Blame view
kernel/rcutree_plugin.h
64.5 KB
f41d911f8 rcu: Merge preemp... |
1 2 3 |
/* * Read-Copy Update mechanism for mutual exclusion (tree-based version) * Internal non-public definitions that provide either classic |
6cc68793e rcu: fix spelling |
4 |
* or preemptible semantics. |
f41d911f8 rcu: Merge preemp... |
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 |
* * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or * (at your option) any later version. * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. * * You should have received a copy of the GNU General Public License * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. * * Copyright Red Hat, 2009 * Copyright IBM Corporation, 2009 * * Author: Ingo Molnar <mingo@elte.hu> * Paul E. McKenney <paulmck@linux.vnet.ibm.com> */ |
d9a3da069 rcu: Add expedite... |
26 |
#include <linux/delay.h> |
7b27d5475 rcu,cleanup: move... |
27 |
#include <linux/stop_machine.h> |
f41d911f8 rcu: Merge preemp... |
28 |
|
5b61b0baa rcu: Wire up RCU_... |
29 30 31 32 33 34 35 |
#define RCU_KTHREAD_PRIO 1 #ifdef CONFIG_RCU_BOOST #define RCU_BOOST_PRIO CONFIG_RCU_BOOST_PRIO #else #define RCU_BOOST_PRIO RCU_KTHREAD_PRIO #endif |
26845c286 rcu: print boot-t... |
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 |
/* * Check the RCU kernel configuration parameters and print informative * messages about anything out of the ordinary. If you like #ifdef, you * will love this function. */ static void __init rcu_bootup_announce_oddness(void) { #ifdef CONFIG_RCU_TRACE printk(KERN_INFO "\tRCU debugfs-based tracing is enabled. "); #endif #if (defined(CONFIG_64BIT) && CONFIG_RCU_FANOUT != 64) || (!defined(CONFIG_64BIT) && CONFIG_RCU_FANOUT != 32) printk(KERN_INFO "\tCONFIG_RCU_FANOUT set to non-default value of %d ", CONFIG_RCU_FANOUT); #endif #ifdef CONFIG_RCU_FANOUT_EXACT printk(KERN_INFO "\tHierarchical RCU autobalancing is disabled. "); #endif #ifdef CONFIG_RCU_FAST_NO_HZ printk(KERN_INFO "\tRCU dyntick-idle grace-period acceleration is enabled. "); #endif #ifdef CONFIG_PROVE_RCU printk(KERN_INFO "\tRCU lockdep checking is enabled. "); #endif #ifdef CONFIG_RCU_TORTURE_TEST_RUNNABLE printk(KERN_INFO "\tRCU torture testing starts during boot. "); #endif |
81a294c44 rcu: fix _oddness... |
69 |
#if defined(CONFIG_TREE_PREEMPT_RCU) && !defined(CONFIG_RCU_CPU_STALL_VERBOSE) |
26845c286 rcu: print boot-t... |
70 71 72 73 74 75 76 77 |
printk(KERN_INFO "\tVerbose stalled-CPUs detection is disabled. "); #endif #if NUM_RCU_LVL_4 != 0 printk(KERN_INFO "\tExperimental four-level hierarchy is enabled. "); #endif } |
f41d911f8 rcu: Merge preemp... |
78 |
#ifdef CONFIG_TREE_PREEMPT_RCU |
e99033c5c rcu: Put names in... |
79 |
struct rcu_state rcu_preempt_state = RCU_STATE_INITIALIZER(rcu_preempt); |
f41d911f8 rcu: Merge preemp... |
80 |
DEFINE_PER_CPU(struct rcu_data, rcu_preempt_data); |
27f4d2805 rcu: priority boo... |
81 |
static struct rcu_state *rcu_state = &rcu_preempt_state; |
f41d911f8 rcu: Merge preemp... |
82 |
|
10f39bb1b rcu: protect __rc... |
83 |
static void rcu_read_unlock_special(struct task_struct *t); |
d9a3da069 rcu: Add expedite... |
84 |
static int rcu_preempted_readers_exp(struct rcu_node *rnp); |
f41d911f8 rcu: Merge preemp... |
85 86 87 |
/* * Tell them what RCU they are running. */ |
0e0fc1c23 rcu: Mark init-ti... |
88 |
static void __init rcu_bootup_announce(void) |
f41d911f8 rcu: Merge preemp... |
89 |
{ |
6cc68793e rcu: fix spelling |
90 91 |
printk(KERN_INFO "Preemptible hierarchical RCU implementation. "); |
26845c286 rcu: print boot-t... |
92 |
rcu_bootup_announce_oddness(); |
f41d911f8 rcu: Merge preemp... |
93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 |
} /* * Return the number of RCU-preempt batches processed thus far * for debug and statistics. */ long rcu_batches_completed_preempt(void) { return rcu_preempt_state.completed; } EXPORT_SYMBOL_GPL(rcu_batches_completed_preempt); /* * Return the number of RCU batches processed thus far for debug & stats. */ long rcu_batches_completed(void) { return rcu_batches_completed_preempt(); } EXPORT_SYMBOL_GPL(rcu_batches_completed); /* |
bf66f18e7 rcu: Add force_qu... |
115 116 117 118 119 120 121 122 123 |
* Force a quiescent state for preemptible RCU. */ void rcu_force_quiescent_state(void) { force_quiescent_state(&rcu_preempt_state, 0); } EXPORT_SYMBOL_GPL(rcu_force_quiescent_state); /* |
6cc68793e rcu: fix spelling |
124 |
* Record a preemptible-RCU quiescent state for the specified CPU. Note |
f41d911f8 rcu: Merge preemp... |
125 126 127 |
* that this just means that the task currently running on the CPU is * not in a quiescent state. There might be any number of tasks blocked * while in an RCU read-side critical section. |
25502a6c1 rcu: refactor RCU... |
128 129 130 131 |
* * Unlike the other rcu_*_qs() functions, callers to this function * must disable irqs in order to protect the assignment to * ->rcu_read_unlock_special. |
f41d911f8 rcu: Merge preemp... |
132 |
*/ |
c3422bea5 rcu: Simplify rcu... |
133 |
static void rcu_preempt_qs(int cpu) |
f41d911f8 rcu: Merge preemp... |
134 135 |
{ struct rcu_data *rdp = &per_cpu(rcu_preempt_data, cpu); |
25502a6c1 rcu: refactor RCU... |
136 |
|
e4cc1f22b rcu: Simplify qui... |
137 |
rdp->passed_quiesce_gpnum = rdp->gpnum; |
c3422bea5 rcu: Simplify rcu... |
138 |
barrier(); |
e4cc1f22b rcu: Simplify qui... |
139 |
if (rdp->passed_quiesce == 0) |
d4c08f2ac rcu: Add grace-pe... |
140 |
trace_rcu_grace_period("rcu_preempt", rdp->gpnum, "cpuqs"); |
e4cc1f22b rcu: Simplify qui... |
141 |
rdp->passed_quiesce = 1; |
25502a6c1 rcu: refactor RCU... |
142 |
current->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS; |
f41d911f8 rcu: Merge preemp... |
143 144 145 |
} /* |
c3422bea5 rcu: Simplify rcu... |
146 147 148 |
* We have entered the scheduler, and the current task might soon be * context-switched away from. If this task is in an RCU read-side * critical section, we will no longer be able to rely on the CPU to |
12f5f524c rcu: merge TREE_P... |
149 150 151 152 153 154 |
* record that fact, so we enqueue the task on the blkd_tasks list. * The task will dequeue itself when it exits the outermost enclosing * RCU read-side critical section. Therefore, the current grace period * cannot be permitted to complete until the blkd_tasks list entries * predating the current grace period drain, in other words, until * rnp->gp_tasks becomes NULL. |
c3422bea5 rcu: Simplify rcu... |
155 156 |
* * Caller must disable preemption. |
f41d911f8 rcu: Merge preemp... |
157 |
*/ |
c3422bea5 rcu: Simplify rcu... |
158 |
static void rcu_preempt_note_context_switch(int cpu) |
f41d911f8 rcu: Merge preemp... |
159 160 |
{ struct task_struct *t = current; |
c3422bea5 rcu: Simplify rcu... |
161 |
unsigned long flags; |
f41d911f8 rcu: Merge preemp... |
162 163 |
struct rcu_data *rdp; struct rcu_node *rnp; |
10f39bb1b rcu: protect __rc... |
164 |
if (t->rcu_read_lock_nesting > 0 && |
f41d911f8 rcu: Merge preemp... |
165 166 167 |
(t->rcu_read_unlock_special & RCU_READ_UNLOCK_BLOCKED) == 0) { /* Possibly blocking in an RCU read-side critical section. */ |
394f99a90 rcu: simplify the... |
168 |
rdp = per_cpu_ptr(rcu_preempt_state.rda, cpu); |
f41d911f8 rcu: Merge preemp... |
169 |
rnp = rdp->mynode; |
1304afb22 rcu: Convert to r... |
170 |
raw_spin_lock_irqsave(&rnp->lock, flags); |
f41d911f8 rcu: Merge preemp... |
171 |
t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BLOCKED; |
868489660 rcu: Changes from... |
172 |
t->rcu_blocked_node = rnp; |
f41d911f8 rcu: Merge preemp... |
173 174 175 176 177 178 179 180 181 |
/* * If this CPU has already checked in, then this task * will hold up the next grace period rather than the * current grace period. Queue the task accordingly. * If the task is queued for the current grace period * (i.e., this CPU has not yet passed through a quiescent * state for the current grace period), then as long * as that task remains queued, the current grace period |
12f5f524c rcu: merge TREE_P... |
182 183 184 185 186 187 |
* cannot end. Note that there is some uncertainty as * to exactly when the current grace period started. * We take a conservative approach, which can result * in unnecessarily waiting on tasks that started very * slightly after the current grace period began. C'est * la vie!!! |
b0e165c03 rcu: Add debug ch... |
188 189 190 |
* * But first, note that the current CPU must still be * on line! |
f41d911f8 rcu: Merge preemp... |
191 |
*/ |
b0e165c03 rcu: Add debug ch... |
192 |
WARN_ON_ONCE((rdp->grpmask & rnp->qsmaskinit) == 0); |
e7d8842ed rcu: Apply result... |
193 |
WARN_ON_ONCE(!list_empty(&t->rcu_node_entry)); |
12f5f524c rcu: merge TREE_P... |
194 195 196 |
if ((rnp->qsmask & rdp->grpmask) && rnp->gp_tasks != NULL) { list_add(&t->rcu_node_entry, rnp->gp_tasks->prev); rnp->gp_tasks = &t->rcu_node_entry; |
27f4d2805 rcu: priority boo... |
197 198 199 200 |
#ifdef CONFIG_RCU_BOOST if (rnp->boost_tasks != NULL) rnp->boost_tasks = rnp->gp_tasks; #endif /* #ifdef CONFIG_RCU_BOOST */ |
12f5f524c rcu: merge TREE_P... |
201 202 203 204 205 |
} else { list_add(&t->rcu_node_entry, &rnp->blkd_tasks); if (rnp->qsmask & rdp->grpmask) rnp->gp_tasks = &t->rcu_node_entry; } |
d4c08f2ac rcu: Add grace-pe... |
206 207 208 209 210 |
trace_rcu_preempt_task(rdp->rsp->name, t->pid, (rnp->qsmask & rdp->grpmask) ? rnp->gpnum : rnp->gpnum + 1); |
1304afb22 rcu: Convert to r... |
211 |
raw_spin_unlock_irqrestore(&rnp->lock, flags); |
10f39bb1b rcu: protect __rc... |
212 213 214 215 216 217 218 219 |
} else if (t->rcu_read_lock_nesting < 0 && t->rcu_read_unlock_special) { /* * Complete exit from RCU read-side critical section on * behalf of preempted instance of __rcu_read_unlock(). */ rcu_read_unlock_special(t); |
f41d911f8 rcu: Merge preemp... |
220 221 222 223 224 225 226 227 228 229 230 |
} /* * Either we were not in an RCU read-side critical section to * begin with, or we have now recorded that critical section * globally. Either way, we can now note a quiescent state * for this CPU. Again, if we were in an RCU read-side critical * section, and if that critical section was blocking the current * grace period, then the fact that the task has been enqueued * means that we continue to block the current grace period. */ |
e7d8842ed rcu: Apply result... |
231 |
local_irq_save(flags); |
25502a6c1 rcu: refactor RCU... |
232 |
rcu_preempt_qs(cpu); |
e7d8842ed rcu: Apply result... |
233 |
local_irq_restore(flags); |
f41d911f8 rcu: Merge preemp... |
234 235 236 |
} /* |
6cc68793e rcu: fix spelling |
237 |
* Tree-preemptible RCU implementation for rcu_read_lock(). |
f41d911f8 rcu: Merge preemp... |
238 239 240 241 242 |
* Just increment ->rcu_read_lock_nesting, shared state will be updated * if we block. */ void __rcu_read_lock(void) { |
80dcf60e6 rcu: apply TINY_P... |
243 |
current->rcu_read_lock_nesting++; |
f41d911f8 rcu: Merge preemp... |
244 245 246 |
barrier(); /* needed if we ever invoke rcu_read_lock in rcutree.c */ } EXPORT_SYMBOL_GPL(__rcu_read_lock); |
fc2219d49 rcu: Clean up cod... |
247 248 249 250 251 |
/* * Check for preempted RCU readers blocking the current grace period * for the specified rcu_node structure. If the caller needs a reliable * answer, it must hold the rcu_node's ->lock. */ |
27f4d2805 rcu: priority boo... |
252 |
static int rcu_preempt_blocked_readers_cgp(struct rcu_node *rnp) |
fc2219d49 rcu: Clean up cod... |
253 |
{ |
12f5f524c rcu: merge TREE_P... |
254 |
return rnp->gp_tasks != NULL; |
fc2219d49 rcu: Clean up cod... |
255 |
} |
b668c9cf3 rcu: Fix grace-pe... |
