1/*-
2 * Copyright (c) 2001 Jake Burkholder <jake@FreeBSD.org>
3 * All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
8 * 1. Redistributions of source code must retain the above copyright
9 *    notice, this list of conditions and the following disclaimer.
10 * 2. Redistributions in binary form must reproduce the above copyright
11 *    notice, this list of conditions and the following disclaimer in the
12 *    documentation and/or other materials provided with the distribution.
13 *
14 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
15 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
16 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
17 * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
18 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
19 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
20 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
21 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
22 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
23 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
24 * SUCH DAMAGE.
25 */
26
27
28#include <sys/cdefs.h>
29__FBSDID("$FreeBSD$");
30
31#include "opt_sched.h"
32
33#include <sys/param.h>
34#include <sys/systm.h>
35#include <sys/kdb.h>
36#include <sys/kernel.h>
37#include <sys/ktr.h>
38#include <sys/lock.h>
39#include <sys/mutex.h>
40#include <sys/proc.h>
41#include <sys/queue.h>
42#include <sys/sched.h>
43#include <sys/smp.h>
44#include <sys/sysctl.h>
45
46#include <machine/cpu.h>
47
48/* Uncomment this to enable logging of critical_enter/exit. */
49#if 0
50#define	KTR_CRITICAL	KTR_SCHED
51#else
52#define	KTR_CRITICAL	0
53#endif
54
55#ifdef FULL_PREEMPTION
56#ifndef PREEMPTION
57#error "The FULL_PREEMPTION option requires the PREEMPTION option"
58#endif
59#endif
60
61CTASSERT((RQB_BPW * RQB_LEN) == RQ_NQS);
62
63/*
64 * kern.sched.preemption allows user space to determine if preemption support
65 * is compiled in or not.  It is not currently a boot or runtime flag that
66 * can be changed.
67 */
68#ifdef PREEMPTION
69static int kern_sched_preemption = 1;
70#else
71static int kern_sched_preemption = 0;
72#endif
73SYSCTL_INT(_kern_sched, OID_AUTO, preemption, CTLFLAG_RD,
74    &kern_sched_preemption, 0, "Kernel preemption enabled");
75
76/*
77 * Support for scheduler stats exported via kern.sched.stats.  All stats may
78 * be reset with kern.sched.stats.reset = 1.  Stats may be defined elsewhere
79 * with SCHED_STAT_DEFINE().
80 */
81#ifdef SCHED_STATS
82SYSCTL_NODE(_kern_sched, OID_AUTO, stats, CTLFLAG_RW, 0, "switch stats");
83
84/* Switch reasons from mi_switch(). */
85DPCPU_DEFINE(long, sched_switch_stats[SWT_COUNT]);
86SCHED_STAT_DEFINE_VAR(uncategorized,
87    &DPCPU_NAME(sched_switch_stats[SWT_NONE]), "");
88SCHED_STAT_DEFINE_VAR(preempt,
89    &DPCPU_NAME(sched_switch_stats[SWT_PREEMPT]), "");
90SCHED_STAT_DEFINE_VAR(owepreempt,
91    &DPCPU_NAME(sched_switch_stats[SWT_OWEPREEMPT]), "");
92SCHED_STAT_DEFINE_VAR(turnstile,
93    &DPCPU_NAME(sched_switch_stats[SWT_TURNSTILE]), "");
94SCHED_STAT_DEFINE_VAR(sleepq,
95    &DPCPU_NAME(sched_switch_stats[SWT_SLEEPQ]), "");
96SCHED_STAT_DEFINE_VAR(sleepqtimo,
97    &DPCPU_NAME(sched_switch_stats[SWT_SLEEPQTIMO]), "");
98SCHED_STAT_DEFINE_VAR(relinquish,
99    &DPCPU_NAME(sched_switch_stats[SWT_RELINQUISH]), "");
100SCHED_STAT_DEFINE_VAR(needresched,
101    &DPCPU_NAME(sched_switch_stats[SWT_NEEDRESCHED]), "");
102SCHED_STAT_DEFINE_VAR(idle,
103    &DPCPU_NAME(sched_switch_stats[SWT_IDLE]), "");
104SCHED_STAT_DEFINE_VAR(iwait,
105    &DPCPU_NAME(sched_switch_stats[SWT_IWAIT]), "");
106SCHED_STAT_DEFINE_VAR(suspend,
107    &DPCPU_NAME(sched_switch_stats[SWT_SUSPEND]), "");
108SCHED_STAT_DEFINE_VAR(remotepreempt,
109    &DPCPU_NAME(sched_switch_stats[SWT_REMOTEPREEMPT]), "");
110SCHED_STAT_DEFINE_VAR(remotewakeidle,
111    &DPCPU_NAME(sched_switch_stats[SWT_REMOTEWAKEIDLE]), "");
112
113static int
114sysctl_stats_reset(SYSCTL_HANDLER_ARGS)
115{
116	struct sysctl_oid *p;
117	uintptr_t counter;
118        int error;
119	int val;
120	int i;
121
122        val = 0;
123        error = sysctl_handle_int(oidp, &val, 0, req);
124        if (error != 0 || req->newptr == NULL)
125                return (error);
126        if (val == 0)
127                return (0);
128	/*
129	 * Traverse the list of children of _kern_sched_stats and reset each
130	 * to 0.  Skip the reset entry.
