1 /* $NetBSD: kern_runq.c,v 1.42 2014/02/25 18:30:11 pooka Exp $ */ 2 3 /* 4 * Copyright (c) 2007, 2008 Mindaugas Rasiukevicius <rmind at NetBSD org> 5 * All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions 9 * are met: 10 * 1. Redistributions of source code must retain the above copyright 11 * notice, this list of conditions and the following disclaimer. 12 * 2. Redistributions in binary form must reproduce the above copyright 13 * notice, this list of conditions and the following disclaimer in the 14 * documentation and/or other materials provided with the distribution. 15 * 16 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 17 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 18 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 19 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 20 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 21 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 22 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 23 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 24 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 25 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 26 * SUCH DAMAGE. 27 */ 28 29 #include <sys/cdefs.h> 30 __KERNEL_RCSID(0, "$NetBSD: kern_runq.c,v 1.42 2014/02/25 18:30:11 pooka Exp $"); 31 32 #include <sys/param.h> 33 #include <sys/kernel.h> 34 #include <sys/bitops.h> 35 #include <sys/cpu.h> 36 #include <sys/idle.h> 37 #include <sys/intr.h> 38 #include <sys/kmem.h> 39 #include <sys/lwp.h> 40 #include <sys/mutex.h> 41 #include <sys/proc.h> 42 #include <sys/sched.h> 43 #include <sys/syscallargs.h> 44 #include <sys/sysctl.h> 45 #include <sys/systm.h> 46 #include <sys/types.h> 47 #include <sys/evcnt.h> 48 49 /* 50 * Priority related defintions. 51 */ 52 #define PRI_TS_COUNT (NPRI_USER) 53 #define PRI_RT_COUNT (PRI_COUNT - PRI_TS_COUNT) 54 #define PRI_HTS_RANGE (PRI_TS_COUNT / 10) 55 56 #define PRI_HIGHEST_TS (MAXPRI_USER) 57 58 /* 59 * Bits per map. 60 */ 61 #define BITMAP_BITS (32) 62 #define BITMAP_SHIFT (5) 63 #define BITMAP_MSB (0x80000000U) 64 #define BITMAP_MASK (BITMAP_BITS - 1) 65 66 /* 67 * Structures, runqueue. 68 */ 69 70 const int schedppq = 1; 71 72 typedef struct { 73 TAILQ_HEAD(, lwp) q_head; 74 } queue_t; 75 76 typedef struct { 77 /* Bitmap */ 78 uint32_t r_bitmap[PRI_COUNT >> BITMAP_SHIFT]; 79 /* Counters */ 80 u_int r_count; /* Count of the threads */ 81 u_int r_avgcount; /* Average count of threads */ 82 u_int r_mcount; /* Count of migratable threads */ 83 /* Runqueues */ 84 queue_t r_rt_queue[PRI_RT_COUNT]; 85 queue_t r_ts_queue[PRI_TS_COUNT]; 86 /* Event counters */ 87 struct evcnt r_ev_pull; 88 struct evcnt r_ev_push; 89 struct evcnt r_ev_stay; 90 struct evcnt r_ev_localize; 91 } runqueue_t; 92 93 static void * sched_getrq(runqueue_t *, const pri_t); 94 #ifdef MULTIPROCESSOR 95 static lwp_t * sched_catchlwp(struct cpu_info *); 96 static void sched_balance(void *); 97 #endif 98 99 /* 100 * Preemption control. 