256 257 258 259 260 261 262 |
/* * Record a quiescent state for all tasks that were previously queued * on the specified rcu_node structure and that were blocking the current * RCU grace period. The caller must hold the specified rnp->lock with * irqs disabled, and this lock is released upon return, but irqs remain * disabled. */ |
d3f6bad39 rcu: Rename "quie... |
263 |
static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp, unsigned long flags) |
b668c9cf3 rcu: Fix grace-pe... |
264 265 266 267 |
__releases(rnp->lock) { unsigned long mask; struct rcu_node *rnp_p; |
27f4d2805 rcu: priority boo... |
268 |
if (rnp->qsmask != 0 || rcu_preempt_blocked_readers_cgp(rnp)) { |
1304afb22 rcu: Convert to r... |
269 |
raw_spin_unlock_irqrestore(&rnp->lock, flags); |
b668c9cf3 rcu: Fix grace-pe... |
270 271 272 273 274 275 276 277 278 279 |
return; /* Still need more quiescent states! */ } rnp_p = rnp->parent; if (rnp_p == NULL) { /* * Either there is only one rcu_node in the tree, * or tasks were kicked up to root rcu_node due to * CPUs going offline. */ |
d3f6bad39 rcu: Rename "quie... |
280 |
rcu_report_qs_rsp(&rcu_preempt_state, flags); |
b668c9cf3 rcu: Fix grace-pe... |
281 282 283 284 285 |
return; } /* Report up the rest of the hierarchy. */ mask = rnp->grpmask; |
1304afb22 rcu: Convert to r... |
286 287 |
raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ raw_spin_lock(&rnp_p->lock); /* irqs already disabled. */ |
d3f6bad39 rcu: Rename "quie... |
288 |
rcu_report_qs_rnp(mask, &rcu_preempt_state, rnp_p, flags); |
b668c9cf3 rcu: Fix grace-pe... |
289 290 291 |
} /* |
12f5f524c rcu: merge TREE_P... |
292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 |
* Advance a ->blkd_tasks-list pointer to the next entry, instead * returning NULL if at the end of the list. */ static struct list_head *rcu_next_node_entry(struct task_struct *t, struct rcu_node *rnp) { struct list_head *np; np = t->rcu_node_entry.next; if (np == &rnp->blkd_tasks) np = NULL; return np; } /* |
b668c9cf3 rcu: Fix grace-pe... |
307 308 309 310 |
* Handle special cases during rcu_read_unlock(), such as needing to * notify RCU core processing or task having blocked during the RCU * read-side critical section. */ |
be0e1e21e rcu: Streamline c... |
311 |
static noinline void rcu_read_unlock_special(struct task_struct *t) |
f41d911f8 rcu: Merge preemp... |
312 313 |
{ int empty; |
d9a3da069 rcu: Add expedite... |
314 |
int empty_exp; |
389abd48e rcu: Avoid RCU-pr... |
315 |
int empty_exp_now; |
f41d911f8 rcu: Merge preemp... |
316 |
unsigned long flags; |
12f5f524c rcu: merge TREE_P... |
317 |
struct list_head *np; |
82e78d80f rcu: Simplify unb... |
318 319 320 |
#ifdef CONFIG_RCU_BOOST struct rt_mutex *rbmp = NULL; #endif /* #ifdef CONFIG_RCU_BOOST */ |
f41d911f8 rcu: Merge preemp... |
321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 |
struct rcu_node *rnp; int special; /* NMI handlers cannot block and cannot safely manipulate state. */ if (in_nmi()) return; local_irq_save(flags); /* * If RCU core is waiting for this CPU to exit critical section, * let it know that we have done so. */ special = t->rcu_read_unlock_special; if (special & RCU_READ_UNLOCK_NEED_QS) { |
c3422bea5 rcu: Simplify rcu... |
336 |
rcu_preempt_qs(smp_processor_id()); |
f41d911f8 rcu: Merge preemp... |
337 338 339 |
} /* Hardware IRQ handlers cannot block. */ |
ec433f0c5 softirq,rcu: Info... |
340 |
if (in_irq() || in_serving_softirq()) { |
f41d911f8 rcu: Merge preemp... |
341 342 343 344 345 346 347 |
local_irq_restore(flags); return; } /* Clean up if blocked during RCU read-side critical section. */ if (special & RCU_READ_UNLOCK_BLOCKED) { t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_BLOCKED; |
dd5d19baf rcu: Create rcutr... |
348 349 350 351 352 353 |
/* * Remove this task from the list it blocked on. The * task can migrate while we acquire the lock, but at * most one time. So at most two passes through loop. */ for (;;) { |
868489660 rcu: Changes from... |
354 |
rnp = t->rcu_blocked_node; |
1304afb22 rcu: Convert to r... |
355 |
raw_spin_lock(&rnp->lock); /* irqs already disabled. */ |
868489660 rcu: Changes from... |
356 |
if (rnp == t->rcu_blocked_node) |
dd5d19baf rcu: Create rcutr... |
357 |
break; |
1304afb22 rcu: Convert to r... |
358 |
raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ |
dd5d19baf rcu: Create rcutr... |
359 |
} |
27f4d2805 rcu: priority boo... |
360 |
empty = !rcu_preempt_blocked_readers_cgp(rnp); |
d9a3da069 rcu: Add expedite... |
361 362 |
empty_exp = !rcu_preempted_readers_exp(rnp); smp_mb(); /* ensure expedited fastpath sees end of RCU c-s. */ |
12f5f524c rcu: merge TREE_P... |
363 |
np = rcu_next_node_entry(t, rnp); |
f41d911f8 rcu: Merge preemp... |
364 |
list_del_init(&t->rcu_node_entry); |
82e78d80f rcu: Simplify unb... |
365 |
t->rcu_blocked_node = NULL; |
d4c08f2ac rcu: Add grace-pe... |
366 367 |
trace_rcu_unlock_preempted_task("rcu_preempt", rnp->gpnum, t->pid); |
12f5f524c rcu: merge TREE_P... |
368 369 370 371 |
if (&t->rcu_node_entry == rnp->gp_tasks) rnp->gp_tasks = np; if (&t->rcu_node_entry == rnp->exp_tasks) rnp->exp_tasks = np; |
27f4d2805 rcu: priority boo... |
372 373 374 |
#ifdef CONFIG_RCU_BOOST if (&t->rcu_node_entry == rnp->boost_tasks) rnp->boost_tasks = np; |
82e78d80f rcu: Simplify unb... |
375 376 377 378 |
/* Snapshot/clear ->rcu_boost_mutex with rcu_node lock held. */ if (t->rcu_boost_mutex) { rbmp = t->rcu_boost_mutex; t->rcu_boost_mutex = NULL; |
7765be2fe rcu: Fix RCU_BOOS... |
379 |
} |
27f4d2805 rcu: priority boo... |
380 |
#endif /* #ifdef CONFIG_RCU_BOOST */ |
f41d911f8 rcu: Merge preemp... |
381 382 383 384 |
/* * If this was the last task on the current list, and if * we aren't waiting on any CPUs, report the quiescent state. |
389abd48e rcu: Avoid RCU-pr... |
385 386 |
* Note that rcu_report_unblock_qs_rnp() releases rnp->lock, * so we must take a snapshot of the expedited state. |
f41d911f8 rcu: Merge preemp... |
387 |
*/ |
389abd48e rcu: Avoid RCU-pr... |
388 |
empty_exp_now = !rcu_preempted_readers_exp(rnp); |
d4c08f2ac rcu: Add grace-pe... |
389 390 391 392 393 394 395 396 |
if (!empty && !rcu_preempt_blocked_readers_cgp(rnp)) { trace_rcu_quiescent_state_report("preempt_rcu", rnp->gpnum, 0, rnp->qsmask, rnp->level, rnp->grplo, rnp->grphi, !!rnp->gp_tasks); |
d3f6bad39 rcu: Rename "quie... |
397 |
rcu_report_unblock_qs_rnp(rnp, flags); |
d4c08f2ac rcu: Add grace-pe... |
398 399 |
} else raw_spin_unlock_irqrestore(&rnp->lock, flags); |
d9a3da069 rcu: Add expedite... |
400 |
|
27f4d2805 rcu: priority boo... |
401 402 |
#ifdef CONFIG_RCU_BOOST /* Unboost if we were boosted. */ |
82e78d80f rcu: Simplify unb... |
403 404 |
if (rbmp) rt_mutex_unlock(rbmp); |
27f4d2805 rcu: priority boo... |
405 |
#endif /* #ifdef CONFIG_RCU_BOOST */ |
d9a3da069 rcu: Add expedite... |
406 407 408 409 |
/* * If this was the last task on the expedited lists, * then we need to report up the rcu_node hierarchy. */ |
389abd48e rcu: Avoid RCU-pr... |
410 |
if (!empty_exp && empty_exp_now) |
b40d293eb rcu: Omit self-aw... |
411 |
rcu_report_exp_rnp(&rcu_preempt_state, rnp, true); |
b668c9cf3 rcu: Fix grace-pe... |
412 413 |
} else { local_irq_restore(flags); |
f41d911f8 rcu: Merge preemp... |
414 |
} |
f41d911f8 rcu: Merge preemp... |
415 416 417 |
} /* |
6cc68793e rcu: fix spelling |
418 |
* Tree-preemptible RCU implementation for rcu_read_unlock(). |
f41d911f8 rcu: Merge preemp... |
419 420 421 422 423 424 425 426 |
* Decrement ->rcu_read_lock_nesting. If the result is zero (outermost * rcu_read_unlock()) and ->rcu_read_unlock_special is non-zero, then * invoke rcu_read_unlock_special() to clean up after a context switch * in an RCU read-side critical section and other special cases. */ void __rcu_read_unlock(void) { struct task_struct *t = current; |
10f39bb1b rcu: protect __rc... |
427 428 429 |
if (t->rcu_read_lock_nesting != 1) --t->rcu_read_lock_nesting; else { |
6206ab9ba rcu: Move __rcu_r... |
430 |
barrier(); /* critical section before exit code. */ |
10f39bb1b rcu: protect __rc... |
431 432 |
t->rcu_read_lock_nesting = INT_MIN; barrier(); /* assign before ->rcu_read_unlock_special load */ |
be0e1e21e rcu: Streamline c... |
433 434 |
if (unlikely(ACCESS_ONCE(t->rcu_read_unlock_special))) rcu_read_unlock_special(t); |
10f39bb1b rcu: protect __rc... |
435 436 |
barrier(); /* ->rcu_read_unlock_special load before assign */ t->rcu_read_lock_nesting = 0; |
be0e1e21e rcu: Streamline c... |
437 |
} |
cba8244a0 rcu: Add debug ch... |
438 |
#ifdef CONFIG_PROVE_LOCKING |
10f39bb1b rcu: protect __rc... |
439 440 441 442 443 |
{ int rrln = ACCESS_ONCE(t->rcu_read_lock_nesting); WARN_ON_ONCE(rrln < 0 && rrln > INT_MIN / 2); } |
cba8244a0 rcu: Add debug ch... |
444 |
#endif /* #ifdef CONFIG_PROVE_LOCKING */ |
f41d911f8 rcu: Merge preemp... |
445 446 |
} EXPORT_SYMBOL_GPL(__rcu_read_unlock); |
1ed509a22 rcu: Add RCU_CPU_... |
447 448 449 450 451 452 453 454 455 |
#ifdef CONFIG_RCU_CPU_STALL_VERBOSE /* * Dump detailed information for all tasks blocking the current RCU * grace period on the specified rcu_node structure. */ static void rcu_print_detail_task_stall_rnp(struct rcu_node *rnp) { unsigned long flags; |
1ed509a22 rcu: Add RCU_CPU_... |
456 |
struct task_struct *t; |
27f4d2805 rcu: priority boo... |
457 |
if (!rcu_preempt_blocked_readers_cgp(rnp)) |
12f5f524c rcu: merge TREE_P... |
458 459 460 461 462 463 464 |
return; raw_spin_lock_irqsave(&rnp->lock, flags); t = list_entry(rnp->gp_tasks, struct task_struct, rcu_node_entry); list_for_each_entry_continue(t, &rnp->blkd_tasks, rcu_node_entry) sched_show_task(t); raw_spin_unlock_irqrestore(&rnp->lock, flags); |
1ed509a22 rcu: Add RCU_CPU_... |
465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 |
} /* * Dump detailed information for all tasks blocking the current RCU * grace period. */ static void rcu_print_detail_task_stall(struct rcu_state *rsp) { struct rcu_node *rnp = rcu_get_root(rsp); rcu_print_detail_task_stall_rnp(rnp); rcu_for_each_leaf_node(rsp, rnp) rcu_print_detail_task_stall_rnp(rnp); } #else /* #ifdef CONFIG_RCU_CPU_STALL_VERBOSE */ static void rcu_print_detail_task_stall(struct rcu_state *rsp) { } #endif /* #else #ifdef CONFIG_RCU_CPU_STALL_VERBOSE */ |
f41d911f8 rcu: Merge preemp... |
487 488 489 490 |
/* * Scan the current list of tasks blocked within RCU read-side critical * sections, printing out the tid of each. */ |
9bc8b5586 rcu: Suppress NMI... |
491 |
static int rcu_print_task_stall(struct rcu_node *rnp) |
f41d911f8 rcu: Merge preemp... |
492 |
{ |
f41d911f8 rcu: Merge preemp... |
493 |
struct task_struct *t; |
9bc8b5586 rcu: Suppress NMI... |
494 |
int ndetected = 0; |
f41d911f8 rcu: Merge preemp... |
495 |
|
27f4d2805 rcu: priority boo... |
496 |
if (!rcu_preempt_blocked_readers_cgp(rnp)) |
9bc8b5586 rcu: Suppress NMI... |
497 |
return 0; |
12f5f524c rcu: merge TREE_P... |
498 499 |
t = list_entry(rnp->gp_tasks, struct task_struct, rcu_node_entry); |
9bc8b5586 rcu: Suppress NMI... |
500 |
list_for_each_entry_continue(t, &rnp->blkd_tasks, rcu_node_entry) { |
12f5f524c rcu: merge TREE_P... |
501 |
printk(" P%d", t->pid); |
9bc8b5586 rcu: Suppress NMI... |
502 503 504 |
ndetected++; } return ndetected; |
f41d911f8 rcu: Merge preemp... |
505 |
} |
53d84e004 rcu: permit suppr... |
506 507 508 509 510 511 512 513 514 |
/* * Suppress preemptible RCU's CPU stall warnings by pushing the * time of the next stall-warning message comfortably far into the * future. */ static void rcu_preempt_stall_reset(void) { rcu_preempt_state.jiffies_stall = jiffies + ULONG_MAX / 2; } |
f41d911f8 rcu: Merge preemp... |