131	 */
132	SLIST_FOREACH(p, oidp->oid_parent, oid_link) {
133		if (p == oidp || p->oid_arg1 == NULL)
134			continue;
135		counter = (uintptr_t)p->oid_arg1;
136		CPU_FOREACH(i) {
137			*(long *)(dpcpu_off[i] + counter) = 0;
138		}
139	}
140	return (0);
141}
142
143SYSCTL_PROC(_kern_sched_stats, OID_AUTO, reset, CTLTYPE_INT | CTLFLAG_WR, NULL,
144    0, sysctl_stats_reset, "I", "Reset scheduler statistics");
145#endif
146
147/************************************************************************
148 * Functions that manipulate runnability from a thread perspective.	*
149 ************************************************************************/
150/*
151 * Select the thread that will be run next.
152 */
153struct thread *
154choosethread(void)
155{
156	struct thread *td;
157
158retry:
159	td = sched_choose();
160
161	/*
162	 * If we are in panic, only allow system threads,
163	 * plus the one we are running in, to be run.
164	 */
165	if (panicstr && ((td->td_proc->p_flag & P_SYSTEM) == 0 &&
166	    (td->td_flags & TDF_INPANIC) == 0)) {
167		/* note that it is no longer on the run queue */
168		TD_SET_CAN_RUN(td);
169		goto retry;
170	}
171
172	TD_SET_RUNNING(td);
173	return (td);
174}
175
176/*
177 * Kernel thread preemption implementation.  Critical sections mark
178 * regions of code in which preemptions are not allowed.
179 *
180 * It might seem a good idea to inline critical_enter() but, in order
181 * to prevent instructions reordering by the compiler, a __compiler_membar()
182 * would have to be used here (the same as sched_pin()).  The performance
183 * penalty imposed by the membar could, then, produce slower code than
184 * the function call itself, for most cases.
185 */
186void
187critical_enter(void)
188{
189	struct thread *td;
190
191	td = curthread;
192	td->td_critnest++;
193	CTR4(KTR_CRITICAL, "critical_enter by thread %p (%ld, %s) to %d", td,
194	    (long)td->td_proc->p_pid, td->td_name, td->td_critnest);
195}
196
197void
198critical_exit(void)
199{
200	struct thread *td;
201	int flags;
202
203	td = curthread;
204	KASSERT(td->td_critnest != 0,
205	    ("critical_exit: td_critnest == 0"));
206
207	if (td->td_critnest == 1) {
208		td->td_critnest = 0;
209		if (td->td_owepreempt && !kdb_active) {
210			td->td_critnest = 1;
211			thread_lock(td);
212			td->td_critnest--;
213			flags = SW_INVOL | SW_PREEMPT;
214			if (TD_IS_IDLETHREAD(td))
215				flags |= SWT_IDLE;
216			else
217				flags |= SWT_OWEPREEMPT;
218			mi_switch(flags, NULL);
219			thread_unlock(td);
220		}
221	} else
222		td->td_critnest--;
223
224	CTR4(KTR_CRITICAL, "critical_exit by thread %p (%ld, %s) to %d", td,
225	    (long)td->td_proc->p_pid, td->td_name, td->td_critnest);
226}
227
228/************************************************************************
229 * SYSTEM RUN QUEUE manipulations and tests				*
230 ************************************************************************/
231/*
232 * Initialize a run structure.
233 */
234void
235runq_init(struct runq *rq)
236{
237	int i;
238
239	bzero(rq, sizeof *rq);
240	for (i = 0; i < RQ_NQS; i++)
241		TAILQ_INIT(&rq->rq_queues[i]);
242}
243
244/*
245 * Clear the status bit of the queue corresponding to priority level pri,
246 * indicating that it is empty.
