* [PATCH 1/6] Generic container system abstracted from cpusets code
2006-10-20 18:38 [PATCH 0/6] Generic Process Containers menage
@ 2006-10-20 18:38 ` menage
2006-10-20 18:38 ` [PATCH 2/6] Cpusets hooked into containers menage
` (4 subsequent siblings)
5 siblings, 0 replies; 12+ messages in thread
From: menage @ 2006-10-20 18:38 UTC (permalink / raw)
To: akpm, pj, sekharan
Cc: linux-kernel, ckrm-tech, jlan, mbligh, rohitseth, winget,
Simon.Derr
[-- Attachment #1: container.patch --]
[-- Type: text/plain, Size: 57484 bytes --]
This patch creates a generic process container system based on (and
parallel top) the cpusets code. At a coarse level it was created by
copying kernel/cpuset.c, doing s/cpuset/container/g, and stripping out any
code that was cpuset-specific rather than applicable to any process
container subsystem.
Signed-off-by: Paul Menage <menage@google.com>
---
Documentation/containers.txt | 229 +++++++
fs/proc/base.c | 7
include/linux/container.h | 96 +++
include/linux/sched.h | 5
init/Kconfig | 9
init/main.c | 3
kernel/Makefile | 1
kernel/container.c | 1343 +++++++++++++++++++++++++++++++++++++++++++
kernel/exit.c | 2
kernel/fork.c | 3
10 files changed, 1697 insertions(+), 1 deletion(-)
Index: container-2.6.19-rc2/fs/proc/base.c
===================================================================
--- container-2.6.19-rc2.orig/fs/proc/base.c
+++ container-2.6.19-rc2/fs/proc/base.c
@@ -69,6 +69,7 @@
#include <linux/ptrace.h>
#include <linux/seccomp.h>
#include <linux/cpuset.h>
+#include <linux/container.h>
#include <linux/audit.h>
#include <linux/poll.h>
#include <linux/nsproxy.h>
@@ -1780,6 +1781,9 @@ static struct pid_entry tgid_base_stuff[
#ifdef CONFIG_SCHEDSTATS
INF("schedstat", S_IRUGO, pid_schedstat),
#endif
+#ifdef CONFIG_CONTAINERS
+ REG("container", S_IRUGO, container),
+#endif
#ifdef CONFIG_CPUSETS
REG("cpuset", S_IRUGO, cpuset),
#endif
@@ -2054,6 +2058,9 @@ static struct pid_entry tid_base_stuff[]
#ifdef CONFIG_SCHEDSTATS
INF("schedstat", S_IRUGO, pid_schedstat),
#endif
+#ifdef CONFIG_CONTAINERS
+ REG("container", S_IRUGO, container),
+#endif
#ifdef CONFIG_CPUSETS
REG("cpuset", S_IRUGO, cpuset),
#endif
Index: container-2.6.19-rc2/include/linux/container.h
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/include/linux/container.h
@@ -0,0 +1,96 @@
+#ifndef _LINUX_CONTAINER_H
+#define _LINUX_CONTAINER_H
+/*
+ * container interface
+ *
+ * Copyright (C) 2003 BULL SA
+ * Copyright (C) 2004-2006 Silicon Graphics, Inc.
+ *
+ */
+
+#include <linux/sched.h>
+#include <linux/cpumask.h>
+#include <linux/nodemask.h>
+
+#ifdef CONFIG_CONTAINERS
+
+extern int number_of_containers; /* How many containers are defined in system? */
+
+extern int container_init_early(void);
+extern int container_init(void);
+extern void container_init_smp(void);
+extern void container_fork(struct task_struct *p);
+extern void container_exit(struct task_struct *p);
+
+extern struct file_operations proc_container_operations;
+
+extern void container_lock(void);
+extern void container_unlock(void);
+
+extern void container_manage_lock(void);
+extern void container_manage_unlock(void);
+
+struct container {
+ unsigned long flags; /* "unsigned long" so bitops work */
+
+ /*
+ * Count is atomic so can incr (fork) or decr (exit) without a lock.
+ */
+ atomic_t count; /* count tasks using this container */
+
+ /*
+ * We link our 'sibling' struct into our parent's 'children'.
+ * Our children link their 'sibling' into our 'children'.
+ */
+ struct list_head sibling; /* my parent's children */
+ struct list_head children; /* my children */
+
+ struct container *parent; /* my parent */
+ struct dentry *dentry; /* container fs entry */
+};
+
+/* struct cftype:
+ *
+ * The files in the container 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.
+ *
+ *
+ * When reading/writing to a file:
+ * - the container to use in file->f_dentry->d_parent->d_fsdata
+ * - the 'cftype' of the file is file->f_dentry->d_fsdata
+ */
+
+struct inode;
+struct cftype {
+ char *name;
+ int private;
+ int (*open) (struct inode *inode, struct file *file);
+ ssize_t (*read) (struct container *cont, struct cftype *cft,
+ struct file *file,
+ char __user *buf, size_t nbytes, loff_t *ppos);
+ ssize_t (*write) (struct container *cont, struct cftype *cft,
+ struct file *file,
+ const char __user *buf, size_t nbytes, loff_t *ppos);
+ int (*release) (struct inode *inode, struct file *file);
+};
+
+int container_add_file(struct container *cont, const struct cftype *cft);
+
+int container_is_removed(const struct container *cont);
+
+#else /* !CONFIG_CONTAINERS */
+
+static inline int container_init_early(void) { return 0; }
+static inline int container_init(void) { return 0; }
+static inline void container_init_smp(void) {}
+static inline void container_fork(struct task_struct *p) {}
+static inline void container_exit(struct task_struct *p) {}
+
+static inline void container_lock(void) {}
+static inline void container_unlock(void) {}
+
+#endif /* !CONFIG_CONTAINERS */
+
+#endif /* _LINUX_CONTAINER_H */
Index: container-2.6.19-rc2/include/linux/sched.h
===================================================================
--- container-2.6.19-rc2.orig/include/linux/sched.h
+++ container-2.6.19-rc2/include/linux/sched.h
@@ -720,8 +720,8 @@ extern unsigned int max_cache_size;
struct io_context; /* See blkdev.h */
+struct container;
struct cpuset;
-
#define NGROUPS_SMALL 32
#define NGROUPS_PER_BLOCK ((int)(PAGE_SIZE / sizeof(gid_t)))
struct group_info {
@@ -1007,6 +1007,9 @@ struct task_struct {
int cpuset_mems_generation;
int cpuset_mem_spread_rotor;
#endif
+#ifdef CONFIG_CONTAINERS
+ struct container *container;
+#endif
struct robust_list_head __user *robust_list;
#ifdef CONFIG_COMPAT
struct compat_robust_list_head __user *compat_robust_list;
Index: container-2.6.19-rc2/init/Kconfig
===================================================================
--- container-2.6.19-rc2.orig/init/Kconfig
+++ container-2.6.19-rc2/init/Kconfig
@@ -237,6 +237,15 @@ config IKCONFIG_PROC
This option enables access to the kernel configuration file
through /proc/config.gz.
+config CONTAINERS
+ bool "Container support"
+ help
+ This option will let you create and manage process containers,
+ which can be used to aggregate multiple processes, e.g. for
+ the purposes of resource tracking.
+
+ Say N if unsure
+
config CPUSETS
bool "Cpuset support"
depends on SMP
Index: container-2.6.19-rc2/init/main.c
===================================================================
--- container-2.6.19-rc2.orig/init/main.c
+++ container-2.6.19-rc2/init/main.c
@@ -38,6 +38,7 @@
#include <linux/writeback.h>
#include <linux/cpu.h>
#include <linux/cpuset.h>
+#include <linux/container.h>
#include <linux/efi.h>
#include <linux/taskstats_kern.h>
#include <linux/delayacct.h>
@@ -567,6 +568,7 @@ asmlinkage void __init start_kernel(void
}
#endif
vfs_caches_init_early();
+ container_init_early();
cpuset_init_early();
mem_init();
kmem_cache_init();
@@ -597,6 +599,7 @@ asmlinkage void __init start_kernel(void
#ifdef CONFIG_PROC_FS
proc_root_init();
#endif
+ container_init();
cpuset_init();
taskstats_init_early();
delayacct_init();
Index: container-2.6.19-rc2/kernel/container.c
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/kernel/container.c
@@ -0,0 +1,1343 @@
+/*
+ * kernel/container.c
+ *
+ * Generic process-grouping system.
+ *
+ * Based originally on the cpuset system, extracted by Paul Menage
+ * Copyright (C) 2006 Google, Inc
+ *
+ * Copyright notices from the original cpuset code:
+ * --------------------------------------------------
+ * Copyright (C) 2003 BULL SA.
+ * Copyright (C) 2004-2006 Silicon Graphics, Inc.
+ *
+ * Portions derived from Patrick Mochel's sysfs code.
+ * sysfs is Copyright (c) 2001-3 Patrick Mochel
+ *
+ * 2003-10-10 Written by Simon Derr.
+ * 2003-10-22 Updates by Stephen Hemminger.
+ * 2004 May-July Rework by Paul Jackson.
+ * ---------------------------------------------------
+ *
+ * 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/cpu.h>
+#include <linux/cpumask.h>
+#include <linux/container.h>
+#include <linux/err.h>
+#include <linux/errno.h>
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/kmod.h>
+#include <linux/list.h>
+#include <linux/mempolicy.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/pagemap.h>
+#include <linux/proc_fs.h>
+#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>
+#include <linux/time.h>
+#include <linux/backing-dev.h>
+#include <linux/sort.h>
+
+#include <asm/uaccess.h>
+#include <asm/atomic.h>
+#include <linux/mutex.h>
+
+#define CONTAINER_SUPER_MAGIC 0x27e0eb
+
+/*
+ * Tracks how many containers are currently defined in system.
+ * When there is only one container (the root container) we can
+ * short circuit some hooks.
+ */
+int number_of_containers __read_mostly;
+
+/* bits in struct container flags field */
+typedef enum {
+ CONT_REMOVED,
+ CONT_NOTIFY_ON_RELEASE,
+} container_flagbits_t;
+
+/* convenient tests for these bits */
+inline int container_is_removed(const struct container *cont)
+{
+ return test_bit(CONT_REMOVED, &cont->flags);
+}
+
+static inline int notify_on_release(const struct container *cont)
+{
+ return test_bit(CONT_NOTIFY_ON_RELEASE, &cont->flags);
+}
+
+static struct container top_container = {
+ .count = ATOMIC_INIT(0),
+ .sibling = LIST_HEAD_INIT(top_container.sibling),
+ .children = LIST_HEAD_INIT(top_container.children),
+};
+
+static struct vfsmount *container_mount;
+static struct super_block *container_sb;
+
+/*
+ * We have two global container mutexes below. They can nest.
+ * It is ok to first take manage_mutex, then nest callback_mutex. We also
+ * require taking task_lock() when dereferencing a tasks container pointer.
+ * See "The task_lock() exception", at the end of this comment.
+ *
+ * A task must hold both mutexes to modify containers. If a task
+ * holds manage_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 containers. It can perform various checks on
+ * the container structure first, knowing nothing will change. It can
+ * also allocate memory while just holding manage_mutex. While it is
+ * performing these checks, various callback routines can briefly
+ * acquire callback_mutex to query containers. 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_mutex, as that would risk double tripping on callback_mutex
+ * from one of the callbacks into the container code from within
+ * __alloc_pages().
+ *
+ * If a task is only holding callback_mutex, then it has read-only
+ * access to containers.
+ *
+ * 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_mutex or callback_mutex can't rely
+ * on the count field not changing. However, if the count goes to
+ * zero, then only attach_task(), which holds both mutexes, 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 container can fork (the other way to increment the count).
+ * So code holding manage_mutex or callback_mutex can safely assume that
+ * if the count is zero, it will stay zero. Similarly, if a task
+ * holds manage_mutex or callback_mutex on a container with zero count, it
+ * knows that the container won't be removed, as container_rmdir() needs
+ * both of those mutexes.
+ *
+ * The container_common_file_write handler for operations that modify
+ * the container hierarchy holds manage_mutex across the entire operation,
+ * single threading all such container modifications across the system.
+ *
+ * The container_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 container_fork() and container_exit(), don't
+ * (usually) take either mutex. These are the two most performance
+ * critical pieces of code here. The exception occurs on container_exit(),
+ * when a task in a notify_on_release container exits. Then manage_mutex
+ * is taken, and if the container count is zero, a usermode call made
+ * to /sbin/container_release_agent with the name of the container (path
+ * relative to the root of container file system) as the argument.
+ *
+ * A container can only be deleted if both its 'count' of using tasks
+ * is zero, and its list of 'children' containers is empty. Since all
+ * tasks in the system use _some_ container, and since there is always at
+ * least one task in the system (init, pid == 1), therefore, top_container
+ * always has either children containers and/or using tasks. So we don't
+ * need a special hack to ensure that top_container 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 container pointer with another. It does
+ * so using both mutexes, however there are several performance
+ * critical places that need to reference task->container without the
+ * expense of grabbing a system global mutex. Therefore except as
+ * noted below, when dereferencing or, as in attach_task(), modifying
+ * a tasks container 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 container pointer by attach_task() and the
+ * access of task->container->mems_generation via that pointer in
+ * the routine container_update_task_memory_state().
+ */
+
+static DEFINE_MUTEX(manage_mutex);
+static DEFINE_MUTEX(callback_mutex);
+
+/*
+ * A couple of forward declarations required, due to cyclic reference loop:
+ * container_mkdir -> container_create -> container_populate_dir -> container_add_file
+ * -> container_create_file -> container_dir_inode_operations -> container_mkdir.
+ */
+
+static int container_mkdir(struct inode *dir, struct dentry *dentry, int mode);
+static int container_rmdir(struct inode *unused_dir, struct dentry *dentry);
+
+static struct backing_dev_info container_backing_dev_info = {
+ .ra_pages = 0, /* No readahead */
+ .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK,
+};
+
+static struct inode *container_new_inode(mode_t mode)
+{
+ struct inode *inode = new_inode(container_sb);
+
+ if (inode) {
+ inode->i_mode = mode;
+ inode->i_uid = current->fsuid;
+ inode->i_gid = current->fsgid;
+ inode->i_blocks = 0;
+ inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
+ inode->i_mapping->backing_dev_info = &container_backing_dev_info;
+ }
+ return inode;
+}
+
+static void container_diput(struct dentry *dentry, struct inode *inode)
+{
+ /* is dentry a directory ? if so, kfree() associated container */
+ if (S_ISDIR(inode->i_mode)) {
+ struct container *cont = dentry->d_fsdata;
+ BUG_ON(!(container_is_removed(cont)));
+ kfree(cont);
+ }
+ iput(inode);
+}
+
+static struct dentry_operations container_dops = {
+ .d_iput = container_diput,
+};
+
+static struct dentry *container_get_dentry(struct dentry *parent, const char *name)
+{
+ struct dentry *d = lookup_one_len(name, parent, strlen(name));
+ if (!IS_ERR(d))
+ d->d_op = &container_dops;
+ return d;
+}
+
+static void remove_dir(struct dentry *d)
+{
+ struct dentry *parent = dget(d->d_parent);
+
+ d_delete(d);
+ simple_rmdir(parent->d_inode, d);
+ dput(parent);
+}
+
+/*
+ * NOTE : the dentry must have been dget()'ed
+ */
+static void container_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 struct super_operations container_ops = {
+ .statfs = simple_statfs,
+ .drop_inode = generic_delete_inode,
+};
+
+static int container_fill_super(struct super_block *sb, void *unused_data,
+ int unused_silent)
+{
+ struct inode *inode;
+ struct dentry *root;
+
+ sb->s_blocksize = PAGE_CACHE_SIZE;
+ sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
+ sb->s_magic = CONTAINER_SUPER_MAGIC;
+ sb->s_op = &container_ops;
+ container_sb = sb;
+
+ inode = container_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;
+ }
+
+ root = d_alloc_root(inode);
+ if (!root) {
+ iput(inode);
+ return -ENOMEM;
+ }
+ sb->s_root = root;
+ return 0;
+}
+
+static int container_get_sb(struct file_system_type *fs_type,
+ int flags, const char *unused_dev_name,
+ void *data, struct vfsmount *mnt)
+{
+ return get_sb_single(fs_type, flags, data, container_fill_super, mnt);
+}
+
+static struct file_system_type container_fs_type = {
+ .name = "container",
+ .get_sb = container_get_sb,
+ .kill_sb = kill_litter_super,
+};
+
+static inline struct container *__d_cont(struct dentry *dentry)
+{
+ return dentry->d_fsdata;
+}
+
+static inline struct cftype *__d_cft(struct dentry *dentry)
+{
+ return dentry->d_fsdata;
+}
+
+/*
+ * Call with manage_mutex held. Writes path of container into buf.
+ * Returns 0 on success, -errno on error.
+ */
+
+static int container_path(const struct container *cont, char *buf, int buflen)
+{
+ char *start;
+
+ start = buf + buflen;
+
+ *--start = '\0';
+ for (;;) {
+ int len = cont->dentry->d_name.len;
+ if ((start -= len) < buf)
+ return -ENAMETOOLONG;
+ memcpy(start, cont->dentry->d_name.name, len);
+ cont = cont->parent;
+ if (!cont)
+ break;
+ if (!cont->parent)
+ continue;
+ if (--start < buf)
+ return -ENAMETOOLONG;
+ *start = '/';
+ }
+ memmove(buf, start, buf + buflen - start);
+ return 0;
+}
+
+/*
+ * Notify userspace when a container is released, by running
+ * /sbin/container_release_agent with the name of the container (path
+ * relative to the root of container file system) as the argument.
+ *
+ * Most likely, this user command will try to rmdir this container.
+ *
+ * This races with the possibility that some other task will be
+ * attached to this container before it is removed, or that some other
+ * user task will 'mkdir' a child container of this container. That's ok.
+ * The presumed 'rmdir' will fail quietly if this container is no longer
+ * unused, and this container 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/container_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/container_release_agent task, so no sense holding
+ * our caller up for that.
+ *
+ * When we had only one container mutex, 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_mutex, but we still don't, so as to minimize
+ * the time manage_mutex is held.
+ */
+
+static void container_release_agent(const char *pathbuf)
+{
+ char *argv[3], *envp[3];
+ int i;
+
+ if (!pathbuf)
+ return;
+
+ i = 0;
+ argv[i++] = "/sbin/container_release_agent";
+ argv[i++] = (char *)pathbuf;
+ argv[i] = NULL;
+
+ i = 0;
+ /* minimal command environment */
+ envp[i++] = "HOME=/";
+ envp[i++] = "PATH=/sbin:/bin:/usr/sbin:/usr/bin";
+ envp[i] = NULL;
+
+ call_usermodehelper(argv[0], argv, envp, 0);
+ kfree(pathbuf);
+}
+
+/*
+ * Either cont->count of using tasks transitioned to zero, or the
+ * cont->children list of child containers just became empty. If this
+ * cont is notify_on_release() and now both the user count is zero and
+ * the list of children is empty, prepare container path in a kmalloc'd
+ * buffer, to be returned via ppathbuf, so that the caller can invoke
+ * container_release_agent() with it later on, once manage_mutex is dropped.
+ * Call here with manage_mutex held.
+ *
+ * This check_for_release() routine is responsible for kmalloc'ing
+ * pathbuf. The above container_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_mutex held and the address
+ * of the pathbuf pointer, then dropping manage_mutex, then calling
+ * container_release_agent() with pathbuf, as set by check_for_release().
+ */
+
+static void check_for_release(struct container *cont, char **ppathbuf)
+{
+ if (notify_on_release(cont) && atomic_read(&cont->count) == 0 &&
+ list_empty(&cont->children)) {
+ char *buf;
+
+ buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+ if (!buf)
+ return;
+ if (container_path(cont, buf, PAGE_SIZE) < 0)
+ kfree(buf);
+ else
+ *ppathbuf = buf;
+ }
+}
+
+
+/*
+ * update_flag - read a 0 or a 1 in a file and update associated flag
+ * bit: the bit to update (CONT_NOTIFY_ON_RELEASE)
+ * cont: the container to update
+ * buf: the buffer where we read the 0 or 1
+ *
+ * Call with manage_mutex held.
+ */
+
+static int update_flag(container_flagbits_t bit, struct container *cont, char *buf)
+{
+ int turning_on;
+
+ turning_on = (simple_strtoul(buf, NULL, 10) != 0);
+
+ mutex_lock(&callback_mutex);
+ if (turning_on)
+ set_bit(bit, &cont->flags);
+ else
+ clear_bit(bit, &cont->flags);
+ mutex_unlock(&callback_mutex);
+
+ return 0;
+}
+
+
+/*
+ * Attack task specified by pid in 'pidbuf' to container 'cont', possibly
+ * writing the path of the old container in 'ppathbuf' if it needs to be
+ * notified on release.
+ *
+ * Call holding manage_mutex. May take callback_mutex and task_lock of
+ * the task 'pid' during call.
+ */
+
+static int attach_task(struct container *cont, char *pidbuf, char **ppathbuf)
+{
+ pid_t pid;
+ struct task_struct *tsk;
+ struct container *oldcont;
+ int retval;
+
+ if (sscanf(pidbuf, "%d", &pid) != 1)
+ return -EIO;
+
+ if (pid) {
+ read_lock(&tasklist_lock);
+
+ 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);
+
+ if ((current->euid) && (current->euid != tsk->uid)
+ && (current->euid != tsk->suid)) {
+ put_task_struct(tsk);
+ return -EACCES;
+ }
+ } else {
+ tsk = current;
+ get_task_struct(tsk);
+ }
+
+ retval = security_task_setscheduler(tsk, 0, NULL);
+ if (retval) {
+ put_task_struct(tsk);
+ return retval;
+ }
+
+ mutex_lock(&callback_mutex);
+
+ task_lock(tsk);
+ oldcont = tsk->container;
+ if (!oldcont) {
+ task_unlock(tsk);
+ mutex_unlock(&callback_mutex);
+ put_task_struct(tsk);
+ return -ESRCH;
+ }
+ atomic_inc(&cont->count);
+ rcu_assign_pointer(tsk->container, cont);
+ task_unlock(tsk);
+
+ mutex_unlock(&callback_mutex);
+
+ put_task_struct(tsk);
+ synchronize_rcu();
+ if (atomic_dec_and_test(&oldcont->count))
+ check_for_release(oldcont, ppathbuf);
+ return 0;
+}
+
+/* The various types of files and directories in a container file system */
+
+typedef enum {
+ FILE_ROOT,
+ FILE_DIR,
+ FILE_NOTIFY_ON_RELEASE,
+ FILE_TASKLIST,
+} container_filetype_t;
+
+static ssize_t container_common_file_write(struct container *cont,
+ struct cftype *cft,
+ struct file *file,
+ const char __user *userbuf,
+ size_t nbytes, loff_t *unused_ppos)
+{
+ container_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 */
+
+ mutex_lock(&manage_mutex);
+
+ if (container_is_removed(cont)) {
+ retval = -ENODEV;
+ goto out2;
+ }
+
+ switch (type) {
+ case FILE_NOTIFY_ON_RELEASE:
+ retval = update_flag(CONT_NOTIFY_ON_RELEASE, cont, buffer);
+ break;
+ case FILE_TASKLIST:
+ retval = attach_task(cont, buffer, &pathbuf);
+ break;
+ default:
+ retval = -EINVAL;
+ goto out2;
+ }
+
+ if (retval == 0)
+ retval = nbytes;
+out2:
+ mutex_unlock(&manage_mutex);
+ container_release_agent(pathbuf);
+out1:
+ kfree(buffer);
+ return retval;
+}
+
+static ssize_t container_file_write(struct file *file, const char __user *buf,
+ size_t nbytes, loff_t *ppos)
+{
+ ssize_t retval = 0;
+ struct cftype *cft = __d_cft(file->f_dentry);
+ struct container *cont = __d_cont(file->f_dentry->d_parent);
+ if (!cft)
+ return -ENODEV;
+
+ /* special function ? */
+ if (cft->write)
+ retval = cft->write(cont, cft, file, buf, nbytes, ppos);
+ else
+ retval = -EINVAL;
+
+ return retval;
+}
+
+static ssize_t container_common_file_read(struct container *cont,
+ struct cftype *cft,
+ struct file *file,
+ char __user *buf,
+ size_t nbytes, loff_t *ppos)
+{
+ container_filetype_t type = cft->private;
+ char *page;
+ ssize_t retval = 0;
+ char *s;
+
+ if (!(page = (char *)__get_free_page(GFP_KERNEL)))
+ return -ENOMEM;
+
+ s = page;
+
+ switch (type) {
+ case FILE_NOTIFY_ON_RELEASE:
+ *s++ = notify_on_release(cont) ? '1' : '0';
+ break;
+ default:
+ retval = -EINVAL;
+ goto out;
+ }
+ *s++ = '\n';
+
+ retval = simple_read_from_buffer(buf, nbytes, ppos, page, s - page);
+out:
+ free_page((unsigned long)page);
+ return retval;
+}
+
+static ssize_t container_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);
+ struct container *cont = __d_cont(file->f_dentry->d_parent);
+ if (!cft)
+ return -ENODEV;
+
+ /* special function ? */
+ if (cft->read)
+ retval = cft->read(cont, cft, file, buf, nbytes, ppos);
+ else
+ retval = -EINVAL;
+
+ return retval;
+}
+
+static int container_file_open(struct inode *inode, struct file *file)
+{
+ 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;
+}
+
+static int container_file_release(struct inode *inode, struct file *file)
+{
+ struct cftype *cft = __d_cft(file->f_dentry);
+ if (cft->release)
+ return cft->release(inode, file);
+ return 0;
+}
+
+/*
+ * container_rename - Only allow simple rename of directories in place.
+ */
+static int container_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 container_file_operations = {
+ .read = container_file_read,
+ .write = container_file_write,
+ .llseek = generic_file_llseek,
+ .open = container_file_open,
+ .release = container_file_release,
+};
+
+static struct inode_operations container_dir_inode_operations = {
+ .lookup = simple_lookup,
+ .mkdir = container_mkdir,
+ .rmdir = container_rmdir,
+ .rename = container_rename,
+};
+
+static int container_create_file(struct dentry *dentry, int mode)
+{
+ struct inode *inode;
+
+ if (!dentry)
+ return -ENOENT;
+ if (dentry->d_inode)
+ return -EEXIST;
+
+ inode = container_new_inode(mode);
+ if (!inode)
+ return -ENOMEM;
+
+ if (S_ISDIR(mode)) {
+ inode->i_op = &container_dir_inode_operations;
+ inode->i_fop = &simple_dir_operations;
+
+ /* start off with i_nlink == 2 (for "." entry) */
+ inode->i_nlink++;
+ } else if (S_ISREG(mode)) {
+ inode->i_size = 0;
+ inode->i_fop = &container_file_operations;
+ }
+
+ d_instantiate(dentry, inode);
+ dget(dentry); /* Extra count - pin the dentry in core */
+ return 0;
+}
+
+/*
+ * container_create_dir - create a directory for an object.
+ * cont: the container 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 container directory. Will be copied.
+ * mode: mode to set on new directory.
+ */
+
+static int container_create_dir(struct container *cont, const char *name, int mode)
+{
+ struct dentry *dentry = NULL;
+ struct dentry *parent;
+ int error = 0;
+
+ parent = cont->parent->dentry;
+ dentry = container_get_dentry(parent, name);
+ if (IS_ERR(dentry))
+ return PTR_ERR(dentry);
+ error = container_create_file(dentry, S_IFDIR | mode);
+ if (!error) {
+ dentry->d_fsdata = cont;
+ parent->d_inode->i_nlink++;
+ cont->dentry = dentry;
+ }
+ dput(dentry);
+
+ return error;
+}
+
+int container_add_file(struct container *cont, const struct cftype *cft)
+{
+ struct dentry *dir = cont->dentry;
+ struct dentry *dentry;
+ int error;
+
+ mutex_lock(&dir->d_inode->i_mutex);
+ dentry = container_get_dentry(dir, cft->name);
+ if (!IS_ERR(dentry)) {
+ error = container_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;
+}
+
+/*
+ * Stuff for reading the 'tasks' file.
+ *
+ * Reading this file can return large amounts of data if a container 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().
+ */
+
+/* containers_tasks_read array */
+
+struct ctr_struct {
+ char *buf;
+ int bufsz;
+};
+
+/*
+ * Load into 'pidarray' up to 'npids' of the tasks using container 'cont'.
+ * Return actual number of pids loaded. No need to task_lock(p)
+ * when reading out p->container, 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 container *cont)
+{
+ int n = 0;
+ struct task_struct *g, *p;
+
+ read_lock(&tasklist_lock);
+
+ do_each_thread(g, p) {
+ if (p->container == cont) {
+ pidarray[n++] = p->pid;
+ if (unlikely(n == npids))
+ goto array_full;
+ }
+ } 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;
+}
+
+/*
+ * 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.
