(); + write_seqcount_begin(&tsk->mems_allowed_seq); } - /* Remaining checks don't apply to root cpuset */ - if ((par = cur->parent) == NULL) - return 0; + nodes_or(tsk->mems_allowed, tsk->mems_allowed, *newmems); + mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP1); - /* We must be a subset of our parent cpuset */ - if (!is_cpuset_subset(trial, par)) - return -EACCES; + mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP2); + tsk->mems_allowed = *newmems; - /* If either I or some sibling (!= me) is exclusive, we can't overlap */ - list_for_each_entry(c, &par->children, sibling) { - if ((is_cpu_exclusive(trial) || is_cpu_exclusive(c)) && - c != cur && - cpus_intersects(trial->cpus_allowed, c->cpus_allowed)) - return -EINVAL; - if ((is_mem_exclusive(trial) || is_mem_exclusive(c)) && - c != cur && - nodes_intersects(trial->mems_allowed, c->mems_allowed)) - return -EINVAL; + if (need_loop) { + write_seqcount_end(&tsk->mems_allowed_seq); + local_irq_enable(); } - return 0; + task_unlock(tsk); } -/* - * For a given cpuset cur, partition the system as follows - * a. All cpus in the parent cpuset's cpus_allowed that are not part of any - * exclusive child cpusets - * b. All cpus in the current cpuset's cpus_allowed that are not part of any - * exclusive child cpusets - * Build these two partitions by calling partition_sched_domains - * - * Call with manage_sem held. May nest a call to the - * lock_cpu_hotplug()/unlock_cpu_hotplug() pair. - */ +static void *cpuset_being_rebound; -static void update_cpu_domains(struct cpuset *cur) +/** + * 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 + * + * Iterate through each task of @cs updating its mems_allowed to the + * effective cpuset's. As this function is called with cpuset_mutex held, + * cpuset membership stays stable. + */ +static void update_tasks_nodemask(struct cpuset *cs) { - struct cpuset *c, *par = cur->parent; - cpumask_t pspan, cspan; + static nodemask_t newmems; /* protected by cpuset_mutex */ + struct cpuset *mems_cs = effective_nodemask_cpuset(cs); + struct css_task_iter it; + struct task_struct *task; - if (par == NULL || cpus_empty(cur->cpus_allowed)) - return; + cpuset_being_rebound = cs; /* causes mpol_dup() rebind */ + + guarantee_online_mems(mems_cs, &newmems); /* - * Get all cpus from parent's cpus_allowed not part of exclusive - * children + * The mpol_rebind_mm() call takes mmap_sem, which we couldn't + * take while holding tasklist_lock. Forks can happen - the + * mpol_dup() cpuset_being_rebound check will catch such forks, + * and rebind their vma mempolicies too. Because we still hold + * the global cpuset_mutex, we know that no other rebind effort + * will be contending for the global variable cpuset_being_rebound. + * It's ok if we rebind the same mm twice; mpol_rebind_mm() + * is idempotent. Also migrate pages in each mm to new nodes. */ - pspan = par->cpus_allowed; - list_for_each_entry(c, &par->children, sibling) { - if (is_cpu_exclusive(c)) - cpus_andnot(pspan, pspan, c->cpus_allowed); - } - if (is_removed(cur) || !is_cpu_exclusive(cur)) { - cpus_or(pspan, pspan, cur->cpus_allowed); - if (cpus_equal(pspan, cur->cpus_allowed)) - return; - cspan = CPU_MASK_NONE; - } else { - if (cpus_empty(pspan)) - return; - cspan = cur->cpus_allowed; - /* - * Get all cpus from current cpuset's cpus_allowed not part - * of exclusive children - */ - list_for_each_entry(c, &cur->children, sibling) { - if (is_cpu_exclusive(c)) - cpus_andnot(cspan, cspan, c->cpus_allowed); - } + css_task_iter_start(&cs->css, &it); + while ((task = css_task_iter_next(&it))) { + struct mm_struct *mm; + bool migrate; + + cpuset_change_task_nodemask(task, &newmems); + + mm = get_task_mm(task); + if (!mm) + continue; + + migrate = is_memory_migrate(cs); + + mpol_rebind_mm(mm, &cs->mems_allowed); + if (migrate) + cpuset_migrate_mm(mm, &cs->old_mems_allowed, &newmems); + mmput(mm); } + css_task_iter_end(&it); + + /* + * All the tasks' nodemasks have been updated, update + * cs->old_mems_allowed. + */ + cs->old_mems_allowed = newmems; - lock_cpu_hotplug(); - partition_sched_domains(&pspan, &cspan); - unlock_cpu_hotplug(); + /* We're done rebinding vmas to this cpuset's new mems_allowed. */ + cpuset_being_rebound = NULL; } /* - * Call with manage_sem held. May take callback_sem during call. + * 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? + * + * This will update nodemasks of tasks in @root_cs and all other empty cpusets + * which take on nodemask of @root_cs. + * + * Called with cpuset_mutex held */ +static void update_tasks_nodemask_hier(struct cpuset *root_cs, bool update_root) +{ + struct cpuset *cp; + struct cgroup_subsys_state *pos_css; + + rcu_read_lock(); + 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_online(&cp->css)) + continue; + rcu_read_unlock(); -static int update_cpumask(struct cpuset *cs, char *buf) -{ - struct cpuset trialcs; - int retval, cpus_unchanged; + update_tasks_nodemask(cp); - trialcs = *cs; - retval = cpulist_parse(buf, trialcs.cpus_allowed); - if (retval < 0) - return retval; - cpus_and(trialcs.cpus_allowed, trialcs.cpus_allowed, cpu_online_map); - if (cpus_empty(trialcs.cpus_allowed)) - return -ENOSPC; - retval = validate_change(cs, &trialcs); - if (retval < 0) - return retval; - cpus_unchanged = cpus_equal(cs->cpus_allowed, trialcs.cpus_allowed); - down(&callback_sem); - cs->cpus_allowed = trialcs.cpus_allowed; - up(&callback_sem); - if (is_cpu_exclusive(cs) && !cpus_unchanged) - update_cpu_domains(cs); - return 0; + rcu_read_lock(); + css_put(&cp->css); + } + rcu_read_unlock(); } /* * Handle user request to change the 'mems' memory placement * of a cpuset. Needs to validate the request, update the - * cpusets mems_allowed and mems_generation, and for each - * task in the cpuset, rebind any vma mempolicies and if - * the cpuset is marked 'memory_migrate', migrate the tasks - * pages to the new memory. + * cpusets mems_allowed, and for each task in the cpuset, + * update mems_allowed and rebind task's mempolicy and any vma + * mempolicies and if the cpuset is marked 'memory_migrate', + * migrate the tasks pages to the new memory. * - * Call with manage_sem held. May take callback_sem during call. + * Call with cpuset_mutex held. May take callback_mutex during call. * Will take tasklist_lock, scan tasklist for tasks in cpuset cs, * lock each such tasks mm->mmap_sem, scan its vma's and rebind * their mempolicies to the cpusets new mems_allowed. */ - -static int update_nodemask(struct cpuset *cs, char *buf) +static int update_nodemask(struct cpuset *cs, struct cpuset *trialcs, + const char *buf) { - struct cpuset trialcs; - nodemask_t oldmem; - struct task_struct *g, *p; - struct mm_struct **mmarray; - int i, n, ntasks; - int migrate; - int fudge; int retval; - trialcs = *cs; - retval = nodelist_parse(buf, trialcs.mems_allowed); - if (retval < 0) - goto done; - nodes_and(trialcs.mems_allowed, trialcs.mems_allowed, node_online_map); - oldmem = cs->mems_allowed; - if (nodes_equal(oldmem, trialcs.mems_allowed)) { - retval = 0; /* Too easy - nothing to do */ + /* + * top_cpuset.mems_allowed tracks node_stats[N_MEMORY]; + * it's read-only + */ + if (cs == &top_cpuset) { + retval = -EACCES; goto done; } - if (nodes_empty(trialcs.mems_allowed)) { - retval = -ENOSPC; + + /* + * An empty mems_allowed is ok iff there are no tasks in the cpuset. + * Since nodelist_parse() fails on an empty mask, we special case + * that parsing. The validate_change() call ensures that cpusets + * with tasks have memory. + */ + if (!*buf) { + nodes_clear(trialcs->mems_allowed); + } else { + retval = nodelist_parse(buf, trialcs->mems_allowed); + if (retval < 0) + goto done; + + if (!nodes_subset(trialcs->mems_allowed, + node_states[N_MEMORY])) { + retval = -EINVAL; + goto done; + } + } + + if (nodes_equal(cs->mems_allowed, trialcs->mems_allowed)) { + retval = 0; /* Too easy - nothing to do */ goto done; } - retval = validate_change(cs, &trialcs); + retval = validate_change(cs, trialcs); if (retval < 0) goto done; - down(&callback_sem); - cs->mems_allowed = trialcs.mems_allowed; - atomic_inc(&cpuset_mems_generation); - cs->mems_generation = atomic_read(&cpuset_mems_generation); - up(&callback_sem); + mutex_lock(&callback_mutex); + cs->mems_allowed = trialcs->mems_allowed; + mutex_unlock(&callback_mutex); - set_cpuset_being_rebound(cs); /* causes mpol_copy() rebind */ - - fudge = 10; /* spare mmarray[] slots */ - fudge += cpus_weight(cs->cpus_allowed); /* imagine one fork-bomb/cpu */ - retval = -ENOMEM; - - /* - * Allocate mmarray[] to hold mm reference for each task - * in cpuset cs. Can't kmalloc GFP_KERNEL while holding - * tasklist_lock. We could use GFP_ATOMIC, but with a - * few more lines of code, we can retry until we get a big - * enough mmarray[] w/o using GFP_ATOMIC. - */ - while (1) { - ntasks = atomic_read(&cs->count); /* guess */ - ntasks += fudge; - mmarray = kmalloc(ntasks * sizeof(*mmarray), GFP_KERNEL); - if (!mmarray) - goto done; - write_lock_irq(&tasklist_lock); /* block fork */ - if (atomic_read(&cs->count) <= ntasks) - break; /* got enough */ - write_unlock_irq(&tasklist_lock); /* try again */ - kfree(mmarray); - } + update_tasks_nodemask_hier(cs, true); +done: + return retval; +} - n = 0; +int current_cpuset_is_being_rebound(void) +{ + int ret; - /* Load up mmarray[] with mm reference for each task in cpuset. */ - do_each_thread(g, p) { - struct mm_struct *mm; + rcu_read_lock(); + ret = task_cs(current) == cpuset_being_rebound; + rcu_read_unlock(); - if (n >= ntasks) { - printk(KERN_WARNING - "Cpuset mempolicy rebind incomplete.