515 |
/* |
b0e165c03 rcu: Add debug ch... |
516 517 518 519 520 |
* Check that the list of blocked tasks for the newly completed grace * period is in fact empty. It is a serious bug to complete a grace * period that still has RCU readers blocked! This function must be * invoked -before- updating this rnp's ->gpnum, and the rnp's ->lock * must be held by the caller. |
12f5f524c rcu: merge TREE_P... |
521 522 523 |
* * Also, if there are blocked tasks on the list, they automatically * block the newly created grace period, so set up ->gp_tasks accordingly. |
b0e165c03 rcu: Add debug ch... |
524 525 526 |
*/ static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp) { |
27f4d2805 rcu: priority boo... |
527 |
WARN_ON_ONCE(rcu_preempt_blocked_readers_cgp(rnp)); |
12f5f524c rcu: merge TREE_P... |
528 529 |
if (!list_empty(&rnp->blkd_tasks)) rnp->gp_tasks = rnp->blkd_tasks.next; |
28ecd5802 rcu: Add WARN_ON_... |
530 |
WARN_ON_ONCE(rnp->qsmask); |
b0e165c03 rcu: Add debug ch... |
531 |
} |
33f76148c rcu: Add CPU-offl... |
532 533 534 |
#ifdef CONFIG_HOTPLUG_CPU /* |
dd5d19baf rcu: Create rcutr... |
535 536 537 538 539 |
* Handle tasklist migration for case in which all CPUs covered by the * specified rcu_node have gone offline. Move them up to the root * rcu_node. The reason for not just moving them to the immediate * parent is to remove the need for rcu_read_unlock_special() to * make more than two attempts to acquire the target rcu_node's lock. |
b668c9cf3 rcu: Fix grace-pe... |
540 541 |
* Returns true if there were tasks blocking the current RCU grace * period. |
dd5d19baf rcu: Create rcutr... |
542 |
* |
237c80c5c rcu: Fix TREE_PRE... |
543 544 545 |
* Returns 1 if there was previously a task blocking the current grace * period on the specified rcu_node structure. * |
dd5d19baf rcu: Create rcutr... |
546 547 |
* The caller must hold rnp->lock with irqs disabled. */ |
237c80c5c rcu: Fix TREE_PRE... |
548 549 550 |
static int rcu_preempt_offline_tasks(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) |
dd5d19baf rcu: Create rcutr... |
551 |
{ |
dd5d19baf rcu: Create rcutr... |
552 553 |
struct list_head *lp; struct list_head *lp_root; |
d9a3da069 rcu: Add expedite... |
554 |
int retval = 0; |
dd5d19baf rcu: Create rcutr... |
555 |
struct rcu_node *rnp_root = rcu_get_root(rsp); |
12f5f524c rcu: merge TREE_P... |
556 |
struct task_struct *t; |
dd5d19baf rcu: Create rcutr... |
557 |
|
868489660 rcu: Changes from... |
558 559 |
if (rnp == rnp_root) { WARN_ONCE(1, "Last CPU thought to be offlined?"); |
237c80c5c rcu: Fix TREE_PRE... |
560 |
return 0; /* Shouldn't happen: at least one CPU online. */ |
868489660 rcu: Changes from... |
561 |
} |
12f5f524c rcu: merge TREE_P... |
562 563 564 |
/* If we are on an internal node, complain bitterly. */ WARN_ON_ONCE(rnp != rdp->mynode); |
dd5d19baf rcu: Create rcutr... |
565 566 |
/* |
12f5f524c rcu: merge TREE_P... |
567 568 569 570 571 572 573 |
* Move tasks up to root rcu_node. Don't try to get fancy for * this corner-case operation -- just put this node's tasks * at the head of the root node's list, and update the root node's * ->gp_tasks and ->exp_tasks pointers to those of this node's, * if non-NULL. This might result in waiting for more tasks than * absolutely necessary, but this is a good performance/complexity * tradeoff. |
dd5d19baf rcu: Create rcutr... |
574 |
*/ |
27f4d2805 rcu: priority boo... |
575 |
if (rcu_preempt_blocked_readers_cgp(rnp)) |
d9a3da069 rcu: Add expedite... |
576 577 578 |
retval |= RCU_OFL_TASKS_NORM_GP; if (rcu_preempted_readers_exp(rnp)) retval |= RCU_OFL_TASKS_EXP_GP; |
12f5f524c rcu: merge TREE_P... |
579 580 581 582 583 584 585 586 587 588 589 590 |
lp = &rnp->blkd_tasks; lp_root = &rnp_root->blkd_tasks; while (!list_empty(lp)) { t = list_entry(lp->next, typeof(*t), rcu_node_entry); raw_spin_lock(&rnp_root->lock); /* irqs already disabled */ list_del(&t->rcu_node_entry); t->rcu_blocked_node = rnp_root; list_add(&t->rcu_node_entry, lp_root); if (&t->rcu_node_entry == rnp->gp_tasks) rnp_root->gp_tasks = rnp->gp_tasks; if (&t->rcu_node_entry == rnp->exp_tasks) rnp_root->exp_tasks = rnp->exp_tasks; |
27f4d2805 rcu: priority boo... |
591 592 593 594 |
#ifdef CONFIG_RCU_BOOST if (&t->rcu_node_entry == rnp->boost_tasks) rnp_root->boost_tasks = rnp->boost_tasks; #endif /* #ifdef CONFIG_RCU_BOOST */ |
12f5f524c rcu: merge TREE_P... |
595 |
raw_spin_unlock(&rnp_root->lock); /* irqs still disabled */ |
dd5d19baf rcu: Create rcutr... |
596 |
} |
27f4d2805 rcu: priority boo... |
597 598 599 600 601 602 603 604 605 |
#ifdef CONFIG_RCU_BOOST /* In case root is being boosted and leaf is not. */ raw_spin_lock(&rnp_root->lock); /* irqs already disabled */ if (rnp_root->boost_tasks != NULL && rnp_root->boost_tasks != rnp_root->gp_tasks) rnp_root->boost_tasks = rnp_root->gp_tasks; raw_spin_unlock(&rnp_root->lock); /* irqs still disabled */ #endif /* #ifdef CONFIG_RCU_BOOST */ |
12f5f524c rcu: merge TREE_P... |
606 607 |
rnp->gp_tasks = NULL; rnp->exp_tasks = NULL; |
237c80c5c rcu: Fix TREE_PRE... |
608 |
return retval; |
dd5d19baf rcu: Create rcutr... |
609 610 611 |
} /* |
6cc68793e rcu: fix spelling |
612 |
* Do CPU-offline processing for preemptible RCU. |
33f76148c rcu: Add CPU-offl... |
613 614 615 616 617 618 619 |
*/ static void rcu_preempt_offline_cpu(int cpu) { __rcu_offline_cpu(cpu, &rcu_preempt_state); } #endif /* #ifdef CONFIG_HOTPLUG_CPU */ |
f41d911f8 rcu: Merge preemp... |
620 621 622 623 624 625 626 627 628 629 630 631 |
/* * Check for a quiescent state from the current CPU. When a task blocks, * the task is recorded in the corresponding CPU's rcu_node structure, * which is checked elsewhere. * * Caller must disable hard irqs. */ static void rcu_preempt_check_callbacks(int cpu) { struct task_struct *t = current; if (t->rcu_read_lock_nesting == 0) { |
c3422bea5 rcu: Simplify rcu... |
632 |
rcu_preempt_qs(cpu); |
f41d911f8 rcu: Merge preemp... |
633 634 |
return; } |
10f39bb1b rcu: protect __rc... |
635 636 |
if (t->rcu_read_lock_nesting > 0 && per_cpu(rcu_preempt_data, cpu).qs_pending) |
c3422bea5 rcu: Simplify rcu... |
637 |
t->rcu_read_unlock_special |= RCU_READ_UNLOCK_NEED_QS; |
f41d911f8 rcu: Merge preemp... |
638 639 640 |
} /* |
6cc68793e rcu: fix spelling |
641 |
* Process callbacks for preemptible RCU. |
f41d911f8 rcu: Merge preemp... |
642 643 644 645 646 647 |
*/ static void rcu_preempt_process_callbacks(void) { __rcu_process_callbacks(&rcu_preempt_state, &__get_cpu_var(rcu_preempt_data)); } |
a46e0899e rcu: use softirq ... |
648 |
#ifdef CONFIG_RCU_BOOST |
09223371d rcu: Use softirq ... |
649 650 651 652 |
static void rcu_preempt_do_callbacks(void) { rcu_do_batch(&rcu_preempt_state, &__get_cpu_var(rcu_preempt_data)); } |
a46e0899e rcu: use softirq ... |
653 |
#endif /* #ifdef CONFIG_RCU_BOOST */ |
f41d911f8 rcu: Merge preemp... |
654 |
/* |
6cc68793e rcu: fix spelling |
655 |
* Queue a preemptible-RCU callback for invocation after a grace period. |
f41d911f8 rcu: Merge preemp... |
656 657 658 659 660 661 |
*/ void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) { __call_rcu(head, func, &rcu_preempt_state); } EXPORT_SYMBOL_GPL(call_rcu); |
6ebb237be rcu: Re-arrange c... |
662 663 664 665 666 |
/** * synchronize_rcu - wait until a grace period has elapsed. * * Control will return to the caller some time after a full grace * period has elapsed, in other words after all currently executing RCU |
77d8485a8 rcu: improve kern... |
667 668 669 670 671 |
* read-side critical sections have completed. Note, however, that * upon return from synchronize_rcu(), the caller might well be executing * concurrently with new RCU read-side critical sections that began while * synchronize_rcu() was waiting. RCU read-side critical sections are * delimited by rcu_read_lock() and rcu_read_unlock(), and may be nested. |
6ebb237be rcu: Re-arrange c... |
672 673 674 |
*/ void synchronize_rcu(void) { |
6ebb237be rcu: Re-arrange c... |
675 676 |
if (!rcu_scheduler_active) return; |
2c42818e9 rcu: Abstract com... |
677 |
wait_rcu_gp(call_rcu); |
6ebb237be rcu: Re-arrange c... |
678 679 |
} EXPORT_SYMBOL_GPL(synchronize_rcu); |
d9a3da069 rcu: Add expedite... |
680 681 682 683 684 685 686 687 688 689 690 691 |
static DECLARE_WAIT_QUEUE_HEAD(sync_rcu_preempt_exp_wq); static long sync_rcu_preempt_exp_count; static DEFINE_MUTEX(sync_rcu_preempt_exp_mutex); /* * Return non-zero if there are any tasks in RCU read-side critical * sections blocking the current preemptible-RCU expedited grace period. * If there is no preemptible-RCU expedited grace period currently in * progress, returns zero unconditionally. */ static int rcu_preempted_readers_exp(struct rcu_node *rnp) { |
12f5f524c rcu: merge TREE_P... |
692 |
return rnp->exp_tasks != NULL; |
d9a3da069 rcu: Add expedite... |
693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 |
} /* * return non-zero if there is no RCU expedited grace period in progress * for the specified rcu_node structure, in other words, if all CPUs and * tasks covered by the specified rcu_node structure have done their bit * for the current expedited grace period. Works only for preemptible * RCU -- other RCU implementation use other means. * * Caller must hold sync_rcu_preempt_exp_mutex. */ static int sync_rcu_preempt_exp_done(struct rcu_node *rnp) { return !rcu_preempted_readers_exp(rnp) && ACCESS_ONCE(rnp->expmask) == 0; } /* * Report the exit from RCU read-side critical section for the last task * that queued itself during or before the current expedited preemptible-RCU * grace period. This event is reported either to the rcu_node structure on * which the task was queued or to one of that rcu_node structure's ancestors, * recursively up the tree. (Calm down, calm down, we do the recursion * iteratively!) * |
b40d293eb rcu: Omit self-aw... |
718 719 720 |
* Most callers will set the "wake" flag, but the task initiating the * expedited grace period need not wake itself. * |
d9a3da069 rcu: Add expedite... |
721 722 |
* Caller must hold sync_rcu_preempt_exp_mutex. */ |
b40d293eb rcu: Omit self-aw... |
723 724 |
static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp, bool wake) |
d9a3da069 rcu: Add expedite... |
725 726 727 |
{ unsigned long flags; unsigned long mask; |
1304afb22 rcu: Convert to r... |
728 |
raw_spin_lock_irqsave(&rnp->lock, flags); |
d9a3da069 rcu: Add expedite... |
729 |
for (;;) { |
131906b00 rcu: decrease rcu... |
730 731 |
if (!sync_rcu_preempt_exp_done(rnp)) { raw_spin_unlock_irqrestore(&rnp->lock, flags); |
d9a3da069 rcu: Add expedite... |
732 |
break; |
131906b00 rcu: decrease rcu... |
733 |
} |
d9a3da069 rcu: Add expedite... |
734 |
if (rnp->parent == NULL) { |
131906b00 rcu: decrease rcu... |
735 |
raw_spin_unlock_irqrestore(&rnp->lock, flags); |
b40d293eb rcu: Omit self-aw... |
736 737 |
if (wake) wake_up(&sync_rcu_preempt_exp_wq); |
d9a3da069 rcu: Add expedite... |
738 739 740 |
break; } mask = rnp->grpmask; |
1304afb22 rcu: Convert to r... |
741 |
raw_spin_unlock(&rnp->lock); /* irqs remain disabled */ |
d9a3da069 rcu: Add expedite... |
742 |
rnp = rnp->parent; |
1304afb22 rcu: Convert to r... |
743 |
raw_spin_lock(&rnp->lock); /* irqs already disabled */ |
d9a3da069 rcu: Add expedite... |
744 745 |
rnp->expmask &= ~mask; } |
d9a3da069 rcu: Add expedite... |
746 747 748 749 750 751 752 753 754 755 756 757 |
} /* * Snapshot the tasks blocking the newly started preemptible-RCU expedited * grace period for the specified rcu_node structure. If there are no such * tasks, report it up the rcu_node hierarchy. * * Caller must hold sync_rcu_preempt_exp_mutex and rsp->onofflock. */ static void sync_rcu_preempt_exp_init(struct rcu_state *rsp, struct rcu_node *rnp) { |