+ */
+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 container being opened.
+ *
+ * Does not require any specific container mutexes, and does not take any.
+ */
+static int container_tasks_open(struct inode *unused, struct file *file)
+{
+ struct container *cont = __d_cont(file->f_dentry->d_parent);
+ struct ctr_struct *ctr;
+ pid_t *pidarray;
+ int npids;
+ char c;
+
+ if (!(file->f_mode & FMODE_READ))
+ return 0;
+
+ ctr = kmalloc(sizeof(*ctr), GFP_KERNEL);
+ if (!ctr)
+ goto err0;
+
+ /*
+ * If container 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 container users didn't
+ * show up until sometime later on.
+ */
+ npids = atomic_read(&cont->count);
+ pidarray = kmalloc(npids * sizeof(pid_t), GFP_KERNEL);
+ if (!pidarray)
+ goto err1;
+
+ npids = pid_array_load(pidarray, npids, cont);
+ 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;
+
+err2:
+ kfree(pidarray);
+err1:
+ kfree(ctr);
+err0:
+ return -ENOMEM;
+}
+
+static ssize_t container_tasks_read(struct container *cont,
+ struct cftype *cft,
+ struct file *file, char __user *buf,
+ size_t nbytes, loff_t *ppos)
+{
+ struct ctr_struct *ctr = file->private_data;
+
+ 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 container_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;
+}
+
+/*
+ * for the common functions, 'private' gives the type of file
+ */
+
+static struct cftype cft_tasks = {
+ .name = "tasks",
+ .open = container_tasks_open,
+ .read = container_tasks_read,
+ .write = container_common_file_write,
+ .release = container_tasks_release,
+ .private = FILE_TASKLIST,
+};
+
+static struct cftype cft_notify_on_release = {
+ .name = "notify_on_release",
+ .read = container_common_file_read,
+ .write = container_common_file_write,
+ .private = FILE_NOTIFY_ON_RELEASE,
+};
+
+static int container_populate_dir(struct container *cont)
+{
+ int err;
+
+ if ((err = container_add_file(cont, &cft_notify_on_release)) < 0)
+ return err;
+ if ((err = container_add_file(cont, &cft_tasks)) < 0)
+ return err;
+ return 0;
+}
+
+/*
+ * container_create - create a container
+ * parent: container that will be parent of the new container.
+ * name: name of the new container. Will be strcpy'ed.
+ * mode: mode to set on new inode
+ *
+ * Must be called with the mutex on the parent inode held
+ */
+
+static long container_create(struct container *parent, const char *name, int mode)
+{
+ struct container *cont;
+ int err;
+
+ cont = kmalloc(sizeof(*cont), GFP_KERNEL);
+ if (!cont)
+ return -ENOMEM;
+
+ mutex_lock(&manage_mutex);
+ cont->flags = 0;
+ if (notify_on_release(parent))
+ set_bit(CONT_NOTIFY_ON_RELEASE, &cont->flags);
+ atomic_set(&cont->count, 0);
+ INIT_LIST_HEAD(&cont->sibling);
+ INIT_LIST_HEAD(&cont->children);
+
+ cont->parent = parent;
+
+ mutex_lock(&callback_mutex);
+ list_add(&cont->sibling, &cont->parent->children);
+ number_of_containers++;
+ mutex_unlock(&callback_mutex);
+
+ err = container_create_dir(cont, name, mode);
+ if (err < 0)
+ goto err_remove;
+
+ /*
+ * Release manage_mutex before container_populate_dir() because it
+ * will down() this new directory's i_mutex and if we race with
+ * another mkdir, we might deadlock.
+ */
+ mutex_unlock(&manage_mutex);
+
+ err = container_populate_dir(cont);
+ /* If err < 0, we have a half-filled directory - oh well ;) */
+ return 0;
+
+ err_remove:
+ mutex_lock(&callback_mutex);
+ list_del(&cont->sibling);
+ number_of_containers--;
+ mutex_unlock(&callback_mutex);
+
+ mutex_unlock(&manage_mutex);
+ kfree(cont);
+ return err;
+}
+
+static int container_mkdir(struct inode *dir, struct dentry *dentry, int mode)
+{
+ struct container *c_parent = dentry->d_parent->d_fsdata;
+
+ /* the vfs holds inode->i_mutex already */
+ return container_create(c_parent, dentry->d_name.name, mode | S_IFDIR);
+}
+
+/*
+ * Locking note on the strange update_flag() call below:
+ *
+ * If the container being removed is marked cpu_exclusive, then simulate
+ * turning cpu_exclusive off, which will call update_cpu_domains().
+ * The lock_cpu_hotplug() call in update_cpu_domains() must not be
+ * made while holding callback_mutex. Elsewhere the kernel nests
+ * callback_mutex inside lock_cpu_hotplug() calls. So the reverse
+ * nesting would risk an ABBA deadlock.
+ */
+
+static int container_rmdir(struct inode *unused_dir, struct dentry *dentry)
+{
+ struct container *cont = dentry->d_fsdata;
+ struct dentry *d;
+ struct container *parent;
+ char *pathbuf = NULL;
+
+ /* the vfs holds both inode->i_mutex already */
+
+ mutex_lock(&manage_mutex);
+ if (atomic_read(&cont->count) > 0) {
+ mutex_unlock(&manage_mutex);
+ return -EBUSY;
+ }
+ if (!list_empty(&cont->children)) {
+ mutex_unlock(&manage_mutex);
+ return -EBUSY;
+ }
+ parent = cont->parent;
+ mutex_lock(&callback_mutex);
+ set_bit(CONT_REMOVED, &cont->flags);
+ list_del(&cont->sibling); /* delete my sibling from parent->children */
+ spin_lock(&cont->dentry->d_lock);
+ d = dget(cont->dentry);
+ cont->dentry = NULL;
+ spin_unlock(&d->d_lock);
+ container_d_remove_dir(d);
+ dput(d);
+ number_of_containers--;
+ mutex_unlock(&callback_mutex);
+ if (list_empty(&parent->children))
+ check_for_release(parent, &pathbuf);
+ mutex_unlock(&manage_mutex);
+ container_release_agent(pathbuf);
+ return 0;
+}
+
+/*
+ * container_init_early - probably not needed yet, but will be needed
+ * once cpusets are hooked into this code
+ */
+
+int __init container_init_early(void)
+{
+ struct task_struct *tsk = current;
+
+ tsk->container = &top_container;
+ return 0;
+}
+
+/**
+ * container_init - initialize containers at system boot
+ *
+ * Description: Initialize top_container and the container internal file system,
+ **/
+
+int __init container_init(void)
+{
+ struct dentry *root;
+ int err;
+
+ init_task.container = &top_container;
+
+ err = register_filesystem(&container_fs_type);
+ if (err < 0)
+ goto out;
+ container_mount = kern_mount(&container_fs_type);
+ if (IS_ERR(container_mount)) {
+ printk(KERN_ERR "container: could not mount!\n");
+ err = PTR_ERR(container_mount);
+ container_mount = NULL;
+ goto out;
+ }
+ root = container_mount->mnt_sb->s_root;
+ root->d_fsdata = &top_container;
+ root->d_inode->i_nlink++;
+ top_container.dentry = root;
+ root->d_inode->i_op = &container_dir_inode_operations;
+ number_of_containers = 1;
+ err = container_populate_dir(&top_container);
+out:
+ return err;
+}
+
+/**
+ * container_fork - attach newly forked task to its parents container.
+ * @tsk: pointer to task_struct of forking parent process.
+ *
+ * Description: A task inherits its parent's container at fork().
+ *
+ * A pointer to the shared container 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 container pointer. attach_task() might have already changed
+ * current->container, allowing the previously referenced container to
+ * be removed and freed. Instead, we task_lock(current) and copy
+ * its present value of current->container for our freshly forked child.
+ *
+ * At the point that container_fork() is called, 'current' is the parent
+ * task, and the passed argument 'child' points to the child task.
+ **/
+
+void container_fork(struct task_struct *child)
+{
+ task_lock(current);
+ child->container = current->container;
+ atomic_inc(&child->container->count);
+ task_unlock(current);
+}
+
+/**
+ * container_exit - detach container from exiting task
+ * @tsk: pointer to task_struct of exiting process
+ *
+ * Description: Detach container from @tsk and release it.
+ *
+ * Note that containers marked notify_on_release force every task in
+ * them to take the global manage_mutex mutex when exiting.
+ * This could impact scaling on very large systems. Be reluctant to
+ * use notify_on_release containers where very high task exit scaling
+ * is required on large systems.
+ *
+ * Don't even think about derefencing 'cont' after the container use count
+ * goes to zero, except inside a critical section guarded by manage_mutex
+ * or callback_mutex. Otherwise a zero container use count is a license to
+ * any other task to nuke the container immediately, via container_rmdir().
+ *
+ * This routine has to take manage_mutex, not callback_mutex, because
+ * it is holding that mutex while calling check_for_release(),
+ * which calls kmalloc(), so can't be called holding callback_mutex().
+ *
+ * We don't need to task_lock() this reference to tsk->container,
+ * 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.
+ *
+ * the_top_container_hack:
+ *
+ * Set the exiting tasks container to the root container (top_container).
+ *
+ * Don't leave a task unable to allocate memory, as that is an
+ * accident waiting to happen should someone add a callout in
+ * do_exit() after the container_exit() call that might allocate.
+ * If a task tries to allocate memory with an invalid container,
+ * it will oops in container_update_task_memory_state().
+ *
+ * We call container_exit() while the task is still competent to
+ * handle notify_on_release(), then leave the task attached to
+ * the root container (top_container) for the remainder of its exit.
+ *
+ * To do this properly, we would increment the reference count on
+ * top_container, and near the very end of the kernel/exit.c do_exit()
+ * code we would add a second container function call, to drop that
+ * reference. This would just create an unnecessary hot spot on
+ * the top_container reference count, to no avail.
+ *
+ * Normally, holding a reference to a container without bumping its
+ * count is unsafe. The container could go away, or someone could
+ * attach us to a different container, decrementing the count on
+ * the first container that we never incremented. But in this case,
+ * top_container isn't going away, and either task has PF_EXITING set,
+ * which wards off any attach_task() attempts, or task is a failed
+ * fork, never visible to attach_task.
+ *
+ * Another way to do this would be to set the container pointer
+ * to NULL here, and check in container_update_task_memory_state()
+ * for a NULL pointer. This hack avoids that NULL check, for no
+ * cost (other than this way too long comment ;).
+ **/
+
+void container_exit(struct task_struct *tsk)
+{
+ struct container *cont;
+
+ cont = tsk->container;
+ tsk->container = &top_container; /* the_top_container_hack - see above */
+
+ if (notify_on_release(cont)) {
+ char *pathbuf = NULL;
+
+ mutex_lock(&manage_mutex);
+ if (atomic_dec_and_test(&cont->count))
+ check_for_release(cont, &pathbuf);
+ mutex_unlock(&manage_mutex);
+ container_release_agent(pathbuf);
+ } else {
+ atomic_dec(&cont->count);
+ }
+}
+
+/**
+ * container_lock - lock out any changes to container structures
+ *
+ * The out of memory (oom) code needs to mutex_lock containers
+ * from being changed while it scans the tasklist looking for a
+ * task in an overlapping container. Expose callback_mutex via this
+ * container_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_mutex.
+ */
+
+void container_lock(void)
+{
+ mutex_lock(&callback_mutex);
+}
+
+/**
+ * container_unlock - release lock on container changes
+ *
+ * Undo the lock taken in a previous container_lock() call.
+ */
+
+void container_unlock(void)
+{
+ mutex_unlock(&callback_mutex);
+}
+
+/*
+ * proc_container_show()
+ * - Print tasks container path into seq_file.
+ * - Used for /proc/<pid>/container.
+ * - No need to task_lock(tsk) on this tsk->container reference, as it
+ * doesn't really matter if tsk->container changes after we read it,
+ * and we take manage_mutex, keeping attach_task() from changing it
+ * anyway. No need to check that tsk->container != NULL, thanks to
+ * the_top_container_hack in container_exit(), which sets an exiting tasks
+ * container to top_container.
+ */
+static int proc_container_show(struct seq_file *m, void *v)
+{
+ struct pid *pid;
+ struct task_struct *tsk;
+ char *buf;
+ int retval;
+
+ retval = -ENOMEM;
+ buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+ if (!buf)
+ goto out;
+
+ retval = -ESRCH;
+ pid = m->private;
+ tsk = get_pid_task(pid, PIDTYPE_PID);
+ if (!tsk)
+ goto out_free;
+
+ retval = -EINVAL;
+ mutex_lock(&manage_mutex);
+
+ retval = container_path(tsk->container, buf, PAGE_SIZE);
+ if (retval < 0)
+ goto out_unlock;
+ seq_puts(m, buf);
+ seq_putc(m, '\n');
+out_unlock:
+ mutex_unlock(&manage_mutex);
+ put_task_struct(tsk);
+out_free:
+ kfree(buf);
+out:
+ return retval;
+}
+
+static int container_open(struct inode *inode, struct file *file)
+{
+ struct pid *pid = PROC_I(inode)->pid;
+ return single_open(file, proc_container_show, pid);
+}
+
+struct file_operations proc_container_operations = {
+ .open = container_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
Index: container-2.6.19-rc2/kernel/exit.c
===================================================================
--- container-2.6.19-rc2.orig/kernel/exit.c
+++ container-2.6.19-rc2/kernel/exit.c
@@ -30,6 +30,7 @@
#include <linux/taskstats_kern.h>
#include <linux/delayacct.h>
#include <linux/cpuset.h>
+#include <linux/container.h>
#include <linux/syscalls.h>
#include <linux/signal.h>
#include <linux/posix-timers.h>
@@ -922,6 +923,7 @@ fastcall NORET_TYPE void do_exit(long co
__exit_fs(tsk);
exit_thread();
cpuset_exit(tsk);
+ container_exit(tsk);
exit_keys(tsk);
if (group_dead && tsk->signal->leader)
Index: container-2.6.19-rc2/kernel/fork.c
===================================================================
--- container-2.6.19-rc2.orig/kernel/fork.c
+++ container-2.6.19-rc2/kernel/fork.c
@@ -31,6 +31,7 @@
#include <linux/capability.h>
#include <linux/cpu.h>
#include <linux/cpuset.h>
+#include <linux/container.h>
#include <linux/security.h>
#include <linux/swap.h>
#include <linux/syscalls.h>
@@ -1053,6 +1054,7 @@ static struct task_struct *copy_process(
p->io_context = NULL;
p->io_wait = NULL;
p->audit_context = NULL;
+ container_fork(p);
cpuset_fork(p);
#ifdef CONFIG_NUMA
p->mempolicy = mpol_copy(p->mempolicy);
@@ -1288,6 +1290,7 @@ bad_fork_cleanup_policy:
bad_fork_cleanup_cpuset:
#endif
cpuset_exit(p);
+ container_exit(p);
bad_fork_cleanup_delays_binfmt:
delayacct_tsk_free(p);
if (p->binfmt)
Index: container-2.6.19-rc2/kernel/Makefile
===================================================================
--- container-2.6.19-rc2.orig/kernel/Makefile
+++ container-2.6.19-rc2/kernel/Makefile
@@ -36,6 +36,7 @@ obj-$(CONFIG_PM) += power/
obj-$(CONFIG_BSD_PROCESS_ACCT) += acct.o
obj-$(CONFIG_KEXEC) += kexec.o
obj-$(CONFIG_COMPAT) += compat.o
+obj-$(CONFIG_CONTAINERS) += container.o
obj-$(CONFIG_CPUSETS) += cpuset.o
obj-$(CONFIG_IKCONFIG) += configs.o
obj-$(CONFIG_STOP_MACHINE) += stop_machine.o
Index: container-2.6.19-rc2/Documentation/containers.txt
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/Documentation/containers.txt
@@ -0,0 +1,229 @@
+ CONTAINERS
+ -------
+
+Written by Paul Menage <menage@google.com> based on Documentation/cpusets.txt
+
+Original copyright in cpusets.txt:
+Portions Copyright (C) 2004 BULL SA.
+Portions Copyright (c) 2004-2006 Silicon Graphics, Inc.
+Modified by Paul Jackson <pj@sgi.com>
+Modified by Christoph Lameter <clameter@sgi.com>
+
+CONTENTS:
+=========
+
+1. Containers
+ 1.1 What are containers ?
+ 1.2 Why are containers needed ?
+ 1.3 How are containers implemented ?
+ 1.4 What does notify_on_release do ?
+ 1.5 How do I use containers ?
+2. Usage Examples and Syntax
+ 2.1 Basic Usage
+ 2.2 Attaching processes
+3. Questions
+4. Contact
+
+1. Containers
+==========
+
+1.1 What are containers ?
+----------------------
+
+Containers provide a mechanism for aggregating sets of tasks, and all
+their children, into hierarchical groups.
+
+Each task has a pointer to a container. Multiple tasks may reference
+the same container. User level code may create and destroy containers
+by name in the container virtual file system, specify and query to
+which container a task is assigned, and list the task pids assigned to
+a container.
+
+On their own, the only use for containers is for simple job
+tracking. The intention is that other subsystems, such as cpusets (see
+Documentation/cpusets.txt) hook into the generic container support to
+provide new attributes for containers, such as accounting/limiting the
+resources which processes in a container can access.
+
+1.2 Why are containers needed ?
+----------------------------
+
+There are multiple efforts to provide process aggregations in the
+Linux kernel, mainly for resource tracking purposes. Such efforts
+include cpusets, CKRM/ResGroups, and UserBeanCounters. These all
+require the basic notion of a grouping of processes, with newly forked
+processes ending in the same group (container) as their parent
+process.
+
+The kernel container patch provides the minimum essential kernel
+mechanisms required to efficiently implement such groups. It has
+minimal impact on the system fast paths, and provides hooks for
+specific subsystems such as cpusets to provide additional behaviour as
+desired.
+
+
+1.3 How are containers implemented ?
+---------------------------------
+
+Containers extends the kernel as follows:
+
+ - Each task in the system is attached to a container, via a pointer
+ in the task structure to a reference counted container structure.
+ - The hierarchy of containers can be mounted at /dev/container (or
+ elsewhere), for browsing and manipulation from user space.
+ - You can list all the tasks (by pid) attached to any container.
+
+The implementation of containers requires a few, simple hooks
+into the rest of the kernel, none in performance critical paths:
+
+ - in init/main.c, to initialize the root container at system boot.
+ - in fork and exit, to attach and detach a task from its container.
+
+In addition a new file system, of type "container" may be mounted,
+typically at /dev/container, to enable browsing and modifying the containers
+presently known to the kernel. No new system calls are added for
+containers - all support for querying and modifying containers is via
+this container file system.
+
+Each task under /proc has an added file named 'container', displaying
+the container name, as the path relative to the root of the container file
+system.
+
+Each container is represented by a directory in the container file system
+containing the following files describing that container:
+
+ - tasks: list of tasks (by pid) attached to that container
+ - notify_on_release flag: run /sbin/container_release_agent on exit?
+
+Other subsystems such as cpusets may add additional files in each
+container dir
+
+New containers are created using the mkdir system call or shell
+command. The properties of a container, such as its flags, are
+modified by writing to the appropriate file in that containers
+directory, as listed above.
+
+The named hierarchical structure of nested containers allows partitioning
+a large system into nested, dynamically changeable, "soft-partitions".
+
+The attachment of each task, automatically inherited at fork by any
+children of that task, to a container allows organizing the work load
+on a system into related sets of tasks. A task may be re-attached to
+any other container, if allowed by the permissions on the necessary
+container file system directories.
+
+The use of a Linux virtual file system (vfs) to represent the
+container hierarchy provides for a familiar permission and name space
+for containers, with a minimum of additional kernel code.
+
+1.4 What does notify_on_release do ?
+------------------------------------
+
+If the notify_on_release flag is enabled (1) in a container, then whenever
+the last task in the container leaves (exits or attaches to some other
+container) and the last child container of that container is removed, then
+the kernel runs the command /sbin/container_release_agent, supplying the
+pathname (relative to the mount point of the container file system) of the
+abandoned container. This enables automatic removal of abandoned containers.
+The default value of notify_on_release in the root container at system
+boot is disabled (0). The default value of other containers at creation
+is the current value of their parents notify_on_release setting.
+
+1.5 How do I use containers ?
+--------------------------
+
+To start a new job that is to be contained within a container, the steps are:
+
+ 1) mkdir /dev/container
+ 2) mount -t container container /dev/container
+ 3) Create the new container by doing mkdir's and write's (or echo's) in
+ the /dev/container virtual file system.
+ 4) Start a task that will be the "founding father" of the new job.
+ 5) Attach that task to the new container by writing its pid to the
+ /dev/container tasks file for that container.
+ 6) fork, exec or clone the job tasks from this founding father task.
+
+For example, the following sequence of commands will setup a container
+named "Charlie", containing just CPUs 2 and 3, and Memory Node 1,
+and then start a subshell 'sh' in that container:
+
+ mount -t container none /dev/container
+ cd /dev/container
+ mkdir Charlie
+ cd Charlie
+ /bin/echo $$ > tasks
+ sh
+ # The subshell 'sh' is now running in container Charlie
+ # The next line should display '/Charlie'
+ cat /proc/self/container
+
+In the future, a C library interface to containers will likely be
+available. For now, the only way to query or modify containers is
+via the container file system, using the various cd, mkdir, echo, cat,
+rmdir commands from the shell, or their equivalent from C.
+
+2. Usage Examples and Syntax
+============================
+
+2.1 Basic Usage
+---------------
+
+Creating, modifying, using the containers can be done through the container
+virtual filesystem.
+
+To mount it, type:
+# mount -t container none /dev/container
+
+Then under /dev/container you can find a tree that corresponds to the
+tree of the containers in the system. For instance, /dev/container
+is the container that holds the whole system.
+
+If you want to create a new container under /dev/container:
+# cd /dev/container
+# mkdir my_container
+
+Now you want to do something with this container.
+# cd my_container
+
+In this directory you can find several files:
+# ls
+notify_on_release tasks
+
+Now attach your shell to this container:
+# /bin/echo $$ > tasks
+
+You can also create containers inside your container by using mkdir in this
+directory.
+# mkdir my_sub_cs
+
+To remove a container, just use rmdir:
+# rmdir my_sub_cs
+This will fail if the container is in use (has containers inside, or has
+processes attached).
+
+2.2 Attaching processes
+-----------------------
+
+# /bin/echo PID > tasks
+
+Note that it is PID, not PIDs. You can only attach ONE task at a time.
+If you have several tasks to attach, you have to do it one after another:
+
+# /bin/echo PID1 > tasks
+# /bin/echo PID2 > tasks
+ ...
+# /bin/echo PIDn > tasks
+
+
+3. Questions
+============
+
+Q: what's up with this '/bin/echo' ?
+A: bash's builtin 'echo' command does not check calls to write() against
+ errors. If you use it in the container file system, you won't be
+ able to tell whether a command succeeded or failed.
+
+Q: When I attach processes, only the first of the line gets really attached !
+A: We can only return one error code per call to write(). So you should also
+ put only ONE pid.
+
--
^ permalink raw reply [flat|nested] 12+ messages in thread* [PATCH 2/6] Cpusets hooked into containers
2006-10-20 18:38 [PATCH 0/6] Generic Process Containers menage
2006-10-20 18:38 ` [PATCH 1/6] Generic container system abstracted from cpusets code menage
@ 2006-10-20 18:38 ` menage
2006-11-06 6:34 ` [ckrm-tech] " Balbir Singh
2006-10-20 18:38 ` [PATCH 3/6] Add generic multi-subsystem API to containers menage
` (3 subsequent siblings)
5 siblings, 1 reply; 12+ messages in thread
From: menage @ 2006-10-20 18:38 UTC (permalink / raw)
To: akpm, pj, sekharan
Cc: linux-kernel, ckrm-tech, jlan, mbligh, rohitseth, winget,
Simon.Derr
[-- Attachment #1: cpusets_using_containers.patch --]
[-- Type: text/plain, Size: 78527 bytes --]
This patch removes the process grouping code from the cpusets code,
instead hooking it into the generic container system. This temporarily
adds cpuset-specific code in kernel/container.c, which is removed by
the next patch in the series.