101 */ 102 int sched_upreempt_pri = 0; 103 #ifdef __HAVE_PREEMPTION 104 # ifdef DEBUG 105 int sched_kpreempt_pri = 0; 106 # else 107 int sched_kpreempt_pri = PRI_USER_RT; 108 # endif 109 #else 110 int sched_kpreempt_pri = 1000; 111 #endif 112 113 /* 114 * Migration and balancing. 115 */ 116 static u_int cacheht_time; /* Cache hotness time */ 117 static u_int min_catch; /* Minimal LWP count for catching */ 118 static u_int balance_period; /* Balance period */ 119 static struct cpu_info *worker_ci; /* Victim CPU */ 120 #ifdef MULTIPROCESSOR 121 static struct callout balance_ch; /* Callout of balancer */ 122 #endif 123 124 void 125 runq_init(void) 126 { 127 128 /* Balancing */ 129 worker_ci = curcpu(); 130 cacheht_time = mstohz(3); /* ~3 ms */ 131 balance_period = mstohz(300); /* ~300 ms */ 132 133 /* Minimal count of LWPs for catching */ 134 min_catch = 1; 135 136 /* Initialize balancing callout and run it */ 137 #ifdef MULTIPROCESSOR 138 callout_init(&balance_ch, CALLOUT_MPSAFE); 139 callout_setfunc(&balance_ch, sched_balance, NULL); 140 callout_schedule(&balance_ch, balance_period); 141 #endif 142 } 143 144 void 145 sched_cpuattach(struct cpu_info *ci) 146 { 147 runqueue_t *ci_rq; 148 void *rq_ptr; 149 u_int i, size; 150 151 if (ci->ci_schedstate.spc_lwplock == NULL) { 152 ci->ci_schedstate.spc_lwplock = 153 mutex_obj_alloc(MUTEX_DEFAULT, IPL_SCHED); 154 } 155 if (ci == lwp0.l_cpu) { 156 /* Initialize the scheduler structure of the primary LWP */ 157 lwp0.l_mutex = ci->ci_schedstate.spc_lwplock; 158 } 159 if (ci->ci_schedstate.spc_mutex != NULL) { 160 /* Already initialized. */ 161 return; 162 } 163 164 /* Allocate the run queue */ 165 size = roundup2(sizeof(runqueue_t), coherency_unit) + coherency_unit; 166 rq_ptr = kmem_zalloc(size, KM_SLEEP); 167 if (rq_ptr == NULL) { 168 panic("sched_cpuattach: could not allocate the runqueue"); 169 } 170 ci_rq = (void *)(roundup2((uintptr_t)(rq_ptr), coherency_unit)); 171 172 /* Initialize run queues */ 173 ci->ci_schedstate.spc_mutex = 174 mutex_obj_alloc(MUTEX_DEFAULT, IPL_SCHED); 175 for (i = 0; i < PRI_RT_COUNT; i++) 176 TAILQ_INIT(&ci_rq->r_rt_queue[i].q_head); 177 for (i = 0; i < PRI_TS_COUNT; i++) 178 TAILQ_INIT(&ci_rq->r_ts_queue[i].q_head); 179 180 ci->ci_schedstate.spc_sched_info = ci_rq; 181 182 evcnt_attach_dynamic(&ci_rq->r_ev_pull, EVCNT_TYPE_MISC, NULL, 183 cpu_name(ci), "runqueue pull"); 184 evcnt_attach_dynamic(&ci_rq->r_ev_push, EVCNT_TYPE_MISC, NULL, 185 cpu_name(ci), "runqueue push"); 186 evcnt_attach_dynamic(&ci_rq->r_ev_stay, EVCNT_TYPE_MISC, NULL, 187 cpu_name(ci), "runqueue stay"); 188 evcnt_attach_dynamic(&ci_rq->r_ev_localize, EVCNT_TYPE_MISC, NULL, 189 cpu_name(ci), "runqueue localize"); 190 } 191 192 /* 193 * Control of the runqueue. 194 */ 195 196 static inline void * 197 sched_getrq(runqueue_t *ci_rq, const pri_t prio) 198 { 199 200 KASSERT(prio < PRI_COUNT); 201 return (prio <= PRI_HIGHEST_TS) ? 