1217ed1ba rcu: permit rcu_r... |
758 |
unsigned long flags; |
12f5f524c rcu: merge TREE_P... |
759 |
int must_wait = 0; |
d9a3da069 rcu: Add expedite... |
760 |
|
1217ed1ba rcu: permit rcu_r... |
761 762 763 764 |
raw_spin_lock_irqsave(&rnp->lock, flags); if (list_empty(&rnp->blkd_tasks)) raw_spin_unlock_irqrestore(&rnp->lock, flags); else { |
12f5f524c rcu: merge TREE_P... |
765 |
rnp->exp_tasks = rnp->blkd_tasks.next; |
1217ed1ba rcu: permit rcu_r... |
766 |
rcu_initiate_boost(rnp, flags); /* releases rnp->lock */ |
12f5f524c rcu: merge TREE_P... |
767 768 |
must_wait = 1; } |
d9a3da069 rcu: Add expedite... |
769 |
if (!must_wait) |
b40d293eb rcu: Omit self-aw... |
770 |
rcu_report_exp_rnp(rsp, rnp, false); /* Don't wake self. */ |
d9a3da069 rcu: Add expedite... |
771 |
} |
f41d911f8 rcu: Merge preemp... |
772 |
/* |
d9a3da069 rcu: Add expedite... |
773 774 |
* Wait for an rcu-preempt grace period, but expedite it. The basic idea * is to invoke synchronize_sched_expedited() to push all the tasks to |
12f5f524c rcu: merge TREE_P... |
775 |
* the ->blkd_tasks lists and wait for this list to drain. |
019129d59 rcu: Stopgap fix ... |
776 777 778 |
*/ void synchronize_rcu_expedited(void) { |
d9a3da069 rcu: Add expedite... |
779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 |
unsigned long flags; struct rcu_node *rnp; struct rcu_state *rsp = &rcu_preempt_state; long snap; int trycount = 0; smp_mb(); /* Caller's modifications seen first by other CPUs. */ snap = ACCESS_ONCE(sync_rcu_preempt_exp_count) + 1; smp_mb(); /* Above access cannot bleed into critical section. */ /* * Acquire lock, falling back to synchronize_rcu() if too many * lock-acquisition failures. Of course, if someone does the * expedited grace period for us, just leave. */ while (!mutex_trylock(&sync_rcu_preempt_exp_mutex)) { if (trycount++ < 10) udelay(trycount * num_online_cpus()); else { synchronize_rcu(); return; } if ((ACCESS_ONCE(sync_rcu_preempt_exp_count) - snap) > 0) goto mb_ret; /* Others did our work for us. */ } if ((ACCESS_ONCE(sync_rcu_preempt_exp_count) - snap) > 0) goto unlock_mb_ret; /* Others did our work for us. */ |
12f5f524c rcu: merge TREE_P... |
806 |
/* force all RCU readers onto ->blkd_tasks lists. */ |
d9a3da069 rcu: Add expedite... |
807 |
synchronize_sched_expedited(); |
1304afb22 rcu: Convert to r... |
808 |
raw_spin_lock_irqsave(&rsp->onofflock, flags); |
d9a3da069 rcu: Add expedite... |
809 810 811 |
/* Initialize ->expmask for all non-leaf rcu_node structures. */ rcu_for_each_nonleaf_node_breadth_first(rsp, rnp) { |
1304afb22 rcu: Convert to r... |
812 |
raw_spin_lock(&rnp->lock); /* irqs already disabled. */ |
d9a3da069 rcu: Add expedite... |
813 |
rnp->expmask = rnp->qsmaskinit; |
1304afb22 rcu: Convert to r... |
814 |
raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ |
d9a3da069 rcu: Add expedite... |
815 |
} |
12f5f524c rcu: merge TREE_P... |
816 |
/* Snapshot current state of ->blkd_tasks lists. */ |
d9a3da069 rcu: Add expedite... |
817 818 819 820 |
rcu_for_each_leaf_node(rsp, rnp) sync_rcu_preempt_exp_init(rsp, rnp); if (NUM_RCU_NODES > 1) sync_rcu_preempt_exp_init(rsp, rcu_get_root(rsp)); |
1304afb22 rcu: Convert to r... |
821 |
raw_spin_unlock_irqrestore(&rsp->onofflock, flags); |
d9a3da069 rcu: Add expedite... |
822 |
|
12f5f524c rcu: merge TREE_P... |
823 |
/* Wait for snapshotted ->blkd_tasks lists to drain. */ |
d9a3da069 rcu: Add expedite... |
824 825 826 827 828 829 830 831 832 833 834 |
rnp = rcu_get_root(rsp); wait_event(sync_rcu_preempt_exp_wq, sync_rcu_preempt_exp_done(rnp)); /* Clean up and exit. */ smp_mb(); /* ensure expedited GP seen before counter increment. */ ACCESS_ONCE(sync_rcu_preempt_exp_count)++; unlock_mb_ret: mutex_unlock(&sync_rcu_preempt_exp_mutex); mb_ret: smp_mb(); /* ensure subsequent action seen after grace period. */ |
019129d59 rcu: Stopgap fix ... |
835 836 837 838 |
} EXPORT_SYMBOL_GPL(synchronize_rcu_expedited); /* |
6cc68793e rcu: fix spelling |
839 |
* Check to see if there is any immediate preemptible-RCU-related work |
f41d911f8 rcu: Merge preemp... |
840 841 842 843 844 845 846 847 848 |
* to be done. */ static int rcu_preempt_pending(int cpu) { return __rcu_pending(&rcu_preempt_state, &per_cpu(rcu_preempt_data, cpu)); } /* |
6cc68793e rcu: fix spelling |
849 |
* Does preemptible RCU need the CPU to stay out of dynticks mode? |
f41d911f8 rcu: Merge preemp... |
850 851 852 853 854 |
*/ static int rcu_preempt_needs_cpu(int cpu) { return !!per_cpu(rcu_preempt_data, cpu).nxtlist; } |
e74f4c456 rcu: Make hot-unp... |
855 856 857 858 859 860 861 862 |
/** * rcu_barrier - Wait until all in-flight call_rcu() callbacks complete. */ void rcu_barrier(void) { _rcu_barrier(&rcu_preempt_state, call_rcu); } EXPORT_SYMBOL_GPL(rcu_barrier); |
f41d911f8 rcu: Merge preemp... |
863 |
/* |
6cc68793e rcu: fix spelling |
864 |
* Initialize preemptible RCU's per-CPU data. |
f41d911f8 rcu: Merge preemp... |
865 866 867 868 869 870 871 |
*/ static void __cpuinit rcu_preempt_init_percpu_data(int cpu) { rcu_init_percpu_data(cpu, &rcu_preempt_state, 1); } /* |
6cc68793e rcu: fix spelling |
872 |
* Move preemptible RCU's callbacks from dying CPU to other online CPU. |
e74f4c456 rcu: Make hot-unp... |
873 |
*/ |
29494be71 rcu,cleanup: simp... |
874 |
static void rcu_preempt_send_cbs_to_online(void) |
e74f4c456 rcu: Make hot-unp... |
875 |
{ |
29494be71 rcu,cleanup: simp... |
876 |
rcu_send_cbs_to_online(&rcu_preempt_state); |
e74f4c456 rcu: Make hot-unp... |
877 878 879 |
} /* |
6cc68793e rcu: fix spelling |
880 |
* Initialize preemptible RCU's state structures. |
1eba8f843 rcu: Clean up cod... |
881 882 883 |
*/ static void __init __rcu_init_preempt(void) { |
394f99a90 rcu: simplify the... |
884 |
rcu_init_one(&rcu_preempt_state, &rcu_preempt_data); |
1eba8f843 rcu: Clean up cod... |
885 886 887 |
} /* |
6cc68793e rcu: fix spelling |
888 |
* Check for a task exiting while in a preemptible-RCU read-side |
f41d911f8 rcu: Merge preemp... |
889 890 891 892 893 894 895 896 897 898 899 |
* critical section, clean up if so. No need to issue warnings, * as debug_check_no_locks_held() already does this if lockdep * is enabled. */ void exit_rcu(void) { struct task_struct *t = current; if (t->rcu_read_lock_nesting == 0) return; t->rcu_read_lock_nesting = 1; |
13491a0ee rcu: call __rcu_r... |
900 |
__rcu_read_unlock(); |
f41d911f8 rcu: Merge preemp... |
901 902 903 |
} #else /* #ifdef CONFIG_TREE_PREEMPT_RCU */ |
27f4d2805 rcu: priority boo... |
904 |
static struct rcu_state *rcu_state = &rcu_sched_state; |
f41d911f8 rcu: Merge preemp... |
905 906 907 |
/* * Tell them what RCU they are running. */ |
0e0fc1c23 rcu: Mark init-ti... |
908 |
static void __init rcu_bootup_announce(void) |
f41d911f8 rcu: Merge preemp... |
909 910 911 |
{ printk(KERN_INFO "Hierarchical RCU implementation. "); |
26845c286 rcu: print boot-t... |
912 |
rcu_bootup_announce_oddness(); |
f41d911f8 rcu: Merge preemp... |
913 914 915 916 917 918 919 920 921 922 923 924 |
} /* * Return the number of RCU batches processed thus far for debug & stats. */ long rcu_batches_completed(void) { return rcu_batches_completed_sched(); } EXPORT_SYMBOL_GPL(rcu_batches_completed); /* |
bf66f18e7 rcu: Add force_qu... |
925 926 927 928 929 930 931 932 933 934 |
* Force a quiescent state for RCU, which, because there is no preemptible * RCU, becomes the same as rcu-sched. */ void rcu_force_quiescent_state(void) { rcu_sched_force_quiescent_state(); } EXPORT_SYMBOL_GPL(rcu_force_quiescent_state); /* |
6cc68793e rcu: fix spelling |
935 |
* Because preemptible RCU does not exist, we never have to check for |
f41d911f8 rcu: Merge preemp... |
936 937 |
* CPUs being in quiescent states. */ |
c3422bea5 rcu: Simplify rcu... |
938 |
static void rcu_preempt_note_context_switch(int cpu) |
f41d911f8 rcu: Merge preemp... |
939 940 |
{ } |
fc2219d49 rcu: Clean up cod... |
941 |
/* |
6cc68793e rcu: fix spelling |
942 |
* Because preemptible RCU does not exist, there are never any preempted |
fc2219d49 rcu: Clean up cod... |
943 944 |
* RCU readers. */ |
27f4d2805 rcu: priority boo... |
945 |
static int rcu_preempt_blocked_readers_cgp(struct rcu_node *rnp) |
fc2219d49 rcu: Clean up cod... |
946 947 948 |
{ return 0; } |
b668c9cf3 rcu: Fix grace-pe... |
949 950 951 |
#ifdef CONFIG_HOTPLUG_CPU /* Because preemptible RCU does not exist, no quieting of tasks. */ |
d3f6bad39 rcu: Rename "quie... |
952 |
static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp, unsigned long flags) |
b668c9cf3 rcu: Fix grace-pe... |
953 |
{ |
1304afb22 rcu: Convert to r... |
954 |
raw_spin_unlock_irqrestore(&rnp->lock, flags); |
b668c9cf3 rcu: Fix grace-pe... |
955 956 957 |
} #endif /* #ifdef CONFIG_HOTPLUG_CPU */ |
f41d911f8 rcu: Merge preemp... |
958 |
/* |
6cc68793e rcu: fix spelling |
959 |
* Because preemptible RCU does not exist, we never have to check for |
f41d911f8 rcu: Merge preemp... |
960 961 |
* tasks blocked within RCU read-side critical sections. */ |
1ed509a22 rcu: Add RCU_CPU_... |
962 963 964 965 966 |
static void rcu_print_detail_task_stall(struct rcu_state *rsp) { } /* |
6cc68793e rcu: fix spelling |
967 |
* Because preemptible RCU does not exist, we never have to check for |
1ed509a22 rcu: Add RCU_CPU_... |
968 969 |
* tasks blocked within RCU read-side critical sections. */ |
9bc8b5586 rcu: Suppress NMI... |
970 |
static int rcu_print_task_stall(struct rcu_node *rnp) |
f41d911f8 rcu: Merge preemp... |
971 |
{ |
9bc8b5586 rcu: Suppress NMI... |
972 |
return 0; |
f41d911f8 rcu: Merge preemp... |
973 |
} |
53d84e004 rcu: permit suppr... |
974 975 976 977 978 979 980 |
/* * Because preemptible RCU does not exist, there is no need to suppress * its CPU stall warnings. */ static void rcu_preempt_stall_reset(void) { } |
f41d911f8 rcu: Merge preemp... |
981 |
/* |
6cc68793e rcu: fix spelling |
982 |
* Because there is no preemptible RCU, there can be no readers blocked, |
49e291266 rcu: Fix thinko, ... |
983 984 |
* so there is no need to check for blocked tasks. So check only for * bogus qsmask values. |
b0e165c03 rcu: Add debug ch... |
985 986 987 |
*/ static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp) { |
49e291266 rcu: Fix thinko, ... |
988 |
WARN_ON_ONCE(rnp->qsmask); |
b0e165c03 rcu: Add debug ch... |
989 |
} |
33f76148c rcu: Add CPU-offl... |
990 991 992 |
#ifdef CONFIG_HOTPLUG_CPU /* |
6cc68793e rcu: fix spelling |
993 |
* Because preemptible RCU does not exist, it never needs to migrate |
237c80c5c rcu: Fix TREE_PRE... |
994 995 996 |
* tasks that were blocked within RCU read-side critical sections, and * such non-existent tasks cannot possibly have been blocking the current * grace period. |
dd5d19baf rcu: Create rcutr... |
997 |
*/ |
237c80c5c rcu: Fix TREE_PRE... |
998 999 1000 |
static int rcu_preempt_offline_tasks(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) |
dd5d19baf rcu: Create rcutr... |
1001 |
{ |
237c80c5c rcu: Fix TREE_PRE... |
1002 |
return 0; |
dd5d19baf rcu: Create rcutr... |
1003 1004 1005 |
} /* |
6cc68793e rcu: fix spelling |
1006 |
* Because preemptible RCU does not exist, it never needs CPU-offline |
33f76148c rcu: Add CPU-offl... |
1007 1008 1009 1010 1011 1012 1013 |
* processing. */ static void rcu_preempt_offline_cpu(int cpu) { } #endif /* #ifdef CONFIG_HOTPLUG_CPU */ |
f41d911f8 rcu: Merge preemp... |
1014 |
/* |
6cc68793e rcu: fix spelling |
1015 |
* Because preemptible RCU does not exist, it never has any callbacks |