Signed-off-by: Paul Menage <menage@google.com>
---
Documentation/cpusets.txt | 80 +-
fs/super.c | 5
include/linux/container.h | 7
include/linux/cpuset.h | 25
include/linux/fs.h | 2
include/linux/sched.h | 4
init/Kconfig | 19
kernel/container.c | 105 +++
kernel/cpuset.c | 1321 +++++-----------------------------------------
kernel/exit.c | 2
kernel/fork.c | 7
mm/oom_kill.c | 6
12 files changed, 317 insertions(+), 1266 deletions(-)
Index: container-2.6.19-rc2/include/linux/container.h
===================================================================
--- container-2.6.19-rc2.orig/include/linux/container.h
+++ container-2.6.19-rc2/include/linux/container.h
@@ -47,6 +47,10 @@ struct container {
struct container *parent; /* my parent */
struct dentry *dentry; /* container fs entry */
+
+#ifdef CONFIG_CPUSETS
+ struct cpuset *cpuset;
+#endif
};
/* struct cftype:
@@ -79,6 +83,9 @@ struct cftype {
int container_add_file(struct container *cont, const struct cftype *cft);
int container_is_removed(const struct container *cont);
+#ifdef CONFIG_CPUSETS_LEGACY_API
+void container_set_release_agent_path(const char *path);
+#endif
#else /* !CONFIG_CONTAINERS */
Index: container-2.6.19-rc2/include/linux/cpuset.h
===================================================================
--- container-2.6.19-rc2.orig/include/linux/cpuset.h
+++ container-2.6.19-rc2/include/linux/cpuset.h
@@ -11,16 +11,15 @@
#include <linux/sched.h>
#include <linux/cpumask.h>
#include <linux/nodemask.h>
+#include <linux/container.h>
#ifdef CONFIG_CPUSETS
-extern int number_of_cpusets; /* How many cpusets are defined in system? */
+extern int number_of_cpusets; /* How many cpusets are defined in system? */
extern int cpuset_init_early(void);
extern int cpuset_init(void);
extern void cpuset_init_smp(void);
-extern void cpuset_fork(struct task_struct *p);
-extern void cpuset_exit(struct task_struct *p);
extern cpumask_t cpuset_cpus_allowed(struct task_struct *p);
extern nodemask_t cpuset_mems_allowed(struct task_struct *p);
void cpuset_init_current_mems_allowed(void);
@@ -47,10 +46,6 @@ extern void __cpuset_memory_pressure_bum
extern struct file_operations proc_cpuset_operations;
extern char *cpuset_task_status_allowed(struct task_struct *task, char *buffer);
-
-extern void cpuset_lock(void);
-extern void cpuset_unlock(void);
-
extern int cpuset_mem_spread_node(void);
static inline int cpuset_do_page_mem_spread(void)
@@ -65,13 +60,22 @@ static inline int cpuset_do_slab_mem_spr
extern void cpuset_track_online_nodes(void);
+extern int cpuset_can_attach_task(struct container *cont,
+ struct task_struct *tsk);
+extern void cpuset_attach_task(struct container *cont,
+ struct task_struct *tsk);
+extern void cpuset_post_attach_task(struct container *cont,
+ struct container *oldcont,
+ struct task_struct *tsk);
+extern int cpuset_populate_dir(struct container *cont);
+extern int cpuset_create(struct container *cont);
+extern void cpuset_destroy(struct container *cont);
+
#else /* !CONFIG_CPUSETS */
static inline int cpuset_init_early(void) { return 0; }
static inline int cpuset_init(void) { return 0; }
static inline void cpuset_init_smp(void) {}
-static inline void cpuset_fork(struct task_struct *p) {}
-static inline void cpuset_exit(struct task_struct *p) {}
static inline cpumask_t cpuset_cpus_allowed(struct task_struct *p)
{
@@ -110,9 +114,6 @@ static inline char *cpuset_task_status_a
return buffer;
}
-static inline void cpuset_lock(void) {}
-static inline void cpuset_unlock(void) {}
-
static inline int cpuset_mem_spread_node(void)
{
return 0;
Index: container-2.6.19-rc2/kernel/exit.c
===================================================================
--- container-2.6.19-rc2.orig/kernel/exit.c
+++ container-2.6.19-rc2/kernel/exit.c
@@ -29,7 +29,6 @@
#include <linux/mempolicy.h>
#include <linux/taskstats_kern.h>
#include <linux/delayacct.h>
-#include <linux/cpuset.h>
#include <linux/container.h>
#include <linux/syscalls.h>
#include <linux/signal.h>
@@ -922,7 +921,6 @@ fastcall NORET_TYPE void do_exit(long co
__exit_files(tsk);
__exit_fs(tsk);
exit_thread();
- cpuset_exit(tsk);
container_exit(tsk);
exit_keys(tsk);
Index: container-2.6.19-rc2/kernel/fork.c
===================================================================
--- container-2.6.19-rc2.orig/kernel/fork.c
+++ container-2.6.19-rc2/kernel/fork.c
@@ -30,7 +30,6 @@
#include <linux/nsproxy.h>
#include <linux/capability.h>
#include <linux/cpu.h>
-#include <linux/cpuset.h>
#include <linux/container.h>
#include <linux/security.h>
#include <linux/swap.h>
@@ -1055,13 +1054,12 @@ static struct task_struct *copy_process(
p->io_wait = NULL;
p->audit_context = NULL;
container_fork(p);
- cpuset_fork(p);
#ifdef CONFIG_NUMA
p->mempolicy = mpol_copy(p->mempolicy);
if (IS_ERR(p->mempolicy)) {
retval = PTR_ERR(p->mempolicy);
p->mempolicy = NULL;
- goto bad_fork_cleanup_cpuset;
+ goto bad_fork_cleanup_container;
}
mpol_fix_fork_child_flag(p);
#endif
@@ -1287,9 +1285,8 @@ bad_fork_cleanup_security:
bad_fork_cleanup_policy:
#ifdef CONFIG_NUMA
mpol_free(p->mempolicy);
-bad_fork_cleanup_cpuset:
+bad_fork_cleanup_container:
#endif
- cpuset_exit(p);
container_exit(p);
bad_fork_cleanup_delays_binfmt:
delayacct_tsk_free(p);
Index: container-2.6.19-rc2/kernel/container.c
===================================================================
--- container-2.6.19-rc2.orig/kernel/container.c
+++ container-2.6.19-rc2/kernel/container.c
@@ -55,6 +55,7 @@
#include <linux/time.h>
#include <linux/backing-dev.h>
#include <linux/sort.h>
+#include <linux/cpuset.h>
#include <asm/uaccess.h>
#include <asm/atomic.h>
@@ -92,6 +93,18 @@ static struct container top_container =
.children = LIST_HEAD_INIT(top_container.children),
};
+/* The path to use for release notifications. No locking between
+ * setting and use - so if userspace updates this while subcontainers
+ * exist, you could miss a notification */
+static char release_agent_path[PATH_MAX] = "/sbin/container_release_agent";
+
+void container_set_release_agent_path(const char *path)
+{
+ container_manage_lock();
+ strcpy(release_agent_path, path);
+ container_manage_unlock();
+}
+
static struct vfsmount *container_mount;
static struct super_block *container_sb;
@@ -397,7 +410,7 @@ static void container_release_agent(cons
return;
i = 0;
- argv[i++] = "/sbin/container_release_agent";
+ argv[i++] = release_agent_path;
argv[i++] = (char *)pathbuf;
argv[i] = NULL;
@@ -438,6 +451,7 @@ static void check_for_release(struct con
buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
if (!buf)
return;
+
if (container_path(cont, buf, PAGE_SIZE) < 0)
kfree(buf);
else
@@ -486,7 +500,7 @@ static int attach_task(struct container
pid_t pid;
struct task_struct *tsk;
struct container *oldcont;
- int retval;
+ int retval = 0;
if (sscanf(pidbuf, "%d", &pid) != 1)
return -EIO;
@@ -513,7 +527,9 @@ static int attach_task(struct container
get_task_struct(tsk);
}
- retval = security_task_setscheduler(tsk, 0, NULL);
+#ifdef CONFIG_CPUSETS
+ retval = cpuset_can_attach_task(cont, tsk);
+#endif
if (retval) {
put_task_struct(tsk);
return retval;
@@ -533,8 +549,16 @@ static int attach_task(struct container
rcu_assign_pointer(tsk->container, cont);
task_unlock(tsk);
+#ifdef CONFIG_CPUSETS
+ cpuset_attach_task(cont, tsk);
+#endif
+
mutex_unlock(&callback_mutex);
+#ifdef CONFIG_CPUSETS
+ cpuset_post_attach_task(cont, oldcont, tsk);
+#endif
+
put_task_struct(tsk);
synchronize_rcu();
if (atomic_dec_and_test(&oldcont->count))
@@ -549,6 +573,7 @@ typedef enum {
FILE_DIR,
FILE_NOTIFY_ON_RELEASE,
FILE_TASKLIST,
+ FILE_RELEASE_AGENT,
} container_filetype_t;
static ssize_t container_common_file_write(struct container *cont,
@@ -562,8 +587,7 @@ static ssize_t container_common_file_wri
char *pathbuf = NULL;
int retval = 0;
- /* Crude upper limit on largest legitimate cpulist user might write. */
- if (nbytes > 100 + 6 * NR_CPUS)
+ if (nbytes >= PATH_MAX)
return -E2BIG;
/* +1 for nul-terminator */
@@ -590,6 +614,20 @@ static ssize_t container_common_file_wri
case FILE_TASKLIST:
retval = attach_task(cont, buffer, &pathbuf);
break;
+ case FILE_RELEASE_AGENT:
+ {
+ if (nbytes < sizeof(release_agent_path)) {
+ /* We never write anything other than '\0'
+ * into the last char of release_agent_path,
+ * so it always remains a NUL-terminated
+ * string */
+ strncpy(release_agent_path, buffer, nbytes);
+ release_agent_path[nbytes] = 0;
+ } else {
+ retval = -ENOSPC;
+ }
+ break;
+ }
default:
retval = -EINVAL;
goto out2;
@@ -643,6 +681,17 @@ static ssize_t container_common_file_rea
case FILE_NOTIFY_ON_RELEASE:
*s++ = notify_on_release(cont) ? '1' : '0';
break;
+ case FILE_RELEASE_AGENT:
+ {
+ size_t n;
+ container_manage_lock();
+ n = strnlen(release_agent_path, sizeof(release_agent_path));
+ n = min(n, (size_t) PAGE_SIZE);
+ strncpy(s, release_agent_path, n);
+ container_manage_unlock();
+ s += n;
+ break;
+ }
default:
retval = -EINVAL;
goto out;
@@ -978,6 +1027,13 @@ static struct cftype cft_notify_on_relea
.private = FILE_NOTIFY_ON_RELEASE,
};
+static struct cftype cft_release_agent = {
+ .name = "release_agent",
+ .read = container_common_file_read,
+ .write = container_common_file_write,
+ .private = FILE_RELEASE_AGENT,
+};
+
static int container_populate_dir(struct container *cont)
{
int err;
@@ -986,6 +1042,13 @@ static int container_populate_dir(struct
return err;
if ((err = container_add_file(cont, &cft_tasks)) < 0)
return err;
+ if ((cont == &top_container) &&
+ (err = container_add_file(cont, &cft_release_agent)) < 0)
+ return err;
+#ifdef CONFIG_CPUSETS
+ if ((err = cpuset_populate_dir(cont)) < 0)
+ return err;
+#endif
return 0;
}
@@ -1017,6 +1080,12 @@ static long container_create(struct cont
cont->parent = parent;
+#ifdef CONFIG_CPUSETS
+ err = cpuset_create(cont);
+ if (err)
+ goto err_unlock_free;
+#endif
+
mutex_lock(&callback_mutex);
list_add(&cont->sibling, &cont->parent->children);
number_of_containers++;
@@ -1038,11 +1107,14 @@ static long container_create(struct cont
return 0;
err_remove:
+#ifdef CONFIG_CPUSETS
+ cpuset_destroy(cont);
+#endif
mutex_lock(&callback_mutex);
list_del(&cont->sibling);
number_of_containers--;
mutex_unlock(&callback_mutex);
-
+ err_unlock_free:
mutex_unlock(&manage_mutex);
kfree(cont);
return err;
@@ -1097,6 +1169,9 @@ static int container_rmdir(struct inode
dput(d);
number_of_containers--;
mutex_unlock(&callback_mutex);
+#ifdef CONFIG_CPUSETS
+ cpuset_destroy(cont);
+#endif
if (list_empty(&parent->children))
check_for_release(parent, &pathbuf);
mutex_unlock(&manage_mutex);
@@ -1283,6 +1358,24 @@ void container_unlock(void)
mutex_unlock(&callback_mutex);
}
+void container_manage_lock(void)
+{
+ mutex_lock(&manage_mutex);
+}
+
+/**
+ * container_manage_unlock - release lock on container changes
+ *
+ * Undo the lock taken in a previous container_manage_lock() call.
+ */
+
+void container_manage_unlock(void)
+{
+ mutex_unlock(&manage_mutex);
+}
+
+
+
/*
* proc_container_show()
* - Print tasks container path into seq_file.
Index: container-2.6.19-rc2/kernel/cpuset.c
===================================================================
--- container-2.6.19-rc2.orig/kernel/cpuset.c
+++ container-2.6.19-rc2/kernel/cpuset.c
@@ -54,8 +54,6 @@
#include <asm/atomic.h>
#include <linux/mutex.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
@@ -77,20 +75,8 @@ struct cpuset {
cpumask_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'.
- */
- struct list_head sibling; /* my parents children */
- struct list_head children; /* my children */
-
+ struct container *container; /* Task container */
struct cpuset *parent; /* my parent */
- struct dentry *dentry; /* cpuset fs entry */
/*
* Copy of global cpuset_mems_generation as of the most
@@ -106,8 +92,6 @@ typedef enum {
CS_CPU_EXCLUSIVE,
CS_MEM_EXCLUSIVE,
CS_MEMORY_MIGRATE,
- CS_REMOVED,
- CS_NOTIFY_ON_RELEASE,
CS_SPREAD_PAGE,
CS_SPREAD_SLAB,
} cpuset_flagbits_t;
@@ -123,16 +107,6 @@ static inline int is_mem_exclusive(const
return test_bit(CS_MEM_EXCLUSIVE, &cs->flags);
}
-static inline int is_removed(const struct cpuset *cs)
-{
- return test_bit(CS_REMOVED, &cs->flags);
-}
-
-static inline int notify_on_release(const struct cpuset *cs)
-{
- return test_bit(CS_NOTIFY_ON_RELEASE, &cs->flags);
-}
-
static inline int is_memory_migrate(const struct cpuset *cs)
{
return test_bit(CS_MEMORY_MIGRATE, &cs->flags);
@@ -173,388 +147,33 @@ 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),
-};
-
-static struct vfsmount *cpuset_mount;
-static struct super_block *cpuset_sb;
-
-/*
- * We have two global cpuset mutexes below. They can nest.
- * It is ok to first take manage_mutex, then nest callback_mutex. We also
- * require taking task_lock() when dereferencing a tasks cpuset pointer.
- * See "The task_lock() exception", at the end of this comment.
- *
- * A task must hold both mutexes to modify cpusets. If a task
- * holds manage_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 manage_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_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_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_mutex or callback_mutex can't rely
- * on the count field not changing. However, if the count goes to
- * zero, then only attach_task(), which holds both mutexes, 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_mutex or callback_mutex can safely assume that
- * if the count is zero, it will stay zero. Similarly, if a task
- * holds manage_mutex or callback_mutex on a cpuset with zero count, it
- * knows that the cpuset won't be removed, as cpuset_rmdir() needs
- * both of those mutexes.
- *
- * The cpuset_common_file_write handler for operations that modify
- * the cpuset hierarchy holds manage_mutex across the entire operation,
- * single threading all such cpuset modifications across the system.
- *
- * 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 mutex. 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_mutex
- * 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), 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 mutexes, however there are several performance
- * critical places that need to reference task->cpuset without the
- * expense of grabbing a system global mutex. 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().
- */
-
-static DEFINE_MUTEX(manage_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.
- */
-
-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 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_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);
- }
- iput(inode);
-}
-
-static struct dentry_operations cpuset_dops = {
- .d_iput = cpuset_diput,
-};
-
-static struct dentry *cpuset_get_dentry(struct dentry *parent, const char *name)
-{
- struct dentry *d = lookup_one_len(name, parent, strlen(name));
- if (!IS_ERR(d))
- d->d_op = &cpuset_dops;
- return d;
-}
-
-static void remove_dir(struct dentry *d)
-{
- struct dentry *parent = dget(d->d_parent);
-
- d_delete(d);
- simple_rmdir(parent->d_inode, d);
- dput(parent);
-}
-
-/*
- * NOTE : the dentry must have been dget()'ed
- */
-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 struct super_operations cpuset_ops = {
- .statfs = simple_statfs,
- .drop_inode = generic_delete_inode,
};
-static int cpuset_fill_super(struct super_block *sb, void *unused_data,
- int unused_silent)
-{
- struct inode *inode;
- struct dentry *root;
-
- 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;
-
- 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) */
- inc_nlink(inode);
- } else {
- return -ENOMEM;
- }
-
- root = d_alloc_root(inode);
- if (!root) {
- iput(inode);
- return -ENOMEM;
- }
- sb->s_root = root;
- return 0;
-}
-
+#ifdef CONFIG_CPUSETS_LEGACY_API
+/* 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 "container" instead */
static int cpuset_get_sb(struct file_system_type *fs_type,
int flags, const char *unused_dev_name,
void *data, struct vfsmount *mnt)
{
- return get_sb_single(fs_type, flags, data, cpuset_fill_super, mnt);
+ struct file_system_type *container_fs = get_fs_type("container");
+ int ret = -ENODEV;
+ container_set_release_agent_path("/sbin/cpuset_release_agent");
+ if (container_fs) {
+ ret = container_fs->get_sb(container_fs, flags,
+ unused_dev_name,
+ data, mnt);
+ put_filesystem(container_fs);
+ }
+ return ret;
}
static struct file_system_type cpuset_fs_type = {
.name = "cpuset",
.get_sb = cpuset_get_sb,
- .kill_sb = kill_litter_super,
-};
-
-/* struct cftype:
- *
- * 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.
- *
- *
- * 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
- */
-
-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)
-{
- return dentry->d_fsdata;
-}
-
-static inline struct cftype *__d_cft(struct dentry *dentry)
-{
- return dentry->d_fsdata;
-}
-
-/*
- * Call with manage_mutex held. Writes path of cpuset into buf.
- * Returns 0 on success, -errno on error.
- */
-
-static int cpuset_path(const struct cpuset *cs, char *buf, int buflen)
-{
- char *start;
-
- start = buf + buflen;
-
- *--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)
- continue;
- if (--start < buf)
- return -ENAMETOOLONG;
- *start = '/';
- }
- memmove(buf, start, buf + buflen - start);
- return 0;
-}
-
-/*
- * 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 mutex, 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_mutex, but we still don't, so as to minimize
- * the time manage_mutex is held.
- */
-
-static void cpuset_release_agent(const char *pathbuf)
-{
- char *argv[3], *envp[3];
- int i;
-
- if (!pathbuf)
- return;
-
- i = 0;
- argv[i++] = "/sbin/cpuset_release_agent";
- argv[i++] = (char *)pathbuf;
- argv[i] = NULL;
-
- i = 0;
- /* minimal command environment */
- envp[i++] = "HOME=/";
- envp[i++] = "PATH=/sbin:/bin:/usr/sbin:/usr/bin";
- envp[i] = NULL;
-
- call_usermodehelper(argv[0], argv, envp, 0);
- kfree(pathbuf);
-}
-
-/*
- * 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_mutex is dropped.
- * Call here with manage_mutex 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_mutex held and the address
- * of the pathbuf pointer, then dropping manage_mutex, then calling
- * cpuset_release_agent() with pathbuf, as set by check_for_release().
- */
-
-static void check_for_release(struct cpuset *cs, char **ppathbuf)
-{
- if (notify_on_release(cs) && atomic_read(&cs->count) == 0 &&
- list_empty(&cs->children)) {
- char *buf;
-
- buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
- if (!buf)
- return;
- if (cpuset_path(cs, buf, PAGE_SIZE) < 0)
- kfree(buf);
- else
- *ppathbuf = buf;
- }
-}
-
+#endif /* CONFIG_CPUSETS_LEGACY_API */
/*
* Return in *pmask the portion of a cpusets's cpus_allowed that
* are online. If none are online, walk up the cpuset hierarchy
@@ -652,20 +271,20 @@ void cpuset_update_task_memory_state(voi
struct task_struct *tsk = current;
struct cpuset *cs;
- if (tsk->cpuset == &top_cpuset) {
+ if (tsk->container->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);
+ cs = rcu_dereference(tsk->container->cpuset);
my_cpusets_mem_gen = cs->mems_generation;
rcu_read_unlock();
}
if (my_cpusets_mem_gen != tsk->cpuset_mems_generation) {
- mutex_lock(&callback_mutex);
+ container_lock();
task_lock(tsk);
- cs = tsk->cpuset; /* Maybe changed when task not locked */
+ cs = tsk->container->cpuset; /* Maybe changed when task not locked */
guarantee_online_mems(cs, &tsk->mems_allowed);
tsk->cpuset_mems_generation = cs->mems_generation;
if (is_spread_page(cs))
@@ -677,7 +296,7 @@ void cpuset_update_task_memory_state(voi
else
tsk->flags &= ~PF_SPREAD_SLAB;
task_unlock(tsk);
- mutex_unlock(&callback_mutex);
+ container_unlock();
mpol_rebind_task(tsk, &tsk->mems_allowed);
}
}
@@ -720,10 +339,12 @@ static int is_cpuset_subset(const struct
static int validate_change(const struct cpuset *cur, const struct cpuset *trial)
{
+ struct container *cont;
struct cpuset *c, *par;
/* Each of our child cpusets must be a subset of us */
- list_for_each_entry(c, &cur->children, sibling) {
+ list_for_each_entry(cont, &cur->container->children, sibling) {
+ c = cont->cpuset;
if (!is_cpuset_subset(c, trial))
return -EBUSY;
}
@@ -737,7 +358,8 @@ static int validate_change(const struct
return -EACCES;
/* If either I or some sibling (!= me) is exclusive, we can't overlap */
- list_for_each_entry(c, &par->children, sibling) {
+ list_for_each_entry(cont, &par->container->children, sibling) {
+ c = cont->cpuset;
if ((is_cpu_exclusive(trial) || is_cpu_exclusive(c)) &&
c != cur &&
cpus_intersects(trial->cpus_allowed, c->cpus_allowed))
@@ -767,6 +389,7 @@ static int validate_change(const struct
static void update_cpu_domains(struct cpuset *cur)
{
+ struct container *cont;
struct cpuset *c, *par = cur->parent;
cpumask_t pspan, cspan;
@@ -778,7 +401,8 @@ static void update_cpu_domains(struct cp
* children
*/
pspan = par->cpus_allowed;
- list_for_each_entry(c, &par->children, sibling) {
+ list_for_each_entry(cont, &par->container->children, sibling) {
+ c = cont->cpuset;
if (is_cpu_exclusive(c))
cpus_andnot(pspan, pspan, c->cpus_allowed);
}
@@ -795,7 +419,8 @@ static void update_cpu_domains(struct cp
* Get all cpus from current cpuset's cpus_allowed not part
* of exclusive children
*/
- list_for_each_entry(c, &cur->children, sibling) {
+ list_for_each_entry(cont, &cur->container->children, sibling) {
+ c = cont->cpuset;
if (is_cpu_exclusive(c))
cpus_andnot(cspan, cspan, c->cpus_allowed);
}
@@ -830,9 +455,9 @@ static int update_cpumask(struct cpuset
if (retval < 0)
return retval;
cpus_unchanged = cpus_equal(cs->cpus_allowed, trialcs.cpus_allowed);
- mutex_lock(&callback_mutex);
+ container_lock();
cs->cpus_allowed = trialcs.cpus_allowed;
- mutex_unlock(&callback_mutex);
+ container_unlock();
if (is_cpu_exclusive(cs) && !cpus_unchanged)
update_cpu_domains(cs);
return 0;
@@ -876,15 +501,15 @@ static void cpuset_migrate_mm(struct mm_
cpuset_update_task_memory_state();
- mutex_lock(&callback_mutex);
+ container_lock();
tsk->mems_allowed = *to;
- mutex_unlock(&callback_mutex);
+ container_unlock();
do_migrate_pages(mm, from, to, MPOL_MF_MOVE_ALL);
- mutex_lock(&callback_mutex);
- guarantee_online_mems(tsk->cpuset, &tsk->mems_allowed);
- mutex_unlock(&callback_mutex);
+ container_lock();
+ guarantee_online_mems(tsk->container->cpuset, &tsk->mems_allowed);
+ container_unlock();
}
/*
@@ -911,12 +536,14 @@ static int update_nodemask(struct cpuset
int migrate;
int fudge;
int retval;
+ struct container *cont;
/* top_cpuset.mems_allowed tracks node_online_map; it's read-only */
if (cs == &top_cpuset)
return -EACCES;
trialcs = *cs;
+ cont = cs->container;
retval = nodelist_parse(buf, trialcs.mems_allowed);
if (retval < 0)
goto done;
@@ -934,10 +561,10 @@ static int update_nodemask(struct cpuset
if (retval < 0)
goto done;
- mutex_lock(&callback_mutex);
+ container_lock();
cs->mems_allowed = trialcs.mems_allowed;
cs->mems_generation = cpuset_mems_generation++;
- mutex_unlock(&callback_mutex);
+ container_unlock();
set_cpuset_being_rebound(cs); /* causes mpol_copy() rebind */
@@ -953,13 +580,13 @@ static int update_nodemask(struct cpuset
* enough mmarray[] w/o using GFP_ATOMIC.
*/
while (1) {
- ntasks = atomic_read(&cs->count); /* guess */
+ ntasks = atomic_read(&cs->container->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)
+ if (atomic_read(&cs->container->count) <= ntasks)
break; /* got enough */
write_unlock_irq(&tasklist_lock); /* try again */
kfree(mmarray);
@@ -976,7 +603,7 @@ static int update_nodemask(struct cpuset
"Cpuset mempolicy rebind incomplete.\n");
continue;
}
- if (p->cpuset != cs)
+ if (p->container != cont)
continue;
mm = get_task_mm(p);
if (!mm)
@@ -1059,15 +686,15 @@ static int update_flag(cpuset_flagbits_t
return err;
cpu_exclusive_changed =
(is_cpu_exclusive(cs) != is_cpu_exclusive(&trialcs));
- mutex_lock(&callback_mutex);
+ container_lock();
if (turning_on)
set_bit(bit, &cs->flags);
else
clear_bit(bit, &cs->flags);
- mutex_unlock(&callback_mutex);
+ container_unlock();
if (cpu_exclusive_changed)
- update_cpu_domains(cs);
+ update_cpu_domains(cs);
return 0;
}
@@ -1169,85 +796,35 @@ static int fmeter_getrate(struct fmeter
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_mutex. May take callback_mutex and task_lock of
- * the task 'pid' during call.
- */
-
-static int attach_task(struct cpuset *cs, char *pidbuf, char **ppathbuf)
+int cpuset_can_attach_task(struct container *cont, struct task_struct *tsk)
{
- pid_t pid;
- struct task_struct *tsk;
- struct cpuset *oldcs;
- cpumask_t cpus;
- nodemask_t from, to;
- struct mm_struct *mm;
- int retval;
+ struct cpuset *cs = cont->cpuset;
- if (sscanf(pidbuf, "%d", &pid) != 1)
- return -EIO;
if (cpus_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed))
return -ENOSPC;
- if (pid) {
- read_lock(&tasklist_lock);
-
- 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);
-
- if ((current->euid) && (current->euid != tsk->uid)
- && (current->euid != tsk->suid)) {
- put_task_struct(tsk);
- return -EACCES;
- }
- } else {
- tsk = current;
- get_task_struct(tsk);
- }
-
- retval = security_task_setscheduler(tsk, 0, NULL);
- if (retval) {
- put_task_struct(tsk);
- return retval;
- }
-
- mutex_lock(&callback_mutex);
-
- task_lock(tsk);
- oldcs = tsk->cpuset;
- /*
- * After getting 'oldcs' cpuset ptr, be sure still not exiting.
- * If 'oldcs' might be the top_cpuset due to the_top_cpuset_hack
- * then fail this attach_task(), to avoid breaking top_cpuset.count.
- */
- if (tsk->flags & PF_EXITING) {
- task_unlock(tsk);
- mutex_unlock(&callback_mutex);
- put_task_struct(tsk);
- return -ESRCH;
- }
- atomic_inc(&cs->count);
- rcu_assign_pointer(tsk->cpuset, cs);
- task_unlock(tsk);
+ return security_task_setscheduler(tsk, 0, NULL);
+}
+void cpuset_attach_task(struct container *cont, struct task_struct *tsk)
+{
+ cpumask_t cpus;
+ struct cpuset *cs = cont->cpuset;
guarantee_online_cpus(cs, &cpus);
set_cpus_allowed(tsk, cpus);
+}
+
+void cpuset_post_attach_task(struct container *cont,
+ struct container *oldcont,
+ struct task_struct *tsk)
+{
+ nodemask_t from, to;
+ struct mm_struct *mm;
+ struct cpuset *cs = cont->cpuset;
+ struct cpuset *oldcs = oldcont->cpuset;
from = oldcs->mems_allowed;
to = cs->mems_allowed;
-
- mutex_unlock(&callback_mutex);
-
mm = get_task_mm(tsk);
if (mm) {
mpol_rebind_mm(mm, &to);
@@ -1256,39 +833,31 @@ static int attach_task(struct cpuset *cs
mmput(mm);
}
- put_task_struct(tsk);
- synchronize_rcu();
- if (atomic_dec_and_test(&oldcs->count))
- check_for_release(oldcs, ppathbuf);
- return 0;
}
/* 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_MEMORY_PRESSURE_ENABLED,
FILE_MEMORY_PRESSURE,
FILE_SPREAD_PAGE,
FILE_SPREAD_SLAB,
- FILE_TASKLIST,
} cpuset_filetype_t;
-static ssize_t cpuset_common_file_write(struct file *file, const char __user *userbuf,
+static ssize_t cpuset_common_file_write(struct container *cont,
+ struct cftype *cft,
+ struct file *file,
+ const char __user *userbuf,
size_t nbytes, loff_t *unused_ppos)
{
- struct cpuset *cs = __d_cs(file->f_dentry->d_parent);
- struct cftype *cft = __d_cft(file->f_dentry);
+ struct cpuset *cs = cont->cpuset;
cpuset_filetype_t type = cft->private;
char *buffer;
- char *pathbuf = NULL;
int retval = 0;
/* Crude upper limit on largest legitimate cpulist user might write. */
@@ -1305,9 +874,9 @@ static ssize_t cpuset_common_file_write(
}
buffer[nbytes] = 0; /* nul-terminate */
- mutex_lock(&manage_mutex);
+ container_manage_lock();
- if (is_removed(cs)) {
+ if (container_is_removed(cont)) {
retval = -ENODEV;
goto out2;
}
@@ -1325,9 +894,6 @@ static ssize_t cpuset_common_file_write(
case FILE_MEM_EXCLUSIVE:
retval = update_flag(CS_MEM_EXCLUSIVE, cs, buffer);
break;
- case FILE_NOTIFY_ON_RELEASE:
- retval = update_flag(CS_NOTIFY_ON_RELEASE, cs, buffer);
- break;
case FILE_MEMORY_MIGRATE:
retval = update_flag(CS_MEMORY_MIGRATE, cs, buffer);
break;
@@ -1345,9 +911,6 @@ static ssize_t cpuset_common_file_write(
retval = update_flag(CS_SPREAD_SLAB, cs, buffer);
cs->mems_generation = cpuset_mems_generation++;
break;
- case FILE_TASKLIST:
- retval = attach_task(cs, buffer, &pathbuf);
- break;
default:
retval = -EINVAL;
goto out2;
@@ -1356,30 +919,12 @@ static ssize_t cpuset_common_file_write(
if (retval == 0)
retval = nbytes;
out2:
- mutex_unlock(&manage_mutex);
- cpuset_release_agent(pathbuf);
+ container_manage_unlock();
out1:
kfree(buffer);
return retval;
}
-static ssize_t cpuset_file_write(struct file *file, const 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->write)
- retval = cft->write(file, buf, nbytes, ppos);
- else
- retval = cpuset_common_file_write(file, buf, nbytes, ppos);
-
- 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
@@ -1396,9 +941,9 @@ static int cpuset_sprintf_cpulist(char *
{
cpumask_t mask;
- mutex_lock(&callback_mutex);
+ container_lock();
mask = cs->cpus_allowed;
- mutex_unlock(&callback_mutex);
+ container_unlock();
return cpulist_scnprintf(page, PAGE_SIZE, mask);
}
@@ -1407,18 +952,20 @@ static int cpuset_sprintf_memlist(char *
{
nodemask_t mask;
- mutex_lock(&callback_mutex);
+ container_lock();
mask = cs->mems_allowed;
- mutex_unlock(&callback_mutex);
+ container_unlock();
return nodelist_scnprintf(page, PAGE_SIZE, mask);
}
-static ssize_t cpuset_common_file_read(struct file *file, char __user *buf,
- size_t nbytes, loff_t *ppos)
+static ssize_t cpuset_common_file_read(struct container *cont,
+ struct cftype *cft,
+ struct file *file,
+ char __user *buf,
+ size_t nbytes, loff_t *ppos)
{
- struct cftype *cft = __d_cft(file->f_dentry);
- struct cpuset *cs = __d_cs(file->f_dentry->d_parent);
+ struct cpuset *cs = cont->cpuset;
cpuset_filetype_t type = cft->private;
char *page;
ssize_t retval = 0;
@@ -1442,9 +989,6 @@ static ssize_t cpuset_common_file_read(s
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;
case FILE_MEMORY_MIGRATE:
*s++ = is_memory_migrate(cs) ? '1' : '0';
break;
@@ -1472,391 +1016,97 @@ out:
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;
-}
-
-static int cpuset_file_open(struct inode *inode, struct file *file)
-{
- 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;
-}
-
-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);
- return 0;
-}
-
-/*
- * cpuset_rename - Only allow simple rename of directories in place.
