Blame view
kernel/cpuset.c
77.9 KB
1da177e4c Linux-2.6.12-rc2 |
1 2 3 4 5 6 |
/* * kernel/cpuset.c * * Processor and Memory placement constraints for sets of tasks. * * Copyright (C) 2003 BULL SA. |
825a46af5 [PATCH] cpuset me... |
7 |
* Copyright (C) 2004-2006 Silicon Graphics, Inc. |
1da177e4c Linux-2.6.12-rc2 |
8 9 10 |
* * Portions derived from Patrick Mochel's sysfs code. * sysfs is Copyright (c) 2001-3 Patrick Mochel |
1da177e4c Linux-2.6.12-rc2 |
11 |
* |
825a46af5 [PATCH] cpuset me... |
12 |
* 2003-10-10 Written by Simon Derr. |
1da177e4c Linux-2.6.12-rc2 |
13 |
* 2003-10-22 Updates by Stephen Hemminger. |
825a46af5 [PATCH] cpuset me... |
14 |
* 2004 May-July Rework by Paul Jackson. |
1da177e4c Linux-2.6.12-rc2 |
15 16 17 18 19 |
* * 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. */ |
1da177e4c Linux-2.6.12-rc2 |
20 21 22 23 24 25 26 27 28 29 30 31 |
#include <linux/cpu.h> #include <linux/cpumask.h> #include <linux/cpuset.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> |
68860ec10 [PATCH] cpusets: ... |
32 |
#include <linux/mempolicy.h> |
1da177e4c Linux-2.6.12-rc2 |
33 34 35 36 37 38 |
#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> |
6b9c2603c [PATCH] cpuset: u... |
39 |
#include <linux/rcupdate.h> |
1da177e4c Linux-2.6.12-rc2 |
40 41 |
#include <linux/sched.h> #include <linux/seq_file.h> |
22fb52dd7 [PATCH] SELinux: ... |
42 |
#include <linux/security.h> |
1da177e4c Linux-2.6.12-rc2 |
43 44 45 46 47 48 49 50 51 52 53 |
#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> |
3d3f26a7b [PATCH] kernel/cp... |
54 |
#include <linux/mutex.h> |
1da177e4c Linux-2.6.12-rc2 |
55 |
|
c5b2aff89 [PATCH] cpuset: m... |
56 |
#define CPUSET_SUPER_MAGIC 0x27e0eb |
1da177e4c Linux-2.6.12-rc2 |
57 |
|
202f72d5d [PATCH] cpuset: n... |
58 59 60 61 62 |
/* * Tracks how many cpusets are currently defined in system. * When there is only one cpuset (the root cpuset) we can * short circuit some hooks. */ |
7edc59628 [PATCH] cpuset: m... |
63 |
int number_of_cpusets __read_mostly; |
202f72d5d [PATCH] cpuset: n... |
64 |
|
3e0d98b9f [PATCH] cpuset: m... |
65 66 67 68 69 70 71 72 |
/* See "Frequency meter" comments, below. */ struct fmeter { int cnt; /* unprocessed events count */ int val; /* most recent output value */ time_t time; /* clock (secs) when val computed */ spinlock_t lock; /* guards read or write of above */ }; |
1da177e4c Linux-2.6.12-rc2 |
73 74 75 76 |
struct cpuset { unsigned long flags; /* "unsigned long" so bitops work */ cpumask_t cpus_allowed; /* CPUs allowed to tasks in cpuset */ nodemask_t mems_allowed; /* Memory Nodes allowed to tasks */ |
053199edf [PATCH] cpusets: ... |
77 78 79 |
/* * Count is atomic so can incr (fork) or decr (exit) without a lock. */ |
1da177e4c Linux-2.6.12-rc2 |
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 |
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 cpuset *parent; /* my parent */ struct dentry *dentry; /* cpuset fs entry */ /* * Copy of global cpuset_mems_generation as of the most * recent time this cpuset changed its mems_allowed. */ |
3e0d98b9f [PATCH] cpuset: m... |
96 97 98 |
int mems_generation; struct fmeter fmeter; /* memory_pressure filter */ |
1da177e4c Linux-2.6.12-rc2 |
99 100 101 102 103 104 |
}; /* bits in struct cpuset flags field */ typedef enum { CS_CPU_EXCLUSIVE, CS_MEM_EXCLUSIVE, |
45b07ef31 [PATCH] cpusets: ... |
105 |
CS_MEMORY_MIGRATE, |
1da177e4c Linux-2.6.12-rc2 |
106 |
CS_REMOVED, |
825a46af5 [PATCH] cpuset me... |
107 108 109 |
CS_NOTIFY_ON_RELEASE, CS_SPREAD_PAGE, CS_SPREAD_SLAB, |
1da177e4c Linux-2.6.12-rc2 |
110 111 112 113 114 |
} cpuset_flagbits_t; /* convenient tests for these bits */ static inline int is_cpu_exclusive(const struct cpuset *cs) { |
7b5b9ef0e [PATCH] cpuset cl... |
115 |
return test_bit(CS_CPU_EXCLUSIVE, &cs->flags); |
1da177e4c Linux-2.6.12-rc2 |
116 117 118 119 |
} static inline int is_mem_exclusive(const struct cpuset *cs) { |
7b5b9ef0e [PATCH] cpuset cl... |
120 |
return test_bit(CS_MEM_EXCLUSIVE, &cs->flags); |
1da177e4c Linux-2.6.12-rc2 |
121 122 123 124 |
} static inline int is_removed(const struct cpuset *cs) { |
7b5b9ef0e [PATCH] cpuset cl... |
125 |
return test_bit(CS_REMOVED, &cs->flags); |
1da177e4c Linux-2.6.12-rc2 |
126 127 128 129 |
} static inline int notify_on_release(const struct cpuset *cs) { |
7b5b9ef0e [PATCH] cpuset cl... |
130 |
return test_bit(CS_NOTIFY_ON_RELEASE, &cs->flags); |
1da177e4c Linux-2.6.12-rc2 |
131 |
} |
45b07ef31 [PATCH] cpusets: ... |
132 133 |
static inline int is_memory_migrate(const struct cpuset *cs) { |
7b5b9ef0e [PATCH] cpuset cl... |
134 |
return test_bit(CS_MEMORY_MIGRATE, &cs->flags); |
45b07ef31 [PATCH] cpusets: ... |
135 |
} |
825a46af5 [PATCH] cpuset me... |
136 137 138 139 140 141 142 143 144 |
static inline int is_spread_page(const struct cpuset *cs) { return test_bit(CS_SPREAD_PAGE, &cs->flags); } static inline int is_spread_slab(const struct cpuset *cs) { return test_bit(CS_SPREAD_SLAB, &cs->flags); } |
1da177e4c Linux-2.6.12-rc2 |
145 |
/* |
151a44202 [PATCH] cpuset: d... |
146 |
* Increment this integer everytime any cpuset changes its |
1da177e4c Linux-2.6.12-rc2 |
147 148 149 150 151 152 153 154 155 156 157 158 159 |
* mems_allowed value. Users of cpusets can track this generation * number, and avoid having to lock and reload mems_allowed unless * the cpuset they're using changes generation. * * A single, global generation is needed because attach_task() could * reattach a task to a different cpuset, which must not have its * generation numbers aliased with those of that tasks previous cpuset. * * Generations are needed for mems_allowed because one task cannot * modify anothers memory placement. So we must enable every task, * on every visit to __alloc_pages(), to efficiently check whether * its current->cpuset->mems_allowed has changed, requiring an update * of its current->mems_allowed. |
151a44202 [PATCH] cpuset: d... |
160 161 162 |
* * Since cpuset_mems_generation is guarded by manage_mutex, * there is no need to mark it atomic. |
1da177e4c Linux-2.6.12-rc2 |
163 |
*/ |
151a44202 [PATCH] cpuset: d... |
164 |
static int cpuset_mems_generation; |
1da177e4c Linux-2.6.12-rc2 |
165 166 167 168 169 170 171 172 |
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), |
1da177e4c Linux-2.6.12-rc2 |
173 174 175 |
}; static struct vfsmount *cpuset_mount; |
3e0d98b9f [PATCH] cpuset: m... |
176 |
static struct super_block *cpuset_sb; |
1da177e4c Linux-2.6.12-rc2 |
177 178 |
/* |
3d3f26a7b [PATCH] kernel/cp... |
179 180 |
* We have two global cpuset mutexes below. They can nest. * It is ok to first take manage_mutex, then nest callback_mutex. We also |
053199edf [PATCH] cpusets: ... |
181 182 183 |
* require taking task_lock() when dereferencing a tasks cpuset pointer. * See "The task_lock() exception", at the end of this comment. * |
3d3f26a7b [PATCH] kernel/cp... |
184 185 186 |
* 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 |
053199edf [PATCH] cpusets: ... |
187 188 |
* and be able to modify cpusets. It can perform various checks on * the cpuset structure first, knowing nothing will change. It can |
3d3f26a7b [PATCH] kernel/cp... |
189 |
* also allocate memory while just holding manage_mutex. While it is |
053199edf [PATCH] cpusets: ... |
190 |
* performing these checks, various callback routines can briefly |
3d3f26a7b [PATCH] kernel/cp... |
191 192 |
* acquire callback_mutex to query cpusets. Once it is ready to make * the changes, it takes callback_mutex, blocking everyone else. |
053199edf [PATCH] cpusets: ... |
193 194 |
* * Calls to the kernel memory allocator can not be made while holding |
3d3f26a7b [PATCH] kernel/cp... |
195 |
* callback_mutex, as that would risk double tripping on callback_mutex |
053199edf [PATCH] cpusets: ... |
196 197 198 |
* from one of the callbacks into the cpuset code from within * __alloc_pages(). * |
3d3f26a7b [PATCH] kernel/cp... |
199 |
* If a task is only holding callback_mutex, then it has read-only |
053199edf [PATCH] cpusets: ... |
200 201 202 203 204 205 |
* 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. |
3d3f26a7b [PATCH] kernel/cp... |
206 |
* So in general, code holding manage_mutex or callback_mutex can't rely |
053199edf [PATCH] cpusets: ... |
207 |
* on the count field not changing. However, if the count goes to |
3d3f26a7b [PATCH] kernel/cp... |
208 |
* zero, then only attach_task(), which holds both mutexes, can |
053199edf [PATCH] cpusets: ... |
209 210 211 |
* 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). |
3d3f26a7b [PATCH] kernel/cp... |
212 |
* So code holding manage_mutex or callback_mutex can safely assume that |
053199edf [PATCH] cpusets: ... |
213 |
* if the count is zero, it will stay zero. Similarly, if a task |
3d3f26a7b [PATCH] kernel/cp... |
214 |
* holds manage_mutex or callback_mutex on a cpuset with zero count, it |
053199edf [PATCH] cpusets: ... |
215 |
* knows that the cpuset won't be removed, as cpuset_rmdir() needs |
3d3f26a7b [PATCH] kernel/cp... |
216 |
* both of those mutexes. |
053199edf [PATCH] cpusets: ... |
217 218 |
* * The cpuset_common_file_write handler for operations that modify |
3d3f26a7b [PATCH] kernel/cp... |
219 |
* the cpuset hierarchy holds manage_mutex across the entire operation, |
053199edf [PATCH] cpusets: ... |
220 221 |
* single threading all such cpuset modifications across the system. * |
3d3f26a7b [PATCH] kernel/cp... |
222 |
* The cpuset_common_file_read() handlers only hold callback_mutex across |
053199edf [PATCH] cpusets: ... |
223 224 225 226 |
* 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 |
3d3f26a7b [PATCH] kernel/cp... |
227 |
* (usually) take either mutex. These are the two most performance |
053199edf [PATCH] cpusets: ... |
228 |
* critical pieces of code here. The exception occurs on cpuset_exit(), |
3d3f26a7b [PATCH] kernel/cp... |
229 |
* when a task in a notify_on_release cpuset exits. Then manage_mutex |
2efe86b80 [PATCH] cpuset ex... |
230 |
* is taken, and if the cpuset count is zero, a usermode call made |
1da177e4c Linux-2.6.12-rc2 |
231 232 233 |
* to /sbin/cpuset_release_agent with the name of the cpuset (path * relative to the root of cpuset file system) as the argument. * |
053199edf [PATCH] cpusets: ... |
234 235 236 |
* 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 |
f400e198b [PATCH] pidspace:... |
237 |
* least one task in the system (init), therefore, top_cpuset |
053199edf [PATCH] cpusets: ... |
238 239 240 241 242 243 244 245 246 |
* 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 |
3d3f26a7b [PATCH] kernel/cp... |
247 |
* so using both mutexes, however there are several performance |
053199edf [PATCH] cpusets: ... |
248 |
* critical places that need to reference task->cpuset without the |
3d3f26a7b [PATCH] kernel/cp... |
249 |
* expense of grabbing a system global mutex. Therefore except as |
053199edf [PATCH] cpusets: ... |
250 251 252 253 |
* 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. |
6b9c2603c [PATCH] cpuset: u... |
254 255 256 257 258 |
* * 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(). |
1da177e4c Linux-2.6.12-rc2 |
259 |
*/ |