f41d911f8 rcu: Merge preemp... |
1016 1017 |
* to check. */ |
1eba8f843 rcu: Clean up cod... |
1018 |
static void rcu_preempt_check_callbacks(int cpu) |
f41d911f8 rcu: Merge preemp... |
1019 1020 1021 1022 |
{ } /* |
6cc68793e rcu: fix spelling |
1023 |
* Because preemptible RCU does not exist, it never has any callbacks |
f41d911f8 rcu: Merge preemp... |
1024 1025 |
* to process. */ |
1eba8f843 rcu: Clean up cod... |
1026 |
static void rcu_preempt_process_callbacks(void) |
f41d911f8 rcu: Merge preemp... |
1027 1028 1029 1030 |
{ } /* |
019129d59 rcu: Stopgap fix ... |
1031 |
* Wait for an rcu-preempt grace period, but make it happen quickly. |
6cc68793e rcu: fix spelling |
1032 |
* But because preemptible RCU does not exist, map to rcu-sched. |
019129d59 rcu: Stopgap fix ... |
1033 1034 1035 1036 1037 1038 |
*/ void synchronize_rcu_expedited(void) { synchronize_sched_expedited(); } EXPORT_SYMBOL_GPL(synchronize_rcu_expedited); |
d9a3da069 rcu: Add expedite... |
1039 1040 1041 |
#ifdef CONFIG_HOTPLUG_CPU /* |
6cc68793e rcu: fix spelling |
1042 |
* Because preemptible RCU does not exist, there is never any need to |
d9a3da069 rcu: Add expedite... |
1043 1044 1045 |
* report on tasks preempted in RCU read-side critical sections during * expedited RCU grace periods. */ |
b40d293eb rcu: Omit self-aw... |
1046 1047 |
static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp, bool wake) |
d9a3da069 rcu: Add expedite... |
1048 |
{ |
d9a3da069 rcu: Add expedite... |
1049 1050 1051 |
} #endif /* #ifdef CONFIG_HOTPLUG_CPU */ |
019129d59 rcu: Stopgap fix ... |
1052 |
/* |
6cc68793e rcu: fix spelling |
1053 |
* Because preemptible RCU does not exist, it never has any work to do. |
f41d911f8 rcu: Merge preemp... |
1054 1055 1056 1057 1058 1059 1060 |
*/ static int rcu_preempt_pending(int cpu) { return 0; } /* |
6cc68793e rcu: fix spelling |
1061 |
* Because preemptible RCU does not exist, it never needs any CPU. |
f41d911f8 rcu: Merge preemp... |
1062 1063 1064 1065 1066 1067 1068 |
*/ static int rcu_preempt_needs_cpu(int cpu) { return 0; } /* |
6cc68793e rcu: fix spelling |
1069 |
* Because preemptible RCU does not exist, rcu_barrier() is just |
e74f4c456 rcu: Make hot-unp... |
1070 1071 1072 1073 1074 1075 1076 1077 1078 |
* another name for rcu_barrier_sched(). */ void rcu_barrier(void) { rcu_barrier_sched(); } EXPORT_SYMBOL_GPL(rcu_barrier); /* |
6cc68793e rcu: fix spelling |
1079 |
* Because preemptible RCU does not exist, there is no per-CPU |
f41d911f8 rcu: Merge preemp... |
1080 1081 1082 1083 1084 |
* data to initialize. */ static void __cpuinit rcu_preempt_init_percpu_data(int cpu) { } |
1eba8f843 rcu: Clean up cod... |
1085 |
/* |
6cc68793e rcu: fix spelling |
1086 |
* Because there is no preemptible RCU, there are no callbacks to move. |
e74f4c456 rcu: Make hot-unp... |
1087 |
*/ |
29494be71 rcu,cleanup: simp... |
1088 |
static void rcu_preempt_send_cbs_to_online(void) |
e74f4c456 rcu: Make hot-unp... |
1089 1090 1091 1092 |
{ } /* |
6cc68793e rcu: fix spelling |
1093 |
* Because preemptible RCU does not exist, it need not be initialized. |
1eba8f843 rcu: Clean up cod... |
1094 1095 1096 1097 |
*/ static void __init __rcu_init_preempt(void) { } |
f41d911f8 rcu: Merge preemp... |
1098 |
#endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */ |
8bd93a2c5 rcu: Accelerate g... |
1099 |
|
27f4d2805 rcu: priority boo... |
1100 1101 1102 |
#ifdef CONFIG_RCU_BOOST #include "rtmutex_common.h" |
0ea1f2ebe rcu: Add boosting... |
1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 |
#ifdef CONFIG_RCU_TRACE static void rcu_initiate_boost_trace(struct rcu_node *rnp) { if (list_empty(&rnp->blkd_tasks)) rnp->n_balk_blkd_tasks++; else if (rnp->exp_tasks == NULL && rnp->gp_tasks == NULL) rnp->n_balk_exp_gp_tasks++; else if (rnp->gp_tasks != NULL && rnp->boost_tasks != NULL) rnp->n_balk_boost_tasks++; else if (rnp->gp_tasks != NULL && rnp->qsmask != 0) rnp->n_balk_notblocked++; else if (rnp->gp_tasks != NULL && |
a9f4793d8 rcu: fix tracing ... |
1116 |
ULONG_CMP_LT(jiffies, rnp->boost_time)) |
0ea1f2ebe rcu: Add boosting... |
1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 |
rnp->n_balk_notyet++; else rnp->n_balk_nos++; } #else /* #ifdef CONFIG_RCU_TRACE */ static void rcu_initiate_boost_trace(struct rcu_node *rnp) { } #endif /* #else #ifdef CONFIG_RCU_TRACE */ |
27f4d2805 rcu: priority boo... |
1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 |
/* * Carry out RCU priority boosting on the task indicated by ->exp_tasks * or ->boost_tasks, advancing the pointer to the next task in the * ->blkd_tasks list. * * Note that irqs must be enabled: boosting the task can block. * Returns 1 if there are more tasks needing to be boosted. */ static int rcu_boost(struct rcu_node *rnp) { unsigned long flags; struct rt_mutex mtx; struct task_struct *t; struct list_head *tb; if (rnp->exp_tasks == NULL && rnp->boost_tasks == NULL) return 0; /* Nothing left to boost. */ raw_spin_lock_irqsave(&rnp->lock, flags); /* * Recheck under the lock: all tasks in need of boosting * might exit their RCU read-side critical sections on their own. */ if (rnp->exp_tasks == NULL && rnp->boost_tasks == NULL) { raw_spin_unlock_irqrestore(&rnp->lock, flags); return 0; } /* * Preferentially boost tasks blocking expedited grace periods. * This cannot starve the normal grace periods because a second * expedited grace period must boost all blocked tasks, including * those blocking the pre-existing normal grace period. */ |
0ea1f2ebe rcu: Add boosting... |
1164 |
if (rnp->exp_tasks != NULL) { |
27f4d2805 rcu: priority boo... |
1165 |
tb = rnp->exp_tasks; |
0ea1f2ebe rcu: Add boosting... |
1166 1167 |
rnp->n_exp_boosts++; } else { |
27f4d2805 rcu: priority boo... |
1168 |
tb = rnp->boost_tasks; |
0ea1f2ebe rcu: Add boosting... |
1169 1170 1171 |
rnp->n_normal_boosts++; } rnp->n_tasks_boosted++; |
27f4d2805 rcu: priority boo... |
1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 |
/* * We boost task t by manufacturing an rt_mutex that appears to * be held by task t. We leave a pointer to that rt_mutex where * task t can find it, and task t will release the mutex when it * exits its outermost RCU read-side critical section. Then * simply acquiring this artificial rt_mutex will boost task * t's priority. (Thanks to tglx for suggesting this approach!) * * Note that task t must acquire rnp->lock to remove itself from * the ->blkd_tasks list, which it will do from exit() if from * nowhere else. We therefore are guaranteed that task t will * stay around at least until we drop rnp->lock. Note that * rnp->lock also resolves races between our priority boosting * and task t's exiting its outermost RCU read-side critical * section. */ t = container_of(tb, struct task_struct, rcu_node_entry); rt_mutex_init_proxy_locked(&mtx, t); t->rcu_boost_mutex = &mtx; |
27f4d2805 rcu: priority boo... |
1192 1193 1194 |
raw_spin_unlock_irqrestore(&rnp->lock, flags); rt_mutex_lock(&mtx); /* Side effect: boosts task t's priority. */ rt_mutex_unlock(&mtx); /* Keep lockdep happy. */ |
4f89b336f rcu: Apply ACCESS... |
1195 1196 |
return ACCESS_ONCE(rnp->exp_tasks) != NULL || ACCESS_ONCE(rnp->boost_tasks) != NULL; |
27f4d2805 rcu: priority boo... |
1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 |
} /* * Timer handler to initiate waking up of boost kthreads that * have yielded the CPU due to excessive numbers of tasks to * boost. We wake up the per-rcu_node kthread, which in turn * will wake up the booster kthread. */ static void rcu_boost_kthread_timer(unsigned long arg) { |
1217ed1ba rcu: permit rcu_r... |
1207 |
invoke_rcu_node_kthread((struct rcu_node *)arg); |
27f4d2805 rcu: priority boo... |
1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 |
} /* * Priority-boosting kthread. One per leaf rcu_node and one for the * root rcu_node. */ static int rcu_boost_kthread(void *arg) { struct rcu_node *rnp = (struct rcu_node *)arg; int spincnt = 0; int more2boost; |
385680a94 rcu: Add event-tr... |
1219 |
trace_rcu_utilization("Start boost kthread@init"); |
27f4d2805 rcu: priority boo... |
1220 |
for (;;) { |
d71df90ea rcu: add tracing ... |
1221 |
rnp->boost_kthread_status = RCU_KTHREAD_WAITING; |
385680a94 rcu: Add event-tr... |
1222 |
trace_rcu_utilization("End boost kthread@rcu_wait"); |
08bca60a6 rcu: Remove waitq... |
1223 |
rcu_wait(rnp->boost_tasks || rnp->exp_tasks); |
385680a94 rcu: Add event-tr... |
1224 |
trace_rcu_utilization("Start boost kthread@rcu_wait"); |
d71df90ea rcu: add tracing ... |
1225 |
rnp->boost_kthread_status = RCU_KTHREAD_RUNNING; |
27f4d2805 rcu: priority boo... |
1226 1227 1228 1229 1230 1231 |
more2boost = rcu_boost(rnp); if (more2boost) spincnt++; else spincnt = 0; if (spincnt > 10) { |
385680a94 rcu: Add event-tr... |
1232 |
trace_rcu_utilization("End boost kthread@rcu_yield"); |
27f4d2805 rcu: priority boo... |
1233 |
rcu_yield(rcu_boost_kthread_timer, (unsigned long)rnp); |
385680a94 rcu: Add event-tr... |
1234 |
trace_rcu_utilization("Start boost kthread@rcu_yield"); |
27f4d2805 rcu: priority boo... |
1235 1236 1237 |
spincnt = 0; } } |
1217ed1ba rcu: permit rcu_r... |
1238 |
/* NOTREACHED */ |
385680a94 rcu: Add event-tr... |
1239 |
trace_rcu_utilization("End boost kthread@notreached"); |
27f4d2805 rcu: priority boo... |
1240 1241 1242 1243 1244 1245 1246 1247 1248 |
return 0; } /* * Check to see if it is time to start boosting RCU readers that are * blocking the current grace period, and, if so, tell the per-rcu_node * kthread to start boosting them. If there is an expedited grace * period in progress, it is always time to boost. * |
1217ed1ba rcu: permit rcu_r... |
1249 1250 1251 |
* The caller must hold rnp->lock, which this function releases, * but irqs remain disabled. The ->boost_kthread_task is immortal, * so we don't need to worry about it going away. |
27f4d2805 rcu: priority boo... |
1252 |
*/ |
1217ed1ba rcu: permit rcu_r... |
1253 |
static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags) |
27f4d2805 rcu: priority boo... |
1254 1255 |
{ struct task_struct *t; |
0ea1f2ebe rcu: Add boosting... |
1256 1257 |
if (!rcu_preempt_blocked_readers_cgp(rnp) && rnp->exp_tasks == NULL) { rnp->n_balk_exp_gp_tasks++; |
1217ed1ba rcu: permit rcu_r... |
1258 |
raw_spin_unlock_irqrestore(&rnp->lock, flags); |
27f4d2805 rcu: priority boo... |
1259 |
return; |
0ea1f2ebe rcu: Add boosting... |
1260 |
} |
27f4d2805 rcu: priority boo... |
1261 1262 1263 1264 1265 1266 1267 |
if (rnp->exp_tasks != NULL || (rnp->gp_tasks != NULL && rnp->boost_tasks == NULL && rnp->qsmask == 0 && ULONG_CMP_GE(jiffies, rnp->boost_time))) { if (rnp->exp_tasks == NULL) rnp->boost_tasks = rnp->gp_tasks; |
1217ed1ba rcu: permit rcu_r... |
1268 |
raw_spin_unlock_irqrestore(&rnp->lock, flags); |
27f4d2805 rcu: priority boo... |
1269 1270 1271 |
t = rnp->boost_kthread_task; if (t != NULL) wake_up_process(t); |
1217ed1ba rcu: permit rcu_r... |
1272 |
} else { |
0ea1f2ebe rcu: Add boosting... |
1273 |
rcu_initiate_boost_trace(rnp); |
1217ed1ba rcu: permit rcu_r... |
1274 1275 |
raw_spin_unlock_irqrestore(&rnp->lock, flags); } |
27f4d2805 rcu: priority boo... |
1276 |
} |
0f962a5e7 rcu: Force per-rc... |
1277 |
/* |
a46e0899e rcu: use softirq ... |
1278 1279 1280 1281 1282 1283 1284 1285 |
* Wake up the per-CPU kthread to invoke RCU callbacks. */ static void invoke_rcu_callbacks_kthread(void) { unsigned long flags; local_irq_save(flags); __this_cpu_write(rcu_cpu_has_work, 1); |
1eb521210 rcu: Avoid unnece... |
1286 1287 1288 |
if (__this_cpu_read(rcu_cpu_kthread_task) != NULL && current != __this_cpu_read(rcu_cpu_kthread_task)) wake_up_process(__this_cpu_read(rcu_cpu_kthread_task)); |