\n"); - continue; - } - if (p->cpuset != cs) - continue; - mm = get_task_mm(p); - if (!mm) - continue; - mmarray[n++] = mm; - } while_each_thread(g, p); - write_unlock_irq(&tasklist_lock); + return ret; +} - /* - * Now that we've dropped the tasklist spinlock, we can - * rebind the vma mempolicies of each mm in mmarray[] to their - * new cpuset, and release that mm. The mpol_rebind_mm() - * call takes mmap_sem, which we couldn't take while holding - * tasklist_lock. Forks can happen again now - the mpol_copy() - * cpuset_being_rebound check will catch such forks, and rebind - * their vma mempolicies too. Because we still hold the global - * cpuset manage_sem, we know that no other rebind effort will - * be contending for the global variable cpuset_being_rebound. - * It's ok if we rebind the same mm twice; mpol_rebind_mm() - * is idempotent. Also migrate pages in each mm to new nodes. - */ - migrate = is_memory_migrate(cs); - for (i = 0; i < n; i++) { - struct mm_struct *mm = mmarray[i]; +static int update_relax_domain_level(struct cpuset *cs, s64 val) +{ +#ifdef CONFIG_SMP + if (val < -1 || val >= sched_domain_level_max) + return -EINVAL; +#endif - mpol_rebind_mm(mm, &cs->mems_allowed); - if (migrate) { - do_migrate_pages(mm, &oldmem, &cs->mems_allowed, - MPOL_MF_MOVE_ALL); - } - mmput(mm); + if (val != cs->relax_domain_level) { + cs->relax_domain_level = val; + if (!cpumask_empty(cs->cpus_allowed) && + is_sched_load_balance(cs)) + rebuild_sched_domains_locked(); } - /* We're done rebinding vma's to this cpusets new mems_allowed. */ - kfree(mmarray); - set_cpuset_being_rebound(NULL); - retval = 0; -done: - return retval; + return 0; } -/* - * Call with manage_sem held. +/** + * 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 + * + * Iterate through each task of @cs updating its spread flags. As this + * function is called with cpuset_mutex held, cpuset membership stays + * stable. */ - -static int update_memory_pressure_enabled(struct cpuset *cs, char *buf) +static void update_tasks_flags(struct cpuset *cs) { - if (simple_strtoul(buf, NULL, 10) != 0) - cpuset_memory_pressure_enabled = 1; - else - cpuset_memory_pressure_enabled = 0; - return 0; + struct css_task_iter it; + struct task_struct *task; + + css_task_iter_start(&cs->css, &it); + while ((task = css_task_iter_next(&it))) + cpuset_update_task_spread_flag(cs, task); + css_task_iter_end(&it); } /* * update_flag - read a 0 or a 1 in a file and update associated flag - * bit: the bit to update (CS_CPU_EXCLUSIVE, CS_MEM_EXCLUSIVE, - * CS_NOTIFY_ON_RELEASE, CS_MEMORY_MIGRATE) - * cs: the cpuset to update - * buf: the buffer where we read the 0 or 1 + * bit: the bit to update (see cpuset_flagbits_t) + * cs: the cpuset to update + * turning_on: whether the flag is being set or cleared * - * Call with manage_sem held. + * Call with cpuset_mutex held. */ -static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, char *buf) +static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, + int turning_on) { - int turning_on; - struct cpuset trialcs; - int err, cpu_exclusive_changed; + struct cpuset *trialcs; + int balance_flag_changed; + int spread_flag_changed; + int err; - turning_on = (simple_strtoul(buf, NULL, 10) != 0); + trialcs = alloc_trial_cpuset(cs); + if (!trialcs) + return -ENOMEM; - trialcs = *cs; if (turning_on) - set_bit(bit, &trialcs.flags); + set_bit(bit, &trialcs->flags); else - clear_bit(bit, &trialcs.flags); + clear_bit(bit, &trialcs->flags); - err = validate_change(cs, &trialcs); + err = validate_change(cs, trialcs); if (err < 0) - return err; - cpu_exclusive_changed = - (is_cpu_exclusive(cs) != is_cpu_exclusive(&trialcs)); - down(&callback_sem); - if (turning_on) - set_bit(bit, &cs->flags); - else - clear_bit(bit, &cs->flags); - up(&callback_sem); + goto out; - if (cpu_exclusive_changed) - update_cpu_domains(cs); - return 0; + balance_flag_changed = (is_sched_load_balance(cs) != + is_sched_load_balance(trialcs)); + + spread_flag_changed = ((is_spread_slab(cs) != is_spread_slab(trialcs)) + || (is_spread_page(cs) != is_spread_page(trialcs))); + + mutex_lock(&callback_mutex); + cs->flags = trialcs->flags; + mutex_unlock(&callback_mutex); + + if (!cpumask_empty(trialcs->cpus_allowed) && balance_flag_changed) + rebuild_sched_domains_locked(); + + if (spread_flag_changed) + update_tasks_flags(cs); +out: + free_trial_cpuset(trialcs); + return err; } /* - * Frequency meter - How fast is some event occuring? + * Frequency meter - How fast is some event occurring? * * These routines manage a digitally filtered, constant time based, * event frequency meter. There are four routines: @@ -1099,932 +1374,917 @@ static int fmeter_getrate(struct fmeter *fmp) return val; } -/* - * Attack task specified by pid in 'pidbuf' to cpuset 'cs', possibly - * writing the path of the old cpuset in 'ppathbuf' if it needs to be - * notified on release. - * - * Call holding manage_sem. May take callback_sem and task_lock of - * the task 'pid' during call. - */ +static struct cpuset *cpuset_attach_old_cs; -static int attach_task(struct cpuset *cs, char *pidbuf, char **ppathbuf) +/* Called by cgroups to determine if a cpuset is usable; cpuset_mutex held */ +static int cpuset_can_attach(struct cgroup_subsys_state *css, + struct cgroup_taskset *tset) { - pid_t pid; - struct task_struct *tsk; - struct cpuset *oldcs; - cpumask_t cpus; - nodemask_t from, to; - struct mm_struct *mm; + struct cpuset *cs = css_cs(css); + struct task_struct *task; + int ret; - if (sscanf(pidbuf, "%d", &pid) != 1) - return -EIO; - if (cpus_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed)) - return -ENOSPC; + /* used later by cpuset_attach() */ + cpuset_attach_old_cs = task_cs(cgroup_taskset_first(tset)); - if (pid) { - read_lock(&tasklist_lock); + mutex_lock(&cpuset_mutex); - tsk = find_task_by_pid(pid); - if (!tsk || tsk->flags & PF_EXITING) { - read_unlock(&tasklist_lock); - return -ESRCH; - } - - get_task_struct(tsk); - read_unlock(&tasklist_lock); + /* + * We allow to move tasks into an empty cpuset if sane_behavior + * flag is set. + */ + ret = -ENOSPC; + if (!cgroup_sane_behavior(css->cgroup) && + (cpumask_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed))) + goto out_unlock; - if ((current->euid) && (current->euid != tsk->uid) - && (current->euid != tsk->suid)) { - put_task_struct(tsk); - return -EACCES; - } - } else { - tsk = current; - get_task_struct(tsk); + cgroup_taskset_for_each(task, tset) { + /* + * Kthreads which disallow setaffinity shouldn't be moved + * to a new cpuset; we don't want to change their cpu + * affinity and isolating such threads by their set of + * allowed nodes is unnecessary. Thus, cpusets are not + * applicable for such threads. This prevents checking for + * success of set_cpus_allowed_ptr() on all attached tasks + * before cpus_allowed may be changed. + */ + ret = -EINVAL; + if (task->flags & PF_NO_SETAFFINITY) + goto out_unlock; + ret = security_task_setscheduler(task); + if (ret) + goto out_unlock; } - down(&callback_sem); + /* + * Mark attach is in progress. This makes validate_change() fail + * changes which zero cpus/mems_allowed. + */ + cs->attach_in_progress++; + ret = 0; +out_unlock: + mutex_unlock(&cpuset_mutex); + return ret; +} - task_lock(tsk); - oldcs = tsk->cpuset; - if (!oldcs) { - task_unlock(tsk); - up(&callback_sem); - put_task_struct(tsk); - return -ESRCH; - } - atomic_inc(&cs->count); - rcu_assign_pointer(tsk->cpuset, cs); - task_unlock(tsk); +static void cpuset_cancel_attach(struct cgroup_subsys_state *css, + struct cgroup_taskset *tset) +{ + mutex_lock(&cpuset_mutex); + css_cs(css)->attach_in_progress--; + mutex_unlock(&cpuset_mutex); +} - guarantee_online_cpus(cs, &cpus); - set_cpus_allowed(tsk, cpus); +/* + * Protected by cpuset_mutex. cpus_attach is used only by cpuset_attach() + * but we can't allocate it dynamically there. Define it global and + * allocate from cpuset_init(). + */ +static cpumask_var_t cpus_attach; - from = oldcs->mems_allowed; - to = cs->mems_allowed; +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 cpuset *cs = css_cs(css); + struct cpuset *oldcs = cpuset_attach_old_cs; + struct cpuset *cpus_cs = effective_cpumask_cpuset(cs); + struct cpuset *mems_cs = effective_nodemask_cpuset(cs); + + mutex_lock(&cpuset_mutex); + + /* prepare for attach */ + if (cs == &top_cpuset) + cpumask_copy(cpus_attach, cpu_possible_mask); + else + guarantee_online_cpus(cpus_cs, cpus_attach); - up(&callback_sem); + guarantee_online_mems(mems_cs, &cpuset_attach_nodemask_to); + + cgroup_taskset_for_each(task, tset) { + /* + * can_attach beforehand should guarantee that this doesn't + * fail. TODO: have a better way to handle failure here + */ + WARN_ON_ONCE(set_cpus_allowed_ptr(task, cpus_attach)); - mm = get_task_mm(tsk); + cpuset_change_task_nodemask(task, &cpuset_attach_nodemask_to); + cpuset_update_task_spread_flag(cs, task); + } + + /* + * Change mm, possibly for multiple threads in a threadgroup. This is + * expensive and may sleep. + */ + cpuset_attach_nodemask_to = cs->mems_allowed; + mm = get_task_mm(leader); if (mm) { - mpol_rebind_mm(mm, &to); + struct cpuset *mems_oldcs = effective_nodemask_cpuset(oldcs); + + mpol_rebind_mm(mm, &cpuset_attach_nodemask_to); + + /* + * old_mems_allowed is the same with mems_allowed here, except + * if this task is being moved automatically due to hotplug. + * In that case @mems_allowed has been updated and is empty, + * so @old_mems_allowed is the right nodesets that we migrate + * mm from. + */ + if (is_memory_migrate(cs)) { + cpuset_migrate_mm(mm, &mems_oldcs->old_mems_allowed, + &cpuset_attach_nodemask_to); + } mmput(mm); } - if (is_memory_migrate(cs)) - do_migrate_pages(tsk->mm, &from, &to, MPOL_MF_MOVE_ALL); - put_task_struct(tsk); - synchronize_rcu(); - if (atomic_dec_and_test(&oldcs->count)) - check_for_release(oldcs, ppathbuf); - return 0; + cs->old_mems_allowed = cpuset_attach_nodemask_to; + + cs->attach_in_progress--; + if (!cs->attach_in_progress) + wake_up(&cpuset_attach_wq); + + mutex_unlock(&cpuset_mutex); } /* The various types of files and directories in a cpuset file system */ typedef enum { - FILE_ROOT, - FILE_DIR, FILE_MEMORY_MIGRATE, FILE_CPULIST, FILE_MEMLIST, FILE_CPU_EXCLUSIVE, FILE_MEM_EXCLUSIVE, - FILE_NOTIFY_ON_RELEASE, + FILE_MEM_HARDWALL, + FILE_SCHED_LOAD_BALANCE, + FILE_SCHED_RELAX_DOMAIN_LEVEL, FILE_MEMORY_PRESSURE_ENABLED, FILE_MEMORY_PRESSURE, - FILE_TASKLIST, + FILE_SPREAD_PAGE, + FILE_SPREAD_SLAB, } cpuset_filetype_t; -static ssize_t cpuset_common_file_write(struct file *file, const char __user *userbuf, - size_t nbytes, loff_t *unused_ppos) +static int cpuset_write_u64(struct cgroup_subsys_state *css, struct cftype *cft, + u64 val) { - struct cpuset *cs = __d_cs(file->f_dentry->d_parent); - struct cftype *cft = __d_cft(file->f_dentry); + struct cpuset *cs = css_cs(css); cpuset_filetype_t type = cft->private; - char *buffer; - char *pathbuf = NULL; int retval = 0; - /* Crude upper limit on largest legitimate cpulist user might write. */ - if (nbytes > 100 + 6 * NR_CPUS) - return -E2BIG; - - /* +1 for nul-terminator */ - if ((buffer = kmalloc(nbytes + 1, GFP_KERNEL)) == 0) - return -ENOMEM; - - if (copy_from_user(buffer, userbuf, nbytes)) { - retval = -EFAULT; - goto out1; - } - buffer[nbytes] = 0; /* nul-terminate */ - - down(&manage_sem); - - if (is_removed(cs)) { + mutex_lock(&cpuset_mutex); + if (!is_cpuset_online(cs)) { retval = -ENODEV; - goto out2; + goto out_unlock; } switch (type) { - case FILE_CPULIST: - retval = update_cpumask(cs, buffer); - break; - case FILE_MEMLIST: - retval = update_nodemask(cs, buffer); - break; case FILE_CPU_EXCLUSIVE: - retval = update_flag(CS_CPU_EXCLUSIVE, cs, buffer); + retval = update_flag(CS_CPU_EXCLUSIVE, cs, val); break; case FILE_MEM_EXCLUSIVE: - retval = update_flag(CS_MEM_EXCLUSIVE, cs, buffer); + retval = update_flag(CS_MEM_EXCLUSIVE, cs, val); break; - case FILE_NOTIFY_ON_RELEASE: - retval = update_flag(CS_NOTIFY_ON_RELEASE, cs, buffer); + case FILE_MEM_HARDWALL: + retval = update_flag(CS_MEM_HARDWALL, cs, val); + break; + case FILE_SCHED_LOAD_BALANCE: + retval = update_flag(CS_SCHED_LOAD_BALANCE, cs, val); break; case FILE_MEMORY_MIGRATE: - retval = update_flag(CS_MEMORY_MIGRATE, cs, buffer); + retval = update_flag(CS_MEMORY_MIGRATE, cs, val); break; case FILE_MEMORY_PRESSURE_ENABLED: - retval = update_memory_pressure_enabled(cs, buffer); + cpuset_memory_pressure_enabled = !!val; break; case FILE_MEMORY_PRESSURE: retval = -EACCES; break; - case FILE_TASKLIST: - retval = attach_task(cs, buffer, &pathbuf); + case FILE_SPREAD_PAGE: + retval = update_flag(CS_SPREAD_PAGE, cs, val); + break; + case FILE_SPREAD_SLAB: + retval = update_flag(CS_SPREAD_SLAB, cs, val); break; default: retval = -EINVAL; - goto out2; + break; } - - if (retval == 0) - retval = nbytes; -out2: - up(&manage_sem); - cpuset_release_agent(pathbuf); -out1: - kfree(buffer); +out_unlock: + mutex_unlock(&cpuset_mutex); return retval; } -static ssize_t cpuset_file_write(struct file *file, const char __user *buf, - size_t nbytes, loff_t *ppos) +static int cpuset_write_s64(struct cgroup_subsys_state *css, struct cftype *cft, + s64 val) { - ssize_t retval = 0; - struct cftype *cft = __d_cft(file->f_dentry); - if (!cft) - return -ENODEV; + struct cpuset *cs = css_cs(css); + cpuset_filetype_t type = cft->private; + int retval = -ENODEV; - /* special function ? */ - if (cft->write) - retval = cft->write(file, buf, nbytes, ppos); - else - retval = cpuset_common_file_write(file, buf, nbytes, ppos); + mutex_lock(&cpuset_mutex); + if (!is_cpuset_online(cs)) + goto out_unlock; + switch (type) { + case FILE_SCHED_RELAX_DOMAIN_LEVEL: + retval = update_relax_domain_level(cs, val); + break; + default: + retval = -EINVAL; + break; + } +out_unlock: + mutex_unlock(&cpuset_mutex); return retval; } /* - * These ascii lists should be read in a single call, by using a user - * buffer large enough to hold the entire map. If read in smaller - * chunks, there is no guarantee of atomicity. Since the display format - * used, list of ranges of sequential numbers, is variable length, - * and since these maps can change value dynamically, one could read - * gibberish by doing partial reads while a list was changing. - * A single large read to a buffer that crosses a page boundary is - * ok, because the result being copied to user land is not recomputed - * across a page fault. + * Common handling for a write to a "cpus" or "mems" file. */ - -static int cpuset_sprintf_cpulist(char *page, struct cpuset *cs) +static ssize_t cpuset_write_resmask(struct kernfs_open_file *of, + char *buf, size_t nbytes, loff_t off) { - cpumask_t mask; + struct cpuset *cs = css_cs(of_css(of)); + struct cpuset *trialcs; + int retval = -ENODEV; - down(&callback_sem); - mask = cs->cpus_allowed; - up(&callback_sem); + buf = strstrip(buf); - return cpulist_scnprintf(page, PAGE_SIZE, mask); -} - -static int cpuset_sprintf_memlist(char *page, struct cpuset *cs) -{ - nodemask_t mask; + /* + * CPU or memory hotunplug may leave @cs w/o any execution + * resources, in which case the hotplug code asynchronously updates + * configuration and transfers all tasks to the nearest ancestor + * which can execute. + * + * As writes to "cpus" or "mems" may restore @cs's execution + * resources, wait for the previously scheduled operations before + * proceeding, so that we don't end up keep removing tasks added + * after execution capability is restored. + * + * cpuset_hotplug_work calls back into cgroup core via + * cgroup_transfer_tasks() and waiting for it from a cgroupfs + * operation like this one can lead to a deadlock through kernfs + * active_ref protection. Let's break the protection. Losing the + * protection is okay as we check whether @cs is online after + * grabbing cpuset_mutex anyway. This only happens on the legacy + * hierarchies. + */ + css_get(&cs->css); + kernfs_break_active_protection(of->kn); + flush_work(&cpuset_hotplug_work); + + mutex_lock(&cpuset_mutex); + if (!is_cpuset_online(cs)) + goto out_unlock; + + trialcs = alloc_trial_cpuset(cs); + if (!trialcs) { + retval = -ENOMEM; + goto out_unlock; + } - down(&callback_sem); - mask = cs->mems_allowed; - up(&callback_sem); + switch (of_cft(of)->private) { + case FILE_CPULIST: + retval = update_cpumask(cs, trialcs, buf); + break; + case FILE_MEMLIST: + retval = update_nodemask(cs, trialcs, buf); + break; + default: + retval = -EINVAL; + break; + } - return nodelist_scnprintf(page, PAGE_SIZE, mask); + free_trial_cpuset(trialcs); +out_unlock: + mutex_unlock(&cpuset_mutex); + kernfs_unbreak_active_protection(of->kn); + css_put(&cs->css); + return retval ?: nbytes; } -static ssize_t cpuset_common_file_read(struct file *file, char __user *buf, - size_t nbytes, loff_t *ppos) +/* + * These ascii lists should be read in a single call, by using a user + * buffer large enough to hold the entire map. If read in smaller + * chunks, there is no guarantee of atomicity. Since the display format + * used, list of ranges of sequential numbers, is variable length, + * and since these maps can change value dynamically, one could read + * gibberish by doing partial reads while a list was changing. + */ +static int cpuset_common_seq_show(struct seq_file *sf, void *v) { - struct cftype *cft = __d_cft(file->f_dentry); - struct cpuset *cs = __d_cs(file->f_dentry->d_parent); - cpuset_filetype_t type = cft->private; - char *page; - ssize_t retval = 0; - char *s; + struct cpuset *cs = css_cs(seq_css(sf)); + cpuset_filetype_t type = seq_cft(sf)->private; + ssize_t count; + char *buf, *s; + int ret = 0; - if (!(page = (char *)__get_free_page(GFP_KERNEL))) - return -ENOMEM; + count = seq_get_buf(sf, &buf); + s = buf; - s = page; + mutex_lock(&callback_mutex); switch (type) { case FILE_CPULIST: - s += cpuset_sprintf_cpulist(s, cs); + s += cpulist_scnprintf(s, count, cs->cpus_allowed); break; case FILE_MEMLIST: - s += cpuset_sprintf_memlist(s, cs); + s += nodelist_scnprintf(s, count, cs->mems_allowed); break; + default: + ret = -EINVAL; + goto out_unlock; + } + + if (s < buf + count - 1) { + *s++ = '\n'; + seq_commit(sf, s - buf); + } else { + seq_commit(sf, -1); + } +out_unlock: + mutex_unlock(&callback_mutex); + return ret; +} + +static u64 cpuset_read_u64(struct cgroup_subsys_state *css, struct cftype *cft) +{ + struct cpuset *cs = css_cs(css); + cpuset_filetype_t type = cft->private; + switch (type) { case FILE_CPU_EXCLUSIVE: - *s++ = is_cpu_exclusive(cs) ? '1' : '0'; - break; + return is_cpu_exclusive(cs); case FILE_MEM_EXCLUSIVE: - *s++ = is_mem_exclusive(cs) ? '1' : '0'; - break; - case FILE_NOTIFY_ON_RELEASE: - *s++ = notify_on_release(cs) ? '1' : '0'; - break; + return is_mem_exclusive(cs); + case FILE_MEM_HARDWALL: + return is_mem_hardwall(cs); + case FILE_SCHED_LOAD_BALANCE: + return is_sched_load_balance(cs); case FILE_MEMORY_MIGRATE: - *s++ = is_memory_migrate(cs) ? '1' : '0'; - break; + return is_memory_migrate(cs); case FILE_MEMORY_PRESSURE_ENABLED: - *s++ = cpuset_memory_pressure_enabled ? '1' : '0'; - break; + return cpuset_memory_pressure_enabled; case FILE_MEMORY_PRESSURE: - s += sprintf(s, "%d", fmeter_getrate(&cs->fmeter)); - break; + return fmeter_getrate(&cs->fmeter); + case FILE_SPREAD_PAGE: + return is_spread_page(cs); + case FILE_SPREAD_SLAB: + return is_spread_slab(cs); default: - retval = -EINVAL; - goto out; + BUG(); } - *s++ = '\n'; - - retval = simple_read_from_buffer(buf, nbytes, ppos, page, s - page); -out: - free_page((unsigned long)page); - return retval; -} - -static ssize_t cpuset_file_read(struct file *file, char __user *buf, size_t nbytes, - loff_t *ppos) -{ - ssize_t retval = 0; - struct cftype *cft = __d_cft(file->f_dentry); - if (!cft) - return -ENODEV; - /* special