202 &ci_rq->r_ts_queue[prio].q_head : 203 &ci_rq->r_rt_queue[prio - PRI_HIGHEST_TS - 1].q_head; 204 } 205 206 void 207 sched_enqueue(struct lwp *l, bool swtch) 208 { 209 runqueue_t *ci_rq; 210 struct schedstate_percpu *spc; 211 TAILQ_HEAD(, lwp) *q_head; 212 const pri_t eprio = lwp_eprio(l); 213 struct cpu_info *ci; 214 int type; 215 216 ci = l->l_cpu; 217 spc = &ci->ci_schedstate; 218 ci_rq = spc->spc_sched_info; 219 KASSERT(lwp_locked(l, l->l_cpu->ci_schedstate.spc_mutex)); 220 221 /* Update the last run time on switch */ 222 if (__predict_true(swtch == true)) 223 l->l_rticksum += (hardclock_ticks - l->l_rticks); 224 else if (l->l_rticks == 0) 225 l->l_rticks = hardclock_ticks; 226 227 /* Enqueue the thread */ 228 q_head = sched_getrq(ci_rq, eprio); 229 if (TAILQ_EMPTY(q_head)) { 230 u_int i; 231 uint32_t q; 232 233 /* Mark bit */ 234 i = eprio >> BITMAP_SHIFT; 235 q = BITMAP_MSB >> (eprio & BITMAP_MASK); 236 KASSERT((ci_rq->r_bitmap[i] & q) == 0); 237 ci_rq->r_bitmap[i] |= q; 238 } 239 TAILQ_INSERT_TAIL(q_head, l, l_runq); 240 ci_rq->r_count++; 241 if ((l->l_pflag & LP_BOUND) == 0) 242 ci_rq->r_mcount++; 243 244 /* 245 * Update the value of highest priority in the runqueue, 246 * if priority of this thread is higher. 247 */ 248 if (eprio > spc->spc_maxpriority) 249 spc->spc_maxpriority = eprio; 250 251 sched_newts(l); 252 253 /* 254 * Wake the chosen CPU or cause a preemption if the newly 255 * enqueued thread has higher priority. Don't cause a 256 * preemption if the thread is yielding (swtch). 257 */ 258 if (!swtch && eprio > spc->spc_curpriority) { 259 if (eprio >= sched_kpreempt_pri) 260 type = RESCHED_KPREEMPT; 261 else if (eprio >= sched_upreempt_pri) 262 type = RESCHED_IMMED; 263 else 264 type = RESCHED_LAZY; 265 cpu_need_resched(ci, type); 266 } 267 } 268 269 void 270 sched_dequeue(struct lwp *l) 271 { 272 runqueue_t *ci_rq; 273 TAILQ_HEAD(, lwp) *q_head; 274 struct schedstate_percpu *spc; 275 const pri_t eprio = lwp_eprio(l); 276 277 spc = & l->l_cpu->ci_schedstate; 278 ci_rq = spc->spc_sched_info; 279 KASSERT(lwp_locked(l, spc->spc_mutex)); 280 281 KASSERT(eprio <= spc->spc_maxpriority); 282 KASSERT(ci_rq->r_bitmap[eprio >> BITMAP_SHIFT] != 0); 283 KASSERT(ci_rq->r_count > 0); 284 285 if (spc->spc_migrating == l) 286 spc->spc_migrating = NULL; 287 288 ci_rq->r_count--; 289 if ((l->l_pflag & LP_BOUND) == 0) 290 ci_rq->r_mcount--; 291 292 q_head = sched_getrq(ci_rq, eprio); 293 TAILQ_REMOVE(q_head, l, l_runq); 294 if (TAILQ_EMPTY(q_head)) { 295 u_int i; 296 uint32_t q; 297 298 /* Unmark bit */ 299 i = eprio >> BITMAP_SHIFT; 300 q = BITMAP_MSB >> (eprio & BITMAP_MASK); 301 KASSERT((ci_rq->r_bitmap[i] & q) != 0); 302 ci_rq->r_bitmap[i] &= ~q; 303 304 /* 305 * Update the value of highest priority in the runqueue, in a 306 * case it was a last thread in the queue of highest priority. 307 */ 308 if (eprio != spc->spc_maxpriority) 309 return; 310 311 do { 312 if (ci_rq->r_bitmap[i] != 0) { 313 q = ffs(ci_rq->r_bitmap[i]); 314 spc->spc_maxpriority = 315 (i << BITMAP_SHIFT) + (BITMAP_BITS - q); 316 return; 317 } 318 } while (i--); 319 320 /* If not found - set the lowest value */ 321 spc->spc_maxpriority = 0; 322 } 323 } 324 325 /* 326 * Migration and balancing. 