247 */
248static __inline void
249runq_clrbit(struct runq *rq, int pri)
250{
251	struct rqbits *rqb;
252
253	rqb = &rq->rq_status;
254	CTR4(KTR_RUNQ, "runq_clrbit: bits=%#x %#x bit=%#x word=%d",
255	    rqb->rqb_bits[RQB_WORD(pri)],
256	    rqb->rqb_bits[RQB_WORD(pri)] & ~RQB_BIT(pri),
257	    RQB_BIT(pri), RQB_WORD(pri));
258	rqb->rqb_bits[RQB_WORD(pri)] &= ~RQB_BIT(pri);
259}
260
261/*
262 * Find the index of the first non-empty run queue.  This is done by
263 * scanning the status bits, a set bit indicates a non-empty queue.
264 */
265static __inline int
266runq_findbit(struct runq *rq)
267{
268	struct rqbits *rqb;
269	int pri;
270	int i;
271
272	rqb = &rq->rq_status;
273	for (i = 0; i < RQB_LEN; i++)
274		if (rqb->rqb_bits[i]) {
275			pri = RQB_FFS(rqb->rqb_bits[i]) + (i << RQB_L2BPW);
276			CTR3(KTR_RUNQ, "runq_findbit: bits=%#x i=%d pri=%d",
277			    rqb->rqb_bits[i], i, pri);
278			return (pri);
279		}
280
281	return (-1);
282}
283
284static __inline int
285runq_findbit_from(struct runq *rq, u_char pri)
286{
287	struct rqbits *rqb;
288	rqb_word_t mask;
289	int i;
290
291	/*
292	 * Set the mask for the first word so we ignore priorities before 'pri'.
293	 */
294	mask = (rqb_word_t)-1 << (pri & (RQB_BPW - 1));
295	rqb = &rq->rq_status;
296again:
297	for (i = RQB_WORD(pri); i < RQB_LEN; mask = -1, i++) {
298		mask = rqb->rqb_bits[i] & mask;
299		if (mask == 0)
300			continue;
301		pri = RQB_FFS(mask) + (i << RQB_L2BPW);
302		CTR3(KTR_RUNQ, "runq_findbit_from: bits=%#x i=%d pri=%d",
303		    mask, i, pri);
304		return (pri);
305	}
306	if (pri == 0)
307		return (-1);
308	/*
309	 * Wrap back around to the beginning of the list just once so we
310	 * scan the whole thing.
311	 */
312	pri = 0;
313	goto again;
314}
315
316/*
317 * Set the status bit of the queue corresponding to priority level pri,
318 * indicating that it is non-empty.
319 */
320static __inline void
321runq_setbit(struct runq *rq, int pri)
322{
323	struct rqbits *rqb;
324
325	rqb = &rq->rq_status;
326	CTR4(KTR_RUNQ, "runq_setbit: bits=%#x %#x bit=%#x word=%d",
327	    rqb->rqb_bits[RQB_WORD(pri)],
328	    rqb->rqb_bits[RQB_WORD(pri)] | RQB_BIT(pri),
329	    RQB_BIT(pri), RQB_WORD(pri));
330	rqb->rqb_bits[RQB_WORD(pri)] |= RQB_BIT(pri);
331}
332
333/*
334 * Add the thread to the queue specified by its priority, and set the
335 * corresponding status bit.
336 */
337void
338runq_add(struct runq *rq, struct thread *td, int flags)
339{
340	struct rqhead *rqh;
341	int pri;
342
343	pri = td->td_priority / RQ_PPQ;
344	td->td_rqindex = pri;
345	runq_setbit(rq, pri);
346	rqh = &rq->rq_queues[pri];
347	CTR4(KTR_RUNQ, "runq_add: td=%p pri=%d %d rqh=%p",
348	    td, td->td_priority, pri, rqh);
349	if (flags & SRQ_PREEMPTED) {
350		TAILQ_INSERT_HEAD(rqh, td, td_runq);
351	} else {
352		TAILQ_INSERT_TAIL(rqh, td, td_runq);
353	}
354}
355
356void
357runq_add_pri(struct runq *rq, struct thread *td, u_char pri, int flags)
358{
359	struct rqhead *rqh;
360
361	KASSERT(pri < RQ_NQS, ("runq_add_pri: %d out of range", pri));
362	td->td_rqindex = pri;
363	runq_setbit(rq, pri);
364	rqh = &rq->rq_queues[pri];
365	CTR4(KTR_RUNQ, "runq_add_pri: td=%p pri=%d idx=%d rqh=%p",
366	    td, td->td_priority, pri, rqh);
367	if (flags & SRQ_PREEMPTED) {
368		TAILQ_INSERT_HEAD(rqh, td, td_runq);
369	} else {
370		TAILQ_INSERT_TAIL(rqh, td, td_runq);
371	}
372}
373/*
374 * Return true if there are runnable processes of any priority on the run
375 * queue, false otherwise.  Has no side effects, does not modify the run
376 * queue structure.