function ? */ - if (cft->read) - retval = cft->read(file, buf, nbytes, ppos); - else - retval = cpuset_common_file_read(file, buf, nbytes, ppos); - - return retval; + /* Unreachable but makes gcc happy */ + return 0; } -static int cpuset_file_open(struct inode *inode, struct file *file) +static s64 cpuset_read_s64(struct cgroup_subsys_state *css, struct cftype *cft) { - int err; - struct cftype *cft; - - err = generic_file_open(inode, file); - if (err) - return err; - - cft = __d_cft(file->f_dentry); - if (!cft) - return -ENODEV; - if (cft->open) - err = cft->open(inode, file); - else - err = 0; - - return err; -} + struct cpuset *cs = css_cs(css); + cpuset_filetype_t type = cft->private; + switch (type) { + case FILE_SCHED_RELAX_DOMAIN_LEVEL: + return cs->relax_domain_level; + default: + BUG(); + } -static int cpuset_file_release(struct inode *inode, struct file *file) -{ - struct cftype *cft = __d_cft(file->f_dentry); - if (cft->release) - return cft->release(inode, file); + /* Unrechable but makes gcc happy */ return 0; } + /* - * cpuset_rename - Only allow simple rename of directories in place. + * for the common functions, 'private' gives the type of file */ -static int cpuset_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry) -{ - if (!S_ISDIR(old_dentry->d_inode->i_mode)) - return -ENOTDIR; - if (new_dentry->d_inode) - return -EEXIST; - if (old_dir != new_dir) - return -EIO; - return simple_rename(old_dir, old_dentry, new_dir, new_dentry); -} - -static struct file_operations cpuset_file_operations = { - .read = cpuset_file_read, - .write = cpuset_file_write, - .llseek = generic_file_llseek, - .open = cpuset_file_open, - .release = cpuset_file_release, -}; -static struct inode_operations cpuset_dir_inode_operations = { - .lookup = simple_lookup, - .mkdir = cpuset_mkdir, - .rmdir = cpuset_rmdir, - .rename = cpuset_rename, +static struct cftype files[] = { + { + .name = "cpus", + .seq_show = cpuset_common_seq_show, + .write = cpuset_write_resmask, + .max_write_len = (100U + 6 * NR_CPUS), + .private = FILE_CPULIST, + }, + + { + .name = "mems", + .seq_show = cpuset_common_seq_show, + .write = cpuset_write_resmask, + .max_write_len = (100U + 6 * MAX_NUMNODES), + .private = FILE_MEMLIST, + }, + + { + .name = "cpu_exclusive", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_CPU_EXCLUSIVE, + }, + + { + .name = "mem_exclusive", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_MEM_EXCLUSIVE, + }, + + { + .name = "mem_hardwall", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_MEM_HARDWALL, + }, + + { + .name = "sched_load_balance", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_SCHED_LOAD_BALANCE, + }, + + { + .name = "sched_relax_domain_level", + .read_s64 = cpuset_read_s64, + .write_s64 = cpuset_write_s64, + .private = FILE_SCHED_RELAX_DOMAIN_LEVEL, + }, + + { + .name = "memory_migrate", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_MEMORY_MIGRATE, + }, + + { + .name = "memory_pressure", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_MEMORY_PRESSURE, + .mode = S_IRUGO, + }, + + { + .name = "memory_spread_page", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_SPREAD_PAGE, + }, + + { + .name = "memory_spread_slab", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_SPREAD_SLAB, + }, + + { + .name = "memory_pressure_enabled", + .flags = CFTYPE_ONLY_ON_ROOT, + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_MEMORY_PRESSURE_ENABLED, + }, + + { } /* terminate */ }; -static int cpuset_create_file(struct dentry *dentry, int mode) -{ - struct inode *inode; - - if (!dentry) - return -ENOENT; - if (dentry->d_inode) - return -EEXIST; +/* + * cpuset_css_alloc - allocate a cpuset css + * cgrp: control group that the new cpuset will be part of + */ - inode = cpuset_new_inode(mode); - if (!inode) - return -ENOMEM; +static struct cgroup_subsys_state * +cpuset_css_alloc(struct cgroup_subsys_state *parent_css) +{ + struct cpuset *cs; - if (S_ISDIR(mode)) { - inode->i_op = &cpuset_dir_inode_operations; - inode->i_fop = &simple_dir_operations; + if (!parent_css) + return &top_cpuset.css; - /* start off with i_nlink == 2 (for "." entry) */ - inode->i_nlink++; - } else if (S_ISREG(mode)) { - inode->i_size = 0; - inode->i_fop = &cpuset_file_operations; + cs = kzalloc(sizeof(*cs), GFP_KERNEL); + if (!cs) + return ERR_PTR(-ENOMEM); + if (!alloc_cpumask_var(&cs->cpus_allowed, GFP_KERNEL)) { + kfree(cs); + return ERR_PTR(-ENOMEM); } - d_instantiate(dentry, inode); - dget(dentry); /* Extra count - pin the dentry in core */ - return 0; -} - -/* - * cpuset_create_dir - create a directory for an object. - * cs: the cpuset we create the directory for. - * It must have a valid ->parent field - * And we are going to fill its ->dentry field. - * name: The name to give to the cpuset directory. Will be copied. - * mode: mode to set on new directory. - */ - -static int cpuset_create_dir(struct cpuset *cs, const char *name, int mode) -{ - struct dentry *dentry = NULL; - struct dentry *parent; - int error = 0; - - parent = cs->parent->dentry; - dentry = cpuset_get_dentry(parent, name); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - error = cpuset_create_file(dentry, S_IFDIR | mode); - if (!error) { - dentry->d_fsdata = cs; - parent->d_inode->i_nlink++; - cs->dentry = dentry; - } - dput(dentry); + set_bit(CS_SCHED_LOAD_BALANCE, &cs->flags); + cpumask_clear(cs->cpus_allowed); + nodes_clear(cs->mems_allowed); + fmeter_init(&cs->fmeter); + cs->relax_domain_level = -1; - return error; + return &cs->css; } -static int cpuset_add_file(struct dentry *dir, const struct cftype *cft) +static int cpuset_css_online(struct cgroup_subsys_state *css) { - struct dentry *dentry; - int error; - - mutex_lock(&dir->d_inode->i_mutex); - dentry = cpuset_get_dentry(dir, cft->name); - if (!IS_ERR(dentry)) { - error = cpuset_create_file(dentry, 0644 | S_IFREG); - if (!error) - dentry->d_fsdata = (void *)cft; - dput(dentry); - } else - error = PTR_ERR(dentry); - mutex_unlock(&dir->d_inode->i_mutex); - return error; -} + struct cpuset *cs = css_cs(css); + struct cpuset *parent = parent_cs(cs); + struct cpuset *tmp_cs; + struct cgroup_subsys_state *pos_css; -/* - * Stuff for reading the 'tasks' file. - * - * Reading this file can return large amounts of data if a cpuset has - * *lots* of attached tasks. So it may need several calls to read(), - * but we cannot guarantee that the information we produce is correct - * unless we produce it entirely atomically. - * - * Upon tasks file open(), a struct ctr_struct is allocated, that - * will have a pointer to an array (also allocated here). The struct - * ctr_struct * is stored in file->private_data. Its resources will - * be freed by release() when the file is closed. The array is used - * to sprintf the PIDs and then used by read(). - */ + if (!parent) + return 0; -/* cpusets_tasks_read array */ + mutex_lock(&cpuset_mutex); -struct ctr_struct { - char *buf; - int bufsz; -}; + set_bit(CS_ONLINE, &cs->flags); + if (is_spread_page(parent)) + set_bit(CS_SPREAD_PAGE, &cs->flags); + if (is_spread_slab(parent)) + set_bit(CS_SPREAD_SLAB, &cs->flags); -/* - * Load into 'pidarray' up to 'npids' of the tasks using cpuset 'cs'. - * Return actual number of pids loaded. No need to task_lock(p) - * when reading out p->cpuset, as we don't really care if it changes - * on the next cycle, and we are not going to try to dereference it. - */ -static int pid_array_load(pid_t *pidarray, int npids, struct cpuset *cs) -{ - int n = 0; - struct task_struct *g, *p; + cpuset_inc(); - read_lock(&tasklist_lock); + if (!test_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags)) + goto out_unlock; - do_each_thread(g, p) { - if (p->cpuset == cs) { - pidarray[n++] = p->pid; - if (unlikely(n == npids)) - goto array_full; + /* + * Clone @parent's configuration if CGRP_CPUSET_CLONE_CHILDREN is + * set. This flag handling is implemented in cgroup core for + * histrical reasons - the flag may be specified during mount. + * + * Currently, if any sibling cpusets have exclusive cpus or mem, we + * refuse to clone the configuration - thereby refusing the task to + * be entered, and as a result refusing the sys_unshare() or + * clone() which initiated it. If this becomes a problem for some + * users who wish to allow that scenario, then this could be + * changed to grant parent->cpus_allowed-sibling_cpus_exclusive + * (and likewise for mems) to the new cgroup. + */ + rcu_read_lock(); + 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; } - } while_each_thread(g, p); - -array_full: - read_unlock(&tasklist_lock); - return n; -} - -static int cmppid(const void *a, const void *b) -{ - return *(pid_t *)a - *(pid_t *)b; + } + rcu_read_unlock(); + + mutex_lock(&callback_mutex); + cs->mems_allowed = parent->mems_allowed; + cpumask_copy(cs->cpus_allowed, parent->cpus_allowed); + mutex_unlock(&callback_mutex); +out_unlock: + mutex_unlock(&cpuset_mutex); + return 0; } /* - * Convert array 'a' of 'npids' pid_t's to a string of newline separated - * decimal pids in 'buf'. Don't write more than 'sz' chars, but return - * count 'cnt' of how many chars would be written if buf were large enough. + * 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 int pid_array_to_buf(char *buf, int sz, pid_t *a, int npids) -{ - int cnt = 0; - int i; - for (i = 0; i < npids; i++) - cnt += snprintf(buf + cnt, max(sz - cnt, 0), "%d\n", a[i]); - return cnt; -} - -/* - * Handle an open on 'tasks' file. Prepare a buffer listing the - * process id's of tasks currently attached to the cpuset being opened. - * - * Does not require any specific cpuset semaphores, and does not take any. - */ -static int cpuset_tasks_open(struct inode *unused, struct file *file) +static void cpuset_css_offline(struct cgroup_subsys_state *css) { - struct cpuset *cs = __d_cs(file->f_dentry->d_parent); - struct ctr_struct *ctr; - pid_t *pidarray; - int npids; - char c; + struct cpuset *cs = css_cs(css); - if (!