3d3f26a7b [PATCH] kernel/cp... |
260 261 |
static DEFINE_MUTEX(manage_mutex); static DEFINE_MUTEX(callback_mutex); |
4247bdc60 [PATCH] cpuset se... |
262 263 |
/* |
1da177e4c Linux-2.6.12-rc2 |
264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 |
* 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; |
1da177e4c Linux-2.6.12-rc2 |
285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 |
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) { |
5f45f1a78 [PATCH] remove du... |
309 |
struct dentry *d = lookup_one_len(name, parent, strlen(name)); |
1da177e4c Linux-2.6.12-rc2 |
310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 |
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) { |
5160ee6fc [PATCH] shrink de... |
334 |
struct dentry *d = list_entry(node, struct dentry, d_u.d_child); |
1da177e4c Linux-2.6.12-rc2 |
335 336 337 338 339 340 341 342 343 344 345 |
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; } |
5160ee6fc [PATCH] shrink de... |
346 |
list_del_init(&dentry->d_u.d_child); |
1da177e4c Linux-2.6.12-rc2 |
347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 |
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) */ |
d8c76e6f4 [PATCH] r/o bind ... |
373 |
inc_nlink(inode); |
1da177e4c Linux-2.6.12-rc2 |
374 375 376 377 378 379 380 381 382 383 384 385 |
} else { return -ENOMEM; } root = d_alloc_root(inode); if (!root) { iput(inode); return -ENOMEM; } sb->s_root = root; return 0; } |
454e2398b [PATCH] VFS: Perm... |
386 387 388 |
static int cpuset_get_sb(struct file_system_type *fs_type, int flags, const char *unused_dev_name, void *data, struct vfsmount *mnt) |
1da177e4c Linux-2.6.12-rc2 |
389 |
{ |
454e2398b [PATCH] VFS: Perm... |
390 |
return get_sb_single(fs_type, flags, data, cpuset_fill_super, mnt); |
1da177e4c Linux-2.6.12-rc2 |
391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 |
} 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: |
a7a005fd1 [PATCH] struct pa... |
408 409 |
* - the cpuset to use in file->f_path.dentry->d_parent->d_fsdata * - the 'cftype' of the file is file->f_path.dentry->d_fsdata |
1da177e4c Linux-2.6.12-rc2 |
410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 |
*/ 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; } /* |
3d3f26a7b [PATCH] kernel/cp... |
434 |
* Call with manage_mutex held. Writes path of cpuset into buf. |
1da177e4c Linux-2.6.12-rc2 |
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 |
* 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. * |
3077a260e [PATCH] cpuset re... |
478 479 480 481 482 483 484 485 |
* 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. * |
3d3f26a7b [PATCH] kernel/cp... |
486 |
* When we had only one cpuset mutex, we had to call this |
053199edf [PATCH] cpusets: ... |
487 488 |
* without holding it, to avoid deadlock when call_usermodehelper() * allocated memory. With two locks, we could now call this while |
3d3f26a7b [PATCH] kernel/cp... |
489 490 |
* holding manage_mutex, but we still don't, so as to minimize * the time manage_mutex is held. |
1da177e4c Linux-2.6.12-rc2 |
491 |
*/ |
3077a260e [PATCH] cpuset re... |
492 |
static void cpuset_release_agent(const char *pathbuf) |
1da177e4c Linux-2.6.12-rc2 |
493 494 495 |
{ char *argv[3], *envp[3]; int i; |
3077a260e [PATCH] cpuset re... |
496 497 |
if (!pathbuf) return; |
1da177e4c Linux-2.6.12-rc2 |
498 499 |
i = 0; argv[i++] = "/sbin/cpuset_release_agent"; |
3077a260e [PATCH] cpuset re... |
500 |
argv[i++] = (char *)pathbuf; |
1da177e4c Linux-2.6.12-rc2 |
501 502 503 504 505 506 507 |
argv[i] = NULL; i = 0; /* minimal command environment */ envp[i++] = "HOME=/"; envp[i++] = "PATH=/sbin:/bin:/usr/sbin:/usr/bin"; envp[i] = NULL; |
3077a260e [PATCH] cpuset re... |
508 509 |
call_usermodehelper(argv[0], argv, envp, 0); kfree(pathbuf); |
1da177e4c Linux-2.6.12-rc2 |
510 511 512 513 514 515 |
} /* * 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 |
3077a260e [PATCH] cpuset re... |
516 517 |
* 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 |
3d3f26a7b [PATCH] kernel/cp... |
518 519 |
* cpuset_release_agent() with it later on, once manage_mutex is dropped. * Call here with manage_mutex held. |
3077a260e [PATCH] cpuset re... |
520 521 522 523 524 |
* * 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 |
3d3f26a7b [PATCH] kernel/cp... |
525 526 |
* calling check_for_release() with manage_mutex held and the address * of the pathbuf pointer, then dropping manage_mutex, then calling |
3077a260e [PATCH] cpuset re... |
527 |
* cpuset_release_agent() with pathbuf, as set by check_for_release(). |
1da177e4c Linux-2.6.12-rc2 |
528 |
*/ |
3077a260e [PATCH] cpuset re... |
529 |
static void check_for_release(struct cpuset *cs, char **ppathbuf) |
1da177e4c Linux-2.6.12-rc2 |
530 531 532 533 534 535 536 537 538 |
{ 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) |
3077a260e [PATCH] cpuset re... |
539 540 541 |
kfree(buf); else *ppathbuf = buf; |
1da177e4c Linux-2.6.12-rc2 |
542 543 544 545 546 547 548 549 550 551 552 553 554 555 |
} } /* * Return in *pmask the portion of a cpusets's cpus_allowed that * are online. If none are online, walk up the cpuset hierarchy * until we find one that does have some online cpus. If we get * all the way to the top and still haven't found any online cpus, * return cpu_online_map. Or if passed a NULL cs from an exit'ing * task, return cpu_online_map. * * One way or another, we guarantee to return some non-empty subset * of cpu_online_map. * |
3d3f26a7b [PATCH] kernel/cp... |
556 |
* Call with callback_mutex held. |
1da177e4c Linux-2.6.12-rc2 |
557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 |
*/ static void guarantee_online_cpus(const struct cpuset *cs, cpumask_t *pmask) { while (cs && !cpus_intersects(cs->cpus_allowed, cpu_online_map)) cs = cs->parent; if (cs) cpus_and(*pmask, cs->cpus_allowed, cpu_online_map); else *pmask = cpu_online_map; BUG_ON(!cpus_intersects(*pmask, cpu_online_map)); } /* * Return in *pmask the portion of a cpusets's mems_allowed that * are online. If none are online, walk up the cpuset hierarchy * until we find one that does have some online mems. If we get * all the way to the top and still haven't found any online mems, * return node_online_map. * * One way or another, we guarantee to return some non-empty subset * of node_online_map. * |
3d3f26a7b [PATCH] kernel/cp... |
580 |
* Call with callback_mutex held. |
1da177e4c Linux-2.6.12-rc2 |
581 582 583 584 585 586 587 588 589 590 591 592 |
*/ static void guarantee_online_mems(const struct cpuset *cs, nodemask_t *pmask) { while (cs && !nodes_intersects(cs->mems_allowed, node_online_map)) cs = cs->parent; if (cs) nodes_and(*pmask, cs->mems_allowed, node_online_map); else *pmask = node_online_map; BUG_ON(!nodes_intersects(*pmask, node_online_map)); } |
cf2a473c4 [PATCH] cpuset: c... |
593 594 595 596 597 598 |
/** * cpuset_update_task_memory_state - update task memory placement * * If the current tasks cpusets mems_allowed changed behind our * backs, update current->mems_allowed, mems_generation and task NUMA * mempolicy to the new value. |
053199edf [PATCH] cpusets: ... |
599 |
* |
cf2a473c4 [PATCH] cpuset: c... |
600 601 602 603 |
* Task mempolicy is updated by rebinding it relative to the * current->cpuset if a task has its memory placement changed. * Do not call this routine if in_interrupt(). * |
4a01c8d5b [PATCH] cpuset: t... |
604 605 606 607 |
* Call without callback_mutex or task_lock() held. May be * called with or without manage_mutex held. Thanks in part to * 'the_top_cpuset_hack', the tasks cpuset pointer will never * be NULL. This routine also might acquire callback_mutex and |
cf2a473c4 [PATCH] cpuset: c... |
608 |
* current->mm->mmap_sem during call. |
053199edf [PATCH] cpusets: ... |
609 |
* |
6b9c2603c [PATCH] cpuset: u... |
610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 |
* Reading current->cpuset->mems_generation doesn't need task_lock * to guard the current->cpuset derefence, because it is guarded * from concurrent freeing of current->cpuset by attach_task(), * using RCU. * * The rcu_dereference() is technically probably not needed, * as I don't actually mind if I see a new cpuset pointer but * an old value of mems_generation. However this really only * matters on alpha systems using cpusets heavily. If I dropped * that rcu_dereference(), it would save them a memory barrier. * For all other arch's, rcu_dereference is a no-op anyway, and for * alpha systems not using cpusets, another planned optimization, * avoiding the rcu critical section for tasks in the root cpuset * which is statically allocated, so can't vanish, will make this * irrelevant. Better to use RCU as intended, than to engage in * some cute trick to save a memory barrier that is impossible to * test, for alpha systems using cpusets heavily, which might not * even exist. |
053199edf [PATCH] cpusets: ... |
628 629 630 631 632 |
* * This routine is needed to update the per-task mems_allowed data, * within the tasks context, when it is trying to allocate memory * (in various mm/mempolicy.c routines) and notices that some other * task has been modifying its cpuset. |
1da177e4c Linux-2.6.12-rc2 |
633 |
*/ |
fe85a998c [PATCH] cpuset: f... |
634 |
void cpuset_update_task_memory_state(void) |
1da177e4c Linux-2.6.12-rc2 |
635 |
{ |
053199edf [PATCH] cpusets: ... |
636 |
int my_cpusets_mem_gen; |
cf2a473c4 [PATCH] cpuset: c... |
637 |
struct task_struct *tsk = current; |
6b9c2603c [PATCH] cpuset: u... |
638 |
struct cpuset *cs; |
053199edf [PATCH] cpusets: ... |
639 |
|
03a285f58 [PATCH] cpuset: s... |
640 641 642 643 644 645 646 647 648 |
if (tsk->cpuset == &top_cpuset) { /* Don't need rcu for top_cpuset. It's never freed. */ my_cpusets_mem_gen = top_cpuset.mems_generation; } else { rcu_read_lock(); cs = rcu_dereference(tsk->cpuset); my_cpusets_mem_gen = cs->mems_generation; rcu_read_unlock(); } |
1da177e4c Linux-2.6.12-rc2 |
649 |
|
cf2a473c4 [PATCH] cpuset: c... |
650 |
if (my_cpusets_mem_gen != tsk->cpuset_mems_generation) { |
3d3f26a7b [PATCH] kernel/cp... |
651 |
mutex_lock(&callback_mutex); |
cf2a473c4 [PATCH] cpuset: c... |
652 653 |
task_lock(tsk); cs = tsk->cpuset; /* Maybe changed when task not locked */ |
cf2a473c4 [PATCH] cpuset: c... |
654 655 |
guarantee_online_mems(cs, &tsk->mems_allowed); tsk->cpuset_mems_generation = cs->mems_generation; |
825a46af5 [PATCH] cpuset me... |
656 657 658 659 660 661 662 663 |
if (is_spread_page(cs)) tsk->flags |= PF_SPREAD_PAGE; else tsk->flags &= ~PF_SPREAD_PAGE; if (is_spread_slab(cs)) tsk->flags |= PF_SPREAD_SLAB; else tsk->flags &= ~PF_SPREAD_SLAB; |
cf2a473c4 [PATCH] cpuset: c... |
664 |
task_unlock(tsk); |
3d3f26a7b [PATCH] kernel/cp... |
665 |
mutex_unlock(&callback_mutex); |
74cb21553 [PATCH] cpuset: n... |
666 |
mpol_rebind_task(tsk, &tsk->mems_allowed); |
1da177e4c Linux-2.6.12-rc2 |
667 668 669 670 671 672 673 674 |
} } /* * is_cpuset_subset(p, q) - Is cpuset p a subset of cpuset q? * * One cpuset is a subset of another if all its allowed CPUs and * Memory Nodes are a subset of the other, and its exclusive flags |
3d3f26a7b [PATCH] kernel/cp... |
675 |