a46e0899e rcu: use softirq ... |
1289 1290 1291 1292 |
local_irq_restore(flags); } /* |
dff1672d9 rcu: Keep invokin... |
1293 1294 1295 1296 1297 1298 1299 1300 1301 |
* Is the current CPU running the RCU-callbacks kthread? * Caller must have preemption disabled. */ static bool rcu_is_callbacks_kthread(void) { return __get_cpu_var(rcu_cpu_kthread_task) == current; } /* |
0f962a5e7 rcu: Force per-rc... |
1302 1303 1304 1305 |
* Set the affinity of the boost kthread. The CPU-hotplug locks are * held, so no one should be messing with the existence of the boost * kthread. */ |
27f4d2805 rcu: priority boo... |
1306 1307 1308 |
static void rcu_boost_kthread_setaffinity(struct rcu_node *rnp, cpumask_var_t cm) { |
27f4d2805 rcu: priority boo... |
1309 |
struct task_struct *t; |
27f4d2805 rcu: priority boo... |
1310 1311 1312 |
t = rnp->boost_kthread_task; if (t != NULL) set_cpus_allowed_ptr(rnp->boost_kthread_task, cm); |
27f4d2805 rcu: priority boo... |
1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 |
} #define RCU_BOOST_DELAY_JIFFIES DIV_ROUND_UP(CONFIG_RCU_BOOST_DELAY * HZ, 1000) /* * Do priority-boost accounting for the start of a new grace period. */ static void rcu_preempt_boost_start_gp(struct rcu_node *rnp) { rnp->boost_time = jiffies + RCU_BOOST_DELAY_JIFFIES; } /* |
27f4d2805 rcu: priority boo... |
1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 |
* Create an RCU-boost kthread for the specified node if one does not * already exist. We only create this kthread for preemptible RCU. * Returns zero if all is well, a negated errno otherwise. */ static int __cpuinit rcu_spawn_one_boost_kthread(struct rcu_state *rsp, struct rcu_node *rnp, int rnp_index) { unsigned long flags; struct sched_param sp; struct task_struct *t; if (&rcu_preempt_state != rsp) return 0; |
a46e0899e rcu: use softirq ... |
1340 |
rsp->boost = 1; |
27f4d2805 rcu: priority boo... |
1341 1342 1343 |
if (rnp->boost_kthread_task != NULL) return 0; t = kthread_create(rcu_boost_kthread, (void *)rnp, |
5b61b0baa rcu: Wire up RCU_... |
1344 |
"rcub/%d", rnp_index); |
27f4d2805 rcu: priority boo... |
1345 1346 1347 1348 1349 |
if (IS_ERR(t)) return PTR_ERR(t); raw_spin_lock_irqsave(&rnp->lock, flags); rnp->boost_kthread_task = t; raw_spin_unlock_irqrestore(&rnp->lock, flags); |
5b61b0baa rcu: Wire up RCU_... |
1350 |
sp.sched_priority = RCU_BOOST_PRIO; |
27f4d2805 rcu: priority boo... |
1351 |
sched_setscheduler_nocheck(t, SCHED_FIFO, &sp); |
9a4327369 rcu: Simplify cur... |
1352 |
wake_up_process(t); /* get to TASK_INTERRUPTIBLE quickly. */ |
27f4d2805 rcu: priority boo... |
1353 1354 |
return 0; } |
f8b7fc6b5 rcu: Move RCU_BOO... |
1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 |
#ifdef CONFIG_HOTPLUG_CPU /* * Stop the RCU's per-CPU kthread when its CPU goes offline,. */ static void rcu_stop_cpu_kthread(int cpu) { struct task_struct *t; /* Stop the CPU's kthread. */ t = per_cpu(rcu_cpu_kthread_task, cpu); if (t != NULL) { per_cpu(rcu_cpu_kthread_task, cpu) = NULL; kthread_stop(t); } } #endif /* #ifdef CONFIG_HOTPLUG_CPU */ static void rcu_kthread_do_work(void) { rcu_do_batch(&rcu_sched_state, &__get_cpu_var(rcu_sched_data)); rcu_do_batch(&rcu_bh_state, &__get_cpu_var(rcu_bh_data)); rcu_preempt_do_callbacks(); } /* * Wake up the specified per-rcu_node-structure kthread. * Because the per-rcu_node kthreads are immortal, we don't need * to do anything to keep them alive. */ static void invoke_rcu_node_kthread(struct rcu_node *rnp) { struct task_struct *t; t = rnp->node_kthread_task; if (t != NULL) wake_up_process(t); } /* * Set the specified CPU's kthread to run RT or not, as specified by * the to_rt argument. The CPU-hotplug locks are held, so the task * is not going away. */ static void rcu_cpu_kthread_setrt(int cpu, int to_rt) { int policy; struct sched_param sp; struct task_struct *t; t = per_cpu(rcu_cpu_kthread_task, cpu); if (t == NULL) return; if (to_rt) { policy = SCHED_FIFO; sp.sched_priority = RCU_KTHREAD_PRIO; } else { policy = SCHED_NORMAL; sp.sched_priority = 0; } sched_setscheduler_nocheck(t, policy, &sp); } /* * Timer handler to initiate the waking up of per-CPU kthreads that * have yielded the CPU due to excess numbers of RCU callbacks. * We wake up the per-rcu_node kthread, which in turn will wake up * the booster kthread. */ static void rcu_cpu_kthread_timer(unsigned long arg) { struct rcu_data *rdp = per_cpu_ptr(rcu_state->rda, arg); struct rcu_node *rnp = rdp->mynode; atomic_or(rdp->grpmask, &rnp->wakemask); invoke_rcu_node_kthread(rnp); } /* * Drop to non-real-time priority and yield, but only after posting a * timer that will cause us to regain our real-time priority if we * remain preempted. Either way, we restore our real-time priority * before returning. */ static void rcu_yield(void (*f)(unsigned long), unsigned long arg) { struct sched_param sp; struct timer_list yield_timer; |
5b61b0baa rcu: Wire up RCU_... |
1444 |
int prio = current->rt_priority; |
f8b7fc6b5 rcu: Move RCU_BOO... |
1445 1446 1447 1448 1449 1450 1451 |
setup_timer_on_stack(&yield_timer, f, arg); mod_timer(&yield_timer, jiffies + 2); sp.sched_priority = 0; sched_setscheduler_nocheck(current, SCHED_NORMAL, &sp); set_user_nice(current, 19); schedule(); |
5b61b0baa rcu: Wire up RCU_... |
1452 1453 |
set_user_nice(current, 0); sp.sched_priority = prio; |
f8b7fc6b5 rcu: Move RCU_BOO... |
1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 |
sched_setscheduler_nocheck(current, SCHED_FIFO, &sp); del_timer(&yield_timer); } /* * Handle cases where the rcu_cpu_kthread() ends up on the wrong CPU. * This can happen while the corresponding CPU is either coming online * or going offline. We cannot wait until the CPU is fully online * before starting the kthread, because the various notifier functions * can wait for RCU grace periods. So we park rcu_cpu_kthread() until * the corresponding CPU is online. * * Return 1 if the kthread needs to stop, 0 otherwise. * * Caller must disable bh. This function can momentarily enable it. */ static int rcu_cpu_kthread_should_stop(int cpu) { while (cpu_is_offline(cpu) || !cpumask_equal(¤t->cpus_allowed, cpumask_of(cpu)) || smp_processor_id() != cpu) { if (kthread_should_stop()) return 1; per_cpu(rcu_cpu_kthread_status, cpu) = RCU_KTHREAD_OFFCPU; per_cpu(rcu_cpu_kthread_cpu, cpu) = raw_smp_processor_id(); local_bh_enable(); schedule_timeout_uninterruptible(1); if (!cpumask_equal(¤t->cpus_allowed, cpumask_of(cpu))) set_cpus_allowed_ptr(current, cpumask_of(cpu)); local_bh_disable(); } per_cpu(rcu_cpu_kthread_cpu, cpu) = cpu; return 0; } /* * Per-CPU kernel thread that invokes RCU callbacks. This replaces the |
e0f23060a rcu: Update comme... |
1491 1492 |
* RCU softirq used in flavors and configurations of RCU that do not * support RCU priority boosting. |
f8b7fc6b5 rcu: Move RCU_BOO... |
1493 1494 1495 1496 1497 1498 1499 1500 1501 |
*/ static int rcu_cpu_kthread(void *arg) { int cpu = (int)(long)arg; unsigned long flags; int spincnt = 0; unsigned int *statusp = &per_cpu(rcu_cpu_kthread_status, cpu); char work; char *workp = &per_cpu(rcu_cpu_has_work, cpu); |
385680a94 rcu: Add event-tr... |
1502 |
trace_rcu_utilization("Start CPU kthread@init"); |
f8b7fc6b5 rcu: Move RCU_BOO... |
1503 1504 |
for (;;) { *statusp = RCU_KTHREAD_WAITING; |
385680a94 rcu: Add event-tr... |
1505 |
trace_rcu_utilization("End CPU kthread@rcu_wait"); |
f8b7fc6b5 rcu: Move RCU_BOO... |
1506 |
rcu_wait(*workp != 0 || kthread_should_stop()); |
385680a94 rcu: Add event-tr... |
1507 |
trace_rcu_utilization("Start CPU kthread@rcu_wait"); |
f8b7fc6b5 rcu: Move RCU_BOO... |
1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 |
local_bh_disable(); if (rcu_cpu_kthread_should_stop(cpu)) { local_bh_enable(); break; } *statusp = RCU_KTHREAD_RUNNING; per_cpu(rcu_cpu_kthread_loops, cpu)++; local_irq_save(flags); work = *workp; *workp = 0; local_irq_restore(flags); if (work) rcu_kthread_do_work(); local_bh_enable(); if (*workp != 0) spincnt++; else spincnt = 0; if (spincnt > 10) { *statusp = RCU_KTHREAD_YIELDING; |
385680a94 rcu: Add event-tr... |
1528 |
trace_rcu_utilization("End CPU kthread@rcu_yield"); |
f8b7fc6b5 rcu: Move RCU_BOO... |
1529 |
rcu_yield(rcu_cpu_kthread_timer, (unsigned long)cpu); |
385680a94 rcu: Add event-tr... |
1530 |
trace_rcu_utilization("Start CPU kthread@rcu_yield"); |
f8b7fc6b5 rcu: Move RCU_BOO... |
1531 1532 1533 1534 |
spincnt = 0; } } *statusp = RCU_KTHREAD_STOPPED; |
385680a94 rcu: Add event-tr... |
1535 |
trace_rcu_utilization("End CPU kthread@term"); |
f8b7fc6b5 rcu: Move RCU_BOO... |
1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 |
return 0; } /* * Spawn a per-CPU kthread, setting up affinity and priority. * Because the CPU hotplug lock is held, no other CPU will be attempting * to manipulate rcu_cpu_kthread_task. There might be another CPU * attempting to access it during boot, but the locking in kthread_bind() * will enforce sufficient ordering. * * Please note that we cannot simply refuse to wake up the per-CPU * kthread because kthreads are created in TASK_UNINTERRUPTIBLE state, * which can result in softlockup complaints if the task ends up being * idle for more than a couple of minutes. * * However, please note also that we cannot bind the per-CPU kthread to its * CPU until that CPU is fully online. We also cannot wait until the * CPU is fully online before we create its per-CPU kthread, as this would * deadlock the system when CPU notifiers tried waiting for grace * periods. So we bind the per-CPU kthread to its CPU only if the CPU * is online. If its CPU is not yet fully online, then the code in * rcu_cpu_kthread() will wait until it is fully online, and then do * the binding. */ static int __cpuinit rcu_spawn_one_cpu_kthread(int cpu) { struct sched_param sp; struct task_struct *t; |
b0d304172 rcu: Prevent RCU ... |
1564 |
if (!rcu_scheduler_fully_active || |
f8b7fc6b5 rcu: Move RCU_BOO... |
1565 1566 |
per_cpu(rcu_cpu_kthread_task, cpu) != NULL) return 0; |
1f2880948 rcu: Use kthread_... |
1567 1568 1569 |
t = kthread_create_on_node(rcu_cpu_kthread, (void *)(long)cpu, cpu_to_node(cpu), |
5b61b0baa rcu: Wire up RCU_... |
1570 |
"rcuc/%d", cpu); |
f8b7fc6b5 rcu: Move RCU_BOO... |
1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 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 1612 1613 1614 1615 1616 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 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 |