(file->f_mode & FMODE_READ)) - return 0; + mutex_lock(&cpuset_mutex); - ctr = kmalloc(sizeof(*ctr), GFP_KERNEL); - if (!ctr) - goto err0; + if (is_sched_load_balance(cs)) + update_flag(CS_SCHED_LOAD_BALANCE, cs, 0); - /* - * If cpuset gets more users after we read count, we won't have - * enough space - tough. This race is indistinguishable to the - * caller from the case that the additional cpuset users didn't - * show up until sometime later on. - */ - npids = atomic_read(&cs->count); - pidarray = kmalloc(npids * sizeof(pid_t), GFP_KERNEL); - if (!pidarray) - goto err1; - - npids = pid_array_load(pidarray, npids, cs); - sort(pidarray, npids, sizeof(pid_t), cmppid, NULL); - - /* Call pid_array_to_buf() twice, first just to get bufsz */ - ctr->bufsz = pid_array_to_buf(&c, sizeof(c), pidarray, npids) + 1; - ctr->buf = kmalloc(ctr->bufsz, GFP_KERNEL); - if (!ctr->buf) - goto err2; - ctr->bufsz = pid_array_to_buf(ctr->buf, ctr->bufsz, pidarray, npids); - - kfree(pidarray); - file->private_data = ctr; - return 0; + cpuset_dec(); + clear_bit(CS_ONLINE, &cs->flags); -err2: - kfree(pidarray); -err1: - kfree(ctr); -err0: - return -ENOMEM; + mutex_unlock(&cpuset_mutex); } -static ssize_t cpuset_tasks_read(struct file *file, char __user *buf, - size_t nbytes, loff_t *ppos) +static void cpuset_css_free(struct cgroup_subsys_state *css) { - struct ctr_struct *ctr = file->private_data; + struct cpuset *cs = css_cs(css); - if (*ppos + nbytes > ctr->bufsz) - nbytes = ctr->bufsz - *ppos; - if (copy_to_user(buf, ctr->buf + *ppos, nbytes)) - return -EFAULT; - *ppos += nbytes; - return nbytes; -} - -static int cpuset_tasks_release(struct inode *unused_inode, struct file *file) -{ - struct ctr_struct *ctr; - - if (file->f_mode & FMODE_READ) { - ctr = file->private_data; - kfree(ctr->buf); - kfree(ctr); - } - return 0; + free_cpumask_var(cs->cpus_allowed); + kfree(cs); } -/* - * for the common functions, 'private' gives the type of file - */ - -static struct cftype cft_tasks = { - .name = "tasks", - .open = cpuset_tasks_open, - .read = cpuset_tasks_read, - .release = cpuset_tasks_release, - .private = FILE_TASKLIST, +struct cgroup_subsys cpuset_cgrp_subsys = { + .css_alloc = cpuset_css_alloc, + .css_online = cpuset_css_online, + .css_offline = cpuset_css_offline, + .css_free = cpuset_css_free, + .can_attach = cpuset_can_attach, + .cancel_attach = cpuset_cancel_attach, + .attach = cpuset_attach, + .base_cftypes = files, + .early_init = 1, }; -static struct cftype cft_cpus = { - .name = "cpus", - .private = FILE_CPULIST, -}; +/** + * cpuset_init - initialize cpusets at system boot + * + * Description: Initialize top_cpuset and the cpuset internal file system, + **/ -static struct cftype cft_mems = { - .name = "mems", - .private = FILE_MEMLIST, -}; +int __init cpuset_init(void) +{ + int err = 0; -static struct cftype cft_cpu_exclusive = { - .name = "cpu_exclusive", - .private = FILE_CPU_EXCLUSIVE, -}; + if (!alloc_cpumask_var(&top_cpuset.cpus_allowed, GFP_KERNEL)) + BUG(); -static struct cftype cft_mem_exclusive = { - .name = "mem_exclusive", - .private = FILE_MEM_EXCLUSIVE, -}; + cpumask_setall(top_cpuset.cpus_allowed); + nodes_setall(top_cpuset.mems_allowed); -static struct cftype cft_notify_on_release = { - .name = "notify_on_release", - .private = FILE_NOTIFY_ON_RELEASE, -}; + fmeter_init(&top_cpuset.fmeter); + set_bit(CS_SCHED_LOAD_BALANCE, &top_cpuset.flags); + top_cpuset.relax_domain_level = -1; -static struct cftype cft_memory_migrate = { - .name = "memory_migrate", - .private = FILE_MEMORY_MIGRATE, -}; + err = register_filesystem(&cpuset_fs_type); + if (err < 0) + return err; -static struct cftype cft_memory_pressure_enabled = { - .name = "memory_pressure_enabled", - .private = FILE_MEMORY_PRESSURE_ENABLED, -}; + if (!alloc_cpumask_var(&cpus_attach, GFP_KERNEL)) + BUG(); -static struct cftype cft_memory_pressure = { - .name = "memory_pressure", - .private = FILE_MEMORY_PRESSURE, -}; + return 0; +} -static int cpuset_populate_dir(struct dentry *cs_dentry) +/* + * If CPU and/or memory hotplug handlers, below, unplug any CPUs + * or memory nodes, we need to walk over the cpuset hierarchy, + * removing that CPU or node from all cpusets. If this removes the + * last CPU or node from a cpuset, then move the tasks in the empty + * cpuset to its next-highest non-empty parent. + */ +static void remove_tasks_in_empty_cpuset(struct cpuset *cs) { - int err; + struct cpuset *parent; - if ((err = cpuset_add_file(cs_dentry, &cft_cpus)) < 0) - return err; - if ((err = cpuset_add_file(cs_dentry, &cft_mems)) < 0) - return err; - if ((err = cpuset_add_file(cs_dentry, &cft_cpu_exclusive)) < 0) - return err; - if ((err = cpuset_add_file(cs_dentry, &cft_mem_exclusive)) < 0) - return err; - if ((err = cpuset_add_file(cs_dentry, &cft_notify_on_release)) < 0) - return err; - if ((err = cpuset_add_file(cs_dentry, &cft_memory_migrate)) < 0) - return err; - if ((err = cpuset_add_file(cs_dentry, &cft_memory_pressure)) < 0) - return err; - if ((err = cpuset_add_file(cs_dentry, &cft_tasks)) < 0) - return err; - return 0; + /* + * Find its next-highest non-empty parent, (top cpuset + * has online cpus, so can't be empty). + */ + parent = parent_cs(cs); + while (cpumask_empty(parent->cpus_allowed) || + nodes_empty(parent->mems_allowed)) + parent = parent_cs(parent); + + if (cgroup_transfer_tasks(parent->css.cgroup, cs->css.cgroup)) { + pr_err("cpuset: failed to transfer tasks out of empty cpuset "); + pr_cont_cgroup_name(cs->css.cgroup); + pr_cont("\n"); + } } -/* - * cpuset_create - create a cpuset - * parent: cpuset that will be parent of the new cpuset. - * name: name of the new cpuset. Will be strcpy'ed. - * mode: mode to set on new inode +/** + * cpuset_hotplug_update_tasks - update tasks in a cpuset for hotunplug + * @cs: cpuset in interest * - * Must be called with the semaphore on the parent inode held + * Compare @cs's cpu and mem masks against top_cpuset and if some have gone + * offline, update @cs accordingly. If @cs ends up with no CPU or memory, + * all its tasks are moved to the nearest ancestor with both resources. */ - -static long cpuset_create(struct cpuset *parent, const char *name, int mode) +static void cpuset_hotplug_update_tasks(struct cpuset *cs) { - struct cpuset *cs; - int err; + static cpumask_t off_cpus; + static nodemask_t off_mems; + bool is_empty; + bool sane = cgroup_sane_behavior(cs->css.cgroup); - cs = kmalloc(sizeof(*cs), GFP_KERNEL); - if (!cs) - return -ENOMEM; +retry: + wait_event(cpuset_attach_wq, cs->attach_in_progress == 0); - down(&manage_sem); - cpuset_update_task_memory_state(); - cs->flags = 0; - if (notify_on_release(parent)) - set_bit(CS_NOTIFY_ON_RELEASE, &cs->flags); - cs->cpus_allowed = CPU_MASK_NONE; - cs->mems_allowed = NODE_MASK_NONE; - atomic_set(&cs->count, 0); - INIT_LIST_HEAD(&cs->sibling); - INIT_LIST_HEAD(&cs->children); - atomic_inc(&cpuset_mems_generation); - cs->mems_generation = atomic_read(&cpuset_mems_generation); - fmeter_init(&cs->fmeter); + mutex_lock(&cpuset_mutex); - cs->parent = parent; + /* + * We have raced with task attaching. We wait until attaching + * is finished, so we won't attach a task to an empty cpuset. + */ + if (cs->attach_in_progress) { + mutex_unlock(&cpuset_mutex); + goto retry; + } - down(&callback_sem); - list_add(&cs->sibling, &cs->parent->children); - number_of_cpusets++; - up(&callback_sem); + cpumask_andnot(&off_cpus, cs->cpus_allowed, top_cpuset.cpus_allowed); + nodes_andnot(off_mems, cs->mems_allowed, top_cpuset.mems_allowed); - err = cpuset_create_dir(cs, name, mode); - if (err < 0) - goto err; + mutex_lock(&callback_mutex); + cpumask_andnot(cs->cpus_allowed, cs->cpus_allowed, &off_cpus); + mutex_unlock(&callback_mutex); /* - * Release manage_sem before cpuset_populate_dir() because it - * will down() this new directory's i_mutex and if we race with - * another mkdir, we might deadlock. + * If sane_behavior flag is set, we need to update tasks' cpumask + * for empty cpuset to take on ancestor's cpumask. Otherwise, don't + * call update_tasks_cpumask() if the cpuset becomes empty, as + * the tasks in it will be migrated to an ancestor. */ - up(&manage_sem); + if ((sane && cpumask_empty(cs->cpus_allowed)) || + (!cpumask_empty(&off_cpus) && !cpumask_empty(cs->cpus_allowed))) + update_tasks_cpumask(cs); - err = cpuset_populate_dir(cs->dentry); - /* If err < 0, we have a half-filled directory - oh well ;) */ - return 0; -err: - list_del(&cs->sibling); - up(&manage_sem); - kfree(cs); - return err; -} + mutex_lock(&callback_mutex); + nodes_andnot(cs->mems_allowed, cs->mems_allowed, off_mems); + mutex_unlock(&callback_mutex); -static int cpuset_mkdir(struct inode *dir, struct dentry *dentry, int mode) -{ - struct cpuset *c_parent = dentry->d_parent->d_fsdata; - - /* the vfs holds inode->i_mutex already */ - return cpuset_create(c_parent, dentry->d_name.name, mode | S_IFDIR); -} + /* + * If sane_behavior flag is set, we need to update tasks' nodemask + * for empty cpuset to take on ancestor's nodemask. Otherwise, don't + * call update_tasks_nodemask() if the cpuset becomes empty, as + * the tasks in it will be migratd to an ancestor. + */ + if ((sane && nodes_empty(cs->mems_allowed)) || + (!nodes_empty(off_mems) && !nodes_empty(cs->mems_allowed))) + update_tasks_nodemask(cs); -static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry) -{ - struct cpuset *cs = dentry->d_fsdata; - struct dentry *d; - struct cpuset *parent; - char *pathbuf = NULL; + is_empty = cpumask_empty(cs->cpus_allowed) || + nodes_empty(cs->mems_allowed); - /* the vfs holds both inode->i_mutex