* are only set if the other's are set. Call holding manage_mutex. |
1da177e4c Linux-2.6.12-rc2 |
676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 |
*/ static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q) { return cpus_subset(p->cpus_allowed, q->cpus_allowed) && nodes_subset(p->mems_allowed, q->mems_allowed) && is_cpu_exclusive(p) <= is_cpu_exclusive(q) && is_mem_exclusive(p) <= is_mem_exclusive(q); } /* * validate_change() - Used to validate that any proposed cpuset change * follows the structural rules for cpusets. * * If we replaced the flag and mask values of the current cpuset * (cur) with those values in the trial cpuset (trial), would * our various subset and exclusive rules still be valid? Presumes |
3d3f26a7b [PATCH] kernel/cp... |
693 |
* manage_mutex held. |
1da177e4c Linux-2.6.12-rc2 |
694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 |
* * 'cur' is the address of an actual, in-use cpuset. Operations * such as list traversal that depend on the actual address of the * cpuset in the list must use cur below, not trial. * * 'trial' is the address of bulk structure copy of cur, with * perhaps one or more of the fields cpus_allowed, mems_allowed, * or flags changed to new, trial values. * * Return 0 if valid, -errno if not. */ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) { struct cpuset *c, *par; /* Each of our child cpusets must be a subset of us */ list_for_each_entry(c, &cur->children, sibling) { if (!is_cpuset_subset(c, trial)) return -EBUSY; } /* Remaining checks don't apply to root cpuset */ |
696040670 [PATCH] cpuset: m... |
717 |
if (cur == &top_cpuset) |
1da177e4c Linux-2.6.12-rc2 |
718 |
return 0; |
696040670 [PATCH] cpuset: m... |
719 |
par = cur->parent; |
1da177e4c Linux-2.6.12-rc2 |
720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 |
/* We must be a subset of our parent cpuset */ if (!is_cpuset_subset(trial, par)) return -EACCES; /* If either I or some sibling (!= me) is exclusive, we can't overlap */ list_for_each_entry(c, &par->children, sibling) { if ((is_cpu_exclusive(trial) || is_cpu_exclusive(c)) && c != cur && cpus_intersects(trial->cpus_allowed, c->cpus_allowed)) return -EINVAL; if ((is_mem_exclusive(trial) || is_mem_exclusive(c)) && c != cur && nodes_intersects(trial->mems_allowed, c->mems_allowed)) return -EINVAL; } return 0; } |
85d7b9498 [PATCH] Dynamic s... |
738 739 740 741 742 743 744 745 |
/* * For a given cpuset cur, partition the system as follows * a. All cpus in the parent cpuset's cpus_allowed that are not part of any * exclusive child cpusets * b. All cpus in the current cpuset's cpus_allowed that are not part of any * exclusive child cpusets * Build these two partitions by calling partition_sched_domains * |
3d3f26a7b [PATCH] kernel/cp... |
746 |
* Call with manage_mutex held. May nest a call to the |
85d7b9498 [PATCH] Dynamic s... |
747 |
* lock_cpu_hotplug()/unlock_cpu_hotplug() pair. |
abb5a5cc6 [PATCH] Cpuset: f... |
748 749 |
* Must not be called holding callback_mutex, because we must * not call lock_cpu_hotplug() while holding callback_mutex. |
85d7b9498 [PATCH] Dynamic s... |
750 |
*/ |
212d6d223 [PATCH] completel... |
751 |
|
85d7b9498 [PATCH] Dynamic s... |
752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 |
static void update_cpu_domains(struct cpuset *cur) { struct cpuset *c, *par = cur->parent; cpumask_t pspan, cspan; if (par == NULL || cpus_empty(cur->cpus_allowed)) return; /* * Get all cpus from parent's cpus_allowed not part of exclusive * children */ pspan = par->cpus_allowed; list_for_each_entry(c, &par->children, sibling) { if (is_cpu_exclusive(c)) cpus_andnot(pspan, pspan, c->cpus_allowed); } |
abb5a5cc6 [PATCH] Cpuset: f... |
769 |
if (!is_cpu_exclusive(cur)) { |
85d7b9498 [PATCH] Dynamic s... |
770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 |
cpus_or(pspan, pspan, cur->cpus_allowed); if (cpus_equal(pspan, cur->cpus_allowed)) return; cspan = CPU_MASK_NONE; } else { if (cpus_empty(pspan)) return; cspan = cur->cpus_allowed; /* * Get all cpus from current cpuset's cpus_allowed not part * of exclusive children */ list_for_each_entry(c, &cur->children, sibling) { if (is_cpu_exclusive(c)) cpus_andnot(cspan, cspan, c->cpus_allowed); } } lock_cpu_hotplug(); partition_sched_domains(&pspan, &cspan); unlock_cpu_hotplug(); } |
053199edf [PATCH] cpusets: ... |
792 |
/* |
3d3f26a7b [PATCH] kernel/cp... |
793 |
* Call with manage_mutex held. May take callback_mutex during call. |
053199edf [PATCH] cpusets: ... |
794 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
795 796 797 |
static int update_cpumask(struct cpuset *cs, char *buf) { struct cpuset trialcs; |
85d7b9498 [PATCH] Dynamic s... |
798 |
int retval, cpus_unchanged; |
1da177e4c Linux-2.6.12-rc2 |
799 |
|
4c4d50f7b [PATCH] cpuset: t... |
800 801 802 |
/* top_cpuset.cpus_allowed tracks cpu_online_map; it's read-only */ if (cs == &top_cpuset) return -EACCES; |
1da177e4c Linux-2.6.12-rc2 |
803 804 805 806 807 808 809 810 |
trialcs = *cs; retval = cpulist_parse(buf, trialcs.cpus_allowed); if (retval < 0) return retval; cpus_and(trialcs.cpus_allowed, trialcs.cpus_allowed, cpu_online_map); if (cpus_empty(trialcs.cpus_allowed)) return -ENOSPC; retval = validate_change(cs, &trialcs); |
85d7b9498 [PATCH] Dynamic s... |
811 812 813 |
if (retval < 0) return retval; cpus_unchanged = cpus_equal(cs->cpus_allowed, trialcs.cpus_allowed); |
3d3f26a7b [PATCH] kernel/cp... |
814 |
mutex_lock(&callback_mutex); |
85d7b9498 [PATCH] Dynamic s... |
815 |
cs->cpus_allowed = trialcs.cpus_allowed; |
3d3f26a7b [PATCH] kernel/cp... |
816 |
mutex_unlock(&callback_mutex); |
85d7b9498 [PATCH] Dynamic s... |
817 818 819 |
if (is_cpu_exclusive(cs) && !cpus_unchanged) update_cpu_domains(cs); return 0; |
1da177e4c Linux-2.6.12-rc2 |
820 |
} |
053199edf [PATCH] cpusets: ... |
821 |
/* |
e4e364e86 [PATCH] cpuset: m... |
822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 |
* cpuset_migrate_mm * * Migrate memory region from one set of nodes to another. * * Temporarilly set tasks mems_allowed to target nodes of migration, * so that the migration code can allocate pages on these nodes. * * Call holding manage_mutex, so our current->cpuset won't change * during this call, as manage_mutex holds off any attach_task() * calls. Therefore we don't need to take task_lock around the * call to guarantee_online_mems(), as we know no one is changing * our tasks cpuset. * * Hold callback_mutex around the two modifications of our tasks * mems_allowed to synchronize with cpuset_mems_allowed(). * * While the mm_struct we are migrating is typically from some * other task, the task_struct mems_allowed that we are hacking * is for our current task, which must allocate new pages for that * migrating memory region. * * We call cpuset_update_task_memory_state() before hacking * our tasks mems_allowed, so that we are assured of being in * sync with our tasks cpuset, and in particular, callbacks to * cpuset_update_task_memory_state() from nested page allocations * won't see any mismatch of our cpuset and task mems_generation * values, so won't overwrite our hacked tasks mems_allowed * nodemask. */ static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from, const nodemask_t *to) { struct task_struct *tsk = current; cpuset_update_task_memory_state(); mutex_lock(&callback_mutex); tsk->mems_allowed = *to; mutex_unlock(&callback_mutex); 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); } /* |
4225399a6 [PATCH] cpuset: r... |
871 872 873 |
* Handle user request to change the 'mems' memory placement * of a cpuset. Needs to validate the request, update the * cpusets mems_allowed and mems_generation, and for each |
04c19fa6f [PATCH] cpuset: m... |
874 875 876 |
* task in the cpuset, rebind any vma mempolicies and if * the cpuset is marked 'memory_migrate', migrate the tasks * pages to the new memory. |
4225399a6 [PATCH] cpuset: r... |
877 |
* |
3d3f26a7b [PATCH] kernel/cp... |
878 |
* Call with manage_mutex held. May take callback_mutex during call. |
4225399a6 [PATCH] cpuset: r... |
879 880 881 |
* Will take tasklist_lock, scan tasklist for tasks in cpuset cs, * lock each such tasks mm->mmap_sem, scan its vma's and rebind * their mempolicies to the cpusets new mems_allowed. |
053199edf [PATCH] cpusets: ... |
882 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
883 884 885 |
static int update_nodemask(struct cpuset *cs, char *buf) { struct cpuset trialcs; |
04c19fa6f [PATCH] cpuset: m... |
886 |
nodemask_t oldmem; |
4225399a6 [PATCH] cpuset: r... |
887 888 889 |
struct task_struct *g, *p; struct mm_struct **mmarray; int i, n, ntasks; |
04c19fa6f [PATCH] cpuset: m... |
890 |
int migrate; |
4225399a6 [PATCH] cpuset: r... |
891 |
int fudge; |
1da177e4c Linux-2.6.12-rc2 |
892 |
int retval; |
38837fc75 [PATCH] cpuset: t... |
893 894 895 |
/* top_cpuset.mems_allowed tracks node_online_map; it's read-only */ if (cs == &top_cpuset) return -EACCES; |
1da177e4c Linux-2.6.12-rc2 |
896 897 898 |
trialcs = *cs; retval = nodelist_parse(buf, trialcs.mems_allowed); if (retval < 0) |
59dac16fb [PATCH] cpuset: u... |
899 |
goto done; |
1da177e4c Linux-2.6.12-rc2 |
900 |
nodes_and(trialcs.mems_allowed, trialcs.mems_allowed, node_online_map); |
04c19fa6f [PATCH] cpuset: m... |
901 902 903 904 905 |
oldmem = cs->mems_allowed; if (nodes_equal(oldmem, trialcs.mems_allowed)) { retval = 0; /* Too easy - nothing to do */ goto done; } |
59dac16fb [PATCH] cpuset: u... |
906 907 908 |
if (nodes_empty(trialcs.mems_allowed)) { retval = -ENOSPC; goto done; |
1da177e4c Linux-2.6.12-rc2 |
909 |
} |
59dac16fb [PATCH] cpuset: u... |
910 911 912 |
retval = validate_change(cs, &trialcs); if (retval < 0) goto done; |
3d3f26a7b [PATCH] kernel/cp... |
913 |
mutex_lock(&callback_mutex); |
59dac16fb [PATCH] cpuset: u... |
914 |
cs->mems_allowed = trialcs.mems_allowed; |
151a44202 [PATCH] cpuset: d... |
915 |
cs->mems_generation = cpuset_mems_generation++; |
3d3f26a7b [PATCH] kernel/cp... |
916 |
mutex_unlock(&callback_mutex); |
59dac16fb [PATCH] cpuset: u... |
917 |
|
4225399a6 [PATCH] cpuset: r... |
918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 |
set_cpuset_being_rebound(cs); /* causes mpol_copy() rebind */ fudge = 10; /* spare mmarray[] slots */ fudge += cpus_weight(cs->cpus_allowed); /* imagine one fork-bomb/cpu */ retval = -ENOMEM; /* * Allocate mmarray[] to hold mm reference for each task * in cpuset cs. Can't kmalloc GFP_KERNEL while holding * tasklist_lock. We could use GFP_ATOMIC, but with a * few more lines of code, we can retry until we get a big * enough mmarray[] w/o using GFP_ATOMIC. */ while (1) { ntasks = atomic_read(&cs->count); /* guess */ ntasks += fudge; mmarray = kmalloc(ntasks * sizeof(*mmarray), GFP_KERNEL); if (!mmarray) goto done; write_lock_irq(&tasklist_lock); /* block fork */ if (atomic_read(&cs->count) <= ntasks) break; /* got enough */ write_unlock_irq(&tasklist_lock); /* try again */ kfree(mmarray); } n = 0; /* Load up mmarray[] with mm reference for each task in cpuset. */ do_each_thread(g, p) { struct mm_struct *mm; if (n >= ntasks) { printk(KERN_WARNING "Cpuset mempolicy rebind incomplete. "); continue; } if (p->cpuset != cs) continue; mm = get_task_mm(p); if (!mm) continue; mmarray[n++] = mm; } while_each_thread(g, p); write_unlock_irq(&tasklist_lock); /* * Now that we've dropped the tasklist spinlock, we can * rebind the vma mempolicies of each mm in mmarray[] to their * new cpuset, and release that mm. The mpol_rebind_mm() * call takes mmap_sem, which we couldn't take while holding * tasklist_lock. Forks can happen again now - the mpol_copy() * cpuset_being_rebound check will catch such forks, and rebind * their vma mempolicies too. Because we still hold the global |