if (IS_ERR(t)) return PTR_ERR(t); if (cpu_online(cpu)) kthread_bind(t, cpu); per_cpu(rcu_cpu_kthread_cpu, cpu) = cpu; WARN_ON_ONCE(per_cpu(rcu_cpu_kthread_task, cpu) != NULL); sp.sched_priority = RCU_KTHREAD_PRIO; sched_setscheduler_nocheck(t, SCHED_FIFO, &sp); per_cpu(rcu_cpu_kthread_task, cpu) = t; wake_up_process(t); /* Get to TASK_INTERRUPTIBLE quickly. */ return 0; } /* * Per-rcu_node kthread, which is in charge of waking up the per-CPU * kthreads when needed. We ignore requests to wake up kthreads * for offline CPUs, which is OK because force_quiescent_state() * takes care of this case. */ static int rcu_node_kthread(void *arg) { int cpu; unsigned long flags; unsigned long mask; struct rcu_node *rnp = (struct rcu_node *)arg; struct sched_param sp; struct task_struct *t; for (;;) { rnp->node_kthread_status = RCU_KTHREAD_WAITING; rcu_wait(atomic_read(&rnp->wakemask) != 0); rnp->node_kthread_status = RCU_KTHREAD_RUNNING; raw_spin_lock_irqsave(&rnp->lock, flags); mask = atomic_xchg(&rnp->wakemask, 0); rcu_initiate_boost(rnp, flags); /* releases rnp->lock. */ for (cpu = rnp->grplo; cpu <= rnp->grphi; cpu++, mask >>= 1) { if ((mask & 0x1) == 0) continue; preempt_disable(); t = per_cpu(rcu_cpu_kthread_task, cpu); if (!cpu_online(cpu) || t == NULL) { preempt_enable(); continue; } per_cpu(rcu_cpu_has_work, cpu) = 1; sp.sched_priority = RCU_KTHREAD_PRIO; sched_setscheduler_nocheck(t, SCHED_FIFO, &sp); preempt_enable(); } } /* NOTREACHED */ rnp->node_kthread_status = RCU_KTHREAD_STOPPED; return 0; } /* * Set the per-rcu_node kthread's affinity to cover all CPUs that are * served by the rcu_node in question. The CPU hotplug lock is still * held, so the value of rnp->qsmaskinit will be stable. * * We don't include outgoingcpu in the affinity set, use -1 if there is * no outgoing CPU. If there are no CPUs left in the affinity set, * this function allows the kthread to execute on any CPU. */ static void rcu_node_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu) { cpumask_var_t cm; int cpu; unsigned long mask = rnp->qsmaskinit; if (rnp->node_kthread_task == NULL) return; if (!alloc_cpumask_var(&cm, GFP_KERNEL)) return; cpumask_clear(cm); for (cpu = rnp->grplo; cpu <= rnp->grphi; cpu++, mask >>= 1) if ((mask & 0x1) && cpu != outgoingcpu) cpumask_set_cpu(cpu, cm); if (cpumask_weight(cm) == 0) { cpumask_setall(cm); for (cpu = rnp->grplo; cpu <= rnp->grphi; cpu++) cpumask_clear_cpu(cpu, cm); WARN_ON_ONCE(cpumask_weight(cm) == 0); } set_cpus_allowed_ptr(rnp->node_kthread_task, cm); rcu_boost_kthread_setaffinity(rnp, cm); free_cpumask_var(cm); } /* * Spawn a per-rcu_node kthread, setting priority and affinity. * Called during boot before online/offline can happen, or, if * during runtime, with the main CPU-hotplug locks held. So only * one of these can be executing at a time. */ static int __cpuinit rcu_spawn_one_node_kthread(struct rcu_state *rsp, struct rcu_node *rnp) { unsigned long flags; int rnp_index = rnp - &rsp->node[0]; struct sched_param sp; struct task_struct *t; |
b0d304172 rcu: Prevent RCU ... |
1673 |
if (!rcu_scheduler_fully_active || |
f8b7fc6b5 rcu: Move RCU_BOO... |
1674 1675 1676 1677 |
rnp->qsmaskinit == 0) return 0; if (rnp->node_kthread_task == NULL) { t = kthread_create(rcu_node_kthread, (void *)rnp, |
5b61b0baa rcu: Wire up RCU_... |
1678 |
"rcun/%d", rnp_index); |
f8b7fc6b5 rcu: Move RCU_BOO... |
1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 |
if (IS_ERR(t)) return PTR_ERR(t); raw_spin_lock_irqsave(&rnp->lock, flags); rnp->node_kthread_task = t; raw_spin_unlock_irqrestore(&rnp->lock, flags); sp.sched_priority = 99; sched_setscheduler_nocheck(t, SCHED_FIFO, &sp); wake_up_process(t); /* get to TASK_INTERRUPTIBLE quickly. */ } return rcu_spawn_one_boost_kthread(rsp, rnp, rnp_index); } /* * Spawn all kthreads -- called as soon as the scheduler is running. */ static int __init rcu_spawn_kthreads(void) { int cpu; struct rcu_node *rnp; |
b0d304172 rcu: Prevent RCU ... |
1698 |
rcu_scheduler_fully_active = 1; |
f8b7fc6b5 rcu: Move RCU_BOO... |
1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 |
for_each_possible_cpu(cpu) { per_cpu(rcu_cpu_has_work, cpu) = 0; if (cpu_online(cpu)) (void)rcu_spawn_one_cpu_kthread(cpu); } rnp = rcu_get_root(rcu_state); (void)rcu_spawn_one_node_kthread(rcu_state, rnp); if (NUM_RCU_NODES > 1) { rcu_for_each_leaf_node(rcu_state, rnp) (void)rcu_spawn_one_node_kthread(rcu_state, rnp); } return 0; } early_initcall(rcu_spawn_kthreads); static void __cpuinit rcu_prepare_kthreads(int cpu) { struct rcu_data *rdp = per_cpu_ptr(rcu_state->rda, cpu); struct rcu_node *rnp = rdp->mynode; /* Fire up the incoming CPU's kthread and leaf rcu_node kthread. */ |
b0d304172 rcu: Prevent RCU ... |
1720 |
if (rcu_scheduler_fully_active) { |
f8b7fc6b5 rcu: Move RCU_BOO... |
1721 1722 1723 1724 1725 |
(void)rcu_spawn_one_cpu_kthread(cpu); if (rnp->node_kthread_task == NULL) (void)rcu_spawn_one_node_kthread(rcu_state, rnp); } } |
27f4d2805 rcu: priority boo... |
1726 |
#else /* #ifdef CONFIG_RCU_BOOST */ |
1217ed1ba rcu: permit rcu_r... |
1727 |
static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags) |
27f4d2805 rcu: priority boo... |
1728 |
{ |
1217ed1ba rcu: permit rcu_r... |
1729 |
raw_spin_unlock_irqrestore(&rnp->lock, flags); |
27f4d2805 rcu: priority boo... |
1730 |
} |
a46e0899e rcu: use softirq ... |
1731 |
static void invoke_rcu_callbacks_kthread(void) |
27f4d2805 rcu: priority boo... |
1732 |
{ |
a46e0899e rcu: use softirq ... |
1733 |
WARN_ON_ONCE(1); |
27f4d2805 rcu: priority boo... |
1734 |
} |
dff1672d9 rcu: Keep invokin... |
1735 1736 1737 1738 |
static bool rcu_is_callbacks_kthread(void) { return false; } |
27f4d2805 rcu: priority boo... |
1739 1740 1741 |
static void rcu_preempt_boost_start_gp(struct rcu_node *rnp) { } |
f8b7fc6b5 rcu: Move RCU_BOO... |
1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 |
#ifdef CONFIG_HOTPLUG_CPU static void rcu_stop_cpu_kthread(int cpu) { } #endif /* #ifdef CONFIG_HOTPLUG_CPU */ static void rcu_node_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu) { } static void rcu_cpu_kthread_setrt(int cpu, int to_rt) { } |
b0d304172 rcu: Prevent RCU ... |
1757 1758 1759 1760 1761 1762 |
static int __init rcu_scheduler_really_started(void) { rcu_scheduler_fully_active = 1; return 0; } early_initcall(rcu_scheduler_really_started); |
f8b7fc6b5 rcu: Move RCU_BOO... |
1763 1764 1765 |
static void __cpuinit rcu_prepare_kthreads(int cpu) { } |
27f4d2805 rcu: priority boo... |
1766 |
#endif /* #else #ifdef CONFIG_RCU_BOOST */ |
7b27d5475 rcu,cleanup: move... |
1767 1768 1769 1770 1771 1772 1773 1774 1775 |
#ifndef CONFIG_SMP void synchronize_sched_expedited(void) { cond_resched(); } EXPORT_SYMBOL_GPL(synchronize_sched_expedited); #else /* #ifndef CONFIG_SMP */ |
e27fc9641 rcu: increase syn... |
1776 1777 |
static atomic_t sync_sched_expedited_started = ATOMIC_INIT(0); static atomic_t sync_sched_expedited_done = ATOMIC_INIT(0); |
7b27d5475 rcu,cleanup: move... |
1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 |
static int synchronize_sched_expedited_cpu_stop(void *data) { /* * There must be a full memory barrier on each affected CPU * between the time that try_stop_cpus() is called and the * time that it returns. * * In the current initial implementation of cpu_stop, the * above condition is already met when the control reaches * this point and the following smp_mb() is not strictly * necessary. Do smp_mb() anyway for documentation and * robustness against future implementation changes. */ smp_mb(); /* See above comment block. */ return 0; } /* * Wait for an rcu-sched grace period to elapse, but use "big hammer" * approach to force grace period to end quickly. This consumes * significant time on all CPUs, and is thus not recommended for * any sort of common-case code. * * Note that it is illegal to call this function while holding any * lock that is acquired by a CPU-hotplug notifier. Failing to * observe this restriction will result in deadlock. |
db3a89209 rcu: fix race con... |
1805 |
* |
e27fc9641 rcu: increase syn... |
1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 |
* This implementation can be thought of as an application of ticket * locking to RCU, with sync_sched_expedited_started and * sync_sched_expedited_done taking on the roles of the halves * of the ticket-lock word. Each task atomically increments * sync_sched_expedited_started upon entry, snapshotting the old value, * then attempts to stop all the CPUs. If this succeeds, then each * CPU will have executed a context switch, resulting in an RCU-sched * grace period. We are then done, so we use atomic_cmpxchg() to * update sync_sched_expedited_done to match our snapshot -- but * only if someone else has not already advanced past our snapshot. * * On the other hand, if try_stop_cpus() fails, we check the value * of sync_sched_expedited_done. If it has advanced past our * initial snapshot, then someone else must have forced a grace period * some time after we took our snapshot. In this case, our work is * done for us, and we can simply return. Otherwise, we try again, * but keep our initial snapshot for purposes of checking for someone * doing our work for us. * * If we fail too many times in a row, we fall back to synchronize_sched(). |
7b27d5475 rcu,cleanup: move... |
1826 1827 1828 |
*/ void synchronize_sched_expedited(void) { |
e27fc9641 rcu: increase syn... |
1829 |
int firstsnap, s, snap, trycount = 0; |
7b27d5475 rcu,cleanup: move... |
1830 |
|
e27fc9641 rcu: increase syn... |
1831 1832 |
/* Note that atomic_inc_return() implies full memory barrier. */ firstsnap = snap = atomic_inc_return(&sync_sched_expedited_started); |
7b27d5475 rcu,cleanup: move... |
1833 |
get_online_cpus(); |
e27fc9641 rcu: increase syn... |
1834 1835 1836 1837 1838 |
/* * Each pass through the following loop attempts to force a * context switch on each CPU. */ |
7b27d5475 rcu,cleanup: move... |
1839 1840 1841 1842 |
while (try_stop_cpus(cpu_online_mask, synchronize_sched_expedited_cpu_stop, NULL) == -EAGAIN) { put_online_cpus(); |
e27fc9641 rcu: increase syn... |
1843 1844 |
/* No joy, try again later. Or just synchronize_sched(). */ |
7b27d5475 rcu,cleanup: move... |
1845 1846 1847 1848 1849 1850 |
if (trycount++ < 10) udelay(trycount * num_online_cpus()); else { synchronize_sched(); return; } |
e27fc9641 rcu: increase syn... |
1851 1852 1853 1854 |
/* Check to see if someone else did our work for us. */ s = atomic_read(&sync_sched_expedited_done); if (UINT_CMP_GE((unsigned)s, (unsigned)firstsnap)) { |
7b27d5475 rcu,cleanup: move... |
1855 1856 1857 |
smp_mb(); /* ensure test happens before caller kfree */ return; } |
e27fc9641 rcu: increase syn... |
1858 1859 1860 1861 1862 1863 1864 1865 1866 |
/* * Refetching sync_sched_expedited_started allows later * callers to piggyback on our grace period. We subtract * 1 to get the same token that the last incrementer got. * We retry after they started, so our grace period works * for them, and they started after our first try, so their * grace period works for us. */ |
7b27d5475 rcu,cleanup: move... |
1867 |
get_online_cpus(); |
7077714ec rcu: Make synchro... |
1868 |
snap = atomic_read(&sync_sched_expedited_started); |
e27fc9641 rcu: increase syn... |
1869 |
smp_mb(); /* ensure read is before try_stop_cpus(). */ |
7b27d5475 rcu,cleanup: move... |
1870 |
} |
e27fc9641 rcu: increase syn... |
1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 |
/* * Everyone up to our most recent fetch is covered by our grace * period. Update the counter, but only if our work is still * relevant -- which it won't be if someone who started later * than we did beat us to the punch. */ do { s = atomic_read(&sync_sched_expedited_done); if (UINT_CMP_GE((unsigned)s, (unsigned)snap)) { smp_mb(); /* ensure test happens before caller kfree */ break; } } while (atomic_cmpxchg(&sync_sched_expedited_done, s, snap) != s); |
7b27d5475 rcu,cleanup: move... |
1885 1886 1887 1888 1889 |
put_online_cpus(); } EXPORT_SYMBOL_GPL(synchronize_sched_expedited); #endif /* #else #ifndef CONFIG_SMP */ |
8bd93a2c5 rcu: Accelerate g... |
1890 1891 1892 1893 1894 1895 1896 1897 |
#if !defined(CONFIG_RCU_FAST_NO_HZ) /* * Check to see if any future RCU-related work will need to be done * by the current CPU, even if none need be done immediately, returning * 1 if so. This function is part of the RCU implementation; it is -not- * an exported member of the RCU API. * |
7cb924990 rcu: Permit dynti... |
1898 1899 |
* Because we not have RCU_FAST_NO_HZ, just check whether this CPU needs * any flavor of RCU. |
8bd93a2c5 rcu: Accelerate g... |
1900 1901 1902 |
*/ int rcu_needs_cpu(int cpu) { |
