Merge tag 'v3.3.7' into 3.3/master
[zen-stable.git] / kernel / sched / cpupri.c
blobd72586fdf6607db63c5f43a2e1fbbb32ff0ac2c7
1 /*
2 * kernel/sched/cpupri.c
4 * CPU priority management
6 * Copyright (C) 2007-2008 Novell
8 * Author: Gregory Haskins <ghaskins@novell.com>
10 * This code tracks the priority of each CPU so that global migration
11 * decisions are easy to calculate. Each CPU can be in a state as follows:
13 * (INVALID), IDLE, NORMAL, RT1, ... RT99
15 * going from the lowest priority to the highest. CPUs in the INVALID state
16 * are not eligible for routing. The system maintains this state with
17 * a 2 dimensional bitmap (the first for priority class, the second for cpus
18 * in that class). Therefore a typical application without affinity
19 * restrictions can find a suitable CPU with O(1) complexity (e.g. two bit
20 * searches). For tasks with affinity restrictions, the algorithm has a
21 * worst case complexity of O(min(102, nr_domcpus)), though the scenario that
22 * yields the worst case search is fairly contrived.
24 * This program is free software; you can redistribute it and/or
25 * modify it under the terms of the GNU General Public License
26 * as published by the Free Software Foundation; version 2
27 * of the License.
30 #include <linux/gfp.h>
31 #include "cpupri.h"
33 /* Convert between a 140 based task->prio, and our 102 based cpupri */
34 static int convert_prio(int prio)
36 int cpupri;
38 if (prio == CPUPRI_INVALID)
39 cpupri = CPUPRI_INVALID;
40 else if (prio == MAX_PRIO)
41 cpupri = CPUPRI_IDLE;
42 else if (prio >= MAX_RT_PRIO)
43 cpupri = CPUPRI_NORMAL;
44 else
45 cpupri = MAX_RT_PRIO - prio + 1;
47 return cpupri;
50 /**
51 * cpupri_find - find the best (lowest-pri) CPU in the system
52 * @cp: The cpupri context
53 * @p: The task
54 * @lowest_mask: A mask to fill in with selected CPUs (or NULL)
56 * Note: This function returns the recommended CPUs as calculated during the
57 * current invocation. By the time the call returns, the CPUs may have in
58 * fact changed priorities any number of times. While not ideal, it is not
59 * an issue of correctness since the normal rebalancer logic will correct
60 * any discrepancies created by racing against the uncertainty of the current
61 * priority configuration.
63 * Returns: (int)bool - CPUs were found
65 int cpupri_find(struct cpupri *cp, struct task_struct *p,
66 struct cpumask *lowest_mask)
68 int idx = 0;
69 int task_pri = convert_prio(p->prio);
71 if (task_pri >= MAX_RT_PRIO)
72 return 0;
74 for (idx = 0; idx < task_pri; idx++) {
75 struct cpupri_vec *vec = &cp->pri_to_cpu[idx];
76 int skip = 0;
78 if (!atomic_read(&(vec)->count))
79 skip = 1;
81 * When looking at the vector, we need to read the counter,
82 * do a memory barrier, then read the mask.
84 * Note: This is still all racey, but we can deal with it.
85 * Ideally, we only want to look at masks that are set.
87 * If a mask is not set, then the only thing wrong is that we
88 * did a little more work than necessary.
90 * If we read a zero count but the mask is set, because of the
91 * memory barriers, that can only happen when the highest prio
92 * task for a run queue has left the run queue, in which case,
93 * it will be followed by a pull. If the task we are processing
94 * fails to find a proper place to go, that pull request will
95 * pull this task if the run queue is running at a lower
96 * priority.