3d3f26a7b [PATCH] kernel/cp... |
973 |
* cpuset manage_mutex, we know that no other rebind effort will |
4225399a6 [PATCH] cpuset: r... |
974 975 |
* be contending for the global variable cpuset_being_rebound. * It's ok if we rebind the same mm twice; mpol_rebind_mm() |
04c19fa6f [PATCH] cpuset: m... |
976 |
* is idempotent. Also migrate pages in each mm to new nodes. |
4225399a6 [PATCH] cpuset: r... |
977 |
*/ |
04c19fa6f [PATCH] cpuset: m... |
978 |
migrate = is_memory_migrate(cs); |
4225399a6 [PATCH] cpuset: r... |
979 980 981 982 |
for (i = 0; i < n; i++) { struct mm_struct *mm = mmarray[i]; mpol_rebind_mm(mm, &cs->mems_allowed); |
e4e364e86 [PATCH] cpuset: m... |
983 984 |
if (migrate) cpuset_migrate_mm(mm, &oldmem, &cs->mems_allowed); |
4225399a6 [PATCH] cpuset: r... |
985 986 987 988 989 990 991 |
mmput(mm); } /* We're done rebinding vma's to this cpusets new mems_allowed. */ kfree(mmarray); set_cpuset_being_rebound(NULL); retval = 0; |
59dac16fb [PATCH] cpuset: u... |
992 |
done: |
1da177e4c Linux-2.6.12-rc2 |
993 994 995 996 |
return retval; } /* |
3d3f26a7b [PATCH] kernel/cp... |
997 |
* Call with manage_mutex held. |
3e0d98b9f [PATCH] cpuset: m... |
998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 |
*/ static int update_memory_pressure_enabled(struct cpuset *cs, char *buf) { if (simple_strtoul(buf, NULL, 10) != 0) cpuset_memory_pressure_enabled = 1; else cpuset_memory_pressure_enabled = 0; return 0; } /* |
1da177e4c Linux-2.6.12-rc2 |
1010 1011 |
* update_flag - read a 0 or a 1 in a file and update associated flag * bit: the bit to update (CS_CPU_EXCLUSIVE, CS_MEM_EXCLUSIVE, |
825a46af5 [PATCH] cpuset me... |
1012 1013 |
* CS_NOTIFY_ON_RELEASE, CS_MEMORY_MIGRATE, * CS_SPREAD_PAGE, CS_SPREAD_SLAB) |
1da177e4c Linux-2.6.12-rc2 |
1014 1015 |
* cs: the cpuset to update * buf: the buffer where we read the 0 or 1 |
053199edf [PATCH] cpusets: ... |
1016 |
* |
3d3f26a7b [PATCH] kernel/cp... |
1017 |
* Call with manage_mutex held. |
1da177e4c Linux-2.6.12-rc2 |
1018 1019 1020 1021 1022 1023 |
*/ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, char *buf) { int turning_on; struct cpuset trialcs; |
85d7b9498 [PATCH] Dynamic s... |
1024 |
int err, cpu_exclusive_changed; |
1da177e4c Linux-2.6.12-rc2 |
1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 |
turning_on = (simple_strtoul(buf, NULL, 10) != 0); trialcs = *cs; if (turning_on) set_bit(bit, &trialcs.flags); else clear_bit(bit, &trialcs.flags); err = validate_change(cs, &trialcs); |
85d7b9498 [PATCH] Dynamic s... |
1035 1036 1037 1038 |
if (err < 0) return err; cpu_exclusive_changed = (is_cpu_exclusive(cs) != is_cpu_exclusive(&trialcs)); |
3d3f26a7b [PATCH] kernel/cp... |
1039 |
mutex_lock(&callback_mutex); |
696040670 [PATCH] cpuset: m... |
1040 |
cs->flags = trialcs.flags; |
3d3f26a7b [PATCH] kernel/cp... |
1041 |
mutex_unlock(&callback_mutex); |
85d7b9498 [PATCH] Dynamic s... |
1042 1043 1044 1045 |
if (cpu_exclusive_changed) update_cpu_domains(cs); return 0; |
1da177e4c Linux-2.6.12-rc2 |
1046 |
} |
053199edf [PATCH] cpusets: ... |
1047 |
/* |
80f7228b5 typo fixes: occur... |
1048 |
* Frequency meter - How fast is some event occurring? |
3e0d98b9f [PATCH] cpuset: m... |
1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 |
* * These routines manage a digitally filtered, constant time based, * event frequency meter. There are four routines: * fmeter_init() - initialize a frequency meter. * fmeter_markevent() - called each time the event happens. * fmeter_getrate() - returns the recent rate of such events. * fmeter_update() - internal routine used to update fmeter. * * A common data structure is passed to each of these routines, * which is used to keep track of the state required to manage the * frequency meter and its digital filter. * * The filter works on the number of events marked per unit time. * The filter is single-pole low-pass recursive (IIR). The time unit * is 1 second. Arithmetic is done using 32-bit integers scaled to * simulate 3 decimal digits of precision (multiplied by 1000). * * With an FM_COEF of 933, and a time base of 1 second, the filter * has a half-life of 10 seconds, meaning that if the events quit * happening, then the rate returned from the fmeter_getrate() * will be cut in half each 10 seconds, until it converges to zero. * * It is not worth doing a real infinitely recursive filter. If more * than FM_MAXTICKS ticks have elapsed since the last filter event, * just compute FM_MAXTICKS ticks worth, by which point the level * will be stable. * * Limit the count of unprocessed events to FM_MAXCNT, so as to avoid * arithmetic overflow in the fmeter_update() routine. * * Given the simple 32 bit integer arithmetic used, this meter works * best for reporting rates between one per millisecond (msec) and * one per 32 (approx) seconds. At constant rates faster than one * per msec it maxes out at values just under 1,000,000. At constant * rates between one per msec, and one per second it will stabilize * to a value N*1000, where N is the rate of events per second. * At constant rates between one per second and one per 32 seconds, * it will be choppy, moving up on the seconds that have an event, * and then decaying until the next event. At rates slower than * about one in 32 seconds, it decays all the way back to zero between * each event. */ #define FM_COEF 933 /* coefficient for half-life of 10 secs */ #define FM_MAXTICKS ((time_t)99) /* useless computing more ticks than this */ #define FM_MAXCNT 1000000 /* limit cnt to avoid overflow */ #define FM_SCALE 1000 /* faux fixed point scale */ /* Initialize a frequency meter */ static void fmeter_init(struct fmeter *fmp) { fmp->cnt = 0; fmp->val = 0; fmp->time = 0; spin_lock_init(&fmp->lock); } /* Internal meter update - process cnt events and update value */ static void fmeter_update(struct fmeter *fmp) { time_t now = get_seconds(); time_t ticks = now - fmp->time; if (ticks == 0) return; ticks = min(FM_MAXTICKS, ticks); while (ticks-- > 0) fmp->val = (FM_COEF * fmp->val) / FM_SCALE; fmp->time = now; fmp->val += ((FM_SCALE - FM_COEF) * fmp->cnt) / FM_SCALE; fmp->cnt = 0; } /* Process any previous ticks, then bump cnt by one (times scale). */ static void fmeter_markevent(struct fmeter *fmp) { spin_lock(&fmp->lock); fmeter_update(fmp); fmp->cnt = min(FM_MAXCNT, fmp->cnt + FM_SCALE); spin_unlock(&fmp->lock); } /* Process any previous ticks, then return current value. */ static int fmeter_getrate(struct fmeter *fmp) { int val; spin_lock(&fmp->lock); fmeter_update(fmp); val = fmp->val; spin_unlock(&fmp->lock); return val; } /* |
053199edf [PATCH] cpusets: ... |
1146 1147 1148 1149 |
* 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. * |
3d3f26a7b [PATCH] kernel/cp... |
1150 |
* Call holding manage_mutex. May take callback_mutex and task_lock of |
053199edf [PATCH] cpusets: ... |
1151 1152 |
* the task 'pid' during call. */ |
3077a260e [PATCH] cpuset re... |
1153 |
static int attach_task(struct cpuset *cs, char *pidbuf, char **ppathbuf) |
1da177e4c Linux-2.6.12-rc2 |
1154 1155 1156 1157 1158 |
{ pid_t pid; struct task_struct *tsk; struct cpuset *oldcs; cpumask_t cpus; |
45b07ef31 [PATCH] cpusets: ... |
1159 |
nodemask_t from, to; |
4225399a6 [PATCH] cpuset: r... |
1160 |
struct mm_struct *mm; |
22fb52dd7 [PATCH] SELinux: ... |
1161 |
int retval; |
1da177e4c Linux-2.6.12-rc2 |
1162 |
|
3077a260e [PATCH] cpuset re... |
1163 |
if (sscanf(pidbuf, "%d", &pid) != 1) |
1da177e4c Linux-2.6.12-rc2 |
1164 1165 1166 1167 1168 1169 1170 1171 |
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); |
053199edf [PATCH] cpusets: ... |
1172 |
if (!tsk || tsk->flags & PF_EXITING) { |
1da177e4c Linux-2.6.12-rc2 |
1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 |
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); } |
22fb52dd7 [PATCH] SELinux: ... |
1189 1190 1191 1192 1193 |
retval = security_task_setscheduler(tsk, 0, NULL); if (retval) { put_task_struct(tsk); return retval; } |
3d3f26a7b [PATCH] kernel/cp... |
1194 |
mutex_lock(&callback_mutex); |
053199edf [PATCH] cpusets: ... |
1195 |
|
1da177e4c Linux-2.6.12-rc2 |
1196 1197 |
task_lock(tsk); oldcs = tsk->cpuset; |
181b64803 [PATCH] cpuset: f... |
1198 1199 1200 1201 1202 1203 |
/* * 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) { |
1da177e4c Linux-2.6.12-rc2 |
1204 |
task_unlock(tsk); |
3d3f26a7b [PATCH] kernel/cp... |
1205 |
mutex_unlock(&callback_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1206 1207 1208 1209 |
put_task_struct(tsk); return -ESRCH; } atomic_inc(&cs->count); |
6b9c2603c [PATCH] cpuset: u... |
1210 |
rcu_assign_pointer(tsk->cpuset, cs); |
1da177e4c Linux-2.6.12-rc2 |
1211 1212 1213 1214 |
task_unlock(tsk); guarantee_online_cpus(cs, &cpus); set_cpus_allowed(tsk, cpus); |
45b07ef31 [PATCH] cpusets: ... |
1215 1216 |
from = oldcs->mems_allowed; to = cs->mems_allowed; |
3d3f26a7b [PATCH] kernel/cp... |
1217 |
mutex_unlock(&callback_mutex); |
4225399a6 [PATCH] cpuset: r... |
1218 1219 1220 1221 |
mm = get_task_mm(tsk); if (mm) { mpol_rebind_mm(mm, &to); |
2741a559a [PATCH] cpuset: u... |
1222 |
if (is_memory_migrate(cs)) |
e4e364e86 [PATCH] cpuset: m... |
1223 |
cpuset_migrate_mm(mm, &from, &to); |
4225399a6 [PATCH] cpuset: r... |
1224 1225 |
mmput(mm); } |
1da177e4c Linux-2.6.12-rc2 |
1226 |
put_task_struct(tsk); |
6b9c2603c [PATCH] cpuset: u... |
1227 |
synchronize_rcu(); |
1da177e4c Linux-2.6.12-rc2 |
1228 |
if (atomic_dec_and_test(&oldcs->count)) |
3077a260e [PATCH] cpuset re... |
1229 |
check_for_release(oldcs, ppathbuf); |
1da177e4c Linux-2.6.12-rc2 |
1230 1231 1232 1233 1234 1235 1236 1237 |
return 0; } /* The various types of files and directories in a cpuset file system */ typedef enum { FILE_ROOT, FILE_DIR, |
45b07ef31 [PATCH] cpusets: ... |
1238 |
FILE_MEMORY_MIGRATE, |
1da177e4c Linux-2.6.12-rc2 |
1239 1240 1241 1242 1243 |
FILE_CPULIST, FILE_MEMLIST, FILE_CPU_EXCLUSIVE, FILE_MEM_EXCLUSIVE, FILE_NOTIFY_ON_RELEASE, |
3e0d98b9f [PATCH] cpuset: m... |
1244 1245 |
FILE_MEMORY_PRESSURE_ENABLED, FILE_MEMORY_PRESSURE, |
825a46af5 [PATCH] cpuset me... |
1246 1247 |
FILE_SPREAD_PAGE, FILE_SPREAD_SLAB, |
1da177e4c Linux-2.6.12-rc2 |
1248 1249 |
FILE_TASKLIST, } cpuset_filetype_t; |
d3ed11c35 [PATCH] cpuset: a... |
1250 1251 |
static ssize_t cpuset_common_file_write(struct file *file, const char __user *userbuf, |
1da177e4c Linux-2.6.12-rc2 |
1252 1253 |
size_t nbytes, loff_t *unused_ppos) { |
a7a005fd1 [PATCH] struct pa... |
1254 1255 |
struct cpuset *cs = __d_cs(file->f_path.dentry->d_parent); struct cftype *cft = __d_cft(file->f_path.dentry); |