- */
-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 int cpuset_create_file(struct dentry *dentry, int mode)
-{
- struct inode *inode;
-
- if (!dentry)
- return -ENOENT;
- if (dentry->d_inode)
- return -EEXIST;
-
- inode = cpuset_new_inode(mode);
- if (!inode)
- return -ENOMEM;
-
- if (S_ISDIR(mode)) {
- inode->i_op = &cpuset_dir_inode_operations;
- inode->i_fop = &simple_dir_operations;
-
- /* start off with i_nlink == 2 (for "." entry) */
- inc_nlink(inode);
- } else if (S_ISREG(mode)) {
- inode->i_size = 0;
- inode->i_fop = &cpuset_file_operations;
- }
-
- 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;
- inc_nlink(parent->d_inode);
- cs->dentry = dentry;
- }
- dput(dentry);
-
- return error;
-}
-
-static int cpuset_add_file(struct dentry *dir, const struct cftype *cft)
-{
- 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;
-}
-
-/*
- * 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().
- */
-
-/* cpusets_tasks_read array */
-
-struct ctr_struct {
- char *buf;
- int bufsz;
-};
-
-/*
- * 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;
-
- read_lock(&tasklist_lock);
-
- do_each_thread(g, p) {
- if (p->cpuset == cs) {
- pidarray[n++] = p->pid;
- if (unlikely(n == npids))
- goto array_full;
- }
- } 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;
-}
-
-/*
- * 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.
- */
-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 mutexes, and does not take any.
- */
-static int cpuset_tasks_open(struct inode *unused, struct file *file)
-{
- struct cpuset *cs = __d_cs(file->f_dentry->d_parent);
- struct ctr_struct *ctr;
- pid_t *pidarray;
- int npids;
- char c;
-
- if (!(file->f_mode & FMODE_READ))
- return 0;
-
- ctr = kmalloc(sizeof(*ctr), GFP_KERNEL);
- if (!ctr)
- goto err0;
-
- /*
- * 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;
-
-err2:
- kfree(pidarray);
-err1:
- kfree(ctr);
-err0:
- return -ENOMEM;
-}
-
-static ssize_t cpuset_tasks_read(struct file *file, char __user *buf,
- size_t nbytes, loff_t *ppos)
-{
- struct ctr_struct *ctr = file->private_data;
-
- 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;
-}
/*
* 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,
-};
-
static struct cftype cft_cpus = {
.name = "cpus",
+ .read = cpuset_common_file_read,
+ .write = cpuset_common_file_write,
.private = FILE_CPULIST,
};
static struct cftype cft_mems = {
.name = "mems",
+ .read = cpuset_common_file_read,
+ .write = cpuset_common_file_write,
.private = FILE_MEMLIST,
};
static struct cftype cft_cpu_exclusive = {
.name = "cpu_exclusive",
+ .read = cpuset_common_file_read,
+ .write = cpuset_common_file_write,
.private = FILE_CPU_EXCLUSIVE,
};
static struct cftype cft_mem_exclusive = {
.name = "mem_exclusive",
+ .read = cpuset_common_file_read,
+ .write = cpuset_common_file_write,
.private = FILE_MEM_EXCLUSIVE,
};
-static struct cftype cft_notify_on_release = {
- .name = "notify_on_release",
- .private = FILE_NOTIFY_ON_RELEASE,
-};
-
static struct cftype cft_memory_migrate = {
.name = "memory_migrate",
+ .read = cpuset_common_file_read,
+ .write = cpuset_common_file_write,
.private = FILE_MEMORY_MIGRATE,
};
static struct cftype cft_memory_pressure_enabled = {
.name = "memory_pressure_enabled",
+ .read = cpuset_common_file_read,
+ .write = cpuset_common_file_write,
.private = FILE_MEMORY_PRESSURE_ENABLED,
};
static struct cftype cft_memory_pressure = {
.name = "memory_pressure",
+ .read = cpuset_common_file_read,
+ .write = cpuset_common_file_write,
.private = FILE_MEMORY_PRESSURE,
};
static struct cftype cft_spread_page = {
.name = "memory_spread_page",
+ .read = cpuset_common_file_read,
+ .write = cpuset_common_file_write,
.private = FILE_SPREAD_PAGE,
};
static struct cftype cft_spread_slab = {
.name = "memory_spread_slab",
+ .read = cpuset_common_file_read,
+ .write = cpuset_common_file_write,
.private = FILE_SPREAD_SLAB,
};
-static int cpuset_populate_dir(struct dentry *cs_dentry)
+int cpuset_populate_dir(struct container *cont)
{
int err;
- if ((err = cpuset_add_file(cs_dentry, &cft_cpus)) < 0)
+ if ((err = container_add_file(cont, &cft_cpus)) < 0)
return err;
- if ((err = cpuset_add_file(cs_dentry, &cft_mems)) < 0)
+ if ((err = container_add_file(cont, &cft_mems)) < 0)
return err;
- if ((err = cpuset_add_file(cs_dentry, &cft_cpu_exclusive)) < 0)
+ if ((err = container_add_file(cont, &cft_cpu_exclusive)) < 0)
return err;
- if ((err = cpuset_add_file(cs_dentry, &cft_mem_exclusive)) < 0)
+ if ((err = container_add_file(cont, &cft_mem_exclusive)) < 0)
return err;
- if ((err = cpuset_add_file(cs_dentry, &cft_notify_on_release)) < 0)
+ if ((err = container_add_file(cont, &cft_memory_migrate)) < 0)
return err;
- if ((err = cpuset_add_file(cs_dentry, &cft_memory_migrate)) < 0)
+ if ((err = container_add_file(cont, &cft_memory_pressure)) < 0)
return err;
- if ((err = cpuset_add_file(cs_dentry, &cft_memory_pressure)) < 0)
+ if ((err = container_add_file(cont, &cft_spread_page)) < 0)
return err;
- if ((err = cpuset_add_file(cs_dentry, &cft_spread_page)) < 0)
- return err;
- if ((err = cpuset_add_file(cs_dentry, &cft_spread_slab)) < 0)
- return err;
- if ((err = cpuset_add_file(cs_dentry, &cft_tasks)) < 0)
+ if ((err = container_add_file(cont, &cft_spread_slab)) < 0)
return err;
+ /* memory_pressure_enabled is in root cpuset only */
+ if (err == 0 && !cont->parent)
+ err = container_add_file(cont, &cft_memory_pressure_enabled);
return 0;
}
@@ -1869,66 +1119,31 @@ static int cpuset_populate_dir(struct de
* Must be called with the mutex on the parent inode held
*/
-static long cpuset_create(struct cpuset *parent, const char *name, int mode)
+int cpuset_create(struct container *cont)
{
struct cpuset *cs;
- int err;
+ struct cpuset *parent = cont->parent->cpuset;
cs = kmalloc(sizeof(*cs), GFP_KERNEL);
if (!cs)
return -ENOMEM;
- mutex_lock(&manage_mutex);
cpuset_update_task_memory_state();
cs->flags = 0;
- if (notify_on_release(parent))
- set_bit(CS_NOTIFY_ON_RELEASE, &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);
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);
cs->mems_generation = cpuset_mems_generation++;
fmeter_init(&cs->fmeter);
cs->parent = parent;
-
- mutex_lock(&callback_mutex);
- list_add(&cs->sibling, &cs->parent->children);
+ cont->cpuset = cs;
+ cs->container = cont;
number_of_cpusets++;
- mutex_unlock(&callback_mutex);
-
- err = cpuset_create_dir(cs, name, mode);
- if (err < 0)
- goto err;
-
- /*
- * Release manage_mutex before cpuset_populate_dir() because it
- * will down() this new directory's i_mutex and if we race with
- * another mkdir, we might deadlock.
- */
- mutex_unlock(&manage_mutex);
-
- err = cpuset_populate_dir(cs->dentry);
- /* If err < 0, we have a half-filled directory - oh well ;) */
return 0;
-err:
- list_del(&cs->sibling);
- mutex_unlock(&manage_mutex);
- kfree(cs);
- return err;
-}
-
-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);
}
/*
@@ -1942,49 +1157,16 @@ static int cpuset_mkdir(struct inode *di
* nesting would risk an ABBA deadlock.
*/
-static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry)
+void cpuset_destroy(struct container *cont)
{
- struct cpuset *cs = dentry->d_fsdata;
- struct dentry *d;
- struct cpuset *parent;
- char *pathbuf = NULL;
-
- /* the vfs holds both inode->i_mutex already */
+ struct cpuset *cs = cont->cpuset;
- mutex_lock(&manage_mutex);
cpuset_update_task_memory_state();
- if (atomic_read(&cs->count) > 0) {
- mutex_unlock(&manage_mutex);
- return -EBUSY;
- }
- if (!list_empty(&cs->children)) {
- mutex_unlock(&manage_mutex);
- return -EBUSY;
- }
if (is_cpu_exclusive(cs)) {
int retval = update_flag(CS_CPU_EXCLUSIVE, cs, "0");
- if (retval < 0) {
- mutex_unlock(&manage_mutex);
- return retval;
- }
+ BUG_ON(retval);
}
- parent = cs->parent;
- mutex_lock(&callback_mutex);
- set_bit(CS_REMOVED, &cs->flags);
- 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--;
- mutex_unlock(&callback_mutex);
- if (list_empty(&parent->children))
- check_for_release(parent, &pathbuf);
- mutex_unlock(&manage_mutex);
- cpuset_release_agent(pathbuf);
- return 0;
}
/*
@@ -1995,10 +1177,10 @@ static int cpuset_rmdir(struct inode *un
int __init cpuset_init_early(void)
{
- struct task_struct *tsk = current;
-
- tsk->cpuset = &top_cpuset;
- tsk->cpuset->mems_generation = cpuset_mems_generation++;
+ struct container *cont = current->container;
+ cont->cpuset = &top_cpuset;
+ top_cpuset.container = cont;
+ cont->cpuset->mems_generation = cpuset_mems_generation++;
return 0;
}
@@ -2010,39 +1192,21 @@ int __init cpuset_init_early(void)
int __init cpuset_init(void)
{
- struct dentry *root;
- int err;
-
+ int err = 0;
top_cpuset.cpus_allowed = CPU_MASK_ALL;
top_cpuset.mems_allowed = NODE_MASK_ALL;
fmeter_init(&top_cpuset.fmeter);
top_cpuset.mems_generation = cpuset_mems_generation++;
- init_task.cpuset = &top_cpuset;
-
+#ifdef CONFIG_CPUSETS_LEGACY_API
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;
- }
- root = cpuset_mount->mnt_sb->s_root;
- root->d_fsdata = &top_cpuset;
- inc_nlink(root->d_inode);
- top_cpuset.dentry = root;
- root->d_inode->i_op = &cpuset_dir_inode_operations;
+ return err;
+#endif
+
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;
+ return 0;
}
#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_MEMORY_HOTPLUG)
@@ -2099,15 +1263,15 @@ static void guarantee_online_cpus_mems_i
static void common_cpu_mem_hotplug_unplug(void)
{
- mutex_lock(&manage_mutex);
- mutex_lock(&callback_mutex);
+ container_manage_lock();
+ container_lock();
guarantee_online_cpus_mems_in_subtree(&top_cpuset);
top_cpuset.cpus_allowed = cpu_online_map;
top_cpuset.mems_allowed = node_online_map;
- mutex_unlock(&callback_mutex);
- mutex_unlock(&manage_mutex);
+ container_unlock();
+ container_manage_unlock();
}
#endif
@@ -2158,111 +1322,6 @@ void __init cpuset_init_smp(void)
}
/**
- * 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)
-{
- task_lock(current);
- child->cpuset = current->cpuset;
- atomic_inc(&child->cpuset->count);
- task_unlock(current);
-}
-
-/**
- * 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_mutex mutex 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_mutex
- * or callback_mutex. 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_mutex, not callback_mutex, because
- * it is holding that mutex while calling check_for_release(),
- * which calls kmalloc(), so can't be called holding callback_mutex().
- *
- * 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.
- *
- * the_top_cpuset_hack:
- *
- * Set the exiting tasks cpuset to the root cpuset (top_cpuset).
- *
- * Don't leave a task unable to allocate memory, as that is an
- * accident waiting to happen should someone add a callout in
- * do_exit() after the cpuset_exit() call that might allocate.
- * If a task tries to allocate memory with an invalid cpuset,
- * it will oops in cpuset_update_task_memory_state().
- *
- * We call cpuset_exit() while the task is still competent to
- * handle notify_on_release(), then leave the task attached to
- * the root cpuset (top_cpuset) for the remainder of its exit.
- *
- * To do this properly, we would increment the reference count on
- * top_cpuset, and near the very end of the kernel/exit.c do_exit()
- * code we would add a second cpuset function call, to drop that
- * reference. This would just create an unnecessary hot spot on
- * the top_cpuset reference count, to no avail.
- *
- * Normally, holding a reference to a cpuset without bumping its
- * count is unsafe. The cpuset could go away, or someone could
- * attach us to a different cpuset, decrementing the count on
- * the first cpuset that we never incremented. But in this case,
- * top_cpuset isn't going away, and either task has PF_EXITING set,
- * which wards off any attach_task() attempts, or task is a failed
- * fork, never visible to attach_task.
- *
- * Another way to do this would be to set the cpuset pointer
- * to NULL here, and check in cpuset_update_task_memory_state()
- * for a NULL pointer. This hack avoids that NULL check, for no
- * cost (other than this way too long comment ;).
- **/
-
-void cpuset_exit(struct task_struct *tsk)
-{
- struct cpuset *cs;
-
- cs = tsk->cpuset;
- tsk->cpuset = &top_cpuset; /* the_top_cpuset_hack - see above */
-
- if (notify_on_release(cs)) {
- char *pathbuf = NULL;
-
- mutex_lock(&manage_mutex);
- if (atomic_dec_and_test(&cs->count))
- check_for_release(cs, &pathbuf);
- mutex_unlock(&manage_mutex);
- cpuset_release_agent(pathbuf);
- } else {
- atomic_dec(&cs->count);
- }
-}
-
-/**
* cpuset_cpus_allowed - return cpus_allowed mask from a tasks cpuset.
* @tsk: pointer to task_struct from which to obtain cpuset->cpus_allowed.
*
@@ -2276,11 +1335,11 @@ cpumask_t cpuset_cpus_allowed(struct tas
{
cpumask_t mask;
- mutex_lock(&callback_mutex);
+ container_lock();
task_lock(tsk);
- guarantee_online_cpus(tsk->cpuset, &mask);
+ guarantee_online_cpus(tsk->container->cpuset, &mask);
task_unlock(tsk);
- mutex_unlock(&callback_mutex);
+ container_unlock();
return mask;
}
@@ -2304,11 +1363,11 @@ nodemask_t cpuset_mems_allowed(struct ta
{
nodemask_t mask;
- mutex_lock(&callback_mutex);
+ container_lock();
task_lock(tsk);
- guarantee_online_mems(tsk->cpuset, &mask);
+ guarantee_online_mems(tsk->container->cpuset, &mask);
task_unlock(tsk);
- mutex_unlock(&callback_mutex);
+ container_unlock();
return mask;
}
@@ -2408,45 +1467,18 @@ int __cpuset_zone_allowed(struct zone *z
return 1;
/* Not hardwall and node outside mems_allowed: scan up cpusets */
- mutex_lock(&callback_mutex);
+ container_lock();
task_lock(current);
- cs = nearest_exclusive_ancestor(current->cpuset);
+ cs = nearest_exclusive_ancestor(current->container->cpuset);
task_unlock(current);
allowed = node_isset(node, cs->mems_allowed);
- mutex_unlock(&callback_mutex);
+ container_unlock();
return allowed;
}
/**
- * cpuset_lock - lock out any changes to cpuset structures
- *
- * The out of memory (oom) code needs to mutex_lock cpusets
- * from being changed while it scans the tasklist looking for a
- * task in an overlapping cpuset. Expose callback_mutex 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_mutex.
- */
-
-void cpuset_lock(void)
-{
- mutex_lock(&callback_mutex);
-}
-
-/**
- * cpuset_unlock - release lock on cpuset changes
- *
- * Undo the lock taken in a previous cpuset_lock() call.
- */
-
-void cpuset_unlock(void)
-{
- mutex_unlock(&callback_mutex);
-}
-
-/**
* cpuset_mem_spread_node() - On which node to begin search for a page
*
* If a task is marked PF_SPREAD_PAGE or PF_SPREAD_SLAB (as for
@@ -2506,7 +1538,7 @@ int cpuset_excl_nodes_overlap(const stru
task_unlock(current);
goto done;
}
- cs1 = nearest_exclusive_ancestor(current->cpuset);
+ cs1 = nearest_exclusive_ancestor(current->container->cpuset);
task_unlock(current);
task_lock((struct task_struct *)p);
@@ -2514,7 +1546,7 @@ int cpuset_excl_nodes_overlap(const stru
task_unlock((struct task_struct *)p);
goto done;
}
- cs2 = nearest_exclusive_ancestor(p->cpuset);
+ cs2 = nearest_exclusive_ancestor(p->container->cpuset);
task_unlock((struct task_struct *)p);
overlap = nodes_intersects(cs1->mems_allowed, cs2->mems_allowed);
@@ -2553,70 +1585,11 @@ void __cpuset_memory_pressure_bump(void)
struct cpuset *cs;
task_lock(current);
- cs = current->cpuset;
+ cs = current->container->cpuset;
fmeter_markevent(&cs->fmeter);
task_unlock(current);
}
-/*
- * 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_mutex, keeping attach_task() from changing it
- * anyway. No need to check that tsk->cpuset != NULL, thanks to
- * the_top_cpuset_hack in cpuset_exit(), which sets an exiting tasks
- * cpuset to top_cpuset.
- */
-static int proc_cpuset_show(struct seq_file *m, void *v)
-{
- struct pid *pid;
- struct task_struct *tsk;
- char *buf;
- int retval;
-
- retval = -ENOMEM;
- buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
- if (!buf)
- goto out;
-
- retval = -ESRCH;
- pid = m->private;
- tsk = get_pid_task(pid, PIDTYPE_PID);
- if (!tsk)
- goto out_free;
-
- retval = -EINVAL;
- mutex_lock(&manage_mutex);
-
- retval = cpuset_path(tsk->cpuset, buf, PAGE_SIZE);
- if (retval < 0)
- goto out_unlock;
- seq_puts(m, buf);
- seq_putc(m, '\n');
-out_unlock:
- mutex_unlock(&manage_mutex);
- put_task_struct(tsk);
-out_free:
- kfree(buf);
-out:
- return retval;
-}
-
-static int cpuset_open(struct inode *inode, struct file *file)
-{
- struct pid *pid = PROC_I(inode)->pid;
- return single_open(file, proc_cpuset_show, pid);
-}
-
-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)
{
Index: container-2.6.19-rc2/init/Kconfig
===================================================================
--- container-2.6.19-rc2.orig/init/Kconfig
+++ container-2.6.19-rc2/init/Kconfig
@@ -238,17 +238,12 @@ config IKCONFIG_PROC
through /proc/config.gz.
config CONTAINERS
- bool "Container support"
- help
- This option will let you create and manage process containers,
- which can be used to aggregate multiple processes, e.g. for
- the purposes of resource tracking.
-
- Say N if unsure
+ bool
config CPUSETS
bool "Cpuset support"
depends on SMP
+ select CONTAINERS
help
This option will let you create and manage CPUSETs which
allow dynamically partitioning a system into sets of CPUs and
@@ -257,6 +252,16 @@ config CPUSETS
Say N if unsure.
+config CPUSETS_LEGACY_API
+ bool "Maintain the existing cpuset API for generic container features"
+ depends on CPUSETS
+ default n
+ help
+ This option preserves the following legacy CPUset behaviour:
+ - /proc/<pid>/cpuset
+ - defaults cpuset subsystem to enabled
+ - defaults container release agent to /sbin/cpuset_release_agent
+
config RELAY
bool "Kernel->user space relay support (formerly relayfs)"
help
Index: container-2.6.19-rc2/mm/oom_kill.c
===================================================================
--- container-2.6.19-rc2.orig/mm/oom_kill.c
+++ container-2.6.19-rc2/mm/oom_kill.c
@@ -394,7 +394,7 @@ void out_of_memory(struct zonelist *zone
show_mem();
}
- cpuset_lock();
+ container_lock();
read_lock(&tasklist_lock);
/*
@@ -428,7 +428,7 @@ retry:
/* Found nothing?!?! Either we hang forever, or we panic. */
if (!p) {
read_unlock(&tasklist_lock);
- cpuset_unlock();
+ container_unlock();
panic("Out of memory and no killable processes...\n");
}
@@ -440,7 +440,7 @@ retry:
out:
read_unlock(&tasklist_lock);
- cpuset_unlock();
+ container_unlock();
/*
* Give "p" a good chance of killing itself before we
Index: container-2.6.19-rc2/include/linux/sched.h
===================================================================
--- container-2.6.19-rc2.orig/include/linux/sched.h
+++ container-2.6.19-rc2/include/linux/sched.h
@@ -721,7 +721,6 @@ extern unsigned int max_cache_size;
struct io_context; /* See blkdev.h */
struct container;
-struct cpuset;
#define NGROUPS_SMALL 32
#define NGROUPS_PER_BLOCK ((int)(PAGE_SIZE / sizeof(gid_t)))
struct group_info {
@@ -1002,7 +1001,6 @@ struct task_struct {
short il_next;
#endif
#ifdef CONFIG_CPUSETS
- struct cpuset *cpuset;
nodemask_t mems_allowed;
int cpuset_mems_generation;
int cpuset_mem_spread_rotor;
@@ -1433,7 +1431,7 @@ static inline int thread_group_empty(str
/*
* Protects ->fs, ->files, ->mm, ->group_info, ->comm, keyring
* subscriptions and synchronises with wait4(). Also used in procfs. Also
- * pins the final release of task.io_context. Also protects ->cpuset.
+ * pins the final release of task.io_context. Also protects ->container.
*
* Nests both inside and outside of read_lock(&tasklist_lock).
* It must not be nested with write_lock_irq(&tasklist_lock),
Index: container-2.6.19-rc2/Documentation/cpusets.txt
===================================================================
--- container-2.6.19-rc2.orig/Documentation/cpusets.txt
+++ container-2.6.19-rc2/Documentation/cpusets.txt
@@ -7,6 +7,7 @@ Written by Simon.Derr@bull.net
Portions Copyright (c) 2004-2006 Silicon Graphics, Inc.
Modified by Paul Jackson <pj@sgi.com>
Modified by Christoph Lameter <clameter@sgi.com>
+Modified by Paul Menage <menage@google.com>
CONTENTS:
=========
@@ -16,10 +17,9 @@ CONTENTS:
1.2 Why are cpusets needed ?
1.3 How are cpusets implemented ?
1.4 What are exclusive cpusets ?
- 1.5 What does notify_on_release do ?
- 1.6 What is memory_pressure ?
- 1.7 What is memory spread ?
- 1.8 How do I use cpusets ?
+ 1.5 What is memory_pressure ?
+ 1.6 What is memory spread ?
+ 1.7 How do I use cpusets ?
2. Usage Examples and Syntax
2.1 Basic Usage
2.2 Adding/removing cpus
@@ -43,18 +43,19 @@ hierarchy visible in a virtual file syst
hooks, beyond what is already present, required to manage dynamic
job placement on large systems.
-Each task has a pointer to a cpuset. Multiple tasks may reference
-the same cpuset. Requests by a task, using the sched_setaffinity(2)
-system call to include CPUs in its CPU affinity mask, and using the
-mbind(2) and set_mempolicy(2) system calls to include Memory Nodes
-in its memory policy, are both filtered through that tasks cpuset,
-filtering out any CPUs or Memory Nodes not in that cpuset. The
-scheduler will not schedule a task on a CPU that is not allowed in
-its cpus_allowed vector, and the kernel page allocator will not
-allocate a page on a node that is not allowed in the requesting tasks
-mems_allowed vector.
+Cpusets use the generic container subsystem described in
+Documentation/container.txt.
-User level code may create and destroy cpusets by name in the cpuset
+Requests by a task, using the sched_setaffinity(2) system call to
+include CPUs in its CPU affinity mask, and using the mbind(2) and
+set_mempolicy(2) system calls to include Memory Nodes in its memory
+policy, are both filtered through that tasks cpuset, filtering out any
+CPUs or Memory Nodes not in that cpuset. The scheduler will not
+schedule a task on a CPU that is not allowed in its cpus_allowed
+vector, and the kernel page allocator will not allocate a page on a
+node that is not allowed in the requesting tasks mems_allowed vector.
+
+User level code may create and destroy cpusets by name in the container
virtual file system, manage the attributes and permissions of these
cpusets and which CPUs and Memory Nodes are assigned to each cpuset,
specify and query to which cpuset a task is assigned, and list the
@@ -117,7 +118,7 @@ Cpusets extends these two mechanisms as
- Cpusets are sets of allowed CPUs and Memory Nodes, known to the
kernel.
- Each task in the system is attached to a cpuset, via a pointer
- in the task structure to a reference counted cpuset structure.
+ in the task structure to a reference counted container structure.
- Calls to sched_setaffinity are filtered to just those CPUs
allowed in that tasks cpuset.
- Calls to mbind and set_mempolicy are filtered to just
@@ -152,15 +153,10 @@ into the rest of the kernel, none in per
- in page_alloc.c, to restrict memory to allowed nodes.
- in vmscan.c, to restrict page recovery to the current cpuset.
-In addition a new file system, of type "cpuset" may be mounted,
-typically at /dev/cpuset, to enable browsing and modifying the cpusets
-presently known to the kernel. No new system calls are added for
-cpusets - all support for querying and modifying cpusets is via
-this cpuset file system.
-
-Each task under /proc has an added file named 'cpuset', displaying
-the cpuset name, as the path relative to the root of the cpuset file
-system.
+You should mount the "container" filesystem type in order to enable
+browsing and modifying the cpusets presently known to the kernel. No
+new system calls are added for cpusets - all support for querying and
+modifying cpusets is via this cpuset file system.
The /proc/<pid>/status file for each task has two added lines,
displaying the tasks cpus_allowed (on which CPUs it may be scheduled)
@@ -170,16 +166,15 @@ in the format seen in the following exam
Cpus_allowed: ffffffff,ffffffff,ffffffff,ffffffff
Mems_allowed: ffffffff,ffffffff
-Each cpuset is represented by a directory in the cpuset file system
-containing the following files describing that cpuset:
+Each cpuset is represented by a directory in the container file system
+containing (on top of the standard container files) the following
+files describing that cpuset:
- cpus: list of CPUs in that cpuset
- mems: list of Memory Nodes in that cpuset
- memory_migrate flag: if set, move pages to cpusets nodes
- cpu_exclusive flag: is cpu placement exclusive?
- mem_exclusive flag: is memory placement exclusive?
- - tasks: list of tasks (by pid) attached to that cpuset
- - notify_on_release flag: run /sbin/cpuset_release_agent on exit?
- memory_pressure: measure of how much paging pressure in cpuset
In addition, the root cpuset only has the following file:
@@ -253,21 +248,7 @@ such as requests from interrupt handlers
outside even a mem_exclusive cpuset.
-1.5 What does notify_on_release do ?
-------------------------------------
-
-If the notify_on_release flag is enabled (1) in a cpuset, then whenever
-the last task in the cpuset leaves (exits or attaches to some other
-cpuset) and the last child cpuset of that cpuset is removed, then
-the kernel runs the command /sbin/cpuset_release_agent, supplying the
-pathname (relative to the mount point of the cpuset file system) of the
-abandoned cpuset. This enables automatic removal of abandoned cpusets.