377 */
378int
379runq_check(struct runq *rq)
380{
381	struct rqbits *rqb;
382	int i;
383
384	rqb = &rq->rq_status;
385	for (i = 0; i < RQB_LEN; i++)
386		if (rqb->rqb_bits[i]) {
387			CTR2(KTR_RUNQ, "runq_check: bits=%#x i=%d",
388			    rqb->rqb_bits[i], i);
389			return (1);
390		}
391	CTR0(KTR_RUNQ, "runq_check: empty");
392
393	return (0);
394}
395
396/*
397 * Find the highest priority process on the run queue.
398 */
399struct thread *
400runq_choose_fuzz(struct runq *rq, int fuzz)
401{
402	struct rqhead *rqh;
403	struct thread *td;
404	int pri;
405
406	while ((pri = runq_findbit(rq)) != -1) {
407		rqh = &rq->rq_queues[pri];
408		/* fuzz == 1 is normal.. 0 or less are ignored */
409		if (fuzz > 1) {
410			/*
411			 * In the first couple of entries, check if
412			 * there is one for our CPU as a preference.
413			 */
414			int count = fuzz;
415			int cpu = PCPU_GET(cpuid);
416			struct thread *td2;
417			td2 = td = TAILQ_FIRST(rqh);
418
419			while (count-- && td2) {
420				if (td2->td_lastcpu == cpu) {
421					td = td2;
422					break;
423				}
424				td2 = TAILQ_NEXT(td2, td_runq);
425			}
426		} else
427			td = TAILQ_FIRST(rqh);
428		KASSERT(td != NULL, ("runq_choose_fuzz: no proc on busy queue"));
429		CTR3(KTR_RUNQ,
430		    "runq_choose_fuzz: pri=%d thread=%p rqh=%p", pri, td, rqh);
431		return (td);
432	}
433	CTR1(KTR_RUNQ, "runq_choose_fuzz: idleproc pri=%d", pri);
434
435	return (NULL);
436}
437
438/*
439 * Find the highest priority process on the run queue.
440 */
441struct thread *
442runq_choose(struct runq *rq)
443{
444	struct rqhead *rqh;
445	struct thread *td;
446	int pri;
447
448	while ((pri = runq_findbit(rq)) != -1) {
449		rqh = &rq->rq_queues[pri];
450		td = TAILQ_FIRST(rqh);
451		KASSERT(td != NULL, ("runq_choose: no thread on busy queue"));
452		CTR3(KTR_RUNQ,
453		    "runq_choose: pri=%d thread=%p rqh=%p", pri, td, rqh);
454		return (td);
455	}
456	CTR1(KTR_RUNQ, "runq_choose: idlethread pri=%d", pri);
457
458	return (NULL);
459}
460
461struct thread *
462runq_choose_from(struct runq *rq, u_char idx)
463{
464	struct rqhead *rqh;
465	struct thread *td;
466	int pri;
467
468	if ((pri = runq_findbit_from(rq, idx)) != -1) {
469		rqh = &rq->rq_queues[pri];
470		td = TAILQ_FIRST(rqh);
471		KASSERT(td != NULL, ("runq_choose: no thread on busy queue"));
472		CTR4(KTR_RUNQ,
473		    "runq_choose_from: pri=%d thread=%p idx=%d rqh=%p",
474		    pri, td, td->td_rqindex, rqh);
475		return (td);
476	}
477	CTR1(KTR_RUNQ, "runq_choose_from: idlethread pri=%d", pri);
478
479	return (NULL);
480}
481/*
482 * Remove the thread from the queue specified by its priority, and clear the
483 * corresponding status bit if the queue becomes empty.
484 * Caller must set state afterwards.
485 */
486void
487runq_remove(struct runq *rq, struct thread *td)
488{
489
490	runq_remove_idx(rq, td, NULL);
491}
492
493void
494runq_remove_idx(struct runq *rq, struct thread *td, u_char *idx)
495{
496	struct rqhead *rqh;
497	u_char pri;
498
499	KASSERT(td->td_flags & TDF_INMEM,
500		("runq_remove_idx: thread swapped out"));
501	pri = td->td_rqindex;
502	KASSERT(pri < RQ_NQS, ("runq_remove_idx: Invalid index %d\n", pri));
503	rqh = &rq->rq_queues[pri];
504	CTR4(KTR_RUNQ, "runq_remove_idx: td=%p, pri=%d %d rqh=%p",
505	    td, td->td_priority, pri, rqh);
506	TAILQ_REMOVE(rqh, td, td_runq);
507	if (TAILQ_EMPTY(rqh)) {
508		CTR0(KTR_RUNQ, "runq_remove_idx: empty");
509		runq_clrbit(rq, pri);
510		if (idx != NULL && *idx == pri)
511			*idx = (pri + 1) % RQ_NQS;
512	}
513}
514