1da177e4c Linux-2.6.12-rc2 |
1256 1257 |
cpuset_filetype_t type = cft->private; char *buffer; |
3077a260e [PATCH] cpuset re... |
1258 |
char *pathbuf = NULL; |
1da177e4c Linux-2.6.12-rc2 |
1259 1260 1261 |
int retval = 0; /* Crude upper limit on largest legitimate cpulist user might write. */ |
d3ed11c35 [PATCH] cpuset: a... |
1262 |
if (nbytes > 100 + 6 * max(NR_CPUS, MAX_NUMNODES)) |
1da177e4c Linux-2.6.12-rc2 |
1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 |
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 */ |
3d3f26a7b [PATCH] kernel/cp... |
1274 |
mutex_lock(&manage_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 |
if (is_removed(cs)) { retval = -ENODEV; goto out2; } switch (type) { case FILE_CPULIST: retval = update_cpumask(cs, buffer); break; case FILE_MEMLIST: retval = update_nodemask(cs, buffer); break; case FILE_CPU_EXCLUSIVE: retval = update_flag(CS_CPU_EXCLUSIVE, cs, buffer); break; 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; |
45b07ef31 [PATCH] cpusets: ... |
1297 1298 1299 |
case FILE_MEMORY_MIGRATE: retval = update_flag(CS_MEMORY_MIGRATE, cs, buffer); break; |
3e0d98b9f [PATCH] cpuset: m... |
1300 1301 1302 1303 1304 1305 |
case FILE_MEMORY_PRESSURE_ENABLED: retval = update_memory_pressure_enabled(cs, buffer); break; case FILE_MEMORY_PRESSURE: retval = -EACCES; break; |
825a46af5 [PATCH] cpuset me... |
1306 1307 |
case FILE_SPREAD_PAGE: retval = update_flag(CS_SPREAD_PAGE, cs, buffer); |
151a44202 [PATCH] cpuset: d... |
1308 |
cs->mems_generation = cpuset_mems_generation++; |
825a46af5 [PATCH] cpuset me... |
1309 1310 1311 |
break; case FILE_SPREAD_SLAB: retval = update_flag(CS_SPREAD_SLAB, cs, buffer); |
151a44202 [PATCH] cpuset: d... |
1312 |
cs->mems_generation = cpuset_mems_generation++; |
825a46af5 [PATCH] cpuset me... |
1313 |
break; |
1da177e4c Linux-2.6.12-rc2 |
1314 |
case FILE_TASKLIST: |
3077a260e [PATCH] cpuset re... |
1315 |
retval = attach_task(cs, buffer, &pathbuf); |
1da177e4c Linux-2.6.12-rc2 |
1316 1317 1318 1319 1320 1321 1322 1323 1324 |
break; default: retval = -EINVAL; goto out2; } if (retval == 0) retval = nbytes; out2: |
3d3f26a7b [PATCH] kernel/cp... |
1325 |
mutex_unlock(&manage_mutex); |
3077a260e [PATCH] cpuset re... |
1326 |
cpuset_release_agent(pathbuf); |
1da177e4c Linux-2.6.12-rc2 |
1327 1328 1329 1330 1331 1332 1333 1334 1335 |
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; |
a7a005fd1 [PATCH] struct pa... |
1336 |
struct cftype *cft = __d_cft(file->f_path.dentry); |
1da177e4c Linux-2.6.12-rc2 |
1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 |
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 * chunks, there is no guarantee of atomicity. Since the display format * used, list of ranges of sequential numbers, is variable length, * and since these maps can change value dynamically, one could read * gibberish by doing partial reads while a list was changing. * A single large read to a buffer that crosses a page boundary is * ok, because the result being copied to user land is not recomputed * across a page fault. */ static int cpuset_sprintf_cpulist(char *page, struct cpuset *cs) { cpumask_t mask; |
3d3f26a7b [PATCH] kernel/cp... |
1364 |
mutex_lock(&callback_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1365 |
mask = cs->cpus_allowed; |
3d3f26a7b [PATCH] kernel/cp... |
1366 |
mutex_unlock(&callback_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1367 1368 1369 1370 1371 1372 1373 |
return cpulist_scnprintf(page, PAGE_SIZE, mask); } static int cpuset_sprintf_memlist(char *page, struct cpuset *cs) { nodemask_t mask; |
3d3f26a7b [PATCH] kernel/cp... |
1374 |
mutex_lock(&callback_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1375 |
mask = cs->mems_allowed; |
3d3f26a7b [PATCH] kernel/cp... |
1376 |
mutex_unlock(&callback_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1377 1378 1379 1380 1381 1382 1383 |
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) { |
a7a005fd1 [PATCH] struct pa... |
1384 1385 |
struct cftype *cft = __d_cft(file->f_path.dentry); struct cpuset *cs = __d_cs(file->f_path.dentry->d_parent); |
1da177e4c Linux-2.6.12-rc2 |
1386 1387 1388 1389 |
cpuset_filetype_t type = cft->private; char *page; ssize_t retval = 0; char *s; |
1da177e4c Linux-2.6.12-rc2 |
1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 |
if (!(page = (char *)__get_free_page(GFP_KERNEL))) return -ENOMEM; s = page; switch (type) { case FILE_CPULIST: s += cpuset_sprintf_cpulist(s, cs); break; case FILE_MEMLIST: s += cpuset_sprintf_memlist(s, cs); break; case FILE_CPU_EXCLUSIVE: *s++ = is_cpu_exclusive(cs) ? '1' : '0'; break; 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; |
45b07ef31 [PATCH] cpusets: ... |
1412 1413 1414 |
case FILE_MEMORY_MIGRATE: *s++ = is_memory_migrate(cs) ? '1' : '0'; break; |
3e0d98b9f [PATCH] cpuset: m... |
1415 1416 1417 1418 1419 1420 |
case FILE_MEMORY_PRESSURE_ENABLED: *s++ = cpuset_memory_pressure_enabled ? '1' : '0'; break; case FILE_MEMORY_PRESSURE: s += sprintf(s, "%d", fmeter_getrate(&cs->fmeter)); break; |
825a46af5 [PATCH] cpuset me... |
1421 1422 1423 1424 1425 1426 |
case FILE_SPREAD_PAGE: *s++ = is_spread_page(cs) ? '1' : '0'; break; case FILE_SPREAD_SLAB: *s++ = is_spread_slab(cs) ? '1' : '0'; break; |
1da177e4c Linux-2.6.12-rc2 |
1427 1428 1429 1430 1431 1432 |
default: retval = -EINVAL; goto out; } *s++ = ' '; |
1da177e4c Linux-2.6.12-rc2 |
1433 |
|
eacaa1f5a [PATCH] cpuset cr... |
1434 |
retval = simple_read_from_buffer(buf, nbytes, ppos, page, s - page); |
1da177e4c Linux-2.6.12-rc2 |
1435 1436 1437 1438 1439 1440 1441 1442 1443 |
out: free_page((unsigned long)page); return retval; } static ssize_t cpuset_file_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos) { ssize_t retval = 0; |
a7a005fd1 [PATCH] struct pa... |
1444 |
struct cftype *cft = __d_cft(file->f_path.dentry); |
1da177e4c Linux-2.6.12-rc2 |
1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 |
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; |
a7a005fd1 [PATCH] struct pa... |
1465 |
cft = __d_cft(file->f_path.dentry); |
1da177e4c Linux-2.6.12-rc2 |
1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 |
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) { |
a7a005fd1 [PATCH] struct pa... |
1478 |
struct cftype *cft = __d_cft(file->f_path.dentry); |
1da177e4c Linux-2.6.12-rc2 |
1479 1480 1481 1482 |
if (cft->release) return cft->release(inode, file); return 0; } |
18a19cb30 [PATCH] cpusets: ... |
1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 |
/* * 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); } |
15ad7cdcf [PATCH] struct se... |
1497 |
static const struct file_operations cpuset_file_operations = { |
1da177e4c Linux-2.6.12-rc2 |
1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 |
.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, |
18a19cb30 [PATCH] cpusets: ... |
1509 |
.rename = cpuset_rename, |
1da177e4c Linux-2.6.12-rc2 |
1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 |
}; 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) */ |
d8c76e6f4 [PATCH] r/o bind ... |
1530 |
inc_nlink(inode); |
1da177e4c Linux-2.6.12-rc2 |
1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 |
} 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. |
c5b2aff89 [PATCH] cpuset: m... |
1543 |
* cs: the cpuset we create the directory for. |
1da177e4c Linux-2.6.12-rc2 |
1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 |
* 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; |
d8c76e6f4 [PATCH] r/o bind ... |
1563 |
inc_nlink(parent->d_inode); |
1da177e4c Linux-2.6.12-rc2 |
1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 |
cs->dentry = dentry; } dput(dentry); return error; } static int cpuset_add_file(struct dentry *dir, const struct cftype *cft) { struct dentry *dentry; int error; |
1b1dcc1b5 [PATCH] mutex sub... |
1575 |
mutex_lock(&dir->d_inode->i_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1576 1577 1578 1579 1580 1581 1582 1583 |
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); |
1b1dcc1b5 [PATCH] mutex sub... |
1584 |
mutex_unlock(&dir->d_inode->i_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 |
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'. |
053199edf [PATCH] cpusets: ... |
1612 1613 1614 |
* 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. |
1da177e4c Linux-2.6.12-rc2 |
1615 |
*/ |
858119e15 [PATCH] Unlinline... |
1616 |
static int pid_array_load(pid_t *pidarray, int npids, struct cpuset *cs) |
1da177e4c Linux-2.6.12-rc2 |
1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 |
{ 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 ", a[i]); return cnt; } |
053199edf [PATCH] cpusets: ... |
1656 1657 1658 1659 |
/* * Handle an open on 'tasks' file. Prepare a buffer listing the * process id's of tasks currently attached to the cpuset being opened. * |
3d3f26a7b [PATCH] kernel/cp... |
1660 |
* Does not require any specific cpuset mutexes, and does not take any. |
053199edf [PATCH] cpusets: ... |
1661 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
1662 1663 |
static int cpuset_tasks_open(struct inode *unused, struct file *file) { |
a7a005fd1 [PATCH] struct pa... |
1664 |
struct cpuset *cs = __d_cs(file->f_path.dentry->d_parent); |
1da177e4c Linux-2.6.12-rc2 |
1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 |
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", .private = FILE_CPULIST, }; static struct cftype cft_mems = { .name = "mems", .private = FILE_MEMLIST, }; static struct cftype cft_cpu_exclusive = { .name = "cpu_exclusive", .private = FILE_CPU_EXCLUSIVE, }; static struct cftype cft_mem_exclusive = { .name = "mem_exclusive", .private = FILE_MEM_EXCLUSIVE, }; static struct cftype cft_notify_on_release = { .name = "notify_on_release", .private = FILE_NOTIFY_ON_RELEASE, }; |
45b07ef31 [PATCH] cpusets: ... |
1771 1772 1773 1774 |
static struct cftype cft_memory_migrate = { .name = "memory_migrate", .private = FILE_MEMORY_MIGRATE, }; |
3e0d98b9f [PATCH] cpuset: m... |
1775 1776 1777 1778 1779 1780 1781 1782 1783 |
static struct cftype cft_memory_pressure_enabled = { .name = "memory_pressure_enabled", .private = FILE_MEMORY_PRESSURE_ENABLED, }; static struct cftype cft_memory_pressure = { .name = "memory_pressure", .private = FILE_MEMORY_PRESSURE, }; |
825a46af5 [PATCH] cpuset me... |
1784 1785 1786 1787 1788 1789 1790 1791 1792 |
static struct cftype cft_spread_page = { .name = "memory_spread_page", .private = FILE_SPREAD_PAGE, }; static struct cftype cft_spread_slab = { .name = "memory_spread_slab", .private = FILE_SPREAD_SLAB, }; |
1da177e4c Linux-2.6.12-rc2 |
1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 |
static int cpuset_populate_dir(struct dentry *cs_dentry) { int err; if ((err = cpuset_add_file(cs_dentry, &cft_cpus)) < 0) return err; if ((err = cpuset_add_file(cs_dentry, &cft_mems)) < 0) return err; if ((err = cpuset_add_file(cs_dentry, &cft_cpu_exclusive)) < 0) return err; if ((err = cpuset_add_file(cs_dentry, &cft_mem_exclusive)) < 0) return err; if ((err = cpuset_add_file(cs_dentry, &cft_notify_on_release)) < 0) return err; |