already */ + mutex_unlock(&cpuset_mutex); - down(&manage_sem); - cpuset_update_task_memory_state(); - if (atomic_read(&cs->count) > 0) { - up(&manage_sem); - return -EBUSY; - } - if (!list_empty(&cs->children)) { - up(&manage_sem); - return -EBUSY; - } - parent = cs->parent; - down(&callback_sem); - set_bit(CS_REMOVED, &cs->flags); - if (is_cpu_exclusive(cs)) - update_cpu_domains(cs); - list_del(&cs->sibling); /* delete my sibling from parent->children */ - spin_lock(&cs->dentry->d_lock); - d = dget(cs->dentry); - cs->dentry = NULL; - spin_unlock(&d->d_lock); - cpuset_d_remove_dir(d); - dput(d); - number_of_cpusets--; - up(&callback_sem); - if (list_empty(&parent->children)) - check_for_release(parent, &pathbuf); - up(&manage_sem); - cpuset_release_agent(pathbuf); - return 0; + /* + * If sane_behavior flag is set, we'll keep tasks in empty cpusets. + * + * Otherwise move tasks to the nearest ancestor with execution + * resources. This is full cgroup operation which will + * also call back into cpuset. Should be done outside any lock. + */ + if (!sane && is_empty) + remove_tasks_in_empty_cpuset(cs); } -/* - * cpuset_init_early - just enough so that the calls to - * cpuset_update_task_memory_state() in early init code - * are harmless. +/** + * cpuset_hotplug_workfn - handle CPU/memory hotunplug for a cpuset + * + * This function is called after either CPU or memory configuration has + * changed and updates cpuset accordingly. The top_cpuset is always + * synchronized to cpu_active_mask and N_MEMORY, which is necessary in + * order to make cpusets transparent (of no affect) on systems that are + * actively using CPU hotplug but making no active use of cpusets. + * + * Non-root cpusets are only affected by offlining. If any CPUs or memory + * nodes have been taken down, cpuset_hotplug_update_tasks() is invoked on + * all descendants. + * + * Note that CPU offlining during suspend is ignored. We don't modify + * cpusets across suspend/resume cycles at all. */ - -int __init cpuset_init_early(void) +static void cpuset_hotplug_workfn(struct work_struct *work) { - struct task_struct *tsk = current; + static cpumask_t new_cpus; + static nodemask_t new_mems; + bool cpus_updated, mems_updated; - tsk->cpuset = &top_cpuset; - tsk->cpuset->mems_generation = atomic_read(&cpuset_mems_generation); - return 0; -} + mutex_lock(&cpuset_mutex); -/** - * cpuset_init - initialize cpusets at system boot - * - * Description: Initialize top_cpuset and the cpuset internal file system, - **/ + /* fetch the available cpus/mems and find out which changed how */ + cpumask_copy(&new_cpus, cpu_active_mask); + new_mems = node_states[N_MEMORY]; -int __init cpuset_init(void) -{ - struct dentry *root; - int err; + cpus_updated = !cpumask_equal(top_cpuset.cpus_allowed, &new_cpus); + mems_updated = !nodes_equal(top_cpuset.mems_allowed, new_mems); - top_cpuset.cpus_allowed = CPU_MASK_ALL; - top_cpuset.mems_allowed = NODE_MASK_ALL; + /* synchronize cpus_allowed to cpu_active_mask */ + if (cpus_updated) { + mutex_lock(&callback_mutex); + cpumask_copy(top_cpuset.cpus_allowed, &new_cpus); + mutex_unlock(&callback_mutex); + /* we don't mess with cpumasks of tasks in top_cpuset */ + } - fmeter_init(&top_cpuset.fmeter); - atomic_inc(&cpuset_mems_generation); - top_cpuset.mems_generation = atomic_read(&cpuset_mems_generation); + /* synchronize mems_allowed to N_MEMORY */ + if (mems_updated) { + mutex_lock(&callback_mutex); + top_cpuset.mems_allowed = new_mems; + mutex_unlock(&callback_mutex); + update_tasks_nodemask(&top_cpuset); + } - init_task.cpuset = &top_cpuset; + mutex_unlock(&cpuset_mutex); - err = register_filesystem(&cpuset_fs_type); - if (err < 0) - goto out; - cpuset_mount = kern_mount(&cpuset_fs_type); - if (IS_ERR(cpuset_mount)) { - printk(KERN_ERR "cpuset: could not mount!\n"); - err = PTR_ERR(cpuset_mount); - cpuset_mount = NULL; - goto out; + /* if cpus or mems changed, we need to propagate to descendants */ + if (cpus_updated || mems_updated) { + struct cpuset *cs; + struct cgroup_subsys_state *pos_css; + + rcu_read_lock(); + cpuset_for_each_descendant_pre(cs, pos_css, &top_cpuset) { + if (cs == &top_cpuset || !css_tryget_online(&cs->css)) + continue; + rcu_read_unlock(); + + cpuset_hotplug_update_tasks(cs); + + rcu_read_lock(); + css_put(&cs->css); + } + rcu_read_unlock(); } - root = cpuset_mount->mnt_sb->s_root; - root->d_fsdata = &top_cpuset; - root->d_inode->i_nlink++; - top_cpuset.dentry = root; - root->d_inode->i_op = &cpuset_dir_inode_operations; - number_of_cpusets = 1; - err = cpuset_populate_dir(root); - /* memory_pressure_enabled is in root cpuset only */ - if (err == 0) - err = cpuset_add_file(root, &cft_memory_pressure_enabled); -out: - return err; -} -/** - * cpuset_init_smp - initialize cpus_allowed - * - * Description: Finish top cpuset after cpu, node maps are initialized - **/ + /* rebuild sched domains if cpus_allowed has changed */ + if (cpus_updated) + rebuild_sched_domains(); +} -void __init cpuset_init_smp(void) +void cpuset_update_active_cpus(bool cpu_online) { - top_cpuset.cpus_allowed = cpu_online_map; - top_cpuset.mems_allowed = node_online_map; + /* + * We're inside cpu hotplug critical region which usually nests + * inside cgroup synchronization. Bounce actual hotplug processing + * to a work item to avoid reverse locking order. + * + * We still need to do partition_sched_domains() synchronously; + * otherwise, the scheduler will get confused and put tasks to the + * dead CPU. Fall back to the default single domain. + * cpuset_hotplug_workfn() will rebuild it as necessary. + */ + partition_sched_domains(1, NULL, NULL); + schedule_work(&cpuset_hotplug_work); } -/** - * cpuset_fork - attach newly forked task to its parents cpuset. - * @tsk: pointer to task_struct of forking parent process. - * - * Description: A task inherits its parent's cpuset at fork(). - * - * A pointer to the shared cpuset was automatically copied in fork.c - * by dup_task_struct(). However, we ignore that copy, since it was - * not made under the protection of task_lock(), so might no longer be - * a valid cpuset pointer. attach_task() might have already changed - * current->cpuset, allowing the previously referenced cpuset to - * be removed and freed. Instead, we task_lock(current) and copy - * its present value of current->cpuset for our freshly forked child. - * - * At the point that cpuset_fork() is called, 'current' is the parent - * task, and the passed argument 'child' points to the child task. - **/ - -void cpuset_fork(struct task_struct *child) +/* + * Keep top_cpuset.mems_allowed tracking node_states[N_MEMORY]. + * Call this routine anytime after node_states[N_MEMORY] changes. + * See cpuset_update_active_cpus() for CPU hotplug handling. + */ +static int cpuset_track_online_nodes(struct notifier_block *self, + unsigned long action, void *arg) { - task_lock(current); - child->cpuset = current->cpuset; - atomic_inc(&child->cpuset->count); - task_unlock(current); + schedule_work(&cpuset_hotplug_work); + return NOTIFY_OK; } +static struct notifier_block cpuset_track_online_nodes_nb = { + .notifier_call = cpuset_track_online_nodes, + .priority = 10, /* ??! */ +}; + /** - * cpuset_exit - detach cpuset from exiting task - * @tsk: pointer to task_struct of exiting process - * - * Description: Detach cpuset from @tsk and release it. - * - * Note that cpusets marked notify_on_release force every task in - * them to take the global manage_sem semaphore when exiting. - * This could impact scaling on very large systems. Be reluctant to - * use notify_on_release cpusets where very high task exit scaling - * is required on large systems. - * - * Don't even think about derefencing 'cs' after the cpuset use count - * goes to zero, except inside a critical section guarded by manage_sem - * or callback_sem. Otherwise a zero cpuset use count is a license to - * any other task to nuke the cpuset immediately, via cpuset_rmdir(). - * - * This routine has to take manage_sem, not callback_sem, because - * it is holding that semaphore while calling check_for_release(), - * which calls kmalloc(), so can't be called holding callback__sem(). + * cpuset_init_smp - initialize cpus_allowed * - * We don't need to task_lock() this reference to tsk->cpuset, - * because tsk is already marked PF_EXITING, so attach_task() won't - * mess with it, or task is a failed fork, never visible to attach_task. - **/ - -void cpuset_exit(struct task_struct *tsk) + * Description: Finish top cpuset after cpu, node maps are initialized + */ +void __init cpuset_init_smp(void) { - struct cpuset *cs; - - cs = tsk->cpuset; - tsk->cpuset = NULL; + cpumask_copy(top_cpuset.cpus_allowed, cpu_active_mask); + top_cpuset.mems_allowed = node_states[N_MEMORY]; + top_cpuset.old_mems_allowed = top_cpuset.mems_allowed; - if (notify_on_release(cs)) { - char *pathbuf = NULL; - - down(&manage_sem); - if (atomic_dec_and_test(&cs->count)) - check_for_release(cs, &pathbuf); - up(&manage_sem); - cpuset_release_agent(pathbuf); - } else { - atomic_dec(&cs->count); - } + register_hotmemory_notifier(&cpuset_track_online_nodes_nb); } /** * cpuset_cpus_allowed - return cpus_allowed mask from a tasks cpuset. * @tsk: pointer to task_struct from which to obtain cpuset->cpus_allowed. + * @pmask: pointer to struct cpumask variable to receive cpus_allowed set. * - * Description: Returns the cpumask_t cpus_allowed of the cpuset + * Description: Returns the cpumask_var_t cpus_allowed of the cpuset * attached to the specified @tsk. Guaranteed to return some non-empty - * subset of cpu_online_map, even if this means going outside the + * subset of cpu_online_mask, even if this means going outside the * tasks cpuset. **/ -cpumask_t cpuset_cpus_allowed(struct task_struct *tsk) +void cpuset_cpus_allowed(struct