-The default value of notify_on_release in the root cpuset at system
-boot is disabled (0). The default value of other cpusets at creation
-is the current value of their parents notify_on_release setting.
-
-
-1.6 What is memory_pressure ?
+1.5 What is memory_pressure ?
-----------------------------
The memory_pressure of a cpuset provides a simple per-cpuset metric
of the rate that the tasks in a cpuset are attempting to free up in
@@ -324,7 +305,7 @@ the tasks in the cpuset, in units of rec
times 1000.
-1.7 What is memory spread ?
+1.6 What is memory spread ?
---------------------------
There are two boolean flag files per cpuset that control where the
kernel allocates pages for the file system buffers and related in
@@ -395,7 +376,7 @@ data set, the memory allocation across t
can become very uneven.
-1.8 How do I use cpusets ?
+1.7 How do I use cpusets ?
--------------------------
In order to minimize the impact of cpusets on critical kernel
@@ -485,7 +466,7 @@ than stress the kernel.
To start a new job that is to be contained within a cpuset, the steps are:
1) mkdir /dev/cpuset
- 2) mount -t cpuset none /dev/cpuset
+ 2) mount -t container none /dev/cpuset
3) Create the new cpuset by doing mkdir's and write's (or echo's) in
the /dev/cpuset virtual file system.
4) Start a task that will be the "founding father" of the new job.
@@ -497,7 +478,7 @@ For example, the following sequence of c
named "Charlie", containing just CPUs 2 and 3, and Memory Node 1,
and then start a subshell 'sh' in that cpuset:
- mount -t cpuset none /dev/cpuset
+ mount -t container none /dev/cpuset
cd /dev/cpuset
mkdir Charlie
cd Charlie
@@ -507,7 +488,7 @@ and then start a subshell 'sh' in that c
sh
# The subshell 'sh' is now running in cpuset Charlie
# The next line should display '/Charlie'
- cat /proc/self/cpuset
+ cat /proc/self/container
In the future, a C library interface to cpusets will likely be
available. For now, the only way to query or modify cpusets is
@@ -529,7 +510,7 @@ Creating, modifying, using the cpusets c
virtual filesystem.
To mount it, type:
-# mount -t cpuset none /dev/cpuset
+# mount -t container none /dev/cpuset
Then under /dev/cpuset you can find a tree that corresponds to the
tree of the cpusets in the system. For instance, /dev/cpuset
Index: container-2.6.19-rc2/fs/super.c
===================================================================
--- container-2.6.19-rc2.orig/fs/super.c
+++ container-2.6.19-rc2/fs/super.c
@@ -39,11 +39,6 @@
#include <linux/mutex.h>
#include <asm/uaccess.h>
-
-void get_filesystem(struct file_system_type *fs);
-void put_filesystem(struct file_system_type *fs);
-struct file_system_type *get_fs_type(const char *name);
-
LIST_HEAD(super_blocks);
DEFINE_SPINLOCK(sb_lock);
Index: container-2.6.19-rc2/include/linux/fs.h
===================================================================
--- container-2.6.19-rc2.orig/include/linux/fs.h
+++ container-2.6.19-rc2/include/linux/fs.h
@@ -1864,6 +1864,8 @@ extern int vfs_fstat(unsigned int, struc
extern int vfs_ioctl(struct file *, unsigned int, unsigned int, unsigned long);
+extern void get_filesystem(struct file_system_type *fs);
+extern void put_filesystem(struct file_system_type *fs);
extern struct file_system_type *get_fs_type(const char *name);
extern struct super_block *get_super(struct block_device *);
extern struct super_block *user_get_super(dev_t);
--
^ permalink raw reply [flat|nested] 12+ messages in thread* Re: [ckrm-tech] [PATCH 2/6] Cpusets hooked into containers
2006-10-20 18:38 ` [PATCH 2/6] Cpusets hooked into containers menage
@ 2006-11-06 6:34 ` Balbir Singh
2006-11-06 20:55 ` Paul Menage
0 siblings, 1 reply; 12+ messages in thread
From: Balbir Singh @ 2006-11-06 6:34 UTC (permalink / raw)
To: menage
Cc: akpm, pj, sekharan, ckrm-tech, jlan, Simon.Derr, linux-kernel,
mbligh, winget, rohitseth
[-- Attachment #1: Type: text/plain, Size: 606 bytes --]
menage@google.com wrote:
> This patch removes the process grouping code from the cpusets code,
> instead hooking it into the generic container system. This temporarily
> adds cpuset-specific code in kernel/container.c, which is removed by
> the next patch in the series.
>
> Signed-off-by: Paul Menage <menage@google.com>
I needed the following patches to get the cpuset code to compile.
Inlining two patches makes it hard to distinguish between the patches
and harder to read them, so I am attaching them along with this email.
--
Regards,
Balbir Singh,
Linux Technology Center,
IBM Software Labs
[-- Attachment #2: fix-cpuset-guarantee-online-cpus-mems-in-subtree.patch --]
[-- Type: text/x-patch, Size: 940 bytes --]
Signed-off-by: Balbir Singh <balbir@in.ibm.com>
---
kernel/cpuset.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff -puN kernel/cpuset.c~fix-cpuset-guarantee-online-cpus-mems-in-subtree kernel/cpuset.c
--- linux-2.6.19-rc2/kernel/cpuset.c~fix-cpuset-guarantee-online-cpus-mems-in-subtree 2006-11-06 11:41:25.000000000 +0530
+++ linux-2.6.19-rc2-balbir/kernel/cpuset.c 2006-11-06 11:43:12.000000000 +0530
@@ -1280,10 +1280,12 @@ int __init cpuset_init(void)
static void guarantee_online_cpus_mems_in_subtree(const struct cpuset *cur)
{
+ struct container *cont;
struct cpuset *c;
/* Each of our child cpusets mems must be online */
- list_for_each_entry(c, &cur->children, sibling) {
+ list_for_each_entry(cont, &cur->container->children, sibling) {
+ c = container_cs(cont);
guarantee_online_cpus_mems_in_subtree(c);
if (!cpus_empty(c->cpus_allowed))
guarantee_online_cpus(c, &c->cpus_allowed);
_
[-- Attachment #3: fix-cpuset-proc-operations.patch --]
[-- Type: text/x-patch, Size: 1225 bytes --]
Signed-off-by: Balbir Singh <balbir@in.ibm.com>
---
fs/proc/base.c | 7 -------
1 file changed, 7 deletions(-)
diff -puN fs/proc/base.c~fix-cpuset-proc-operations fs/proc/base.c
--- linux-2.6.19-rc2/fs/proc/base.c~fix-cpuset-proc-operations 2006-11-06 11:47:35.000000000 +0530
+++ linux-2.6.19-rc2-balbir/fs/proc/base.c 2006-11-06 11:48:27.000000000 +0530
@@ -68,7 +68,6 @@
#include <linux/security.h>
#include <linux/ptrace.h>
#include <linux/seccomp.h>
-#include <linux/cpuset.h>
#include <linux/container.h>
#include <linux/audit.h>
#include <linux/poll.h>
@@ -1788,9 +1787,6 @@ static struct pid_entry tgid_base_stuff[
#ifdef CONFIG_CONTAINERS
REG("container", S_IRUGO, container),
#endif
-#ifdef CONFIG_CPUSETS
- REG("cpuset", S_IRUGO, cpuset),
-#endif
INF("oom_score", S_IRUGO, oom_score),
REG("oom_adj", S_IRUGO|S_IWUSR, oom_adjust),
#ifdef CONFIG_AUDITSYSCALL
@@ -2065,9 +2061,6 @@ static struct pid_entry tid_base_stuff[]
#ifdef CONFIG_CONTAINERS
REG("container", S_IRUGO, container),
#endif
-#ifdef CONFIG_CPUSETS
- REG("cpuset", S_IRUGO, cpuset),
-#endif
INF("oom_score", S_IRUGO, oom_score),
REG("oom_adj", S_IRUGO|S_IWUSR, oom_adjust),
#ifdef CONFIG_AUDITSYSCALL
_
^ permalink raw reply [flat|nested] 12+ messages in thread* Re: [ckrm-tech] [PATCH 2/6] Cpusets hooked into containers
2006-11-06 6:34 ` [ckrm-tech] " Balbir Singh
@ 2006-11-06 20:55 ` Paul Menage
2006-11-06 21:09 ` Paul Jackson
2006-11-07 14:06 ` Balbir Singh
0 siblings, 2 replies; 12+ messages in thread
From: Paul Menage @ 2006-11-06 20:55 UTC (permalink / raw)
To: balbir
Cc: sekharan, ckrm-tech, jlan, Simon.Derr, linux-kernel, pj, mbligh,
winget, rohitseth
On 11/5/06, Balbir Singh <balbir@in.ibm.com> wrote:
>
> I needed the following patches to get the cpuset code to compile.
> Inlining two patches makes it hard to distinguish between the patches
> and harder to read them, so I am attaching them along with this email.
The first I missed due to not compiling with CONFIG_HOTPLUG_* - thanks
for the patch.
For the second, the following change to fs/proc/base.c should have
appeared in cpusets_using_containers.patch, but got left out due to
quilt misusage. It basically makes "cpuset" an alias for "container"
in the relevant /proc directories if CONFIG_CPUSETS_LEGACY_API is
defined.
--- container-2.6.19-rc2.orig/fs/proc/base.c
+++ container-2.6.19-rc2/fs/proc/base.c
@@ -69,7 +69,6 @@
#include <linux/ptrace.h>
#include <linux/seccomp.h>
#include <linux/container.h>
-#include <linux/cpuset.h>
#include <linux/audit.h>
#include <linux/poll.h>
#include <linux/nsproxy.h>
@@ -1784,8 +1783,8 @@ static struct pid_entry tgid_base_stuff[
#ifdef CONFIG_CONTAINERS
REG("container", S_IRUGO, container),
#endif
-#ifdef CONFIG_CPUSETS
- REG("cpuset", S_IRUGO, cpuset),
+#ifdef CONFIG_CPUSETS_LEGACY_API
+ REG("cpuset", S_IRUGO, container),
#endif
INF("oom_score", S_IRUGO, oom_score),
REG("oom_adj", S_IRUGO|S_IWUSR, oom_adjust),
@@ -2061,8 +2060,8 @@ static struct pid_entry tid_base_stuff[]
#ifdef CONFIG_CONTAINERS
REG("container", S_IRUGO, container),
#endif
-#ifdef CONFIG_CPUSETS
- REG("cpuset", S_IRUGO, cpuset),
+#ifdef CONFIG_CPUSETS_LEGACY_API
+ REG("cpuset", S_IRUGO, container),
#endif
INF("oom_score", S_IRUGO, oom_score),
REG("oom_adj", S_IRUGO|S_IWUSR, oom_adjust),
Paul
^ permalink raw reply [flat|nested] 12+ messages in thread* Re: [ckrm-tech] [PATCH 2/6] Cpusets hooked into containers
2006-11-06 20:55 ` Paul Menage
@ 2006-11-06 21:09 ` Paul Jackson
2006-11-06 21:22 ` Paul Menage
2006-11-07 14:06 ` Balbir Singh
1 sibling, 1 reply; 12+ messages in thread
From: Paul Jackson @ 2006-11-06 21:09 UTC (permalink / raw)
To: Paul Menage
Cc: balbir, sekharan, ckrm-tech, jlan, Simon.Derr, linux-kernel,
mbligh, winget, rohitseth
Paul M wrote:
> It basically makes "cpuset" an alias for "container"
> in the relevant /proc directories if CONFIG_CPUSETS_LEGACY_API is
> defined.
Paul M - I never replied to your initial CONFIG_CPUSETS_LEGACY_API
patch proposal - sorry.
An aspect of this proposal never made sense to me, so I put it aside
and went on to other things.
It is important to me that the current cpuset API be maintained. The
cpuset API seems to be working well, for a number of users.
Occassionally I will agree to subtle API changes (see another thread
concerning cpu_exclusive and sched_domain cpuset flags), but not
anything likely to break user code outright, except under duress.
But I presume this CONFIG_CPUSETS_LEGACY_API option means I either
get to build a kernel that supports the new container API, or a kernel
that supports the old cpuset API. That does not seem useful to me.
We need to support both API's, at runtime, at the same time. Not a choice
of API's at build time with a kernel CONFIG option.
Perhaps I am missing something ...
--
I won't rest till it's the best ...
Programmer, Linux Scalability
Paul Jackson <pj@sgi.com> 1.925.600.0401
^ permalink raw reply [flat|nested] 12+ messages in thread* Re: [ckrm-tech] [PATCH 2/6] Cpusets hooked into containers
2006-11-06 21:09 ` Paul Jackson
@ 2006-11-06 21:22 ` Paul Menage
0 siblings, 0 replies; 12+ messages in thread
From: Paul Menage @ 2006-11-06 21:22 UTC (permalink / raw)
To: Paul Jackson
Cc: balbir, sekharan, ckrm-tech, jlan, Simon.Derr, linux-kernel,
mbligh, winget, rohitseth
On 11/6/06, Paul Jackson <pj@sgi.com> wrote:
>
> But I presume this CONFIG_CPUSETS_LEGACY_API option means I either
> get to build a kernel that supports the new container API, or a kernel
> that supports the old cpuset API. That does not seem useful to me.
No. The changes triggered by CONFIG_CPUSETS_LEGACY_API are:
- /proc/<pid>/cpuset is an alias for /proc/<pid>/containers
- a dummy "cpuset" filesystem exists
- mounting the "cpuset" filesystem causes:
- the container filesystem to be mounted in its place
- the release agent to be switched to /sbin/cpuset_release_agent
- defaults the "cpuset" container type to be enabled
Paul
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [ckrm-tech] [PATCH 2/6] Cpusets hooked into containers
2006-11-06 20:55 ` Paul Menage
2006-11-06 21:09 ` Paul Jackson
@ 2006-11-07 14:06 ` Balbir Singh
1 sibling, 0 replies; 12+ messages in thread
From: Balbir Singh @ 2006-11-07 14:06 UTC (permalink / raw)
To: Paul Menage
Cc: sekharan, ckrm-tech, jlan, Simon.Derr, linux-kernel, pj, mbligh,
winget, rohitseth
Paul Menage wrote:
[snip]
> For the second, the following change to fs/proc/base.c should have
> appeared in cpusets_using_containers.patch, but got left out due to
> quilt misusage. It basically makes "cpuset" an alias for "container"
> in the relevant /proc directories if CONFIG_CPUSETS_LEGACY_API is
> defined.
>
[snip]
Yeah, this looks much better and more like the fix
--
Balbir Singh,
Linux Technology Center,
IBM Software Labs
^ permalink raw reply [flat|nested] 12+ messages in thread
* [PATCH 3/6] Add generic multi-subsystem API to containers
2006-10-20 18:38 [PATCH 0/6] Generic Process Containers menage
2006-10-20 18:38 ` [PATCH 1/6] Generic container system abstracted from cpusets code menage
2006-10-20 18:38 ` [PATCH 2/6] Cpusets hooked into containers menage
@ 2006-10-20 18:38 ` menage
2006-10-20 18:38 ` [PATCH 4/6] Simple CPU accounting container subsystem menage
` (2 subsequent siblings)
5 siblings, 0 replies; 12+ messages in thread
From: menage @ 2006-10-20 18:38 UTC (permalink / raw)
To: akpm, pj, sekharan
Cc: linux-kernel, ckrm-tech, jlan, mbligh, rohitseth, winget,
Simon.Derr
[-- Attachment #1: multiuser_container.patch --]
[-- Type: text/plain, Size: 29383 bytes --]
This patch removes all cpuset-specific knowlege from the container
system, replacing it with a generic API that can be used by multiple
subsystems. Cpusets is adapted to be a container subsystem.
Signed-off-by: Paul Menage <menage@google.com>
---
Documentation/containers.txt | 160 ++++++++++++++++++++++++++++++++++++++-
include/linux/container.h | 43 +++++++++-
include/linux/cpuset.h | 11 --
kernel/container.c | 175 ++++++++++++++++++++++++++++++++++++-------
kernel/cpuset.c | 126 ++++++++++++++++++++----------
5 files changed, 427 insertions(+), 88 deletions(-)
Index: container-2.6.19-rc2/include/linux/container.h
===================================================================
--- container-2.6.19-rc2.orig/include/linux/container.h
+++ container-2.6.19-rc2/include/linux/container.h
@@ -30,6 +30,8 @@ extern void container_unlock(void);
extern void container_manage_lock(void);
extern void container_manage_unlock(void);
+#define MAX_CONTAINER_SUBSYS 8
+
struct container {
unsigned long flags; /* "unsigned long" so bitops work */
@@ -48,9 +50,8 @@ struct container {
struct container *parent; /* my parent */
struct dentry *dentry; /* container fs entry */
-#ifdef CONFIG_CPUSETS
- struct cpuset *cpuset;
-#endif
+ /* Private pointers for each registered subsystem */
+ void *subsys[MAX_CONTAINER_SUBSYS];
};
/* struct cftype:
@@ -87,6 +88,42 @@ int container_is_removed(const struct co
void container_set_release_agent_path(const char *path);
#endif
+/* Container subsystem type. See Documentation/containers.txt for details */
+
+struct container_subsys {
+ int (*create)(struct container_subsys *ss, struct container *cont);
+ void (*destroy)(struct container_subsys *ss, struct container *cont);
+ int (*can_attach)(struct container_subsys *ss,
+ struct container *cont, struct task_struct *tsk);
+ void (*attach)(struct container_subsys *ss, struct container *cont,
+ struct container *old_cont, struct task_struct *tsk);
+ void (*post_attach)(struct container_subsys *ss,
+ struct container *cont,
+ struct container *old_cont,
+ struct task_struct *tsk);
+ int (*populate)(struct container_subsys *ss,
+ struct container *cont);
+
+ int subsys_id;
+#define MAX_CONTAINER_TYPE_NAMELEN 32
+ const char *name;
+
+ /* enabled/disabled flag - managed by container.c, but can be
+ * set to a default value before registration */
+ int enabled;
+
+ /* Fields below here are managed by container.c and should not
+ * be accessed by subsystems. */
+
+ /* file handler for enable/disable */
+ struct cftype enable_cft;
+ char enable_cft_filename[MAX_CONTAINER_TYPE_NAMELEN + 10];
+
+
+};
+
+int container_register_subsys(struct container_subsys *subsys);
+
#else /* !CONFIG_CONTAINERS */
static inline int container_init_early(void) { return 0; }
Index: container-2.6.19-rc2/include/linux/cpuset.h
===================================================================
--- container-2.6.19-rc2.orig/include/linux/cpuset.h
+++ container-2.6.19-rc2/include/linux/cpuset.h
@@ -60,17 +60,6 @@ static inline int cpuset_do_slab_mem_spr
extern void cpuset_track_online_nodes(void);
-extern int cpuset_can_attach_task(struct container *cont,
- struct task_struct *tsk);
-extern void cpuset_attach_task(struct container *cont,
- struct task_struct *tsk);
-extern void cpuset_post_attach_task(struct container *cont,
- struct container *oldcont,
- struct task_struct *tsk);
-extern int cpuset_populate_dir(struct container *cont);
-extern int cpuset_create(struct container *cont);
-extern void cpuset_destroy(struct container *cont);
-
#else /* !CONFIG_CPUSETS */
static inline int cpuset_init_early(void) { return 0; }
Index: container-2.6.19-rc2/kernel/container.c
===================================================================
--- container-2.6.19-rc2.orig/kernel/container.c
+++ container-2.6.19-rc2/kernel/container.c
@@ -55,7 +55,6 @@
#include <linux/time.h>
#include <linux/backing-dev.h>
#include <linux/sort.h>
-#include <linux/cpuset.h>
#include <asm/uaccess.h>
#include <asm/atomic.h>
@@ -70,6 +69,9 @@
*/
int number_of_containers __read_mostly;
+static struct container_subsys *subsys[MAX_CONTAINER_SUBSYS];
+static int subsys_count = 0;
+
/* bits in struct container flags field */
typedef enum {
CONT_REMOVED,
@@ -501,6 +503,7 @@ static int attach_task(struct container
struct task_struct *tsk;
struct container *oldcont;
int retval = 0;
+ int s;
if (sscanf(pidbuf, "%d", &pid) != 1)
return -EIO;
@@ -527,12 +530,15 @@ static int attach_task(struct container
get_task_struct(tsk);
}
-#ifdef CONFIG_CPUSETS
- retval = cpuset_can_attach_task(cont, tsk);
-#endif
- if (retval) {
- put_task_struct(tsk);
- return retval;
+ for (s = 0; s < subsys_count; s++) {
+ struct container_subsys *ss = subsys[s];
+ if (ss->enabled && ss->can_attach) {
+ retval = ss->can_attach(ss, cont, tsk);
+ if (retval) {
+ put_task_struct(tsk);
+ return retval;
+ }
+ }
}
mutex_lock(&callback_mutex);
@@ -549,15 +555,21 @@ static int attach_task(struct container
rcu_assign_pointer(tsk->container, cont);
task_unlock(tsk);
-#ifdef CONFIG_CPUSETS
- cpuset_attach_task(cont, tsk);
-#endif
+ for (s = 0; s < subsys_count; s++) {
+ struct container_subsys *ss = subsys[s];
+ if (ss->enabled && ss->attach) {
+ ss->attach(ss, cont, oldcont, tsk);
+ }
+ }
mutex_unlock(&callback_mutex);
-#ifdef CONFIG_CPUSETS
- cpuset_post_attach_task(cont, oldcont, tsk);
-#endif
+ for (s = 0; s < subsys_count; s++) {
+ struct container_subsys *ss = subsys[s];
+ if (ss->enabled && ss->post_attach) {
+ ss->post_attach(ss, cont, oldcont, tsk);
+ }
+ }
put_task_struct(tsk);
synchronize_rcu();
@@ -574,6 +586,7 @@ typedef enum {
FILE_NOTIFY_ON_RELEASE,
FILE_TASKLIST,
FILE_RELEASE_AGENT,
+ FILE_ENABLED,
} container_filetype_t;
static ssize_t container_common_file_write(struct container *cont,
@@ -628,6 +641,18 @@ static ssize_t container_common_file_wri
}
break;
}
+ case FILE_ENABLED: {
+ struct container_subsys *ss;
+ if (number_of_containers != 1) {
+ retval = -EBUSY;
+ goto out2;
+ }
+ /* The cftype object is embedded in the subsys */
+ ss = container_of(cft, struct container_subsys, enable_cft);
+ ss->enabled = simple_strtoul(buffer, NULL, 10) != 0;
+ retval = 0;
+ break;
+ }
default:
retval = -EINVAL;
goto out2;
@@ -692,6 +717,13 @@ static ssize_t container_common_file_rea
s += n;
break;
}
+ case FILE_ENABLED: {
+ struct container_subsys *ss;
+ /* The cftype object is embedded in the subsys */
+ ss = container_of(cft, struct container_subsys, enable_cft);
+ *s++ = ss->enabled ? '1' : '0';
+ break;
+ }
default:
retval = -EINVAL;
goto out;
@@ -1034,9 +1066,17 @@ static struct cftype cft_release_agent =
.private = FILE_RELEASE_AGENT,
};
+/* This is just a template for the per-subsystem xxx_enabled file */
+static struct cftype cft_enabled_template = {
+ .read = container_common_file_read,
+ .write = container_common_file_write,
+ .private = FILE_ENABLED,
+};
+
static int container_populate_dir(struct container *cont)
{
int err;
+ int s;
if ((err = container_add_file(cont, &cft_notify_on_release)) < 0)
return err;
@@ -1045,10 +1085,19 @@ static int container_populate_dir(struct
if ((cont == &top_container) &&
(err = container_add_file(cont, &cft_release_agent)) < 0)
return err;
-#ifdef CONFIG_CPUSETS
- if ((err = cpuset_populate_dir(cont)) < 0)
- return err;
-#endif
+ for (s = 0; s < subsys_count; s++) {
+ struct container_subsys *ss = subsys[s];
+ /* All subsystems have an "xxx_enabled" file in the top dir */
+ if (cont == &top_container &&
+ (err = container_add_file(cont, &ss->enable_cft)) < 0)
+ return err;
+ /* All subsystems live in the top dir; only enabled
+ * subsystems live in subdirs */
+ if (ss->enabled || (cont == &top_container)) {
+ if (ss->populate && (err = ss->populate(ss, cont)) < 0)
+ return err;
+ }
+ }
return 0;
}
@@ -1065,6 +1114,7 @@ static long container_create(struct cont
{
struct container *cont;
int err;
+ int s = 0;
cont = kmalloc(sizeof(*cont), GFP_KERNEL);
if (!cont)
@@ -1080,11 +1130,23 @@ static long container_create(struct cont
cont->parent = parent;
-#ifdef CONFIG_CPUSETS
- err = cpuset_create(cont);
- if (err)
- goto err_unlock_free;
-#endif
+ for (s = 0; s < subsys_count; s++) {
+ struct container_subsys *ss = subsys[s];
+ if (ss->enabled) {
+ err = ss->create(ss, cont);
+ if (err) {
+ for (s--; s >= 0; s--) {
+ ss = subsys[s];
+ if (ss->enabled)
+ ss->destroy(ss, cont);
+ }
+ goto err_unlock_free;
+ }
+ } else {
+ /* Just copy subsys object from parent */
+ cont->subsys[s] = parent->subsys[s];
+ }
+ }
mutex_lock(&callback_mutex);
list_add(&cont->sibling, &cont->parent->children);
@@ -1107,9 +1169,12 @@ static long container_create(struct cont
return 0;
err_remove:
-#ifdef CONFIG_CPUSETS
- cpuset_destroy(cont);
-#endif
+ for (s = subsys_count - 1; s >= 0; s--) {
+ struct container_subsys *ss = subsys[s];
+ if (ss->enabled)
+ ss->destroy(ss, cont);
+ }
+
mutex_lock(&callback_mutex);
list_del(&cont->sibling);
number_of_containers--;
@@ -1145,6 +1210,7 @@ static int container_rmdir(struct inode
struct dentry *d;
struct container *parent;
char *pathbuf = NULL;
+ int s;
/* the vfs holds both inode->i_mutex already */
@@ -1169,9 +1235,11 @@ static int container_rmdir(struct inode
dput(d);
number_of_containers--;
mutex_unlock(&callback_mutex);
-#ifdef CONFIG_CPUSETS
- cpuset_destroy(cont);
-#endif
+ for (s = 0; s < subsys_count; s++) {
+ struct container_subsys *ss = subsys[s];
+ if (ss->enabled)
+ ss->destroy(ss, cont);
+ }
if (list_empty(&parent->children))
check_for_release(parent, &pathbuf);
mutex_unlock(&manage_mutex);
@@ -1226,6 +1294,57 @@ out:
return err;
}
+int container_register_subsys(struct container_subsys *new_subsys) {
+ int retval = 0;
+ int i;
+ mutex_lock(&manage_mutex);
+ if (number_of_containers > 1) {
+ retval = -EBUSY;
+ goto out;
+ }
+ if (subsys_count == MAX_CONTAINER_SUBSYS) {
+ retval = -ENOSPC;
+ goto out;
+ }
+ if (!new_subsys->name ||
+ (strlen(new_subsys->name) > MAX_CONTAINER_TYPE_NAMELEN) ||
+ !new_subsys->create || !new_subsys->destroy) {
+ retval = -EINVAL;
+ goto out;
+ }
+ for (i = 0; i < subsys_count; i++) {
+ if (!strcmp(subsys[i]->name, new_subsys->name)) {
+ retval = -EEXIST;
+ goto out;
+ }
+ }
+
+ subsys[subsys_count] = new_subsys;
+ new_subsys->subsys_id = subsys_count++;
+ retval = new_subsys->create(new_subsys, &top_container);
+ if (retval) {
+ new_subsys->subsys_id = -1;
+ subsys_count--;
+ goto out;
+ }
+
+ /* Set up the per-container "enabled" file */
+ strcpy(new_subsys->enable_cft_filename, new_subsys->name);
+ strcat(new_subsys->enable_cft_filename, "_enabled");
+ new_subsys->enable_cft = cft_enabled_template;
+ new_subsys->enable_cft.name = new_subsys->enable_cft_filename;
+
+ /* Only populate the top container if we've done
+ * container_init() */
+ if (container_mount && new_subsys->populate) {
+ new_subsys->populate(new_subsys, &top_container);
+ container_add_file(&top_container, &new_subsys->enable_cft);
+ }
+ out:
+ mutex_unlock(&manage_mutex);
+ return retval;
+}
+
/**
* container_fork - attach newly forked task to its parents container.
* @tsk: pointer to task_struct of forking parent process.
Index: container-2.6.19-rc2/kernel/cpuset.c
===================================================================
--- container-2.6.19-rc2.orig/kernel/cpuset.c
+++ container-2.6.19-rc2/kernel/cpuset.c
@@ -5,6 +5,7 @@
*
* Copyright (C) 2003 BULL SA.