45b07ef31 [PATCH] cpusets: ... |
1807 1808 |
if ((err = cpuset_add_file(cs_dentry, &cft_memory_migrate)) < 0) return err; |
3e0d98b9f [PATCH] cpuset: m... |
1809 1810 |
if ((err = cpuset_add_file(cs_dentry, &cft_memory_pressure)) < 0) return err; |
825a46af5 [PATCH] cpuset me... |
1811 1812 1813 1814 |
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; |
1da177e4c Linux-2.6.12-rc2 |
1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 |
if ((err = cpuset_add_file(cs_dentry, &cft_tasks)) < 0) return err; return 0; } /* * cpuset_create - create a cpuset * parent: cpuset that will be parent of the new cpuset. * name: name of the new cpuset. Will be strcpy'ed. * mode: mode to set on new inode * |
3d3f26a7b [PATCH] kernel/cp... |
1826 |
* Must be called with the mutex on the parent inode held |
1da177e4c Linux-2.6.12-rc2 |
1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 |
*/ static long cpuset_create(struct cpuset *parent, const char *name, int mode) { struct cpuset *cs; int err; cs = kmalloc(sizeof(*cs), GFP_KERNEL); if (!cs) return -ENOMEM; |
3d3f26a7b [PATCH] kernel/cp... |
1837 |
mutex_lock(&manage_mutex); |
cf2a473c4 [PATCH] cpuset: c... |
1838 |
cpuset_update_task_memory_state(); |
1da177e4c Linux-2.6.12-rc2 |
1839 1840 1841 |
cs->flags = 0; if (notify_on_release(parent)) set_bit(CS_NOTIFY_ON_RELEASE, &cs->flags); |
825a46af5 [PATCH] cpuset me... |
1842 1843 1844 1845 |
if (is_spread_page(parent)) set_bit(CS_SPREAD_PAGE, &cs->flags); if (is_spread_slab(parent)) set_bit(CS_SPREAD_SLAB, &cs->flags); |
1da177e4c Linux-2.6.12-rc2 |
1846 1847 1848 1849 1850 |
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); |
151a44202 [PATCH] cpuset: d... |
1851 |
cs->mems_generation = cpuset_mems_generation++; |
3e0d98b9f [PATCH] cpuset: m... |
1852 |
fmeter_init(&cs->fmeter); |
1da177e4c Linux-2.6.12-rc2 |
1853 1854 |
cs->parent = parent; |
3d3f26a7b [PATCH] kernel/cp... |
1855 |
mutex_lock(&callback_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1856 |
list_add(&cs->sibling, &cs->parent->children); |
202f72d5d [PATCH] cpuset: n... |
1857 |
number_of_cpusets++; |
3d3f26a7b [PATCH] kernel/cp... |
1858 |
mutex_unlock(&callback_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1859 1860 1861 1862 1863 1864 |
err = cpuset_create_dir(cs, name, mode); if (err < 0) goto err; /* |
3d3f26a7b [PATCH] kernel/cp... |
1865 |
* Release manage_mutex before cpuset_populate_dir() because it |
1b1dcc1b5 [PATCH] mutex sub... |
1866 |
* will down() this new directory's i_mutex and if we race with |
1da177e4c Linux-2.6.12-rc2 |
1867 1868 |
* another mkdir, we might deadlock. */ |
3d3f26a7b [PATCH] kernel/cp... |
1869 |
mutex_unlock(&manage_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1870 1871 1872 1873 1874 1875 |
err = cpuset_populate_dir(cs->dentry); /* If err < 0, we have a half-filled directory - oh well ;) */ return 0; err: list_del(&cs->sibling); |
3d3f26a7b [PATCH] kernel/cp... |
1876 |
mutex_unlock(&manage_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1877 1878 1879 1880 1881 1882 1883 |
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; |
1b1dcc1b5 [PATCH] mutex sub... |
1884 |
/* the vfs holds inode->i_mutex already */ |
1da177e4c Linux-2.6.12-rc2 |
1885 1886 |
return cpuset_create(c_parent, dentry->d_name.name, mode | S_IFDIR); } |
abb5a5cc6 [PATCH] Cpuset: f... |
1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 |
/* * Locking note on the strange update_flag() call below: * * If the cpuset 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. */ |
1da177e4c Linux-2.6.12-rc2 |
1897 1898 1899 1900 1901 |
static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry) { struct cpuset *cs = dentry->d_fsdata; struct dentry *d; struct cpuset *parent; |
3077a260e [PATCH] cpuset re... |
1902 |
char *pathbuf = NULL; |
1da177e4c Linux-2.6.12-rc2 |
1903 |
|
1b1dcc1b5 [PATCH] mutex sub... |
1904 |
/* the vfs holds both inode->i_mutex already */ |
1da177e4c Linux-2.6.12-rc2 |
1905 |
|
3d3f26a7b [PATCH] kernel/cp... |
1906 |
mutex_lock(&manage_mutex); |
cf2a473c4 [PATCH] cpuset: c... |
1907 |
cpuset_update_task_memory_state(); |
1da177e4c Linux-2.6.12-rc2 |
1908 |
if (atomic_read(&cs->count) > 0) { |
3d3f26a7b [PATCH] kernel/cp... |
1909 |
mutex_unlock(&manage_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1910 1911 1912 |
return -EBUSY; } if (!list_empty(&cs->children)) { |
3d3f26a7b [PATCH] kernel/cp... |
1913 |
mutex_unlock(&manage_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1914 1915 |
return -EBUSY; } |
abb5a5cc6 [PATCH] Cpuset: f... |
1916 1917 1918 1919 1920 1921 1922 |
if (is_cpu_exclusive(cs)) { int retval = update_flag(CS_CPU_EXCLUSIVE, cs, "0"); if (retval < 0) { mutex_unlock(&manage_mutex); return retval; } } |
1da177e4c Linux-2.6.12-rc2 |
1923 |
parent = cs->parent; |
3d3f26a7b [PATCH] kernel/cp... |
1924 |
mutex_lock(&callback_mutex); |
1da177e4c Linux-2.6.12-rc2 |
1925 1926 |
set_bit(CS_REMOVED, &cs->flags); list_del(&cs->sibling); /* delete my sibling from parent->children */ |
85d7b9498 [PATCH] Dynamic s... |
1927 |
spin_lock(&cs->dentry->d_lock); |
1da177e4c Linux-2.6.12-rc2 |
1928 1929 1930 1931 1932 |
d = dget(cs->dentry); cs->dentry = NULL; spin_unlock(&d->d_lock); cpuset_d_remove_dir(d); dput(d); |
202f72d5d [PATCH] cpuset: n... |
1933 |
number_of_cpusets--; |
3d3f26a7b [PATCH] kernel/cp... |
1934 |
mutex_unlock(&callback_mutex); |
053199edf [PATCH] cpusets: ... |
1935 1936 |
if (list_empty(&parent->children)) check_for_release(parent, &pathbuf); |
3d3f26a7b [PATCH] kernel/cp... |
1937 |
mutex_unlock(&manage_mutex); |
3077a260e [PATCH] cpuset re... |
1938 |
cpuset_release_agent(pathbuf); |
1da177e4c Linux-2.6.12-rc2 |
1939 1940 |
return 0; } |
c417f0242 [PATCH] cpuset: r... |
1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 |
/* * cpuset_init_early - just enough so that the calls to * cpuset_update_task_memory_state() in early init code * are harmless. */ int __init cpuset_init_early(void) { struct task_struct *tsk = current; tsk->cpuset = &top_cpuset; |
151a44202 [PATCH] cpuset: d... |
1952 |
tsk->cpuset->mems_generation = cpuset_mems_generation++; |
c417f0242 [PATCH] cpuset: r... |
1953 1954 |
return 0; } |
1da177e4c Linux-2.6.12-rc2 |
1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 |
/** * cpuset_init - initialize cpusets at system boot * * Description: Initialize top_cpuset and the cpuset internal file system, **/ int __init cpuset_init(void) { struct dentry *root; int err; top_cpuset.cpus_allowed = CPU_MASK_ALL; top_cpuset.mems_allowed = NODE_MASK_ALL; |
3e0d98b9f [PATCH] cpuset: m... |
1968 |
fmeter_init(&top_cpuset.fmeter); |
151a44202 [PATCH] cpuset: d... |
1969 |
top_cpuset.mems_generation = cpuset_mems_generation++; |
1da177e4c Linux-2.6.12-rc2 |
1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 |
init_task.cpuset = &top_cpuset; 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! "); err = PTR_ERR(cpuset_mount); cpuset_mount = NULL; goto out; } root = cpuset_mount->mnt_sb->s_root; root->d_fsdata = &top_cpuset; |
d8c76e6f4 [PATCH] r/o bind ... |
1986 |
inc_nlink(root->d_inode); |
1da177e4c Linux-2.6.12-rc2 |
1987 1988 |
top_cpuset.dentry = root; root->d_inode->i_op = &cpuset_dir_inode_operations; |
202f72d5d [PATCH] cpuset: n... |
1989 |
number_of_cpusets = 1; |
1da177e4c Linux-2.6.12-rc2 |
1990 |
err = cpuset_populate_dir(root); |
3e0d98b9f [PATCH] cpuset: m... |
1991 1992 1993 |
/* memory_pressure_enabled is in root cpuset only */ if (err == 0) err = cpuset_add_file(root, &cft_memory_pressure_enabled); |
1da177e4c Linux-2.6.12-rc2 |
1994 1995 1996 |
out: return err; } |
b1aac8bb8 [PATCH] cpuset: h... |
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 |
/* * If common_cpu_mem_hotplug_unplug(), below, unplugs any CPUs * or memory nodes, we need to walk over the cpuset hierarchy, * removing that CPU or node from all cpusets. If this removes the * last CPU or node from a cpuset, then the guarantee_online_cpus() * or guarantee_online_mems() code will use that emptied cpusets * parent online CPUs or nodes. Cpusets that were already empty of * CPUs or nodes are left empty. * * This routine is intentionally inefficient in a couple of regards. * It will check all cpusets in a subtree even if the top cpuset of * the subtree has no offline CPUs or nodes. It checks both CPUs and * nodes, even though the caller could have been coded to know that * only one of CPUs or nodes needed to be checked on a given call. * This was done to minimize text size rather than cpu cycles. * * Call with both manage_mutex and callback_mutex held. * * Recursive, on depth of cpuset subtree. */ static void guarantee_online_cpus_mems_in_subtree(const struct cpuset *cur) { struct cpuset *c; /* Each of our child cpusets mems must be online */ list_for_each_entry(c, &cur->children, sibling) { guarantee_online_cpus_mems_in_subtree(c); if (!cpus_empty(c->cpus_allowed)) guarantee_online_cpus(c, &c->cpus_allowed); if (!nodes_empty(c->mems_allowed)) guarantee_online_mems(c, &c->mems_allowed); } } /* * The cpus_allowed and mems_allowed nodemasks in the top_cpuset track * cpu_online_map and node_online_map. Force the top cpuset to track * whats online after any CPU or memory node hotplug or unplug event. * * To ensure that we don't remove a CPU or node from the top cpuset * that is currently in use by a child cpuset (which would violate * the rule that cpusets must be subsets of their parent), we first * call the recursive routine guarantee_online_cpus_mems_in_subtree(). * * Since there are two callers of this routine, one for CPU hotplug * events and one for memory node hotplug events, we could have coded * two separate routines here. We code it as a single common routine * in order to minimize text size. */ static void common_cpu_mem_hotplug_unplug(void) { mutex_lock(&manage_mutex); mutex_lock(&callback_mutex); 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); } |
b1aac8bb8 [PATCH] cpuset: h... |
2060 |
|
4c4d50f7b [PATCH] cpuset: t... |
2061 2062 2063 2064 2065 2066 |
/* * The top_cpuset tracks what CPUs and Memory Nodes are online, * period. This is necessary in order to make cpusets transparent * (of no affect) on systems that are actively using CPU hotplug * but making no active use of cpusets. * |
38837fc75 [PATCH] cpuset: t... |
2067 2068 |
* This routine ensures that top_cpuset.cpus_allowed tracks * cpu_online_map on each CPU hotplug (cpuhp) event. |
4c4d50f7b [PATCH] cpuset: t... |
2069 |
*/ |
4c4d50f7b [PATCH] cpuset: t... |
2070 2071 2072 |
static int cpuset_handle_cpuhp(struct notifier_block *nb, unsigned long phase, void *cpu) { |
b1aac8bb8 [PATCH] cpuset: h... |
2073 |
common_cpu_mem_hotplug_unplug(); |
4c4d50f7b [PATCH] cpuset: t... |
2074 2075 |
return 0; } |
4c4d50f7b [PATCH] cpuset: t... |
2076 |
|
b1aac8bb8 [PATCH] cpuset: h... |
2077 |
#ifdef CONFIG_MEMORY_HOTPLUG |
38837fc75 [PATCH] cpuset: t... |
2078 2079 2080 2081 2082 |
/* * Keep top_cpuset.mems_allowed tracking node_online_map. * Call this routine anytime after you change node_online_map. * See also the previous routine cpuset_handle_cpuhp(). */ |
1af989281 [PATCH] cpuset AN... |
2083 |