task_struct *tsk, struct cpumask *pmask) { - cpumask_t mask; + struct cpuset *cpus_cs; - down(&callback_sem); - task_lock(tsk); - guarantee_online_cpus(tsk->cpuset, &mask); - task_unlock(tsk); - up(&callback_sem); + mutex_lock(&callback_mutex); + rcu_read_lock(); + cpus_cs = effective_cpumask_cpuset(task_cs(tsk)); + guarantee_online_cpus(cpus_cs, pmask); + rcu_read_unlock(); + mutex_unlock(&callback_mutex); +} - return mask; +void cpuset_cpus_allowed_fallback(struct task_struct *tsk) +{ + struct cpuset *cpus_cs; + + rcu_read_lock(); + cpus_cs = effective_cpumask_cpuset(task_cs(tsk)); + do_set_cpus_allowed(tsk, cpus_cs->cpus_allowed); + rcu_read_unlock(); + + /* + * We own tsk->cpus_allowed, nobody can change it under us. + * + * But we used cs && cs->cpus_allowed lockless and thus can + * race with cgroup_attach_task() or update_cpumask() and get + * the wrong tsk->cpus_allowed. However, both cases imply the + * subsequent cpuset_change_cpumask()->set_cpus_allowed_ptr() + * which takes task_rq_lock(). + * + * If we are called after it dropped the lock we must see all + * changes in tsk_cs()->cpus_allowed. Otherwise we can temporary + * set any mask even if it is not right from task_cs() pov, + * the pending set_cpus_allowed_ptr() will fix things. + * + * select_fallback_rq() will fix things ups and set cpu_possible_mask + * if required. + */ } void cpuset_init_current_mems_allowed(void) { - current->mems_allowed = NODE_MASK_ALL; + nodes_setall(current->mems_allowed); } /** @@ -2033,104 +2293,126 @@ void cpuset_init_current_mems_allowed(void) * * Description: Returns the nodemask_t mems_allowed of the cpuset * attached to the specified @tsk. Guaranteed to return some non-empty - * subset of node_online_map, even if this means going outside the + * subset of node_states[N_MEMORY], even if this means going outside the * tasks cpuset. **/ nodemask_t cpuset_mems_allowed(struct task_struct *tsk) { + struct cpuset *mems_cs; nodemask_t mask; - down(&callback_sem); - task_lock(tsk); - guarantee_online_mems(tsk->cpuset, &mask); - task_unlock(tsk); - up(&callback_sem); + mutex_lock(&callback_mutex); + rcu_read_lock(); + mems_cs = effective_nodemask_cpuset(task_cs(tsk)); + guarantee_online_mems(mems_cs, &mask); + rcu_read_unlock(); + mutex_unlock(&callback_mutex); return mask; } /** - * cpuset_zonelist_valid_mems_allowed - check zonelist vs. curremt mems_allowed - * @zl: the zonelist to be checked + * cpuset_nodemask_valid_mems_allowed - check nodemask vs. curremt mems_allowed + * @nodemask: the nodemask to be checked * - * Are any of the nodes on zonelist zl allowed in current->mems_allowed? + * Are any of the nodes in the nodemask allowed in current->mems_allowed? */ -int cpuset_zonelist_valid_mems_allowed(struct zonelist *zl) +int cpuset_nodemask_valid_mems_allowed(nodemask_t *nodemask) { - int i; - - for (i = 0; zl->zones[i]; i++) { - int nid = zl->zones[i]->zone_pgdat->node_id; - - if (node_isset(nid, current->mems_allowed)) - return 1; - } - return 0; + return nodes_intersects(*nodemask, current->mems_allowed); } /* - * nearest_exclusive_ancestor() - Returns the nearest mem_exclusive - * ancestor to the specified cpuset. Call holding callback_sem. - * If no ancestor is mem_exclusive (an unusual configuration), then - * returns the root cpuset. + * nearest_hardwall_ancestor() - Returns the nearest mem_exclusive or + * mem_hardwall ancestor to the specified cpuset. Call holding + * callback_mutex. If no ancestor is mem_exclusive or mem_hardwall + * (an unusual configuration), then returns the root cpuset. */ -static const struct cpuset *nearest_exclusive_ancestor(const struct cpuset *cs) +static struct cpuset *nearest_hardwall_ancestor(struct cpuset *cs) { - while (!is_mem_exclusive(cs) && cs->parent) - cs = cs->parent; + while (!(is_mem_exclusive(cs) || is_mem_hardwall(cs)) && parent_cs(cs)) + cs = parent_cs(cs); return cs; } /** - * cpuset_zone_allowed - Can we allocate memory on zone z's memory node? - * @z: is this zone on an allowed node? - * @gfp_mask: memory allocation flags (we use __GFP_HARDWALL) - * - * If we're in interrupt, yes, we can always allocate. If zone - * z's node is in our tasks mems_allowed, yes. If it's not a - * __GFP_HARDWALL request and this zone's nodes is in the nearest - * mem_exclusive cpuset ancestor to this tasks cpuset, yes. + * cpuset_node_allowed_softwall - Can we allocate on a memory node? + * @node: is this an allowed node? + * @gfp_mask: memory allocation flags + * + * If we're in interrupt, yes, we can always allocate. If __GFP_THISNODE is + * set, yes, we can always allocate. If node is in our task's mems_allowed, + * yes. If it's not a __GFP_HARDWALL request and this node is in the nearest + * hardwalled cpuset ancestor to this task's cpuset, yes. If the task has been + * OOM killed and has access to memory reserves as specified by the TIF_MEMDIE + * flag, yes. * Otherwise, no. * + * If __GFP_HARDWALL is set, cpuset_node_allowed_softwall() reduces to + * cpuset_node_allowed_hardwall(). Otherwise, cpuset_node_allowed_softwall() + * might sleep, and might allow a node from an enclosing cpuset. + * + * cpuset_node_allowed_hardwall() only handles the simpler case of hardwall + * cpusets, and never sleeps. + * + * The __GFP_THISNODE placement logic is really handled elsewhere, + * by forcibly using a zonelist starting at a specified node, and by + * (in get_page_from_freelist()) refusing to consider the zones for + * any node on the zonelist except the first. By the time any such + * calls get to this routine, we should just shut up and say 'yes'. + * * GFP_USER allocations are marked with the __GFP_HARDWALL bit, - * and do not allow allocations outside the current tasks cpuset. + * and do not allow allocations outside the current tasks cpuset + * unless the task has been OOM killed as is marked TIF_MEMDIE. * GFP_KERNEL allocations are not so marked, so can escape to the - * nearest mem_exclusive ancestor cpuset. - * - * Scanning up parent cpusets requires callback_sem. The __alloc_pages() - * routine only calls here with __GFP_HARDWALL bit _not_ set if - * it's a GFP_KERNEL allocation, and all nodes in the current tasks - * mems_allowed came up empty on the first pass over the zonelist. - * So only GFP_KERNEL allocations, if all nodes in the cpuset are - * short of memory, might require taking the callback_sem semaphore. - * - * The first loop over the zonelist in mm/page_alloc.c:__alloc_pages() - * calls here with __GFP_HARDWALL always set in gfp_mask, enforcing - * hardwall cpusets - no allocation on a node outside the cpuset is - * allowed (unless in interrupt, of course). - * - * The second loop doesn't even call here for GFP_ATOMIC requests - * (if the __alloc_pages() local variable 'wait' is set). That check - * and the checks below have the combined affect in the second loop of - * the __alloc_pages() routine that: + * nearest enclosing hardwalled ancestor cpuset. + * + * Scanning up parent cpusets requires callback_mutex. The + * __alloc_pages() routine only calls here with __GFP_HARDWALL bit + * _not_ set if it's a GFP_KERNEL allocation, and all nodes in the + * current tasks mems_allowed came up empty on the first pass over + * the zonelist. So only GFP_KERNEL allocations, if all nodes in the + * cpuset are short of memory, might require taking the callback_mutex + * mutex. + * + * The first call here from mm/page_alloc:get_page_from_freelist() + * has __GFP_HARDWALL set in gfp_mask, enforcing hardwall cpusets, + * so no allocation on a node outside the cpuset is allowed (unless + * in interrupt, of course). + * + * The second pass through get_page_from_freelist() doesn't even call + * here for GFP_ATOMIC calls. For those calls, the __alloc_pages() + * variable 'wait' is not set, and the bit ALLOC_CPUSET is not set + * in alloc_flags. That logic and the checks below have the combined + * affect that: * in_interrupt - any node ok (current task context irrelevant) * GFP_ATOMIC - any node ok - * GFP_KERNEL - any node in enclosing mem_exclusive cpuset ok + * TIF_MEMDIE - any node ok + * GFP_KERNEL - any node in enclosing hardwalled cpuset ok * GFP_USER - only nodes in current tasks mems allowed ok. - **/ - -int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask) + * + * Rule: + * Don't call cpuset_node_allowed_softwall if you can't sleep, unless you + * pass in the __GFP_HARDWALL flag set in gfp_flag, which disables + * the code that might scan up ancestor cpusets and sleep. + */ +int __cpuset_node_allowed_softwall(int node, gfp_t gfp_mask) { - int node; /* node that zone z is on */ - const struct cpuset *cs; /* current cpuset ancestors */ - int allowed = 1; /* is allocation in zone z allowed? */ + struct cpuset *cs; /* current cpuset ancestors */ + int allowed; /* is allocation in zone z allowed? */ - if (in_interrupt()) + if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) return 1; - node = z->zone_pgdat->node_id; + might_sleep_if(!