98 smp_rmb();
100 /* Need to do the rmb for every iteration */
101 if (skip)
102 continue;
104 if (cpumask_any_and(&p->cpus_allowed, vec->mask) >= nr_cpu_ids)
105 continue;
107 if (lowest_mask) {
108 cpumask_and(lowest_mask, &p->cpus_allowed, vec->mask);
111 * We have to ensure that we have at least one bit
112 * still set in the array, since the map could have
113 * been concurrently emptied between the first and
114 * second reads of vec->mask. If we hit this
115 * condition, simply act as though we never hit this
116 * priority level and continue on.
118 if (cpumask_any(lowest_mask) >= nr_cpu_ids)
119 continue;
122 return 1;
125 return 0;
129 * cpupri_set - update the cpu priority setting
130 * @cp: The cpupri context
131 * @cpu: The target cpu
132 * @newpri: The priority (INVALID-RT99) to assign to this CPU
134 * Note: Assumes cpu_rq(cpu)->lock is locked
136 * Returns: (void)
138 void cpupri_set(struct cpupri *cp, int cpu, int newpri)
140 int *currpri = &cp->cpu_to_pri[cpu];
141 int oldpri = *currpri;
142 int do_mb = 0;
144 newpri = convert_prio(newpri);
146 BUG_ON(newpri >= CPUPRI_NR_PRIORITIES);
148 if (newpri == oldpri)
149 return;
152 * If the cpu was currently mapped to a different value, we
153 * need to map it to the new value then remove the old value.
154 * Note, we must add the new value first, otherwise we risk the
155 * cpu being missed by the priority loop in cpupri_find.
157 if (likely(newpri != CPUPRI_INVALID)) {
158 struct cpupri_vec *vec = &cp->pri_to_cpu[newpri];
160 cpumask_set_cpu(cpu, vec->mask);
162 * When adding a new vector, we update the mask first,
163 * do a write memory barrier, and then update the count, to
164 * make sure the vector is visible when count is set.
166 smp_mb__before_atomic_inc();
167 atomic_inc(&(vec)->count);
168 do_mb = 1;
170 if (likely(oldpri != CPUPRI_INVALID)) {
171 struct cpupri_vec *vec = &cp->pri_to_cpu[oldpri];
174 * Because the order of modification of the vec->count
175 * is important, we must make sure that the update
176 * of the new prio is seen before we decrement the
177 * old prio. This makes sure that the loop sees
178 * one or the other when we raise the priority of
179 * the run queue. We don't care about when we lower the
180 * priority, as that will trigger an rt pull anyway.
182 * We only need to do a memory barrier if we updated
183 * the new priority vec.
185 if (do_mb)
186 smp_mb__after_atomic_inc();
189 * When removing from the vector, we decrement the counter first
190 * do a memory barrier and then clear the mask.
192 atomic_dec(&(vec)->count);
193 smp_mb__after_atomic_inc();
194 cpumask_clear_cpu(cpu, vec->mask);
197 *currpri = newpri;
201 * cpupri_init - initialize the cpupri structure
202 * @cp: The cpupri context
204 * Returns: -ENOMEM if memory fails.
206 int cpupri_init(struct cpupri *cp)
208 int i;
210 memset(cp, 0, sizeof(*cp));
212 for (i = 0; i < CPUPRI_NR_PRIORITIES; i++) {
213 struct cpupri_vec *vec = &cp->pri_to_cpu[i];
215 atomic_set(&vec->count, 0);
216 if (!zalloc_cpumask_var(&vec->mask, GFP_KERNEL))
217 goto cleanup;
220 for_each_possible_cpu(i)
221 cp->cpu_to_pri[i] = CPUPRI_INVALID;
222 return 0;
224 cleanup:
225 for (i--; i >= 0; i--)
226 free_cpumask_var(cp->pri_to_cpu[i].mask);
227 return -ENOMEM;
231 * cpupri_cleanup - clean up the cpupri structure
232 * @cp: The cpupri context
234 void cpupri_cleanup(struct cpupri *cp)
236 int i;
238 for (i = 0; i < CPUPRI_NR_PRIORITIES; i++)
239 free_cpumask_var(cp->pri_to_cpu[i].mask);