aea1b35e2 rcu: Allow dyntic... |
1903 1904 1905 1906 |
return rcu_cpu_has_callbacks(cpu); } /* |
7cb924990 rcu: Permit dynti... |
1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 |
* Because we do not have RCU_FAST_NO_HZ, don't bother initializing for it. */ static void rcu_prepare_for_idle_init(int cpu) { } /* * Because we do not have RCU_FAST_NO_HZ, don't bother cleaning up * after it. */ static void rcu_cleanup_after_idle(int cpu) { } /* |
aea1b35e2 rcu: Allow dyntic... |
1922 1923 1924 1925 1926 1927 |
* Do the idle-entry grace-period work, which, because CONFIG_RCU_FAST_NO_HZ=y, * is nothing. */ static void rcu_prepare_for_idle(int cpu) { } |
8bd93a2c5 rcu: Accelerate g... |
1928 |
#else /* #if !defined(CONFIG_RCU_FAST_NO_HZ) */ |
f23f7fa1c rcu: Adaptive dyn... |
1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 |
/* * This code is invoked when a CPU goes idle, at which point we want * to have the CPU do everything required for RCU so that it can enter * the energy-efficient dyntick-idle mode. This is handled by a * state machine implemented by rcu_prepare_for_idle() below. * * The following three proprocessor symbols control this state machine: * * RCU_IDLE_FLUSHES gives the maximum number of times that we will attempt * to satisfy RCU. Beyond this point, it is better to incur a periodic * scheduling-clock interrupt than to loop through the state machine * at full power. * RCU_IDLE_OPT_FLUSHES gives the number of RCU_IDLE_FLUSHES that are * optional if RCU does not need anything immediately from this * CPU, even if this CPU still has RCU callbacks queued. The first * times through the state machine are mandatory: we need to give * the state machine a chance to communicate a quiescent state * to the RCU core. * RCU_IDLE_GP_DELAY gives the number of jiffies that a CPU is permitted * to sleep in dyntick-idle mode with RCU callbacks pending. This * is sized to be roughly one RCU grace period. Those energy-efficiency * benchmarkers who might otherwise be tempted to set this to a large * number, be warned: Setting RCU_IDLE_GP_DELAY too high can hang your * system. And if you are -that- concerned about energy efficiency, * just power the system down and be done with it! * * The values below work well in practice. If future workloads require * adjustment, they can be converted into kernel config parameters, though * making the state machine smarter might be a better option. */ #define RCU_IDLE_FLUSHES 5 /* Number of dyntick-idle tries. */ #define RCU_IDLE_OPT_FLUSHES 3 /* Optional dyntick-idle tries. */ |
7cb924990 rcu: Permit dynti... |
1961 |
#define RCU_IDLE_GP_DELAY 6 /* Roughly one grace period. */ |
f23f7fa1c rcu: Adaptive dyn... |
1962 |
|
a47cd880b rcu: Fix accelera... |
1963 |
static DEFINE_PER_CPU(int, rcu_dyntick_drain); |
71da81324 rcu: Fix accelera... |
1964 |
static DEFINE_PER_CPU(unsigned long, rcu_dyntick_holdoff); |
7cb924990 rcu: Permit dynti... |
1965 1966 |
static DEFINE_PER_CPU(struct hrtimer, rcu_idle_gp_timer); static ktime_t rcu_idle_gp_wait; |
8bd93a2c5 rcu: Accelerate g... |
1967 1968 |
/* |
aea1b35e2 rcu: Allow dyntic... |
1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 |
* Allow the CPU to enter dyntick-idle mode if either: (1) There are no * callbacks on this CPU, (2) this CPU has not yet attempted to enter * dyntick-idle mode, or (3) this CPU is in the process of attempting to * enter dyntick-idle mode. Otherwise, if we have recently tried and failed * to enter dyntick-idle mode, we refuse to try to enter it. After all, * it is better to incur scheduling-clock interrupts than to spin * continuously for the same time duration! */ int rcu_needs_cpu(int cpu) { /* If no callbacks, RCU doesn't need the CPU. */ if (!rcu_cpu_has_callbacks(cpu)) return 0; /* Otherwise, RCU needs the CPU only if it recently tried and failed. */ return per_cpu(rcu_dyntick_holdoff, cpu) == jiffies; } /* |
7cb924990 rcu: Permit dynti... |
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 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 |
* Timer handler used to force CPU to start pushing its remaining RCU * callbacks in the case where it entered dyntick-idle mode with callbacks * pending. The hander doesn't really need to do anything because the * real work is done upon re-entry to idle, or by the next scheduling-clock * interrupt should idle not be re-entered. */ static enum hrtimer_restart rcu_idle_gp_timer_func(struct hrtimer *hrtp) { trace_rcu_prep_idle("Timer"); return HRTIMER_NORESTART; } /* * Initialize the timer used to pull CPUs out of dyntick-idle mode. */ static void rcu_prepare_for_idle_init(int cpu) { static int firsttime = 1; struct hrtimer *hrtp = &per_cpu(rcu_idle_gp_timer, cpu); hrtimer_init(hrtp, CLOCK_MONOTONIC, HRTIMER_MODE_REL); hrtp->function = rcu_idle_gp_timer_func; if (firsttime) { unsigned int upj = jiffies_to_usecs(RCU_IDLE_GP_DELAY); rcu_idle_gp_wait = ns_to_ktime(upj * (u64)1000); firsttime = 0; } } /* * Clean up for exit from idle. Because we are exiting from idle, there * is no longer any point to rcu_idle_gp_timer, so cancel it. This will * do nothing if this timer is not active, so just cancel it unconditionally. */ static void rcu_cleanup_after_idle(int cpu) { hrtimer_cancel(&per_cpu(rcu_idle_gp_timer, cpu)); } /* |
aea1b35e2 rcu: Allow dyntic... |
2028 2029 2030 |
* Check to see if any RCU-related work can be done by the current CPU, * and if so, schedule a softirq to get it done. This function is part * of the RCU implementation; it is -not- an exported member of the RCU API. |
8bd93a2c5 rcu: Accelerate g... |
2031 |
* |
aea1b35e2 rcu: Allow dyntic... |
2032 2033 2034 2035 2036 2037 |
* The idea is for the current CPU to clear out all work required by the * RCU core for the current grace period, so that this CPU can be permitted * to enter dyntick-idle mode. In some cases, it will need to be awakened * at the end of the grace period by whatever CPU ends the grace period. * This allows CPUs to go dyntick-idle more quickly, and to reduce the * number of wakeups by a modest integer factor. |
a47cd880b rcu: Fix accelera... |
2038 2039 2040 |
* * Because it is not legal to invoke rcu_process_callbacks() with irqs * disabled, we do one pass of force_quiescent_state(), then do a |
a46e0899e rcu: use softirq ... |
2041 |
* invoke_rcu_core() to cause rcu_process_callbacks() to be invoked |
27f4d2805 rcu: priority boo... |
2042 |
* later. The per-cpu rcu_dyntick_drain variable controls the sequencing. |
aea1b35e2 rcu: Allow dyntic... |
2043 2044 |
* * The caller must have disabled interrupts. |
8bd93a2c5 rcu: Accelerate g... |
2045 |
*/ |
aea1b35e2 rcu: Allow dyntic... |
2046 |
static void rcu_prepare_for_idle(int cpu) |
8bd93a2c5 rcu: Accelerate g... |
2047 |
{ |
84ad00cb6 rcu: Avoid needle... |
2048 2049 2050 |
unsigned long flags; local_irq_save(flags); |
8bd93a2c5 rcu: Accelerate g... |
2051 |
|
3084f2f80 rcu: Go dyntick-i... |
2052 |
/* |
f535a607c rcu: Eliminate RC... |
2053 2054 |
* If there are no callbacks on this CPU, enter dyntick-idle mode. * Also reset state to avoid prejudicing later attempts. |
3084f2f80 rcu: Go dyntick-i... |
2055 |
*/ |
aea1b35e2 rcu: Allow dyntic... |
2056 2057 |
if (!rcu_cpu_has_callbacks(cpu)) { per_cpu(rcu_dyntick_holdoff, cpu) = jiffies - 1; |
3084f2f80 rcu: Go dyntick-i... |
2058 |
per_cpu(rcu_dyntick_drain, cpu) = 0; |
84ad00cb6 rcu: Avoid needle... |
2059 |
local_irq_restore(flags); |
433cdddcd rcu: Add tracing ... |
2060 |
trace_rcu_prep_idle("No callbacks"); |
aea1b35e2 rcu: Allow dyntic... |
2061 |
return; |
77e38ed34 rcu: RCU_FAST_NO_... |
2062 |
} |
3084f2f80 rcu: Go dyntick-i... |
2063 2064 2065 2066 2067 |
/* * If in holdoff mode, just return. We will presumably have * refrained from disabling the scheduling-clock tick. */ |
433cdddcd rcu: Add tracing ... |
2068 |
if (per_cpu(rcu_dyntick_holdoff, cpu) == jiffies) { |
84ad00cb6 rcu: Avoid needle... |
2069 |
local_irq_restore(flags); |
433cdddcd rcu: Add tracing ... |
2070 |
trace_rcu_prep_idle("In holdoff"); |
aea1b35e2 rcu: Allow dyntic... |
2071 |
return; |
433cdddcd rcu: Add tracing ... |
2072 |
} |
a47cd880b rcu: Fix accelera... |
2073 2074 2075 2076 |
/* Check and update the rcu_dyntick_drain sequencing. */ if (per_cpu(rcu_dyntick_drain, cpu) <= 0) { /* First time through, initialize the counter. */ |
f23f7fa1c rcu: Adaptive dyn... |
2077 2078 2079 |
per_cpu(rcu_dyntick_drain, cpu) = RCU_IDLE_FLUSHES; } else if (per_cpu(rcu_dyntick_drain, cpu) <= RCU_IDLE_OPT_FLUSHES && !rcu_pending(cpu)) { |
7cb924990 rcu: Permit dynti... |
2080 |
/* Can we go dyntick-idle despite still having callbacks? */ |
f23f7fa1c rcu: Adaptive dyn... |
2081 2082 2083 2084 2085 2086 2087 |
trace_rcu_prep_idle("Dyntick with callbacks"); per_cpu(rcu_dyntick_drain, cpu) = 0; per_cpu(rcu_dyntick_holdoff, cpu) = jiffies - 1; hrtimer_start(&per_cpu(rcu_idle_gp_timer, cpu), rcu_idle_gp_wait, HRTIMER_MODE_REL); return; /* Nothing more to do immediately. */ } else if (--per_cpu(rcu_dyntick_drain, cpu) <= 0) { |
a47cd880b rcu: Fix accelera... |
2088 |
/* We have hit the limit, so time to give up. */ |
71da81324 rcu: Fix accelera... |
2089 |
per_cpu(rcu_dyntick_holdoff, cpu) = jiffies; |
84ad00cb6 rcu: Avoid needle... |
2090 |
local_irq_restore(flags); |
433cdddcd rcu: Add tracing ... |
2091 |
trace_rcu_prep_idle("Begin holdoff"); |
aea1b35e2 rcu: Allow dyntic... |
2092 2093 |
invoke_rcu_core(); /* Force the CPU out of dyntick-idle. */ return; |
a47cd880b rcu: Fix accelera... |
2094 |
} |
aea1b35e2 rcu: Allow dyntic... |
2095 2096 2097 2098 2099 2100 |
/* * Do one step of pushing the remaining RCU callbacks through * the RCU core state machine. */ #ifdef CONFIG_TREE_PREEMPT_RCU if (per_cpu(rcu_preempt_data, cpu).nxtlist) { |
3ad0decf9 rcu: Reduce laten... |
2101 |
local_irq_restore(flags); |
aea1b35e2 rcu: Allow dyntic... |
2102 2103 |
rcu_preempt_qs(cpu); force_quiescent_state(&rcu_preempt_state, 0); |
3ad0decf9 rcu: Reduce laten... |
2104 |
local_irq_save(flags); |
aea1b35e2 rcu: Allow dyntic... |
2105 2106 |
} #endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ |
a47cd880b rcu: Fix accelera... |
2107 |
if (per_cpu(rcu_sched_data, cpu).nxtlist) { |
3ad0decf9 rcu: Reduce laten... |
2108 |
local_irq_restore(flags); |
a47cd880b rcu: Fix accelera... |
2109 2110 |
rcu_sched_qs(cpu); force_quiescent_state(&rcu_sched_state, 0); |
3ad0decf9 rcu: Reduce laten... |
2111 |
local_irq_save(flags); |
a47cd880b rcu: Fix accelera... |
2112 2113 |
} if (per_cpu(rcu_bh_data, cpu).nxtlist) { |
3ad0decf9 rcu: Reduce laten... |
2114 |
local_irq_restore(flags); |
a47cd880b rcu: Fix accelera... |
2115 2116 |
rcu_bh_qs(cpu); force_quiescent_state(&rcu_bh_state, 0); |
3ad0decf9 rcu: Reduce laten... |
2117 |
local_irq_save(flags); |
8bd93a2c5 rcu: Accelerate g... |
2118 |
} |
433cdddcd rcu: Add tracing ... |
2119 2120 2121 2122 |
/* * If RCU callbacks are still pending, RCU still needs this CPU. * So try forcing the callbacks through the grace period. */ |
3ad0decf9 rcu: Reduce laten... |
2123 |
if (rcu_cpu_has_callbacks(cpu)) { |
84ad00cb6 rcu: Avoid needle... |
2124 |
local_irq_restore(flags); |
433cdddcd rcu: Add tracing ... |
2125 |
trace_rcu_prep_idle("More callbacks"); |
a46e0899e rcu: use softirq ... |
2126 |
invoke_rcu_core(); |
84ad00cb6 rcu: Avoid needle... |
2127 2128 |
} else { local_irq_restore(flags); |
433cdddcd rcu: Add tracing ... |
2129 |
trace_rcu_prep_idle("Callbacks drained"); |
84ad00cb6 rcu: Avoid needle... |
2130 |
} |
8bd93a2c5 rcu: Accelerate g... |
2131 2132 2133 |
} #endif /* #else #if !defined(CONFIG_RCU_FAST_NO_HZ) */ |