(gfp_mask & __GFP_HARDWALL)); if (node_isset(node, current->mems_allowed)) return 1; + /* + * Allow tasks that have access to memory reserves because they have + * been OOM killed to get memory anywhere. + */ + if (unlikely(test_thread_flag(TIF_MEMDIE))) + return 1; if (gfp_mask & __GFP_HARDWALL) /* If hardwall request, stop here */ return 0; @@ -2138,80 +2420,158 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask) return 1; /* Not hardwall and node outside mems_allowed: scan up cpusets */ - down(&callback_sem); - - task_lock(current); - cs = nearest_exclusive_ancestor(current->cpuset); - task_unlock(current); + mutex_lock(&callback_mutex); + rcu_read_lock(); + cs = nearest_hardwall_ancestor(task_cs(current)); allowed = node_isset(node, cs->mems_allowed); - up(&callback_sem); + rcu_read_unlock(); + + mutex_unlock(&callback_mutex); return allowed; } -/** - * cpuset_lock - lock out any changes to cpuset structures - * - * The out of memory (oom) code needs to lock down cpusets - * from being changed while it scans the tasklist looking for a - * task in an overlapping cpuset. Expose callback_sem via this - * cpuset_lock() routine, so the oom code can lock it, before - * locking the task list. The tasklist_lock is a spinlock, so - * must be taken inside callback_sem. +/* + * cpuset_node_allowed_hardwall - Can we allocate on a memory node? + * @node: is this an allowed node? + * @gfp_mask: memory allocation flags + * + * If we're in interrupt, yes, we can always allocate. If __GFP_THISNODE is + * set, yes, we can always allocate. If node is in our task's mems_allowed, + * yes. If the task has been OOM killed and has access to memory reserves as + * specified by the TIF_MEMDIE flag, yes. + * Otherwise, no. + * + * The __GFP_THISNODE placement logic is really handled elsewhere, + * by forcibly using a zonelist starting at a specified node, and by + * (in get_page_from_freelist()) refusing to consider the zones for + * any node on the zonelist except the first. By the time any such + * calls get to this routine, we should just shut up and say 'yes'. + * + * Unlike the cpuset_node_allowed_softwall() variant, above, + * this variant requires that the node be in the current task's + * mems_allowed or that we're in interrupt. It does not scan up the + * cpuset hierarchy for the nearest enclosing mem_exclusive cpuset. + * It never sleeps. */ - -void cpuset_lock(void) +int __cpuset_node_allowed_hardwall(int node, gfp_t gfp_mask) { - down(&callback_sem); + if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) + return 1; + if (node_isset(node, current->mems_allowed)) + return 1; + /* + * Allow tasks that have access to memory reserves because they have + * been OOM killed to get memory anywhere. + */ + if (unlikely(test_thread_flag(TIF_MEMDIE))) + return 1; + return 0; } /** - * cpuset_unlock - release lock on cpuset changes - * - * Undo the lock taken in a previous cpuset_lock() call. + * cpuset_mem_spread_node() - On which node to begin search for a file page + * cpuset_slab_spread_node() - On which node to begin search for a slab page + * + * If a task is marked PF_SPREAD_PAGE or PF_SPREAD_SLAB (as for + * tasks in a cpuset with is_spread_page or is_spread_slab set), + * and if the memory allocation used cpuset_mem_spread_node() + * to determine on which node to start looking, as it will for + * certain page cache or slab cache pages such as used for file + * system buffers and inode caches, then instead of starting on the + * local node to look for a free page, rather spread the starting + * node around the tasks mems_allowed nodes. + * + * We don't have to worry about the returned node being offline + * because "it can't happen", and even if it did, it would be ok. + * + * The routines calling guarantee_online_mems() are careful to + * only set nodes in task->mems_allowed that are online. So it + * should not be possible for the following code to return an + * offline node. But if it did, that would be ok, as this routine + * is not returning the node where the allocation must be, only + * the node where the search should start. The zonelist passed to + * __alloc_pages() will include all nodes. If the slab allocator + * is passed an offline node, it will fall back to the local node. + * See kmem_cache_alloc_node(). */ -void cpuset_unlock(void) +static int cpuset_spread_node(int *rotor) +{ + int node; + + node = next_node(*rotor, current->mems_allowed); + if (node == MAX_NUMNODES) + node = first_node(current->mems_allowed); + *rotor = node; + return node; +} + +int cpuset_mem_spread_node(void) +{ + if (current->cpuset_mem_spread_rotor == NUMA_NO_NODE) + current->cpuset_mem_spread_rotor = + node_random(¤t->mems_allowed); + + return cpuset_spread_node(¤t->cpuset_mem_spread_rotor); +} + +int cpuset_slab_spread_node(void) { - up(&callback_sem); + if (current->cpuset_slab_spread_rotor == NUMA_NO_NODE) + current->cpuset_slab_spread_rotor = + node_random(¤t->mems_allowed); + + return cpuset_spread_node(¤t->cpuset_slab_spread_rotor); } +EXPORT_SYMBOL_GPL(cpuset_mem_spread_node); + /** - * cpuset_excl_nodes_overlap - Do we overlap @p's mem_exclusive ancestors? - * @p: pointer to task_struct of some other task. - * - * Description: Return true if the nearest mem_exclusive ancestor - * cpusets of tasks @p and current overlap. Used by oom killer to - * determine if task @p's memory usage might impact the memory - * available to the current task. - * - * Call while holding callback_sem. + * cpuset_mems_allowed_intersects - Does @tsk1's mems_allowed intersect @tsk2's? + * @tsk1: pointer to task_struct of some task. + * @tsk2: pointer to task_struct of some other task. + * + * Description: Return true if @tsk1's mems_allowed intersects the + * mems_allowed of @tsk2. Used by the OOM killer to determine if + * one of the task's memory usage might impact the memory available + * to the other. **/ -int cpuset_excl_nodes_overlap(const struct task_struct *p) +int cpuset_mems_allowed_intersects(const struct task_struct *tsk1, + const struct task_struct *tsk2) { - const struct cpuset *cs1, *cs2; /* my and p's cpuset ancestors */ - int overlap = 0; /* do cpusets overlap? */ + return nodes_intersects(tsk1->mems_allowed, tsk2->mems_allowed); +} - task_lock(current); - if (current->flags & PF_EXITING) { - task_unlock(current); - goto done; - } - cs1 = nearest_exclusive_ancestor(current->cpuset); - task_unlock(current); +#define CPUSET_NODELIST_LEN (256) - task_lock((struct task_struct *)p); - if (p->flags & PF_EXITING) { - task_unlock((struct task_struct *)p); - goto done; - } - cs2 = nearest_exclusive_ancestor(p->cpuset); - task_unlock((struct task_struct *)p); +/** + * cpuset_print_task_mems_allowed - prints task's cpuset and mems_allowed + * @tsk: pointer to task_struct of some task. + * + * Description: Prints @task's name, cpuset name, and cached copy of its + * mems_allowed to the kernel log. + */ +void cpuset_print_task_mems_allowed(struct task_struct *tsk) +{ + /* Statically allocated to prevent using excess stack. */ + static char cpuset_nodelist[CPUSET_NODELIST_LEN]; + static DEFINE_SPINLOCK(cpuset_buffer_lock); + struct cgroup *cgrp; - overlap = nodes_intersects(cs1->mems_allowed, cs2->mems_allowed); -done: - return overlap; + spin_lock(&cpuset_buffer_lock); + rcu_read_lock(); + + cgrp = task_cs(tsk)->css.cgroup; + nodelist_scnprintf(cpuset_nodelist, CPUSET_NODELIST_LEN, + tsk->mems_allowed); + pr_info("%s cpuset=", tsk->comm); + pr_cont_cgroup_name(cgrp); + pr_cont(" mems_allowed=%s\n", cpuset_nodelist); + + rcu_read_unlock(); + spin_unlock(&cpuset_buffer_lock); } /* @@ -2242,75 +2602,66 @@ int cpuset_memory_pressure_enabled __read_mostly; void __cpuset_memory_pressure_bump(void) { - struct cpuset *cs; - - task_lock(current); - cs = current->cpuset; - fmeter_markevent(&cs->fmeter); - task_unlock(current); + rcu_read_lock(); + fmeter_markevent(&task_cs(current)->fmeter); + rcu_read_unlock(); } +#ifdef CONFIG_PROC_PID_CPUSET /* * proc_cpuset_show() * - Print tasks cpuset path into seq_file. * - Used for /proc/<pid>/cpuset. * - No need to task_lock(tsk) on this tsk->cpuset reference, as it * doesn't really matter if tsk->cpuset changes after we read it, - * and we take manage_sem, keeping attach_task() from changing it + * and we take cpuset_mutex, keeping cpuset_attach() from changing it * anyway. */ - -static int proc_cpuset_show(struct seq_file *m, void *v) +int proc_cpuset_show(struct seq_file *m, void *unused_v) { - struct cpuset *cs; + struct pid *pid; struct task_struct *tsk; - char *buf; - int retval = 0; + char *buf, *p; + struct cgroup_subsys_state *css; + int retval; - buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + retval = -ENOMEM; + buf = kmalloc(PATH_MAX, GFP_KERNEL); if (!buf) - return -ENOMEM; - - tsk = m->private; - down(&manage_sem); - cs = tsk->cpuset; - if (!cs) { - retval = -EINVAL; goto out; - } - retval = cpuset_path(cs, buf, PAGE_SIZE); - if (retval < 0) - goto out; - seq_puts(m, buf); + retval = -ESRCH; + pid = m->private; + tsk = get_pid_task(pid, PIDTYPE_PID); + if (!tsk) + goto out_free; + + retval = -ENAMETOOLONG; + rcu_read_lock(); + css = task_css(tsk, cpuset_cgrp_id); + p = cgroup_path(css->cgroup, buf, PATH_MAX); + rcu_read_unlock(); + if (!p) + goto out_put_task; + seq_puts(m, p); seq_putc(m, '\n'); -out: - up(&manage_sem); + retval = 0; +out_put_task: + put_task_struct(tsk); +out_free: kfree(buf); +out: return retval; } +#endif /* CONFIG_PROC_PID_CPUSET */ -static int cpuset_open(struct inode *inode, struct file *file) -{ - struct task_struct *tsk = PROC_I(inode)->task; - return single_open(file, proc_cpuset_show, tsk); -} - -struct file_operations proc_cpuset_operations = { - .open = cpuset_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -/* Display task cpus_allowed, mems_allowed in /proc/<pid>/status file. */ -char *cpuset_task_status_allowed(struct task_struct *task, char *buffer) +/* Display task mems_allowed in /proc/<pid>/status file. */ +void cpuset_task_status_allowed(struct seq_file *m, struct task_struct *task) { - buffer += sprintf(buffer, "Cpus_allowed:\t"); - buffer += cpumask_scnprintf(buffer, PAGE_SIZE, task->cpus_allowed); - buffer += sprintf(buffer, "\n"); - buffer += sprintf(buffer, "Mems_allowed:\t"); - buffer += nodemask_scnprintf(buffer, PAGE_SIZE, task->mems_allowed); - buffer += sprintf(buffer, "\n"); - return buffer; + seq_puts(m, "Mems_allowed:\t"); + seq_nodemask(m, &task->mems_allowed); + seq_puts(m, "\n"); + seq_puts(m, "Mems_allowed_list:\t"); + seq_nodemask_list(m, &task->mems_allowed); + seq_puts(m, "\n"); } |