* Copyright (C) 2004-2006 Silicon Graphics, Inc.
+ * Copyright (C) 2006 Google, Inc
*
* Portions derived from Patrick Mochel's sysfs code.
* sysfs is Copyright (c) 2001-3 Patrick Mochel
@@ -12,6 +13,7 @@
* 2003-10-10 Written by Simon Derr.
* 2003-10-22 Updates by Stephen Hemminger.
* 2004 May-July Rework by Paul Jackson.
+ * 2006 Rework by Paul Menage to use generic containers
*
* 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
@@ -61,6 +63,25 @@
*/
int number_of_cpusets __read_mostly;
+/* Retrieve the cpuset from a container */
+static struct container_subsys cpuset_subsys;
+static inline struct cpuset *container_cs(struct container *cont)
+{
+ return (struct cpuset *)cont->subsys[cpuset_subsys.subsys_id];
+}
+
+/* Update the cpuset for a container */
+static inline void set_container_cs(struct container *cont, struct cpuset *cs)
+{
+ cont->subsys[cpuset_subsys.subsys_id] = cs;
+}
+
+/* Retrieve the cpuset for a task */
+static inline struct cpuset *task_cs(struct task_struct *task)
+{
+ return container_cs(task->container);
+}
+
/* See "Frequency meter" comments, below. */
struct fmeter {
@@ -166,6 +187,12 @@ static int cpuset_get_sb(struct file_sys
data, mnt);
put_filesystem(container_fs);
}
+ container_manage_lock();
+ if (number_of_containers == 1) {
+ /* For legacy compatibility, cpuset starts enabled */
+ cpuset_subsys.enabled = 1;
+ }
+ container_manage_unlock();
return ret;
}
@@ -271,20 +298,21 @@ void cpuset_update_task_memory_state(voi
struct task_struct *tsk = current;
struct cpuset *cs;
- if (tsk->container->cpuset == &top_cpuset) {
+ if (task_cs(tsk) == &top_cpuset) {
/* Don't need rcu for top_cpuset. It's never freed. */
my_cpusets_mem_gen = top_cpuset.mems_generation;
} else {
+ struct container *cont;
rcu_read_lock();
- cs = rcu_dereference(tsk->container->cpuset);
- my_cpusets_mem_gen = cs->mems_generation;
+ cont = rcu_dereference(tsk->container);
+ my_cpusets_mem_gen = container_cs(cont)->mems_generation;
rcu_read_unlock();
}
if (my_cpusets_mem_gen != tsk->cpuset_mems_generation) {
container_lock();
task_lock(tsk);
- cs = tsk->container->cpuset; /* Maybe changed when task not locked */
+ cs = task_cs(tsk); /* Maybe changed when task not locked */
guarantee_online_mems(cs, &tsk->mems_allowed);
tsk->cpuset_mems_generation = cs->mems_generation;
if (is_spread_page(cs))
@@ -344,8 +372,7 @@ static int validate_change(const struct
/* Each of our child cpusets must be a subset of us */
list_for_each_entry(cont, &cur->container->children, sibling) {
- c = cont->cpuset;
- if (!is_cpuset_subset(c, trial))
+ if (!is_cpuset_subset(container_cs(cont), trial))
return -EBUSY;
}
@@ -359,7 +386,7 @@ static int validate_change(const struct
/* If either I or some sibling (!= me) is exclusive, we can't overlap */
list_for_each_entry(cont, &par->container->children, sibling) {
- c = cont->cpuset;
+ c = container_cs(cont);
if ((is_cpu_exclusive(trial) || is_cpu_exclusive(c)) &&
c != cur &&
cpus_intersects(trial->cpus_allowed, c->cpus_allowed))
@@ -402,7 +429,7 @@ static void update_cpu_domains(struct cp
*/
pspan = par->cpus_allowed;
list_for_each_entry(cont, &par->container->children, sibling) {
- c = cont->cpuset;
+ c = container_cs(cont);
if (is_cpu_exclusive(c))
cpus_andnot(pspan, pspan, c->cpus_allowed);
}
@@ -420,7 +447,7 @@ static void update_cpu_domains(struct cp
* of exclusive children
*/
list_for_each_entry(cont, &cur->container->children, sibling) {
- c = cont->cpuset;
+ c = container_cs(cont);
if (is_cpu_exclusive(c))
cpus_andnot(cspan, cspan, c->cpus_allowed);
}
@@ -508,7 +535,7 @@ static void cpuset_migrate_mm(struct mm_
do_migrate_pages(mm, from, to, MPOL_MF_MOVE_ALL);
container_lock();
- guarantee_online_mems(tsk->container->cpuset, &tsk->mems_allowed);
+ guarantee_online_mems(task_cs(tsk),&tsk->mems_allowed);
container_unlock();
}
@@ -796,9 +823,10 @@ static int fmeter_getrate(struct fmeter
return val;
}
-int cpuset_can_attach_task(struct container *cont, struct task_struct *tsk)
+int cpuset_can_attach(struct container_subsys *ss,
+ struct container *cont, struct task_struct *tsk)
{
- struct cpuset *cs = cont->cpuset;
+ struct cpuset *cs = container_cs(cont);
if (cpus_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed))
return -ENOSPC;
@@ -806,22 +834,23 @@ int cpuset_can_attach_task(struct contai
return security_task_setscheduler(tsk, 0, NULL);
}
-void cpuset_attach_task(struct container *cont, struct task_struct *tsk)
+void cpuset_attach(struct container_subsys *ss, struct container *cont,
+ struct container *old_cont, struct task_struct *tsk)
{
cpumask_t cpus;
- struct cpuset *cs = cont->cpuset;
- guarantee_online_cpus(cs, &cpus);
+ guarantee_online_cpus(container_cs(cont), &cpus);
set_cpus_allowed(tsk, cpus);
}
-void cpuset_post_attach_task(struct container *cont,
- struct container *oldcont,
- struct task_struct *tsk)
+void cpuset_post_attach(struct container_subsys *ss,
+ struct container *cont,
+ struct container *oldcont,
+ struct task_struct *tsk)
{
nodemask_t from, to;
struct mm_struct *mm;
- struct cpuset *cs = cont->cpuset;
- struct cpuset *oldcs = oldcont->cpuset;
+ struct cpuset *cs = container_cs(cont);
+ struct cpuset *oldcs = container_cs(oldcont);
from = oldcs->mems_allowed;
to = cs->mems_allowed;
@@ -855,7 +884,7 @@ static ssize_t cpuset_common_file_write(
const char __user *userbuf,
size_t nbytes, loff_t *unused_ppos)
{
- struct cpuset *cs = cont->cpuset;
+ struct cpuset *cs = container_cs(cont);
cpuset_filetype_t type = cft->private;
char *buffer;
int retval = 0;
@@ -965,7 +994,7 @@ static ssize_t cpuset_common_file_read(s
char __user *buf,
size_t nbytes, loff_t *ppos)
{
- struct cpuset *cs = cont->cpuset;
+ struct cpuset *cs = container_cs(cont);
cpuset_filetype_t type = cft->private;
char *page;
ssize_t retval = 0;
@@ -1084,7 +1113,7 @@ static struct cftype cft_spread_slab = {
.private = FILE_SPREAD_SLAB,
};
-int cpuset_populate_dir(struct container *cont)
+int cpuset_populate(struct container_subsys *ss, struct container *cont)
{
int err;
@@ -1119,11 +1148,21 @@ int cpuset_populate_dir(struct container
* Must be called with the mutex on the parent inode held
*/
-int cpuset_create(struct container *cont)
+int cpuset_create(struct container_subsys *ss, struct container *cont)
{
struct cpuset *cs;
- struct cpuset *parent = cont->parent->cpuset;
+ struct cpuset *parent;
+
+ if (!cont->parent) {
+ /* This is early initialization for the top container */
+ set_container_cs(cont, &top_cpuset);
+ top_cpuset.container = cont;
+ top_cpuset.mems_generation = cpuset_mems_generation++;
+
+ return 0;
+ }
+ parent = container_cs(cont->parent);
cs = kmalloc(sizeof(*cs), GFP_KERNEL);
if (!cs)
return -ENOMEM;
@@ -1140,7 +1179,7 @@ int cpuset_create(struct container *cont
fmeter_init(&cs->fmeter);
cs->parent = parent;
- cont->cpuset = cs;
+ set_container_cs(cont, cs);
cs->container = cont;
number_of_cpusets++;
return 0;
@@ -1157,9 +1196,9 @@ int cpuset_create(struct container *cont
* nesting would risk an ABBA deadlock.
*/
-void cpuset_destroy(struct container *cont)
+void cpuset_destroy(struct container_subsys *ss, struct container *cont)
{
- struct cpuset *cs = cont->cpuset;
+ struct cpuset *cs = container_cs(cont);
cpuset_update_task_memory_state();
if (is_cpu_exclusive(cs)) {
@@ -1169,6 +1208,16 @@ void cpuset_destroy(struct container *co
number_of_cpusets--;
}
+static struct container_subsys cpuset_subsys = {
+ .name = "cpuset",
+ .create = cpuset_create,
+ .destroy = cpuset_destroy,
+ .can_attach = cpuset_can_attach,
+ .attach = cpuset_attach,
+ .post_attach = cpuset_post_attach,
+ .populate = cpuset_populate,
+};
+
/*
* cpuset_init_early - just enough so that the calls to
* cpuset_update_task_memory_state() in early init code
@@ -1177,10 +1226,8 @@ void cpuset_destroy(struct container *co
int __init cpuset_init_early(void)
{
- struct container *cont = current->container;
- cont->cpuset = &top_cpuset;
- top_cpuset.container = cont;
- cont->cpuset->mems_generation = cpuset_mems_generation++;
+ if (container_register_subsys(&cpuset_subsys) < 0)
+ panic("Couldn't register cpuset subsystem");
return 0;
}
@@ -1337,7 +1384,7 @@ cpumask_t cpuset_cpus_allowed(struct tas
container_lock();
task_lock(tsk);
- guarantee_online_cpus(tsk->container->cpuset, &mask);
+ guarantee_online_cpus(task_cs(tsk), &mask);
task_unlock(tsk);
container_unlock();
@@ -1365,7 +1412,7 @@ nodemask_t cpuset_mems_allowed(struct ta
container_lock();
task_lock(tsk);
- guarantee_online_mems(tsk->container->cpuset, &mask);
+ guarantee_online_mems(task_cs(tsk), &mask);
task_unlock(tsk);
container_unlock();
@@ -1470,7 +1517,7 @@ int __cpuset_zone_allowed(struct zone *z
container_lock();
task_lock(current);
- cs = nearest_exclusive_ancestor(current->container->cpuset);
+ cs = nearest_exclusive_ancestor(task_cs(current));
task_unlock(current);
allowed = node_isset(node, cs->mems_allowed);
@@ -1538,7 +1585,7 @@ int cpuset_excl_nodes_overlap(const stru
task_unlock(current);
goto done;
}
- cs1 = nearest_exclusive_ancestor(current->container->cpuset);
+ cs1 = nearest_exclusive_ancestor(task_cs(current));
task_unlock(current);
task_lock((struct task_struct *)p);
@@ -1546,7 +1593,7 @@ int cpuset_excl_nodes_overlap(const stru
task_unlock((struct task_struct *)p);
goto done;
}
- cs2 = nearest_exclusive_ancestor(p->container->cpuset);
+ cs2 = nearest_exclusive_ancestor(task_cs((struct task_struct *)p));
task_unlock((struct task_struct *)p);
overlap = nodes_intersects(cs1->mems_allowed, cs2->mems_allowed);
@@ -1582,11 +1629,8 @@ int cpuset_memory_pressure_enabled __rea
void __cpuset_memory_pressure_bump(void)
{
- struct cpuset *cs;
-
task_lock(current);
- cs = current->container->cpuset;
- fmeter_markevent(&cs->fmeter);
+ fmeter_markevent(&task_cs(current)->fmeter);
task_unlock(current);
}
Index: container-2.6.19-rc2/Documentation/containers.txt
===================================================================
--- container-2.6.19-rc2.orig/Documentation/containers.txt
+++ container-2.6.19-rc2/Documentation/containers.txt
@@ -17,12 +17,16 @@ CONTENTS:
1.2 Why are containers needed ?
1.3 How are containers implemented ?
1.4 What does notify_on_release do ?
- 1.5 How do I use containers ?
+ 1.5 What do the xxx_enabled files do ?
+ 1.6 How do I use containers ?
2. Usage Examples and Syntax
2.1 Basic Usage
2.2 Attaching processes
-3. Questions
-4. Contact
+3. Kernel API
+ 3.1 Overview
+ 3.2 Synchronization
+ 3.3 Subsystem API
+4. Questions
1. Containers
==========
@@ -129,7 +133,18 @@ The default value of notify_on_release i
boot is disabled (0). The default value of other containers at creation
is the current value of their parents notify_on_release setting.
-1.5 How do I use containers ?
+1.5 What do the xxx_enabled files do ?
+--------------------------------------
+
+In the top-level container directory there are a series of
+<subsys>_enabled files, one for each registered subsystem. Each of
+these files contains 0 or 1 to indicate whether the named container
+subsystem is enabled, and can only be modified when there are no
+subcontainers. Disabled container subsystems don't get new instances
+created when a subcontainer is created; the subsystem-specific state
+is simply inherited from the parent container.
+
+1.6 How do I use containers ?
--------------------------
To start a new job that is to be contained within a container, the steps are:
@@ -214,8 +229,143 @@ If you have several tasks to attach, you
...
# /bin/echo PIDn > tasks
+3. Kernel API
+=============
+
+3.1 Overview
+------------
+
+Each kernel subsystem that wants to hook into the generic container
+system needs to create a container_subsys object. This contains
+various methods, which are callbacks from the container system, along
+with a subsystem id which will be assigned by the container system.
+
+Other fields in the container_subsys object include:
+
+- subsys_id: a unique array index for the subsystem, indicating which
+ entry in container->subsys[] this subsystem should be
+ managing. Initialized by container_register_subsys();
+
+- name: should be initialized to a unique subsystem name prior to
+ calling container_register_subsystem. Should be no longer than
+ MAX_CONTAINER_TYPE_NAMELEN
+
+- enabled: determines whether that container type is in use when
+ creating subcontainers. It's managed by the container system, but
+ may be initialized to either 0 or 1 prior to calling
+ container_register_subsystem
+
+Each container object created by the system has an array of pointers,
+indexed by subsystem id; this pointer is entirely managed by the
+subsystem; the generic container code will never touch this pointer.
+
+Note that all subsystems share the same hierarchy of containers; it's
+not currently possible to have independent hierarchies and container
+memberships for different subsystems.
+
+3.2 Synchronization
+-------------------
+
+There are two global mutexes used by the container system. The first
+is the manage_mutex, which should be taken by anything that wants to
+modify a container; The second if the callback_mutex, which should be
+taken by holders of the manage_mutex at the point when they actually
+make changes, and by callbacks from lower-level subsystems that want
+to ensure that no container changes occur. Note that memory
+allocations cannot be made while holding callback_mutex.
+
+The callback_mutex nests inside the manage_mutex.
+
+In general, the pattern of use is:
+
+1) take manage_mutex
+2) verify that the change is valid and do any necessary allocations\
+3) take callback_mutex
+4) make changes
+5) release callback_mutex
+6) release manage_mutex
+
+See kernel/container.c for more details.
+
+Subsystems can take/release the manage_mutex via the functions
+container_manage_lock()/container_manage_unlock(), and can
+take/release the callback_mutex via the functions
+container_lock()/container_unlock().
+
+Accessing a task's container pointer may be done in the following ways:
+- while holding manage_mutex
+- while holding callback_mutex
+- while holding the task's alloc_lock (via task_lock())
+- inside an rcu_read_lock() section via rcu_dereference()
+
+3.3 Subsystem API
+--------------------------
+
+Each subsystem should call container_register_subsys() with a pointer
+to its subsystem object. This will store the new subsystem id in the
+subsystem subsys_id field and return 0, or a negative error. There's
+currently no facility for deregestering a subsystem nor for
+registering a subsystem after any containers (other than the default
+"top_container") have been created.
+
+Each subsystem may export the following methods. The only mandatory
+methods are create/destroy. Any others that are null are presumed to
+be successful no-ops.
+
+int create(struct container *cont)
+LL=manage_mutex
+
+The subsystem should appropriately initialize its subsystem pointer
+for the passed container, returning 0 on success or a negative error
+code. Typically this will involve allocating a new per-container
+structure and storing a reference to it in the container, but there's
+nothing to stop a subsystem having multiple containers with pointers
+to the same subsystem object. Note that this will be called during
+container_register_subsys() to initialize this subsystem on the root
+container.
+
+void destroy(struct container *cont)
+LL=manage_mutex
+
+The container system is about to destroy the passed container; the
+subsystem should do any necessary cleanup
+
+int can_attach(struct container_subsys *ss, struct container *cont,
+ struct task_struct *task)
+LL=manage_mutex
+
+Called prior to moving a task into a container; if the subsystem
+returns an error, this will abort the attach operation. Note that
+this isn't called on a fork.
+
+void attach(struct container_subsys *ss, struct container *cont,
+ struct container *old_cont, struct task_struct *task)
+LL=manage_mutex & callback_mutex
+
+Called during the attach operation. The subsystem should do any
+necessary work that can be accomplished without memory allocations or
+sleeping.
+
+void post_attach(struct container_subsys *ss, struct container *cont,
+ struct container *old_cont, struct task_struct *task)
+LL=manage_mutex
+
+Called after the task has been attached to the container, to allow any
+post-attachment activity that requires memory allocations or blocking.
+
+int populate(struct container_subsys *ss, struct container *cont)
+LL=none
+
+Called after creation of a container to allow a subsystem to populate
+the container directory with file entries. The subsystem should make
+calls to container_add_file() with objects of type cftype (see
+include/linux/container.h for details). Called during
+container_register_subsys() to populate the root container. Note that
+although this method can return an error code, the error code is
+currently not always handled well.
+
-3. Questions
+4. Questions
============
Q: what's up with this '/bin/echo' ?
--
^ permalink raw reply [flat|nested] 12+ messages in thread* [PATCH 4/6] Simple CPU accounting container subsystem
2006-10-20 18:38 [PATCH 0/6] Generic Process Containers menage
` (2 preceding siblings ...)
2006-10-20 18:38 ` [PATCH 3/6] Add generic multi-subsystem API to containers menage
@ 2006-10-20 18:38 ` menage
2006-10-20 18:38 ` [PATCH 5/6] Extension to container system to allow fork/exit callbacks menage
2006-10-20 18:38 ` [PATCH 6/6] Resource Groups over generic containers menage
5 siblings, 0 replies; 12+ messages in thread
From: menage @ 2006-10-20 18:38 UTC (permalink / raw)
To: akpm, pj, sekharan
Cc: linux-kernel, ckrm-tech, jlan, mbligh, rohitseth, winget,
Simon.Derr
[-- Attachment #1: cpu_acct.patch --]
[-- Type: text/plain, Size: 5886 bytes --]
This demonstrates how to use the generic container subsystem for a
simple resource tracker that counts the total CPU time used by all
processes in a container, during the time that they're members of the
container.
Signed-off-by: Paul Menage <menage@google.com>
---
include/linux/cpu_acct.h | 14 +++++
init/Kconfig | 7 ++
kernel/Makefile | 1
kernel/cpu_acct.c | 110 +++++++++++++++++++++++++++++++++++++++++++++++
kernel/sched.c | 6 ++
5 files changed, 138 insertions(+)
Index: container-2.6.19-rc2/include/linux/cpu_acct.h
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/include/linux/cpu_acct.h
@@ -0,0 +1,14 @@
+
+#ifndef _LINUX_CPU_ACCT_H
+#define _LINUX_CPU_ACCT_H
+
+#include <linux/container.h>
+#include <asm/cputime.h>
+
+#ifdef CONFIG_CONTAINER_CPUACCT
+extern void cpuacct_charge(struct task_struct *, cputime_t cputime);
+#else
+static void inline cpuacct_charge(struct task_struct *p, cputime_t cputime) {}
+#endif
+
+#endif
Index: container-2.6.19-rc2/init/Kconfig
===================================================================
--- container-2.6.19-rc2.orig/init/Kconfig
+++ container-2.6.19-rc2/init/Kconfig
@@ -261,6 +261,13 @@ config CPUSETS_LEGACY_API
- /proc/<pid>/cpuset
- defaults cpuset subsystem to enabled
+config CONTAINER_CPUACCT
+ bool "Simple CPU accounting container subsystem"
+ select CONTAINERS
+ help
+ Provides a simple Resource Controller for monitoring the
+ total CPU consumed by the tasks in a container
+
config RELAY
bool "Kernel->user space relay support (formerly relayfs)"
help
Index: container-2.6.19-rc2/kernel/cpu_acct.c
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/kernel/cpu_acct.c
@@ -0,0 +1,110 @@
+/*
+ * kernel/cpu_acct.c - CPU accounting container subsystem
+ *
+ * Copyright (C) Google Inc, 2006
+ *
+ */
+
+/*
+ * Container subsystem for reporting total CPU usage of tasks in a
+ * container.
+ */
+
+#include <linux/module.h>
+#include <linux/container.h>
+#include <linux/fs.h>
+#include <asm/div64.h>
+
+struct cpuacct {
+ spinlock_t lock;
+ cputime64_t time; // total time used by this class
+};
+
+static struct container_subsys cpuacct_subsys;
+static inline struct cpuacct *container_ca(struct container *cont)
+{
+ return (struct cpuacct *)cont->subsys[cpuacct_subsys.subsys_id];
+}
+
+static int cpuacct_create(struct container_subsys *ss, struct container *cont)
+{
+ struct cpuacct *ca = kzalloc(sizeof(*ca), GFP_KERNEL);
+ if (!ca) return -ENOMEM;
+ spin_lock_init(&ca->lock);
+ cont->subsys[cpuacct_subsys.subsys_id] = ca;
+ return 0;
+}
+
+static void cpuacct_destroy(struct container_subsys *ss,
+ struct container *cont)
+{
+ kfree(container_ca(cont));
+}
+
+static ssize_t cpuusage_read(struct container *cont,
+ struct cftype *cft,
+ struct file *file,
+ char __user *buf,
+ size_t nbytes, loff_t *ppos)
+{
+ struct cpuacct *ca = container_ca(cont);
+ cputime64_t time;
+ char usagebuf[64];
+ char *s = usagebuf;
+
+ spin_lock_irq(&ca->lock);
+ time = ca->time;
+ spin_unlock_irq(&ca->lock);
+
+ time *= 1000;
+ do_div(time, HZ);
+ s += sprintf(s, "%llu", (unsigned long long) time);
+
+ return simple_read_from_buffer(buf, nbytes, ppos, usagebuf, s - usagebuf);
+}
+
+static struct cftype cft_usage = {
+ .name = "cpu_usage",
+ .read = cpuusage_read,
+};
+
+static int cpuacct_populate(struct container_subsys *ss,
+ struct container *cont)
+{
+ return container_add_file(cont, &cft_usage);
+}
+
+
+void cpuacct_charge(struct task_struct *task, cputime_t cputime) {
+
+ struct cpuacct *ca;
+ struct container *cont;
+ unsigned long flags;
+
+ if (cpuacct_subsys.subsys_id < 0) return;
+ rcu_read_lock();
+ cont = rcu_dereference(task->container);
+ ca = container_ca(cont);
+ if (ca) {
+ spin_lock_irqsave(&ca->lock, flags);
+ ca->time = cputime64_add(ca->time, cputime);
+ spin_unlock_irqrestore(&ca->lock, flags);
+ }
+ rcu_read_unlock();
+}
+
+static struct container_subsys cpuacct_subsys = {
+ .name = "cpuacct",
+ .create = cpuacct_create,
+ .destroy = cpuacct_destroy,
+ .populate = cpuacct_populate,
+ .subsys_id = -1,
+};
+
+int __init init_cpuacct(void)
+{
+ int id = container_register_subsys(&cpuacct_subsys);
+ return id < 0 ? id : 0;
+}
+
+module_init(init_cpuacct)
Index: container-2.6.19-rc2/kernel/Makefile
===================================================================
--- container-2.6.19-rc2.orig/kernel/Makefile
+++ container-2.6.19-rc2/kernel/Makefile
@@ -38,6 +38,7 @@ obj-$(CONFIG_KEXEC) += kexec.o
obj-$(CONFIG_COMPAT) += compat.o
obj-$(CONFIG_CONTAINERS) += container.o
obj-$(CONFIG_CPUSETS) += cpuset.o
+obj-$(CONFIG_CONTAINER_CPUACCT) += cpu_acct.o
obj-$(CONFIG_IKCONFIG) += configs.o
obj-$(CONFIG_STOP_MACHINE) += stop_machine.o
obj-$(CONFIG_AUDIT) += audit.o auditfilter.o
Index: container-2.6.19-rc2/kernel/sched.c
===================================================================
--- container-2.6.19-rc2.orig/kernel/sched.c
+++ container-2.6.19-rc2/kernel/sched.c
@@ -52,6 +52,7 @@
#include <linux/tsacct_kern.h>
#include <linux/kprobes.h>
#include <linux/delayacct.h>
+#include <linux/cpu_acct.h>
#include <asm/tlb.h>
#include <asm/unistd.h>
@@ -2977,6 +2978,8 @@ void account_user_time(struct task_struc
p->utime = cputime_add(p->utime, cputime);
+ cpuacct_charge(p, cputime);
+
/* Add user time to cpustat. */
tmp = cputime_to_cputime64(cputime);
if (TASK_NICE(p) > 0)
@@ -3000,6 +3003,9 @@ void account_system_time(struct task_str
p->stime = cputime_add(p->stime, cputime);
+ if (p != rq->idle)
+ cpuacct_charge(p, cputime);
+
/* Add system time to cpustat. */
tmp = cputime_to_cputime64(cputime);
if (hardirq_count() - hardirq_offset)
--
^ permalink raw reply [flat|nested] 12+ messages in thread* [PATCH 5/6] Extension to container system to allow fork/exit callbacks
2006-10-20 18:38 [PATCH 0/6] Generic Process Containers menage
` (3 preceding siblings ...)
2006-10-20 18:38 ` [PATCH 4/6] Simple CPU accounting container subsystem menage
@ 2006-10-20 18:38 ` menage
2006-10-20 18:38 ` [PATCH 6/6] Resource Groups over generic containers menage
5 siblings, 0 replies; 12+ messages in thread
From: menage @ 2006-10-20 18:38 UTC (permalink / raw)
To: akpm, pj, sekharan
Cc: linux-kernel, ckrm-tech, jlan, mbligh, rohitseth, winget,
Simon.Derr
[-- Attachment #1: container_forkexit_callback.patch --]
[-- Type: text/plain, Size: 7414 bytes --]
This patch adds fork/exit callbacks to container subsystems, and
ensures that every registered container has received one fork callback
for each task running int the system, and one exit callback for each
task that exited since it was registered.
Since the fork/exit path is performance sensitive, an RCU-protected
flag indicates to the fork/exit hooks whether they need to take the
callback mutex and scan the list of registered subsystems to look for
fork/exit handlers.
Documentation/containers.txt | 13 +++++
include/linux/container.h | 4 +
kernel/container.c | 108 +++++++++++++++++++++++++++++++++++++++----
3 files changed, 116 insertions(+), 9 deletions(-)
Index: container-2.6.19-rc2/include/linux/container.h
===================================================================
--- container-2.6.19-rc2.orig/include/linux/container.h
+++ container-2.6.19-rc2/include/linux/container.h
@@ -101,6 +101,10 @@ struct container_subsys {
struct container *cont,
struct container *old_cont,
struct task_struct *tsk);
+ void (*fork)(struct container_subsys *ss, struct container *cont,
+ struct task_struct *task);
+ void (*exit)(struct container_subsys *ss, struct container *cont,
+ struct task_struct *task);
int (*populate)(struct container_subsys *ss,
struct container *cont);
Index: container-2.6.19-rc2/kernel/container.c
===================================================================
--- container-2.6.19-rc2.orig/kernel/container.c
+++ container-2.6.19-rc2/kernel/container.c
@@ -72,6 +72,21 @@ int number_of_containers __read_mostly;
static struct container_subsys *subsys[MAX_CONTAINER_SUBSYS];
static int subsys_count = 0;
+/* This flag indicates whether tasks in the fork and exit paths should
+ * take callback_mutex and check for fork/exit handlers to call. This
+ * avoids us having to take locks in the fork/exit path if none of the
+ * subsystems need to be called.
+ *
+ * It is protected via RCU, with the invariant that a process in an
+ * rcu_read_lock() section will never see this as 0 if there are
+ * actually registered subsystems with a fork or exit
+ * handler. (Sometimes it may be 1 without there being any registered
+ * subsystems with such a handler, but such periods are safe and of
+ * short duration).