void cpuset_track_online_nodes(void) |
38837fc75 [PATCH] cpuset: t... |
2084 |
{ |
b1aac8bb8 [PATCH] cpuset: h... |
2085 |
common_cpu_mem_hotplug_unplug(); |
38837fc75 [PATCH] cpuset: t... |
2086 2087 |
} #endif |
1da177e4c Linux-2.6.12-rc2 |
2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 |
/** * cpuset_init_smp - initialize cpus_allowed * * Description: Finish top cpuset after cpu, node maps are initialized **/ void __init cpuset_init_smp(void) { top_cpuset.cpus_allowed = cpu_online_map; top_cpuset.mems_allowed = node_online_map; |
4c4d50f7b [PATCH] cpuset: t... |
2098 2099 |
hotcpu_notifier(cpuset_handle_cpuhp, 0); |
1da177e4c Linux-2.6.12-rc2 |
2100 2101 2102 2103 |
} /** * cpuset_fork - attach newly forked task to its parents cpuset. |
d9fd8a6d4 [PATCH] kernel/cp... |
2104 |
* @tsk: pointer to task_struct of forking parent process. |
1da177e4c Linux-2.6.12-rc2 |
2105 |
* |
053199edf [PATCH] cpusets: ... |
2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 |
* 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. |
1da177e4c Linux-2.6.12-rc2 |
2118 |
**/ |
053199edf [PATCH] cpusets: ... |
2119 |
void cpuset_fork(struct task_struct *child) |
1da177e4c Linux-2.6.12-rc2 |
2120 |
{ |
053199edf [PATCH] cpusets: ... |
2121 2122 2123 2124 |
task_lock(current); child->cpuset = current->cpuset; atomic_inc(&child->cpuset->count); task_unlock(current); |
1da177e4c Linux-2.6.12-rc2 |
2125 2126 2127 2128 2129 2130 2131 2132 |
} /** * cpuset_exit - detach cpuset from exiting task * @tsk: pointer to task_struct of exiting process * * Description: Detach cpuset from @tsk and release it. * |
053199edf [PATCH] cpusets: ... |
2133 |
* Note that cpusets marked notify_on_release force every task in |
3d3f26a7b [PATCH] kernel/cp... |
2134 |
* them to take the global manage_mutex mutex when exiting. |
053199edf [PATCH] cpusets: ... |
2135 2136 2137 2138 2139 |
* 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 |
3d3f26a7b [PATCH] kernel/cp... |
2140 2141 |
* 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 |
053199edf [PATCH] cpusets: ... |
2142 2143 |
* any other task to nuke the cpuset immediately, via cpuset_rmdir(). * |
3d3f26a7b [PATCH] kernel/cp... |
2144 2145 2146 |
* 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(). |
053199edf [PATCH] cpusets: ... |
2147 2148 2149 |
* * We don't need to task_lock() this reference to tsk->cpuset, * because tsk is already marked PF_EXITING, so attach_task() won't |
b4b264184 [PATCH] cpuset: f... |
2150 |
* mess with it, or task is a failed fork, never visible to attach_task. |
06fed3384 [PATCH] cpuset: o... |
2151 |
* |
8488bc359 [PATCH] cpuset: r... |
2152 |
* the_top_cpuset_hack: |
06fed3384 [PATCH] cpuset: o... |
2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 |
* * 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 ;). |
1da177e4c Linux-2.6.12-rc2 |
2184 2185 2186 2187 2188 |
**/ void cpuset_exit(struct task_struct *tsk) { struct cpuset *cs; |
1da177e4c Linux-2.6.12-rc2 |
2189 |
cs = tsk->cpuset; |
8488bc359 [PATCH] cpuset: r... |
2190 |
tsk->cpuset = &top_cpuset; /* the_top_cpuset_hack - see above */ |
1da177e4c Linux-2.6.12-rc2 |
2191 |
|
2efe86b80 [PATCH] cpuset ex... |
2192 |
if (notify_on_release(cs)) { |
3077a260e [PATCH] cpuset re... |
2193 |
char *pathbuf = NULL; |
3d3f26a7b [PATCH] kernel/cp... |
2194 |
mutex_lock(&manage_mutex); |
2efe86b80 [PATCH] cpuset ex... |
2195 |
if (atomic_dec_and_test(&cs->count)) |
3077a260e [PATCH] cpuset re... |
2196 |
check_for_release(cs, &pathbuf); |
3d3f26a7b [PATCH] kernel/cp... |
2197 |
mutex_unlock(&manage_mutex); |
3077a260e [PATCH] cpuset re... |
2198 |
cpuset_release_agent(pathbuf); |
2efe86b80 [PATCH] cpuset ex... |
2199 2200 |
} else { atomic_dec(&cs->count); |
1da177e4c Linux-2.6.12-rc2 |
2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 |
} } /** * cpuset_cpus_allowed - return cpus_allowed mask from a tasks cpuset. * @tsk: pointer to task_struct from which to obtain cpuset->cpus_allowed. * * Description: Returns the cpumask_t cpus_allowed of the cpuset * attached to the specified @tsk. Guaranteed to return some non-empty * subset of cpu_online_map, even if this means going outside the * tasks cpuset. **/ |
909d75a3b [PATCH] cpuset: i... |
2213 |
cpumask_t cpuset_cpus_allowed(struct task_struct *tsk) |
1da177e4c Linux-2.6.12-rc2 |
2214 2215 |
{ cpumask_t mask; |
3d3f26a7b [PATCH] kernel/cp... |
2216 |
mutex_lock(&callback_mutex); |
909d75a3b [PATCH] cpuset: i... |
2217 |
task_lock(tsk); |
1da177e4c Linux-2.6.12-rc2 |
2218 |
guarantee_online_cpus(tsk->cpuset, &mask); |
909d75a3b [PATCH] cpuset: i... |
2219 |
task_unlock(tsk); |
3d3f26a7b [PATCH] kernel/cp... |
2220 |
mutex_unlock(&callback_mutex); |
1da177e4c Linux-2.6.12-rc2 |
2221 2222 2223 2224 2225 2226 2227 2228 |
return mask; } void cpuset_init_current_mems_allowed(void) { current->mems_allowed = NODE_MASK_ALL; } |
d9fd8a6d4 [PATCH] kernel/cp... |
2229 |
/** |
909d75a3b [PATCH] cpuset: i... |
2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 |
* cpuset_mems_allowed - return mems_allowed mask from a tasks cpuset. * @tsk: pointer to task_struct from which to obtain cpuset->mems_allowed. * * Description: Returns the nodemask_t mems_allowed of the cpuset * attached to the specified @tsk. Guaranteed to return some non-empty * subset of node_online_map, even if this means going outside the * tasks cpuset. **/ nodemask_t cpuset_mems_allowed(struct task_struct *tsk) { nodemask_t mask; |
3d3f26a7b [PATCH] kernel/cp... |
2242 |
mutex_lock(&callback_mutex); |
909d75a3b [PATCH] cpuset: i... |
2243 2244 2245 |
task_lock(tsk); guarantee_online_mems(tsk->cpuset, &mask); task_unlock(tsk); |
3d3f26a7b [PATCH] kernel/cp... |
2246 |
mutex_unlock(&callback_mutex); |
909d75a3b [PATCH] cpuset: i... |
2247 2248 2249 2250 2251 |
return mask; } /** |
d9fd8a6d4 [PATCH] kernel/cp... |
2252 2253 2254 |
* cpuset_zonelist_valid_mems_allowed - check zonelist vs. curremt mems_allowed * @zl: the zonelist to be checked * |
1da177e4c Linux-2.6.12-rc2 |
2255 2256 2257 2258 2259 2260 2261 |
* Are any of the nodes on zonelist zl allowed in current->mems_allowed? */ int cpuset_zonelist_valid_mems_allowed(struct zonelist *zl) { int i; for (i = 0; zl->zones[i]; i++) { |
89fa30242 [PATCH] NUMA: Add... |
2262 |
int nid = zone_to_nid(zl->zones[i]); |
1da177e4c Linux-2.6.12-rc2 |
2263 2264 2265 2266 2267 2268 |
if (node_isset(nid, current->mems_allowed)) return 1; } return 0; } |
9bf2229f8 [PATCH] cpusets: ... |
2269 2270 |
/* * nearest_exclusive_ancestor() - Returns the nearest mem_exclusive |
3d3f26a7b [PATCH] kernel/cp... |
2271 |
* ancestor to the specified cpuset. Call holding callback_mutex. |
9bf2229f8 [PATCH] cpusets: ... |
2272 2273 2274 2275 2276 2277 2278 2279 2280 |
* If no ancestor is mem_exclusive (an unusual configuration), then * returns the root cpuset. */ static const struct cpuset *nearest_exclusive_ancestor(const struct cpuset *cs) { while (!is_mem_exclusive(cs) && cs->parent) cs = cs->parent; return cs; } |
d9fd8a6d4 [PATCH] kernel/cp... |
2281 |
/** |
02a0e53d8 [PATCH] cpuset: r... |
2282 |
* cpuset_zone_allowed_softwall - Can we allocate on zone z's memory node? |
9bf2229f8 [PATCH] cpusets: ... |
2283 |
* @z: is this zone on an allowed node? |
02a0e53d8 [PATCH] cpuset: r... |
2284 |
* @gfp_mask: memory allocation flags |
d9fd8a6d4 [PATCH] kernel/cp... |
2285 |
* |
02a0e53d8 [PATCH] cpuset: r... |
2286 2287 |
* If we're in interrupt, yes, we can always allocate. If * __GFP_THISNODE is set, yes, we can always allocate. If zone |
9bf2229f8 [PATCH] cpusets: ... |
2288 2289 2290 2291 2292 |
* z's node is in our tasks mems_allowed, yes. If it's not a * __GFP_HARDWALL request and this zone's nodes is in the nearest * mem_exclusive cpuset ancestor to this tasks cpuset, yes. * Otherwise, no. * |
02a0e53d8 [PATCH] cpuset: r... |
2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 |
* If __GFP_HARDWALL is set, cpuset_zone_allowed_softwall() * reduces to cpuset_zone_allowed_hardwall(). Otherwise, * cpuset_zone_allowed_softwall() might sleep, and might allow a zone * from an enclosing cpuset. * * cpuset_zone_allowed_hardwall() only handles the simpler case of * hardwall cpusets, and never sleeps. * * The __GFP_THISNODE placement logic is really handled elsewhere, * by forcibly using a zonelist starting at a specified node, and by * (in get_page_from_freelist()) refusing to consider the zones for * any node on the zonelist except the first. By the time any such * calls get to this routine, we should just shut up and say 'yes'. * |
9bf2229f8 [PATCH] cpusets: ... |
2307 2308 2309 |
* GFP_USER allocations are marked with the __GFP_HARDWALL bit, * and do not allow allocations outside the current tasks cpuset. * GFP_KERNEL allocations are not so marked, so can escape to the |
02a0e53d8 [PATCH] cpuset: r... |
2310 |
* nearest enclosing mem_exclusive ancestor cpuset. |
9bf2229f8 [PATCH] cpusets: ... |
2311 |
* |
02a0e53d8 [PATCH] cpuset: r... |
2312 2313 2314 2315 2316 2317 2318 |
* Scanning up parent cpusets requires callback_mutex. The * __alloc_pages() routine only calls here with __GFP_HARDWALL bit * _not_ set if it's a GFP_KERNEL allocation, and all nodes in the * current tasks mems_allowed came up empty on the first pass over * the zonelist. So only GFP_KERNEL allocations, if all nodes in the * cpuset are short of memory, might require taking the callback_mutex * mutex. |
9bf2229f8 [PATCH] cpusets: ... |
2319 |
* |
36be57ffe [PATCH] cpuset: u... |
2320 |
* The first call here from mm/page_alloc:get_page_from_freelist() |
02a0e53d8 [PATCH] cpuset: r... |
2321 2322 2323 |
* has __GFP_HARDWALL set in gfp_mask, enforcing hardwall cpusets, * so no allocation on a node outside the cpuset is allowed (unless * in interrupt, of course). |
36be57ffe [PATCH] cpuset: u... |
2324 2325 2326 2327 2328 2329 |
* * The second pass through get_page_from_freelist() doesn't even call * here for GFP_ATOMIC calls. For those calls, the __alloc_pages() * variable 'wait' is not set, and the bit ALLOC_CPUSET is not set * in alloc_flags. That logic and the checks below have the combined * affect that: |
9bf2229f8 [PATCH] cpusets: ... |
2330 2331 2332 2333 |
* in_interrupt - any node ok (current task context irrelevant) * GFP_ATOMIC - any node ok * GFP_KERNEL - any node in enclosing mem_exclusive cpuset ok * GFP_USER - only nodes in current tasks mems allowed ok. |
36be57ffe [PATCH] cpuset: u... |
2334 2335 |
* * Rule: |
02a0e53d8 [PATCH] cpuset: r... |
2336 |
* Don't call cpuset_zone_allowed_softwall if you can't sleep, unless you |
36be57ffe [PATCH] cpuset: u... |
2337 2338 |
* pass in the __GFP_HARDWALL flag set in gfp_flag, which disables * the code that might scan up ancestor cpusets and sleep. |
02a0e53d8 [PATCH] cpuset: r... |
2339 |
*/ |
9bf2229f8 [PATCH] cpusets: ... |
2340 |
|
02a0e53d8 [PATCH] cpuset: r... |
2341 |
int __cpuset_zone_allowed_softwall(struct zone *z, gfp_t gfp_mask) |
1da177e4c Linux-2.6.12-rc2 |
2342 |
{ |
9bf2229f8 [PATCH] cpusets: ... |
2343 2344 |
int node; /* node that zone z is on */ const struct cpuset *cs; /* current cpuset ancestors */ |
29afd49b7 [PATCH] cpuset: r... |
2345 |