+ */
+
+static int need_forkexit_callback = 0;
+
/* bits in struct container flags field */
typedef enum {
CONT_REMOVED,
@@ -487,7 +502,6 @@ static int update_flag(container_flagbit
return 0;
}
-
/*
* Attack task specified by pid in 'pidbuf' to container 'cont', possibly
* writing the path of the old container in 'ppathbuf' if it needs to be
@@ -1254,9 +1268,7 @@ static int container_rmdir(struct inode
int __init container_init_early(void)
{
- struct task_struct *tsk = current;
-
- tsk->container = &top_container;
+ current->container = &top_container;
return 0;
}
@@ -1297,6 +1309,7 @@ out:
int container_register_subsys(struct container_subsys *new_subsys) {
int retval = 0;
int i;
+
mutex_lock(&manage_mutex);
if (number_of_containers > 1) {
retval = -EBUSY;
@@ -1319,8 +1332,7 @@ int container_register_subsys(struct con
}
}
- subsys[subsys_count] = new_subsys;
- new_subsys->subsys_id = subsys_count++;
+ new_subsys->subsys_id = subsys_count;
retval = new_subsys->create(new_subsys, &top_container);
if (retval) {
new_subsys->subsys_id = -1;
@@ -1328,6 +1340,38 @@ int container_register_subsys(struct con
goto out;
}
+ mutex_lock(&callback_mutex);
+ /* If this is the first subsystem that requested a fork or
+ * exit callback, tell our fork/exit hooks that they need to
+ * grab callback_mutex on every invocation. If they are
+ * running concurrently with this code, they will either not
+ * see the change now and go straight on, or they will see it
+ * and grab callback_mutex, which will deschedule them. Either
+ * way once synchronize_rcu() returns we know that all current
+ * and future forks will make the callbacks. */
+ if (!need_forkexit_callback &&
+ (new_subsys->fork || new_subsys->exit)) {
+ need_forkexit_callback = 1;
+ synchronize_rcu();
+ }
+
+ /* If this subsystem requested that it be notified with fork
+ * events, we should send it one now for every process in the
+ * system */
+ if (new_subsys->fork) {
+ struct task_struct *g, *p;
+
+ read_lock(&tasklist_lock);
+ do_each_thread(g, p) {
+ new_subsys->fork(new_subsys, &top_container, p);
+ } while_each_thread(g, p);
+ read_unlock(&tasklist_lock);
+ }
+
+ subsys[subsys_count] = new_subsys;
+ subsys_count++;
+ mutex_unlock(&callback_mutex);
+
/* Set up the per-container "enabled" file */
strcpy(new_subsys->enable_cft_filename, new_subsys->name);
strcat(new_subsys->enable_cft_filename, "_enabled");
@@ -1365,10 +1409,35 @@ int container_register_subsys(struct con
void container_fork(struct task_struct *child)
{
+ struct container *cont;
+ int s, need_callback;
+
+ rcu_read_lock();
+ /* need_forkexit_callback will be true if we might need to do
+ * a callback */
+ need_callback = rcu_dereference(need_forkexit_callback);
+ if (need_callback) {
+ rcu_read_unlock();
+ mutex_lock(&callback_mutex);
+ }
task_lock(current);
- child->container = current->container;
- atomic_inc(&child->container->count);
+ cont = current->container;
+ child->container = cont;
+ atomic_inc(&cont->count);
+ if (need_callback) {
+ for (s = 0; s < subsys_count; s++) {
+ struct container_subsys *ss = subsys[s];
+ if (ss->fork) {
+ ss->fork(ss, cont, child);
+ }
+ }
+ }
task_unlock(current);
+ if (need_callback) {
+ mutex_unlock(&callback_mutex);
+ } else {
+ rcu_read_unlock();
+ }
}
/**
@@ -1433,9 +1502,30 @@ void container_fork(struct task_struct *
void container_exit(struct task_struct *tsk)
{
struct container *cont;
+ int s, need_callback;
cont = tsk->container;
- tsk->container = &top_container; /* the_top_container_hack - see above */
+
+ rcu_read_lock();
+ need_callback = rcu_dereference(need_forkexit_callback);
+ if (need_callback) {
+ rcu_read_unlock();
+ mutex_lock(&callback_mutex);
+ for (s = 0; s < subsys_count; s++) {
+ struct container_subsys *ss = subsys[s];
+ if (ss->exit) {
+ ss->exit(ss, cont, tsk);
+ }
+ }
+ mutex_unlock(&callback_mutex);
+ } else {
+ rcu_read_unlock();
+ }
+
+ if (cont != &top_container) {
+ /* the_top_container_hack - see above */
+ tsk->container = &top_container;
+ }
if (notify_on_release(cont)) {
char *pathbuf = NULL;
Index: container-2.6.19-rc2/Documentation/containers.txt
===================================================================
--- container-2.6.19-rc2.orig/Documentation/containers.txt
+++ container-2.6.19-rc2/Documentation/containers.txt
@@ -353,6 +353,19 @@ LL=manage_mutex
Called after the task has been attached to the container, to allow any
post-attachment activity that requires memory allocations or blocking.
+void fork(struct container_subsy *ss, struct container *cont,
+ struct task_struct *task)
+LL=callback_mutex, maybe read_lock(tasklist_lock)
+
+Called when a task is forked into a container. Also called during
+registration for all existing tasks.
+
+void exit(struct container_subsys *ss, struct container *cont,
+ struct task_struct *task)
+LL=callback_mutex
+
+Called during task exit
+
int populate(struct container_subsys *ss, struct container *cont)
LL=none
--
^ permalink raw reply [flat|nested] 12+ messages in thread* [PATCH 6/6] Resource Groups over generic containers
2006-10-20 18:38 [PATCH 0/6] Generic Process Containers menage
` (4 preceding siblings ...)
2006-10-20 18:38 ` [PATCH 5/6] Extension to container system to allow fork/exit callbacks menage
@ 2006-10-20 18:38 ` menage
5 siblings, 0 replies; 12+ messages in thread
From: menage @ 2006-10-20 18:38 UTC (permalink / raw)
To: akpm, pj, sekharan
Cc: linux-kernel, ckrm-tech, jlan, mbligh, rohitseth, winget,
Simon.Derr
[-- Attachment #1: res_group_f0.8_single_2618.patch --]
[-- Type: text/plain, Size: 46204 bytes --]
This patch provides the RG core and numtasks controller as container
subsystems. The changes to the core involve primarily removing the
group management, task membership and configfs support and adding
interface layers to talk to the generic container layer instead.
Each resource controller becomes an independent container subsystem;
the RG core is essentially a library that the resource controllers can
use to provide the RG API to userspace. Rather than a single shares
and stats file in each group, there's a <controller>_shares and
a <controller>_stats file, each linked to the appropriate resource
controller.
The example numtasks resource controller is almost unchanged; the
primary differences involve locking, the details of which are still to
be finalized.
include/linux/moduleparam.h | 12 -
include/linux/numtasks.h | 28 ++
include/linux/res_group.h | 82 +++++++
include/linux/res_group_rc.h | 97 +++++++++
init/Kconfig | 22 ++
kernel/Makefile | 1
kernel/fork.c | 7
kernel/res_group/Makefile | 2
kernel/res_group/local.h | 38 +++
kernel/res_group/numtasks.c | 461 +++++++++++++++++++++++++++++++++++++++++++
kernel/res_group/res_group.c | 161 +++++++++++++++
kernel/res_group/rgcs.c | 302 ++++++++++++++++++++++++++++
kernel/res_group/shares.c | 228 +++++++++++++++++++++
13 files changed, 1437 insertions(+), 4 deletions(-)
Index: container-2.6.19-rc2/include/linux/moduleparam.h
===================================================================
--- container-2.6.19-rc2.orig/include/linux/moduleparam.h
+++ container-2.6.19-rc2/include/linux/moduleparam.h
@@ -75,11 +75,17 @@ struct kparam_array
/* Helper functions: type is byte, short, ushort, int, uint, long,
ulong, charp, bool or invbool, or XXX if you define param_get_XXX,
param_set_XXX and param_check_XXX. */
-#define module_param_named(name, value, type, perm) \
- param_check_##type(name, &(value)); \
- module_param_call(name, param_set_##type, param_get_##type, &value, perm); \
+#define module_param_named_call(name, value, type, set, perm) \
+ param_check_##type(name, &(value)); \
+ module_param_call(name, set, param_get_##type, &(value), perm); \
__MODULE_PARM_TYPE(name, #type)
+#define module_param_named(name, value, type, perm) \
+ module_param_named_call(name, value, type, param_set_##type, perm)
+
+#define module_param_set_call(name, type, setfn, perm) \
+ module_param_named_call(name, name, type, setfn, perm)
+
#define module_param(name, type, perm) \
module_param_named(name, name, type, perm)
Index: container-2.6.19-rc2/include/linux/numtasks.h
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/include/linux/numtasks.h
@@ -0,0 +1,28 @@
+/* numtasks.h - No. of tasks resource controller for Resource Groups
+ *
+ * Copyright (C) Chandra Seetharaman, IBM Corp. 2003, 2004, 2005
+ *
+ * Provides No. of tasks resource controller for Resource Groups
+ *
+ * Latest version, more details at http://ckrm.sf.net
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ */
+#ifndef _LINUX_NUMTASKS_H
+#define _LINUX_NUMTASKS_H
+
+#ifdef CONFIG_RES_GROUPS_NUMTASKS
+#include <linux/res_group_rc.h>
+
+extern int numtasks_allow_fork(struct task_struct *);
+
+#else /* CONFIG_RES_GROUPS_NUMTASKS */
+
+#define numtasks_allow_fork(task) (0)
+
+#endif /* CONFIG_RES_GROUPS_NUMTASKS */
+#endif /* _LINUX_NUMTASKS_H */
Index: container-2.6.19-rc2/include/linux/res_group.h
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/include/linux/res_group.h
@@ -0,0 +1,82 @@
+/*
+ * res_group.h - Header file to be used by Resource Groups
+ *
+ * Copyright (C) Hubertus Franke, IBM Corp. 2003, 2004
+ * (C) Shailabh Nagar, IBM Corp. 2003, 2004
+ * (C) Chandra Seetharaman, IBM Corp. 2003, 2004, 2005
+ *
+ * Provides data structures, macros and kernel APIs
+ *
+ * More details at http://ckrm.sf.net
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ */
+
+#ifndef _LINUX_RES_GROUP_H
+#define _LINUX_RES_GROUP_H
+
+#ifdef CONFIG_RES_GROUPS
+#include <linux/spinlock.h>
+#include <linux/list.h>
+#include <linux/kref.h>
+#include <linux/container.h>
+
+#define SHARE_UNCHANGED (-1) /* implicitly specified by userspace,
+ * never stored in a resource group'
+ * shares struct; never displayed */
+#define SHARE_UNSUPPORTED (-2) /* If the resource controller doesn't
+ * support user changing a shares value
+ * it sets the corresponding share
+ * value to UNSUPPORTED when it returns
+ * the newly allocated shares data
+ * structure */
+#define SHARE_DONT_CARE (-3)
+
+#define SHARE_DEFAULT_DIVISOR (100)
+
+#define MAX_RES_CTLRS MAX_CONTAINER_SUBSYS /* max # of resource controllers */
+#define MAX_DEPTH 5 /* max depth of hierarchy supported */
+
+#define NO_RES_GROUP NULL
+#define NO_SHARE NULL
+#define NO_RES_ID MAX_RES_CTLRS /* Invalid ID */
+
+/*
+ * Share quantities are a child's fraction of the parent's resource
+ * specified by a divisor in the parent and a dividend in the child.
+ *
+ * Shares are represented as a relative quantity between parent and child
+ * to simplify locking when propagating modifications to the shares of a
+ * resource group. Only the parent and the children of the modified
+ * resource group need to be locked.
+*/
+struct res_shares {
+ /* shares only set by userspace */
+ int min_shares; /* minimun fraction of parent's resources allowed */
+ int max_shares; /* maximum fraction of parent's resources allowed */
+ int child_shares_divisor; /* >= 1, may not be DONT_CARE */
+
+ /*
+ * share values invisible to userspace. adjusted when userspace
+ * sets shares
+ */
+ int unused_min_shares;
+ /* 0 <= unused_min_shares <= (child_shares_divisor -
+ * Sum of min_shares of children)
+ */
+ int cur_max_shares; /* max(children's max_shares). need better name */
+};
+
+/*
+ * Class is the grouping of tasks with shares of each resource that has
+ * registered a resource controller (see include/linux/res_group_rc.h).
+ */
+
+#define resource_group container
+
+#endif /* CONFIG_RES_GROUPS */
+#endif /* _LINUX_RES_GROUP_H */
Index: container-2.6.19-rc2/include/linux/res_group_rc.h
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/include/linux/res_group_rc.h
@@ -0,0 +1,97 @@
+/*
+ * res_group_rc.h - Header file to be used by Resource controllers of
+ * Resource Groups
+ *
+ * Copyright (C) Hubertus Franke, IBM Corp. 2003
+ * (C) Shailabh Nagar, IBM Corp. 2003
+ * (C) Chandra Seetharaman, IBM Corp. 2003, 2004, 2005
+ * (C) Vivek Kashyap , IBM Corp. 2004
+ *
+ * Provides data structures, macros and kernel API of Resource Groups for
+ * resource controllers.
+ *
+ * More details at http://ckrm.sf.net
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ */
+
+#ifndef _LINUX_RES_GROUP_RC_H
+#define _LINUX_RES_GROUP_RC_H
+
+#include <linux/res_group.h>
+#include <linux/container.h>
+
+struct res_group_cft {
+ struct cftype cft;
+ struct res_controller *ctlr;
+ char name[MAX_CONTAINER_TYPE_NAMELEN + 10];
+};
+
+struct res_controller {
+ struct container_subsys subsys;
+ struct res_group_cft shares_cft;
+ struct res_group_cft stats_cft;
+
+ const char *name;
+ unsigned int ctlr_id;
+
+ /*
+ * Keeps number of references to this controller structure. kref
+ * does not work as we want to be able to allow removal of a
+ * controller even when some resource group are still defined.
+ */
+ atomic_t count;
+
+ /*
+ * Allocate a new shares struct for this resource controller.
+ * Called when registering a resource controller with pre-existing
+ * resource groups and when new resource group is created by the user.
+ */
+ struct res_shares *(*alloc_shares_struct)(struct container *);
+ /* Corresponding free of shares struct for this resource controller */
+ void (*free_shares_struct)(struct res_shares *);
+
+ /* Notifies the controller when the shares are changed */
+ void (*shares_changed)(struct res_shares *);
+
+ /* resource statistics */
+ ssize_t (*show_stats)(struct res_shares *, char *, size_t);
+ int (*reset_stats)(struct res_shares *, const char *);
+
+ /*
+ * move_task is called when a task moves from one resource group to
+ * another. First parameter is the task that is moving, the second
+ * is the resource specific shares of the resource group the task
+ * was in, and the third is the shares of the resource group the
+ * task has moved to.
+ */
+ void (*move_task)(struct task_struct *, struct res_shares *,
+ struct res_shares *);
+};
+
+extern int register_controller(struct res_controller *);
+extern int unregister_controller(struct res_controller *);
+extern struct resource_group default_res_group;
+static inline int is_res_group_root(const struct resource_group *rgroup)
+{
+ return (rgroup->parent == NULL);
+}
+
+#define for_each_child(child, parent) \
+ list_for_each_entry(child, &parent->children, sibling)
+
+/* Get controller specific shares structure for the given resource group */
+static inline struct res_shares *get_controller_shares(
+ struct container *rgroup, struct res_controller *ctlr)
+{
+ if (rgroup && ctlr)
+ return rgroup->subsys[ctlr->subsys.subsys_id];
+ else
+ return NO_SHARE;
+}
+
+#endif /* _LINUX_RES_GROUP_RC_H */
Index: container-2.6.19-rc2/init/Kconfig
===================================================================
--- container-2.6.19-rc2.orig/init/Kconfig
+++ container-2.6.19-rc2/init/Kconfig
@@ -303,6 +303,28 @@ config TASK_XACCT
Say N if unsure.
+menu "Resource Groups"
+
+config RES_GROUPS
+ bool "Resource Groups"
+ depends on EXPERIMENTAL
+ select CONTAINERS
+ help
+ Resource Groups is a framework for controlling and monitoring
+ resource allocation of user-defined groups of tasks. For more
+ information, please visit http://ckrm.sf.net.
+
+config RES_GROUPS_NUMTASKS
+ bool "Number of Tasks Resource Controller"
+ depends on RES_GROUPS
+ default y
+ help
+ Provides a Resource Controller for Resource Groups that allows
+ limiting number of tasks a resource group can have.
+
+ Say N if unsure, Y to use the feature.
+
+endmenu
config SYSCTL
bool
Index: container-2.6.19-rc2/kernel/Makefile
===================================================================
--- container-2.6.19-rc2.orig/kernel/Makefile
+++ container-2.6.19-rc2/kernel/Makefile
@@ -53,6 +53,7 @@ obj-$(CONFIG_RELAY) += relay.o
obj-$(CONFIG_UTS_NS) += utsname.o
obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o
obj-$(CONFIG_TASKSTATS) += taskstats.o tsacct.o
+obj-$(CONFIG_RES_GROUPS) += res_group/
ifneq ($(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER),y)
# According to Alan Modra <alan@linuxcare.com.au>, the -fno-omit-frame-pointer is
Index: container-2.6.19-rc2/kernel/fork.c
===================================================================
--- container-2.6.19-rc2.orig/kernel/fork.c
+++ container-2.6.19-rc2/kernel/fork.c
@@ -48,6 +48,7 @@
#include <linux/delayacct.h>
#include <linux/taskstats_kern.h>
#include <linux/random.h>
+#include <linux/numtasks.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
@@ -1353,7 +1354,7 @@ long do_fork(unsigned long clone_flags,
int __user *child_tidptr)
{
struct task_struct *p;
- int trace = 0;
+ int trace = 0, rc;
struct pid *pid = alloc_pid();
long nr;
@@ -1366,6 +1367,10 @@ long do_fork(unsigned long clone_flags,
clone_flags |= CLONE_PTRACE;
}
+ rc = numtasks_allow_fork(current);
+ if (rc)
+ return rc;
+
p = copy_process(clone_flags, stack_start, regs, stack_size, parent_tidptr, child_tidptr, nr);
/*
* Do this prior waking up the new thread - the thread pointer
Index: container-2.6.19-rc2/kernel/res_group/Makefile
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/kernel/res_group/Makefile
@@ -0,0 +1,2 @@
+obj-y = res_group.o shares.o rgcs.o
+obj-$(CONFIG_RES_GROUPS_NUMTASKS) += numtasks.o
Index: container-2.6.19-rc2/kernel/res_group/local.h
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/kernel/res_group/local.h
@@ -0,0 +1,38 @@
+/*
+ * Contains function definitions that are local to the Resource Groups.
+ * NOT to be included by controllers.
+ */
+
+#include <linux/res_group_rc.h>
+
+extern struct res_controller *get_controller_by_name(const char *);
+extern struct res_controller *get_controller_by_id(unsigned int);
+extern void put_controller(struct res_controller *);
+extern struct resource_group *alloc_res_group(struct resource_group *,
+ const char *);
+extern int free_res_group(struct resource_group *);
+extern void release_res_group(struct kref *);
+extern int set_controller_shares(struct resource_group *,
+ struct res_controller *, const struct res_shares *);
+/* Set shares for the given resource group and resource to default values */
+extern void set_shares_to_default(struct resource_group *,
+ struct res_controller *);
+extern void res_group_teardown(void);
+extern int set_res_group(pid_t, struct resource_group *);
+extern void move_tasks_to_parent(struct resource_group *);
+
+ssize_t res_group_file_read(struct container *cont,
+ struct cftype *cft,
+ struct file *file,
+ char __user *buf,
+ size_t nbytes, loff_t *ppos);
+ssize_t res_group_file_write(struct container *cont,
+ struct cftype *cft,
+ struct file *file,
+ const char __user *userbuf,
+ size_t nbytes, loff_t *ppos);
+
+enum {
+ RG_FILE_SHARES,
+ RG_FILE_STATS,
+};
Index: container-2.6.19-rc2/kernel/res_group/numtasks.c
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/kernel/res_group/numtasks.c
@@ -0,0 +1,461 @@
+/* numtasks.c - "Number of tasks" resource controller for Resource Groups
+ *
+ * Copyright (C) Chandra Seetharaman, IBM Corp. 2003-2006
+ * (C) Matt Helsley, IBM Corp. 2006
+ *
+ * Latest version, more details at http://ckrm.sf.net
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ */
+
+/*
+ * Resource controller for tracking number of tasks in a resource group.
+ */
+#include <linux/module.h>
+#include <linux/res_group_rc.h>
+#include <linux/numtasks.h>
+
+static const char res_ctlr_name[] = "numtasks";
+
+#define UNLIMITED INT_MAX
+#define DEF_TOTAL_NUM_TASKS UNLIMITED
+static int total_numtasks __read_mostly = DEF_TOTAL_NUM_TASKS;
+
+static struct resource_group *root_rgroup;
+static int total_cnt_alloc = 0;
+
+#define DEF_FORKRATE UNLIMITED
+#define DEF_FORKRATE_INTERVAL (1)
+static int forkrate __read_mostly = DEF_FORKRATE;
+static int forkrate_interval __read_mostly = DEF_FORKRATE_INTERVAL;
+
+struct numtasks {
+ struct resource_group *rgroup;/* resource group i am part of... */
+ struct res_shares shares;
+ int cnt_min_shares; /* num_tasks min_shares in local units */
+ int cnt_unused; /* has to borrow if more than this is needed */
+ int cnt_max_shares; /* no tasks over this limit. */
+ /* Three above cnt_* fields are protected
+ * by resource group's group_lock */
+ atomic_t cnt_cur_alloc; /* current alloc from self */
+ atomic_t cnt_borrowed; /* borrowed from the parent */
+
+ /* stats */
+ int successes;
+ int failures;
+ int forkrate_failures;
+
+ /* Fork rate fields */
+ int forks_in_period;
+ unsigned long period_start;
+};
+
+struct res_controller numtasks_ctlr;
+
+static struct numtasks *get_shares_numtasks(struct res_shares *shares)
+{
+ if (shares)
+ return container_of(shares, struct numtasks, shares);
+ return NULL;
+}
+
+static struct numtasks *get_numtasks(struct resource_group *rgroup)
+{
+ return get_shares_numtasks(get_controller_shares(rgroup,
+ &numtasks_ctlr));
+}
+
+static inline int check_forkrate(struct numtasks *res)
+{
+ if (time_after(jiffies, res->period_start + forkrate_interval * HZ)) {
+ res->period_start = jiffies;
+ res->forks_in_period = 0;
+ }
+
+ if (res->forks_in_period >= forkrate) {
+ res->forkrate_failures++;
+ return -ENOSPC;
+ }
+ res->forks_in_period++;
+ return 0;
+}
+
+int numtasks_allow_fork(struct task_struct *task)
+{
+ int rc = 0;
+ struct numtasks *res;
+ struct resource_group *rgroup;
+
+ /* task->container won't change during an RCU critical section */
+ rcu_read_lock();
+
+ rgroup = rcu_dereference(task->container);
+ /* controller is not registered; no resource group is given */
+ if ((numtasks_ctlr.ctlr_id == NO_RES_ID) || (rgroup == NULL))
+ goto out;
+ res = get_numtasks(rgroup);
+
+ /* numtasks not available for this resource group */
+ if (!res)
+ goto out;
+
+ /* Check forkrate before checking resource group's usage */
+ rc = check_forkrate(res);
+ if (rc)
+ goto out;
+
+ if (res->cnt_max_shares == SHARE_DONT_CARE)
+ goto out;
+
+ /* Over the limit ? */
+ if (atomic_read(&res->cnt_cur_alloc) >= res->cnt_max_shares) {
+ res->failures++;
+ rc = -ENOSPC;
+ goto out;
+ }
+ out:
+ rcu_read_unlock();
+ return rc;
+}
+
+static void inc_usage_count(struct numtasks *res)
+{
+ atomic_inc(&res->cnt_cur_alloc);
+
+ if (is_res_group_root(res->rgroup)) {
+ total_cnt_alloc++;
+ res->successes++;
+ return;
+ }
+ /* Do we need to borrow from our parent ? */
+ if ((res->cnt_unused == SHARE_DONT_CARE) ||
+ (atomic_read(&res->cnt_cur_alloc) > res->cnt_unused)) {
+ inc_usage_count(get_numtasks(res->rgroup->parent));
+ atomic_inc(&res->cnt_borrowed);
+ } else {
+ total_cnt_alloc++;
+ res->successes++;
+ }
+}
+
+static void dec_usage_count(struct numtasks *res)
+{
+ if (atomic_read(&res->cnt_cur_alloc) == 0)
+ return;
+ atomic_dec(&res->cnt_cur_alloc);
+ if (atomic_read(&res->cnt_borrowed) > 0) {
+ atomic_dec(&res->cnt_borrowed);
+ dec_usage_count(get_numtasks(res->rgroup->parent));
+ } else
+ total_cnt_alloc--;
+
+}
+
+static void numtasks_move_task(struct task_struct *task,
+ struct res_shares *old, struct res_shares *new)
+{
+ struct numtasks *oldres, *newres;
+
+ if (old == new)
+ return;
+
+ /* Decrement usage count of old resource group */
+ oldres = get_shares_numtasks(old);
+ if (oldres)
+ dec_usage_count(oldres);
+
+ /* Increment usage count of new resource group */
+ newres = get_shares_numtasks(new);
+ if (newres)
+ inc_usage_count(newres);
+}
+
+/* Initialize share struct values */
+static void numtasks_res_init_one(struct numtasks *numtasks_res)
+{
+ numtasks_res->shares.min_shares = SHARE_DONT_CARE;
+ numtasks_res->shares.max_shares = SHARE_DONT_CARE;
+ numtasks_res->shares.child_shares_divisor = SHARE_DEFAULT_DIVISOR;
+ numtasks_res->shares.unused_min_shares = SHARE_DEFAULT_DIVISOR;
+
+ numtasks_res->cnt_min_shares = SHARE_DONT_CARE;
+ numtasks_res->cnt_unused = SHARE_DONT_CARE;
+ numtasks_res->cnt_max_shares = SHARE_DONT_CARE;
+ numtasks_res->period_start = jiffies;
+}
+
+static struct res_shares *numtasks_alloc_shares_struct(
+ struct resource_group *rgroup)
+{
+ struct numtasks *res;
+
+ res = kzalloc(sizeof(struct numtasks), GFP_KERNEL);
+ if (!res)
+ return NULL;
+ res->rgroup = rgroup;
+ numtasks_res_init_one(res);
+ if (is_res_group_root(rgroup))
+ root_rgroup = rgroup; /* store root's resource group. */
+ return &res->shares;
+}
+
+/*
+ * No locking of this resource group object necessary as we are not
+ * supposed to be assigned (or used) when/after this function is called.
+ */
+static void numtasks_free_shares_struct(struct res_shares *my_res)
+{
+ struct numtasks *res, *parres;
+ int i, borrowed;
+
+ res = get_shares_numtasks(my_res);
+ if (!is_res_group_root(res->rgroup)) {
+ parres = get_numtasks(res->rgroup->parent);
+ borrowed = atomic_read(&res->cnt_borrowed);
+ for (i = 0; i < borrowed; i++)
+ dec_usage_count(parres);
+ }
+ kfree(res);
+}
+
+static int recalc_shares(int self_shares, int parent_shares, int parent_divisor)
+{
+ u64 numerator;
+
+ if ((self_shares == SHARE_DONT_CARE) ||
+ (parent_shares == SHARE_DONT_CARE))
+ return SHARE_DONT_CARE;
+ if (parent_divisor == 0)
+ return 0;
+ numerator = (u64) self_shares * parent_shares;
+ do_div(numerator, parent_divisor);
+ return numerator;
+}
+
+static int recalc_unused_shares(int self_cnt_min_shares,
+ int self_unused_min_shares, int self_divisor)
+{
+ u64 numerator;
+
+ if (self_cnt_min_shares == SHARE_DONT_CARE)
+ return SHARE_DONT_CARE;
+ if (self_divisor == 0)
+ return 0;
+ numerator = (u64) self_unused_min_shares * self_cnt_min_shares;
+ do_div(numerator, self_divisor);
+ return numerator;
+}
+
+static void recalc_self(struct numtasks *res,
+ struct numtasks *parres)
+{
+ struct res_shares *par = &parres->shares;
+ struct res_shares *self = &res->shares;
+
+ res->cnt_min_shares = recalc_shares(self->min_shares,
+ parres->cnt_min_shares,
+ par->child_shares_divisor);
+ res->cnt_max_shares = recalc_shares(self->max_shares,
+ parres->cnt_max_shares,
+ par->child_shares_divisor);
+
+ /*
+ * Now that we know the new cnt_min/cnt_max boundaries we can update
+ * the unused quantity.