int allowed; /* is allocation in zone z allowed? */ |
9bf2229f8 [PATCH] cpusets: ... |
2346 |
|
9b819d204 [PATCH] Add __GFP... |
2347 |
if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) |
9bf2229f8 [PATCH] cpusets: ... |
2348 |
return 1; |
89fa30242 [PATCH] NUMA: Add... |
2349 |
node = zone_to_nid(z); |
92d1dbd27 [PATCH] cpuset: m... |
2350 |
might_sleep_if(!(gfp_mask & __GFP_HARDWALL)); |
9bf2229f8 [PATCH] cpusets: ... |
2351 2352 2353 2354 |
if (node_isset(node, current->mems_allowed)) return 1; if (gfp_mask & __GFP_HARDWALL) /* If hardwall request, stop here */ return 0; |
5563e7707 [PATCH] cpuset: f... |
2355 2356 |
if (current->flags & PF_EXITING) /* Let dying task have memory */ return 1; |
9bf2229f8 [PATCH] cpusets: ... |
2357 |
/* Not hardwall and node outside mems_allowed: scan up cpusets */ |
3d3f26a7b [PATCH] kernel/cp... |
2358 |
mutex_lock(&callback_mutex); |
053199edf [PATCH] cpusets: ... |
2359 |
|
053199edf [PATCH] cpusets: ... |
2360 2361 2362 |
task_lock(current); cs = nearest_exclusive_ancestor(current->cpuset); task_unlock(current); |
9bf2229f8 [PATCH] cpusets: ... |
2363 |
allowed = node_isset(node, cs->mems_allowed); |
3d3f26a7b [PATCH] kernel/cp... |
2364 |
mutex_unlock(&callback_mutex); |
9bf2229f8 [PATCH] cpusets: ... |
2365 |
return allowed; |
1da177e4c Linux-2.6.12-rc2 |
2366 |
} |
02a0e53d8 [PATCH] cpuset: r... |
2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 |
/* * cpuset_zone_allowed_hardwall - Can we allocate on zone z's memory node? * @z: is this zone on an allowed node? * @gfp_mask: memory allocation flags * * If we're in interrupt, yes, we can always allocate. * If __GFP_THISNODE is set, yes, we can always allocate. If zone * z's node is in our tasks mems_allowed, yes. Otherwise, no. * * The __GFP_THISNODE placement logic is really handled elsewhere, * by forcibly using a zonelist starting at a specified node, and by * (in get_page_from_freelist()) refusing to consider the zones for * any node on the zonelist except the first. By the time any such * calls get to this routine, we should just shut up and say 'yes'. * * Unlike the cpuset_zone_allowed_softwall() variant, above, * this variant requires that the zone be in the current tasks * mems_allowed or that we're in interrupt. It does not scan up the * cpuset hierarchy for the nearest enclosing mem_exclusive cpuset. * It never sleeps. */ int __cpuset_zone_allowed_hardwall(struct zone *z, gfp_t gfp_mask) { int node; /* node that zone z is on */ if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) return 1; node = zone_to_nid(z); if (node_isset(node, current->mems_allowed)) return 1; return 0; } |
ef08e3b49 [PATCH] cpusets: ... |
2400 |
/** |
505970b96 [PATCH] cpuset oo... |
2401 2402 |
* cpuset_lock - lock out any changes to cpuset structures * |
3d3f26a7b [PATCH] kernel/cp... |
2403 |
* The out of memory (oom) code needs to mutex_lock cpusets |
505970b96 [PATCH] cpuset oo... |
2404 |
* from being changed while it scans the tasklist looking for a |
3d3f26a7b [PATCH] kernel/cp... |
2405 |
* task in an overlapping cpuset. Expose callback_mutex via this |
505970b96 [PATCH] cpuset oo... |
2406 2407 |
* cpuset_lock() routine, so the oom code can lock it, before * locking the task list. The tasklist_lock is a spinlock, so |
3d3f26a7b [PATCH] kernel/cp... |
2408 |
* must be taken inside callback_mutex. |
505970b96 [PATCH] cpuset oo... |
2409 2410 2411 2412 |
*/ void cpuset_lock(void) { |
3d3f26a7b [PATCH] kernel/cp... |
2413 |
mutex_lock(&callback_mutex); |
505970b96 [PATCH] cpuset oo... |
2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 |
} /** * cpuset_unlock - release lock on cpuset changes * * Undo the lock taken in a previous cpuset_lock() call. */ void cpuset_unlock(void) { |
3d3f26a7b [PATCH] kernel/cp... |
2424 |
mutex_unlock(&callback_mutex); |
505970b96 [PATCH] cpuset oo... |
2425 2426 2427 |
} /** |
825a46af5 [PATCH] cpuset me... |
2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 |
* 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 * tasks in a cpuset with is_spread_page or is_spread_slab set), * and if the memory allocation used cpuset_mem_spread_node() * to determine on which node to start looking, as it will for * certain page cache or slab cache pages such as used for file * system buffers and inode caches, then instead of starting on the * local node to look for a free page, rather spread the starting * node around the tasks mems_allowed nodes. * * We don't have to worry about the returned node being offline * because "it can't happen", and even if it did, it would be ok. * * The routines calling guarantee_online_mems() are careful to * only set nodes in task->mems_allowed that are online. So it * should not be possible for the following code to return an * offline node. But if it did, that would be ok, as this routine * is not returning the node where the allocation must be, only * the node where the search should start. The zonelist passed to * __alloc_pages() will include all nodes. If the slab allocator * is passed an offline node, it will fall back to the local node. * See kmem_cache_alloc_node(). */ int cpuset_mem_spread_node(void) { int node; node = next_node(current->cpuset_mem_spread_rotor, current->mems_allowed); if (node == MAX_NUMNODES) node = first_node(current->mems_allowed); current->cpuset_mem_spread_rotor = node; return node; } EXPORT_SYMBOL_GPL(cpuset_mem_spread_node); /** |
ef08e3b49 [PATCH] cpusets: ... |
2466 2467 2468 2469 2470 2471 2472 2473 |
* cpuset_excl_nodes_overlap - Do we overlap @p's mem_exclusive ancestors? * @p: pointer to task_struct of some other task. * * Description: Return true if the nearest mem_exclusive ancestor * cpusets of tasks @p and current overlap. Used by oom killer to * determine if task @p's memory usage might impact the memory * available to the current task. * |
3d3f26a7b [PATCH] kernel/cp... |
2474 |
* Call while holding callback_mutex. |
ef08e3b49 [PATCH] cpusets: ... |
2475 2476 2477 2478 2479 |
**/ int cpuset_excl_nodes_overlap(const struct task_struct *p) { const struct cpuset *cs1, *cs2; /* my and p's cpuset ancestors */ |
0d673a5a4 [PATCH] cpuset: o... |
2480 |
int overlap = 1; /* do cpusets overlap? */ |
ef08e3b49 [PATCH] cpusets: ... |
2481 |
|
053199edf [PATCH] cpusets: ... |
2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 |
task_lock(current); if (current->flags & PF_EXITING) { task_unlock(current); goto done; } cs1 = nearest_exclusive_ancestor(current->cpuset); task_unlock(current); task_lock((struct task_struct *)p); if (p->flags & PF_EXITING) { task_unlock((struct task_struct *)p); goto done; } cs2 = nearest_exclusive_ancestor(p->cpuset); task_unlock((struct task_struct *)p); |
ef08e3b49 [PATCH] cpusets: ... |
2497 2498 |
overlap = nodes_intersects(cs1->mems_allowed, cs2->mems_allowed); done: |
ef08e3b49 [PATCH] cpusets: ... |
2499 2500 |
return overlap; } |
1da177e4c Linux-2.6.12-rc2 |
2501 |
/* |
3e0d98b9f [PATCH] cpuset: m... |
2502 2503 2504 2505 |
* Collection of memory_pressure is suppressed unless * this flag is enabled by writing "1" to the special * cpuset file 'memory_pressure_enabled' in the root cpuset. */ |
c5b2aff89 [PATCH] cpuset: m... |
2506 |
int cpuset_memory_pressure_enabled __read_mostly; |
3e0d98b9f [PATCH] cpuset: m... |
2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 |
/** * cpuset_memory_pressure_bump - keep stats of per-cpuset reclaims. * * Keep a running average of the rate of synchronous (direct) * page reclaim efforts initiated by tasks in each cpuset. * * This represents the rate at which some task in the cpuset * ran low on memory on all nodes it was allowed to use, and * had to enter the kernels page reclaim code in an effort to * create more free memory by tossing clean pages or swapping * or writing dirty pages. * * Display to user space in the per-cpuset read-only file * "memory_pressure". Value displayed is an integer * representing the recent rate of entry into the synchronous * (direct) page reclaim by any task attached to the cpuset. **/ void __cpuset_memory_pressure_bump(void) { struct cpuset *cs; task_lock(current); cs = current->cpuset; fmeter_markevent(&cs->fmeter); task_unlock(current); } /* |
1da177e4c Linux-2.6.12-rc2 |
2537 2538 2539 |
* proc_cpuset_show() * - Print tasks cpuset path into seq_file. * - Used for /proc/<pid>/cpuset. |
053199edf [PATCH] cpusets: ... |
2540 2541 |
* - 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, |
3d3f26a7b [PATCH] kernel/cp... |
2542 |
* and we take manage_mutex, keeping attach_task() from changing it |
8488bc359 [PATCH] cpuset: r... |
2543 2544 2545 |
* 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. |
1da177e4c Linux-2.6.12-rc2 |
2546 |
*/ |
1da177e4c Linux-2.6.12-rc2 |
2547 2548 |
static int proc_cpuset_show(struct seq_file *m, void *v) { |
13b41b094 [PATCH] proc: Use... |
2549 |
struct pid *pid; |
1da177e4c Linux-2.6.12-rc2 |
2550 2551 |
struct task_struct *tsk; char *buf; |
99f895518 [PATCH] proc: don... |
2552 |
int retval; |
1da177e4c Linux-2.6.12-rc2 |
2553 |
|
99f895518 [PATCH] proc: don... |
2554 |
retval = -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
2555 2556 |
buf = kmalloc(PAGE_SIZE, GFP_KERNEL); if (!buf) |
99f895518 [PATCH] proc: don... |
2557 2558 2559 |
goto out; retval = -ESRCH; |
13b41b094 [PATCH] proc: Use... |
2560 2561 |
pid = m->private; tsk = get_pid_task(pid, PIDTYPE_PID); |
99f895518 [PATCH] proc: don... |
2562 2563 |
if (!tsk) goto out_free; |
1da177e4c Linux-2.6.12-rc2 |
2564 |
|
99f895518 [PATCH] proc: don... |
2565 |
retval = -EINVAL; |
3d3f26a7b [PATCH] kernel/cp... |
2566 |
mutex_lock(&manage_mutex); |
99f895518 [PATCH] proc: don... |
2567 |
|
8488bc359 [PATCH] cpuset: r... |
2568 |
retval = cpuset_path(tsk->cpuset, buf, PAGE_SIZE); |
1da177e4c Linux-2.6.12-rc2 |
2569 |
if (retval < 0) |
99f895518 [PATCH] proc: don... |
2570 |
goto out_unlock; |
1da177e4c Linux-2.6.12-rc2 |
2571 2572 2573 |
seq_puts(m, buf); seq_putc(m, ' '); |
99f895518 [PATCH] proc: don... |
2574 |
out_unlock: |
3d3f26a7b [PATCH] kernel/cp... |
2575 |
mutex_unlock(&manage_mutex); |
99f895518 [PATCH] proc: don... |
2576 2577 |
put_task_struct(tsk); out_free: |
1da177e4c Linux-2.6.12-rc2 |
2578 |
kfree(buf); |
99f895518 [PATCH] proc: don... |
2579 |
out: |
1da177e4c Linux-2.6.12-rc2 |
2580 2581 2582 2583 2584 |
return retval; } static int cpuset_open(struct inode *inode, struct file *file) { |
13b41b094 [PATCH] proc: Use... |
2585 2586 |
struct pid *pid = PROC_I(inode)->pid; return single_open(file, proc_cpuset_show, pid); |
1da177e4c Linux-2.6.12-rc2 |
2587 |
} |
089e34b60 [PATCH] cpuset pr... |
2588 |
struct file_operations proc_cpuset_operations = { |
1da177e4c Linux-2.6.12-rc2 |
2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 |
.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) { buffer += sprintf(buffer, "Cpus_allowed:\t"); buffer += cpumask_scnprintf(buffer, PAGE_SIZE, task->cpus_allowed); buffer += sprintf(buffer, " "); buffer += sprintf(buffer, "Mems_allowed:\t"); buffer += nodemask_scnprintf(buffer, PAGE_SIZE, task->mems_allowed); buffer += sprintf(buffer, " "); return buffer; } |