+ */
+ res->cnt_unused = recalc_unused_shares(res->cnt_min_shares,
+ self->unused_min_shares,
+ self->child_shares_divisor);
+}
+
+
+/*
+ * Recalculate the min_shares and max_shares in real units and propagate the
+ * same to children.
+ * Called with container_manage_lock() held.
+ */
+static void recalc_and_propagate(struct numtasks *res,
+ struct numtasks *parres)
+{
+ struct resource_group *child = NULL;
+ struct numtasks *childres;
+
+ if (parres)
+ recalc_self(res, parres);
+
+ /* propagate to children */
+ for_each_child(child, res->rgroup) {
+ childres = get_numtasks(child);
+ BUG_ON(!childres);
+ recalc_and_propagate(childres, res);
+ }
+}
+
+static void numtasks_shares_changed(struct res_shares *my_res)
+{
+ struct numtasks *parres, *res;
+ struct res_shares *cur, *par;
+
+ res = get_shares_numtasks(my_res);
+ if (!res)
+ return;
+ cur = &res->shares;
+
+ if (!is_res_group_root(res->rgroup)) {
+ parres = get_numtasks(res->rgroup->parent);
+ par = &parres->shares;
+ } else {
+ parres = NULL;
+ par = NULL;
+ }
+ if (parres)
+ parres->cnt_unused = recalc_unused_shares(
+ parres->cnt_min_shares,
+ par->unused_min_shares,
+ par->child_shares_divisor);
+ recalc_and_propagate(res, parres);
+}
+
+static ssize_t numtasks_show_stats(struct res_shares *my_res,
+ char *buf, size_t buf_size)
+{
+ ssize_t i, j = 0;
+ struct numtasks *res;
+
+ res = get_shares_numtasks(my_res);
+ if (!res)
+ return -EINVAL;
+
+ i = snprintf(buf, buf_size, "%s: Current usage %d\n",
+ res_ctlr_name,
+ atomic_read(&res->cnt_cur_alloc));
+ buf += i; j += i; buf_size -= i;
+ i = snprintf(buf, buf_size, "%s: Number of successes %d\n",
+ res_ctlr_name, res->successes);
+ buf += i; j += i; buf_size -= i;
+ i = snprintf(buf, buf_size, "%s: Number of failures %d\n",
+ res_ctlr_name, res->failures);
+ buf += i; j += i; buf_size -= i;
+ i = snprintf(buf, buf_size, "%s: Number of forkrate failures %d\n",
+ res_ctlr_name, res->forkrate_failures);
+ j += i;
+ return j;
+}
+
+struct res_controller numtasks_ctlr = {
+ .name = res_ctlr_name,
+ .ctlr_id = NO_RES_ID,
+ .alloc_shares_struct = numtasks_alloc_shares_struct,
+ .free_shares_struct = numtasks_free_shares_struct,
+ .move_task = numtasks_move_task,
+ .shares_changed = numtasks_shares_changed,
+ .show_stats = numtasks_show_stats,
+};
+
+/*
+ * Writeable module parameters use these set_<parameter> functions to respond
+ * to changes. Otherwise the values can be read and used any time.
+ */
+static int set_numtasks_config_val(int *var, int old_value, const char *val,
+ struct kernel_param *kp)
+{
+ int rc = param_set_int(val, kp);
+
+ if (rc < 0)
+ return rc;
+ if (*var < 1) {
+ *var = old_value;
+ return -EINVAL;
+ }
+ return 0;
+}
+
+static int set_total_numtasks(const char *val, struct kernel_param *kp)
+{
+ int prev = total_numtasks;
+ int rc = set_numtasks_config_val(&total_numtasks, prev, val, kp);
+ struct numtasks *res = NULL;
+
+ if (!root_rgroup)
+ return 0;
+ if (rc < 0)
+ return rc;
+ if (total_numtasks <= total_cnt_alloc) {
+ total_numtasks = prev;
+ return -EINVAL;
+ }
+ container_lock();
+ res = get_numtasks(root_rgroup);
+ res->cnt_min_shares = total_numtasks;
+ res->cnt_unused = total_numtasks;
+ res->cnt_max_shares = total_numtasks;
+ recalc_and_propagate(res, NULL);
+ container_unlock();
+ return 0;
+}
+module_param_set_call(total_numtasks, int, set_total_numtasks,
+ S_IRUGO | S_IWUSR);
+
+static void reset_forkrates(struct resource_group *rgroup, unsigned long now)
+{
+ struct numtasks *res;
+ struct resource_group *child = NULL;
+
+ res = get_numtasks(rgroup);
+ if (!res)
+ return;
+ res->forks_in_period = 0;
+ res->period_start = now;
+
+ for_each_child(child, rgroup)
+ reset_forkrates(child, now);
+}
+
+static int set_forkrate(const char *val, struct kernel_param *kp)
+{
+ int prev = forkrate;
+ int rc = set_numtasks_config_val(&forkrate, prev, val, kp);
+ if (rc < 0)
+ return rc;
+ container_lock();
+ reset_forkrates(root_rgroup, jiffies);
+ container_unlock();
+ return 0;
+}
+module_param_set_call(forkrate, int, set_forkrate, S_IRUGO | S_IWUSR);
+
+static int set_forkrate_interval(const char *val, struct kernel_param *kp)
+{
+ int prev = forkrate_interval;
+ int rc = set_numtasks_config_val(&forkrate_interval, prev, val, kp);
+ if (rc < 0)
+ return rc;
+ container_lock();
+ reset_forkrates(root_rgroup, jiffies);
+ container_unlock();
+ return 0;
+}
+module_param_set_call(forkrate_interval, int, set_forkrate_interval,
+ S_IRUGO | S_IWUSR);
+
+int __init init_numtasks_res(void)
+{
+ if (numtasks_ctlr.ctlr_id != NO_RES_ID)
+ return -EBUSY; /* already registered */
+ return register_controller(&numtasks_ctlr);
+}
+
+void __exit exit_numtasks_res(void)
+{
+ int rc;
+ do {
+ rc = unregister_controller(&numtasks_ctlr);
+ } while (rc == -EBUSY);
+ BUG_ON(rc != 0);
+}
+module_init(init_numtasks_res)
+module_exit(exit_numtasks_res)
Index: container-2.6.19-rc2/kernel/res_group/res_group.c
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/kernel/res_group/res_group.c
@@ -0,0 +1,161 @@
+/* res_group.c - Resource Groups: Resource management through grouping of
+ * unrelated tasks.
+ *
+ * Copyright (C) Hubertus Franke, IBM Corp. 2003, 2004
+ * (C) Shailabh Nagar, IBM Corp. 2003, 2004
+ * (C) Chandra Seetharaman, IBM Corp. 2003, 2004, 2005
+ * (C) Vivek Kashyap, IBM Corp. 2004
+ * (C) Matt Helsley, IBM Corp. 2006
+ *
+ * Provides kernel API of Resource Groups for in-kernel,per-resource
+ * controllers (one each for cpu, memory and io).
+ *
+ * Latest version, more details at http://ckrm.sf.net
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ */
+
+#include <linux/module.h>
+#include <asm/uaccess.h>
+#include <linux/fs.h>
+#include "local.h"
+
+
+static int res_group_create(struct container_subsys *ss,
+ struct container *cont)
+{
+ struct res_controller *ctlr = container_of(ss, struct res_controller, subsys);
+ struct res_shares *shares = ctlr->alloc_shares_struct(cont);
+ cont->subsys[ss->subsys_id] = shares;
+ return 0;
+}
+
+static void res_group_destroy(struct container_subsys *ss,
+ struct container *cont)
+{
+ struct res_controller *ctlr = container_of(ss, struct res_controller, subsys);
+ struct res_shares *shares = cont->subsys[ss->subsys_id];
+ ctlr->free_shares_struct(shares);
+}
+
+static int res_group_populate(struct container_subsys *ss,
+ struct container *cont) {
+ int err;
+ struct res_controller *ctlr = container_of(ss, struct res_controller, subsys);
+ if ((err = container_add_file(cont, &ctlr->shares_cft.cft)) < 0)
+ return err;
+ if ((err = container_add_file(cont, &ctlr->stats_cft.cft)) < 0)
+ return err;
+
+ return 0;
+}
+
+static void res_group_attach(struct container_subsys *ss,
+ struct container *cont,
+ struct container *old_cont,
+ struct task_struct *tsk) {
+ struct res_controller *ctlr = container_of(ss, struct res_controller, subsys);
+ struct res_shares *oldshares = old_cont->subsys[ss->subsys_id];
+ struct res_shares *newshares = cont->subsys[ss->subsys_id];
+
+ if (ctlr->move_task) {
+ ctlr->move_task(tsk, oldshares, newshares);
+ }
+}
+
+static void res_group_fork(struct container_subsys *ss, struct container *cont,
+ struct task_struct *task) {
+ struct res_controller *ctlr =
+ container_of(ss, struct res_controller, subsys);
+ struct res_shares *shares = cont->subsys[ss->subsys_id];
+ if (ctlr->move_task) {
+ ctlr->move_task(task, NULL, shares);
+ }
+}
+
+static void res_group_exit(struct container_subsys *ss, struct container *cont,
+ struct task_struct *task) {
+ struct res_controller *ctlr =
+ container_of(ss, struct res_controller, subsys);
+ struct res_shares *shares = cont->subsys[ss->subsys_id];
+ if (ctlr->move_task) {
+ ctlr->move_task(task, shares, NULL);
+ }
+}
+
+/*
+ * Interface for registering a resource controller.
+ *
+ * Returns the 0 on success, -errno for failure.
+ * Fills ctlr->ctlr_id with a valid controller id on success.
+ */
+int register_controller(struct res_controller *ctlr)
+{
+ int ret;
+
+ struct container_subsys *ss = &ctlr->subsys;
+
+ if (!ctlr)
+ return -EINVAL;
+
+ /* Make sure there is an alloc and a free */
+ if (!ctlr->alloc_shares_struct || !ctlr->free_shares_struct)
+ return -EINVAL;
+
+ ss->create = res_group_create;
+ ss->destroy = res_group_destroy;
+ ss->populate = res_group_populate;
+ if (ctlr->move_task) {
+ ss->attach = res_group_attach;
+ ss->fork = res_group_fork;
+ ss->exit = res_group_exit;
+ }
+
+ ctlr->shares_cft.ctlr = ctlr;
+ strcpy(ctlr->shares_cft.name, ctlr->name);
+ strcat(ctlr->shares_cft.name, "_shares");
+ ctlr->shares_cft.cft.name = ctlr->shares_cft.name;
+ ctlr->shares_cft.cft.private = RG_FILE_SHARES;
+ ctlr->shares_cft.cft.read = res_group_file_read;
+ ctlr->shares_cft.cft.write = res_group_file_write;
+
+ ctlr->stats_cft.ctlr = ctlr;
+ strcpy(ctlr->stats_cft.name, ctlr->name);
+ strcat(ctlr->stats_cft.name, "_stats");
+ ctlr->stats_cft.cft.name = ctlr->stats_cft.name;
+ ctlr->stats_cft.cft.private = RG_FILE_STATS;
+ ctlr->stats_cft.cft.read = res_group_file_read;
+ ctlr->stats_cft.cft.write = res_group_file_write;
+
+ ss->name = ctlr->name;
+ ss->enabled = 1;
+
+ ret = container_register_subsys(ss);
+
+ if (ret < 0)
+ return ret;
+
+ ctlr->ctlr_id = ss->subsys_id;
+
+ return 0;
+}
+
+/*
+ * Unregistering resource controller.
+ *
+ * Returns 0 on success -errno for failure.
+ */
+int unregister_controller(struct res_controller *ctlr)
+{
+ BUG();
+ return 0;
+}
+
+
+EXPORT_SYMBOL_GPL(register_controller);
+EXPORT_SYMBOL_GPL(unregister_controller);
+EXPORT_SYMBOL_GPL(set_controller_shares);
Index: container-2.6.19-rc2/kernel/res_group/rgcs.c
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/kernel/res_group/rgcs.c
@@ -0,0 +1,302 @@
+/*
+ * kernel/res_group/rgcs.c
+ *
+ * Copyright (C) Shailabh Nagar, IBM Corp. 2005
+ * Chandra Seetharaman, IBM Corp. 2005, 2006
+ *
+ * Resource Group Configfs Subsystem (rgcs) provides the user interface
+ * for Resource groups.
+ *
+ * Latest version, more details at http://ckrm.sf.net
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License
+ * as published by the Free Software Foundation.
+ *
+ */
+#include <linux/ctype.h>
+#include <linux/module.h>
+#include <linux/configfs.h>
+#include <linux/parser.h>
+#include <linux/fs.h>
+#include <asm/uaccess.h>
+
+#include "local.h"
+
+#define RES_STRING "res"
+#define MIN_SHARES_STRING "min_shares"
+#define MAX_SHARES_STRING "max_shares"
+#define CHILD_SHARES_DIVISOR_STRING "child_shares_divisor"
+
+static ssize_t show_stats(struct resource_group *rgroup,
+ struct res_controller *ctlr,
+ char *buf)
+{
+ int j = 0, rc = 0;
+ size_t buf_size = PAGE_SIZE-1; /* allow only PAGE_SIZE # of bytes */
+ struct res_shares *shares;
+
+ shares = get_controller_shares(rgroup, ctlr);
+ if (shares && ctlr->show_stats)
+ j = ctlr->show_stats(shares, buf, buf_size);
+ rc += j;
+ buf += j;
+ buf_size -= j;
+ return rc;
+}
+
+enum parse_token_t {
+ parse_res_type, parse_err
+};
+
+static match_table_t parse_tokens = {
+ {parse_res_type, RES_STRING"=%s"},
+ {parse_err, NULL}
+};
+
+static int stats_parse(const char *options,
+ char **resname, char **remaining_line)
+{
+ char *p, *str;
+ int rc = -EINVAL;
+
+ if (!options)
+ return -EINVAL;
+
+ while ((p = strsep((char **)&options, ",")) != NULL) {
+ substring_t args[MAX_OPT_ARGS];
+ int token;
+
+ if (!*p)
+ continue;
+ token = match_token(p, parse_tokens, args);
+ if (token == parse_res_type) {
+ *resname = match_strdup(args);
+ str = p + strlen(p) + 1;
+ *remaining_line = kmalloc(strlen(str) + 1, GFP_KERNEL);
+ if (*remaining_line == NULL) {
+ kfree(*resname);
+ *resname = NULL;
+ rc = -ENOMEM;
+ } else {
+ strcpy(*remaining_line, str);
+ rc = 0;
+ }
+ break;
+ }
+ }
+ return rc;
+}
+
+static int reset_stats(struct resource_group *rgroup, struct res_controller *ctlr, const char *str)
+{
+ int rc;
+ char *resname = NULL, *statstr = NULL;
+ struct res_shares *shares;
+
+ rc = stats_parse(str, &resname, &statstr);
+ if (rc)
+ return rc;
+
+ shares = get_controller_shares(rgroup, ctlr);
+ if (shares && ctlr->reset_stats)
+ rc = ctlr->reset_stats(shares, statstr);
+
+ kfree(resname);
+ kfree(statstr);
+ return rc;
+}
+
+
+enum share_token_t {
+ MIN_SHARES_TOKEN,
+ MAX_SHARES_TOKEN,
+ CHILD_SHARES_DIVISOR_TOKEN,
+ RESOURCE_TYPE_TOKEN,
+ ERROR_TOKEN
+};
+
+/* Token matching for parsing input to this magic file */
+static match_table_t shares_tokens = {
+ {RESOURCE_TYPE_TOKEN, RES_STRING"=%s"},
+ {MIN_SHARES_TOKEN, MIN_SHARES_STRING"=%d"},
+ {MAX_SHARES_TOKEN, MAX_SHARES_STRING"=%d"},
+ {CHILD_SHARES_DIVISOR_TOKEN, CHILD_SHARES_DIVISOR_STRING"=%d"},
+ {ERROR_TOKEN, NULL}
+};
+
+static int shares_parse(const char *options, char **resname,
+ struct res_shares *shares)
+{
+ char *p;
+ int option, rc = -EINVAL;
+
+ *resname = NULL;
+ if (!options)
+ goto done;
+
+ while ((p = strsep((char **)&options, ",")) != NULL) {
+ substring_t args[MAX_OPT_ARGS];
+ int token;
+
+ if (!*p)
+ continue;
+
+ token = match_token(p, shares_tokens, args);
+ switch (token) {
+ case RESOURCE_TYPE_TOKEN:
+ if (*resname)
+ goto done;
+ *resname = match_strdup(args);
+ break;
+ case MIN_SHARES_TOKEN:
+ if (match_int(args, &option))
+ goto done;
+ shares->min_shares = option;
+ break;
+ case MAX_SHARES_TOKEN:
+ if (match_int(args, &option))
+ goto done;
+ shares->max_shares = option;
+ break;
+ case CHILD_SHARES_DIVISOR_TOKEN:
+ if (match_int(args, &option))
+ goto done;
+ shares->child_shares_divisor = option;
+ break;
+ default:
+ goto done;
+ }
+ }
+ rc = 0;
+done:
+ if (rc) {
+ kfree(*resname);
+ *resname = NULL;
+ }
+ return rc;
+}
+
+static int set_shares(struct resource_group *rgroup,
+ struct res_controller *ctlr,
+ const char *str)
+{
+ char *resname = NULL;
+ int rc;
+ struct res_shares shares = {
+ .min_shares = SHARE_UNCHANGED,
+ .max_shares = SHARE_UNCHANGED,
+ .child_shares_divisor = SHARE_UNCHANGED,
+ };
+
+ rc = shares_parse(str, &resname, &shares);
+ if (!rc) {
+ rc = set_controller_shares(rgroup, ctlr, &shares);
+ kfree(resname);
+ }
+ return rc;
+}
+
+static ssize_t show_shares(struct resource_group *rgroup,
+ struct res_controller *ctlr,
+ char *buf)
+{
+ ssize_t j, rc = 0, bufsize = PAGE_SIZE;
+ struct res_shares *shares;
+
+ shares = get_controller_shares(rgroup, ctlr);
+ if (shares) {
+ j = snprintf(buf, bufsize, "%s=%s,%s=%d,%s=%d,%s=%d\n",
+ RES_STRING, ctlr->name,
+ MIN_SHARES_STRING, shares->min_shares,
+ MAX_SHARES_STRING, shares->max_shares,
+ CHILD_SHARES_DIVISOR_STRING,
+ shares->child_shares_divisor);
+ rc += j; buf += j; bufsize -= j;
+ }
+ return rc;
+}
+
+ssize_t res_group_file_write(struct container *cont,
+ struct cftype *cft,
+ struct file *file,
+ const char __user *userbuf,
+ size_t nbytes, loff_t *ppos)
+{
+ struct res_group_cft *rgcft = container_of(cft, struct res_group_cft, cft);
+ struct res_controller *ctlr = rgcft->ctlr;
+
+ char *buf;
+ ssize_t retval;
+ int filetype = cft->private;
+
+ if (nbytes >= PAGE_SIZE)
+ return -E2BIG;
+
+ buf = kmalloc(nbytes + 1, GFP_USER);
+ if (!buf) return -ENOMEM;
+ if (copy_from_user(buf, userbuf, nbytes)) {
+ retval = -EFAULT;
+ goto out1;
+ }
+ buf[nbytes] = 0; /* nul-terminate */
+
+ container_manage_lock();
+
+ if (container_is_removed(cont)) {
+ retval = -ENODEV;
+ goto out2;
+ }
+
+ switch(filetype) {
+ case RG_FILE_SHARES:
+ retval = set_shares(cont, ctlr, buf);
+ break;
+ case RG_FILE_STATS:
+ retval = reset_stats(cont, ctlr, buf);
+ break;
+ default:
+ retval = -EINVAL;
+ }
+ if (!retval) retval = nbytes;
+
+ out2:
+ container_manage_unlock();
+ out1:
+ kfree(buf);
+ return retval;
+}
+
+ssize_t res_group_file_read(struct container *cont,
+ struct cftype *cft,
+ struct file *file,
+ char __user *buf,
+ size_t nbytes, loff_t *ppos)
+{
+ struct res_group_cft *rgcft = container_of(cft, struct res_group_cft, cft);
+ struct res_controller *ctlr = rgcft->ctlr;
+
+ char *page = kmalloc(PAGE_SIZE, GFP_USER);
+ ssize_t retval;
+ int filetype = cft->private;
+
+ if (!page) return -ENOMEM;
+
+ switch(filetype) {
+ case RG_FILE_SHARES:
+ retval = show_shares(cont, ctlr, page);
+ break;
+ case RG_FILE_STATS:
+ retval = show_stats(cont, ctlr, page);
+ break;
+ default:
+ retval = -EINVAL;
+ }
+
+ if (retval >= 0) {
+ retval = simple_read_from_buffer(buf, nbytes,
+ ppos, page, retval);
+ }
+ kfree(page);
+ return retval;
+}
Index: container-2.6.19-rc2/kernel/res_group/shares.c
===================================================================
--- /dev/null
+++ container-2.6.19-rc2/kernel/res_group/shares.c
@@ -0,0 +1,228 @@
+/*
+ * shares.c - Share management functions for Resource Groups
+ *
+ * Copyright (C) Chandra Seetharaman, IBM Corp. 2003, 2004, 2005, 2006
+ * (C) Hubertus Franke, IBM Corp. 2004
+ * (C) Matt Helsley, IBM Corp. 2006
+ *
+ * Latest version, more details at http://ckrm.sf.net
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/errno.h>
+#include <linux/res_group_rc.h>
+#include <linux/container.h>
+
+/*
+ * Share values can be quantitative (quantity of memory for instance) or
+ * symbolic. The symbolic value DONT_CARE allows for any quantity of a resource
+ * to be substituted in its place. The symbolic value UNCHANGED is only used
+ * when setting share values and means that the old value should be used.
+ */
+
+/* Is the share a quantity (as opposed to "symbols" DONT_CARE or UNCHANGED) */
+static inline int is_share_quantitative(int share)
+{
+ return (share >= 0);
+}
+
+static inline int is_share_symbolic(int share)
+{
+ return !is_share_quantitative(share);
+}
+
+static inline int is_share_valid(int share)
+{
+ return ((share == SHARE_DONT_CARE) ||
+ (share == SHARE_UNSUPPORTED) ||
+ is_share_quantitative(share));
+}
+
+static inline int did_share_change(int share)
+{
+ return (share != SHARE_UNCHANGED);
+}
+
+static inline int change_supported(int share)
+{
+ return (share != SHARE_UNSUPPORTED);
+}
+
+/*
+ * Caller is responsible for protecting 'parent'
+ * Caller is responsible for making sure that the sum of sibling min_shares
+ * doesn't exceed parent's total min_shares.
+ */
+static inline void child_min_shares_changed(struct res_shares *parent,
+ int child_cur_min_shares,
+ int child_new_min_shares)
+{
+ if (is_share_quantitative(child_new_min_shares))
+ parent->unused_min_shares -= child_new_min_shares;
+ if (is_share_quantitative(child_cur_min_shares))
+ parent->unused_min_shares += child_cur_min_shares;
+}
+
+/*
+ * Set parent's cur_max_shares to the largest 'max_shares' of all
+ * of its children.
+ */
+static inline void set_cur_max_shares(struct resource_group *parent,
+ struct res_controller *ctlr)
+{
+ int max_shares = 0;
+ struct resource_group *child = NULL;
+ struct res_shares *child_shares, *parent_shares;
+
+ for_each_child(child, parent) {
+ child_shares = get_controller_shares(child, ctlr);
+ max_shares = max(max_shares, child_shares->max_shares);
+ }
+
+ parent_shares = get_controller_shares(parent, ctlr);
+ parent_shares->cur_max_shares = max_shares;
+}
+
+/*
+ * Return -EINVAL if the child's shares violate self-consistency or
+ * parent-imposed restrictions. Otherwise return 0.
+ *
+ * This involves checking shares between the child and its parent;
+ * the child and itself (userspace can't be trusted).
+ */
+static inline int are_shares_valid(struct res_shares *child,
+ struct res_shares *parent,
+ int current_usage,
+ int min_shares_increase)
+{
+ /*
+ * CHILD <-> PARENT validation
+ * Increases in child's min_shares or max_shares can't exceed
+ * limitations imposed by the parent resource group.
+ * Only validate this if we have a parent.
+ */
+ if (parent &&
+ ((is_share_quantitative(child->min_shares) &&
+ (min_shares_increase > parent->unused_min_shares)) ||
+ (is_share_quantitative(child->max_shares) &&
+ (child->max_shares > parent->child_shares_divisor))))
+ return -EINVAL;
+
+ /* CHILD validation: is min valid */
+ if (!is_share_valid(child->min_shares))
+ return -EINVAL;
+
+ /* CHILD validation: is max valid */
+ if (!is_share_valid(child->max_shares))
+ return -EINVAL;
+
+ /*
+ * CHILD validation: is divisor quantitative & current_usage
+ * is not more than the new divisor
+ */
+ if (!is_share_quantitative(child->child_shares_divisor) ||
+ (current_usage > child->child_shares_divisor))
+ return -EINVAL;
+
+ /*
+ * CHILD validation: is the new child_shares_divisor large
+ * enough to accomodate largest max_shares of any of my child
+ */
+ if (child->child_shares_divisor < child->cur_max_shares)
+ return -EINVAL;
+
+ /* CHILD validation: min <= max */
+ if (is_share_quantitative(child->min_shares) &&
+ is_share_quantitative(child->max_shares) &&
+ (child->min_shares > child->max_shares))
+ return -EINVAL;
+
+ return 0;
+}
+
+/*
+ * Set the resource shares of a child resource group given the new shares
+ * specified by userspace, the child's current shares, and the parent
+ * resource group's shares.
+ *
+ * Caller is responsible for holding group_lock of child and parent
+ * resource groups to protect the shares structures passed to this function.
+ */
+static int set_shares(const struct res_shares *new,
+ struct res_shares *child_shares,
+ struct res_shares *parent_shares)
+{
+ int rc, current_usage, min_shares_increase;
+ struct res_shares final_shares;
+
+ BUG_ON(!new || !child_shares);
+
+ final_shares = *child_shares;
+ if (did_share_change(new->child_shares_divisor) &&
+ change_supported(child_shares->child_shares_divisor))
+ final_shares.child_shares_divisor = new->child_shares_divisor;
+ if (did_share_change(new->min_shares) &&
+ change_supported(child_shares->min_shares))
+ final_shares.min_shares = new->min_shares;
+ if (did_share_change(new->max_shares) &&
+ change_supported(child_shares->max_shares))
+ final_shares.max_shares = new->max_shares;
+
+ current_usage = child_shares->child_shares_divisor -
+ child_shares->unused_min_shares;
+ min_shares_increase = final_shares.min_shares;
+ if (is_share_quantitative(child_shares->min_shares))
+ min_shares_increase -= child_shares->min_shares;
+
+ rc = are_shares_valid(&final_shares, parent_shares, current_usage,
+ min_shares_increase);
+ if (rc)
+ return rc; /* new shares would violate restrictions */
+
+ if (did_share_change(new->child_shares_divisor))
+ final_shares.unused_min_shares =
+ (final_shares.child_shares_divisor - current_usage);
+ *child_shares = final_shares;
+ return 0;
+}
+
+int set_controller_shares(struct resource_group *rgroup,
+ struct res_controller *ctlr,
+ const struct res_shares *new_shares)
+{
+ struct res_shares *shares, *parent_shares;
+ int prev_min, prev_max, rc;
+
+ if (!ctlr->shares_changed)
+ return -EINVAL;
+
+ shares = get_controller_shares(rgroup, ctlr);
+ if (!shares)
+ return -EINVAL;
+
+ prev_min = shares->min_shares;
+ prev_max = shares->max_shares;
+
+ container_lock(); /* XXX */
+ //spin_lock(&rgroup->group_lock);
+ parent_shares = get_controller_shares(rgroup->parent, ctlr);
+ rc = set_shares(new_shares, shares, parent_shares);
+
+ if (rc || is_res_group_root(rgroup))
+ goto done;
+
+ /* Notify parent about changes in my shares */
+ child_min_shares_changed(parent_shares, prev_min,
+ shares->min_shares);
+ if (prev_max != shares->max_shares)
+ set_cur_max_shares(rgroup->parent, ctlr);
+
+done:
+ container_unlock(); /* XXX */
+ if (!rc)
+ ctlr->shares_changed(shares);
+ return rc;
+}
--
^ permalink raw reply [flat|nested] 12+ messages in thread