327 */ 328 329 #ifdef MULTIPROCESSOR 330 331 /* Estimate if LWP is cache-hot */ 332 static inline bool 333 lwp_cache_hot(const struct lwp *l) 334 { 335 336 if (__predict_false(l->l_slptime || l->l_rticks == 0)) 337 return false; 338 339 return (hardclock_ticks - l->l_rticks <= cacheht_time); 340 } 341 342 /* Check if LWP can migrate to the chosen CPU */ 343 static inline bool 344 sched_migratable(const struct lwp *l, struct cpu_info *ci) 345 { 346 const struct schedstate_percpu *spc = &ci->ci_schedstate; 347 KASSERT(lwp_locked(__UNCONST(l), NULL)); 348 349 /* Is CPU offline? */ 350 if (__predict_false(spc->spc_flags & SPCF_OFFLINE)) 351 return false; 352 353 /* Is affinity set? */ 354 if (__predict_false(l->l_affinity)) 355 return kcpuset_isset(l->l_affinity, cpu_index(ci)); 356 357 /* Is there a processor-set? */ 358 return (spc->spc_psid == l->l_psid); 359 } 360 361 /* 362 * Estimate the migration of LWP to the other CPU. 363 * Take and return the CPU, if migration is needed. 364 */ 365 struct cpu_info * 366 sched_takecpu(struct lwp *l) 367 { 368 struct cpu_info *ci, *tci, *pivot, *next; 369 struct schedstate_percpu *spc; 370 runqueue_t *ci_rq, *ici_rq; 371 pri_t eprio, lpri, pri; 372 373 KASSERT(lwp_locked(l, NULL)); 374 375 /* If thread is strictly bound, do not estimate other CPUs */ 376 ci = l->l_cpu; 377 if (l->l_pflag & LP_BOUND) 378 return ci; 379 380 spc = &ci->ci_schedstate; 381 ci_rq = spc->spc_sched_info; 382 383 /* Make sure that thread is in appropriate processor-set */ 384 if (__predict_true(spc->spc_psid == l->l_psid)) { 385 /* If CPU of this thread is idling - run there */ 386 if (ci_rq->r_count == 0) { 387 ci_rq->r_ev_stay.ev_count++; 388 return ci; 389 } 390 /* Stay if thread is cache-hot */ 391 eprio = lwp_eprio(l); 392 if (__predict_true(l->l_stat != LSIDL) && 393 lwp_cache_hot(l) && eprio >= spc->spc_curpriority) { 394 ci_rq->r_ev_stay.ev_count++; 395 return ci; 396 } 397 } else { 398 eprio = lwp_eprio(l); 399 } 400 401 /* Run on current CPU if priority of thread is higher */ 402 ci = curcpu(); 403 spc = &ci->ci_schedstate; 404 if (eprio > spc->spc_curpriority && sched_migratable(l, ci)) { 405 ci_rq = spc->spc_sched_info; 406 ci_rq->r_ev_localize.ev_count++; 407 return ci; 408 } 409 410 /* 411 * Look for the CPU with the lowest priority thread. In case of 412 * equal priority, choose the CPU with the fewest of threads. 413 */ 414 pivot = l->l_cpu; 415 ci = pivot; 416 tci = pivot; 417 lpri = PRI_COUNT; 418 do { 419 if ((next = cpu_lookup(cpu_index(ci) + 1)) == NULL) { 420 /* Reached the end, start from the beginning. */ 421 next = cpu_lookup(0); 422 } 423 spc = &ci->ci_schedstate; 424 ici_rq = spc->spc_sched_info; 425 pri = MAX(spc->spc_curpriority, spc->spc_maxpriority); 426 if (pri > lpri) 427 continue; 428 429 if (pri == lpri && ci_rq->r_count < ici_rq->r_count) 430 continue; 431 432 if (!sched_migratable(l, ci)) 433 continue; 434 435 lpri = pri; 436 tci = ci; 437 ci_rq = ici_rq; 438 } while (ci = next, ci != pivot); 439 440 ci_rq = tci->ci_schedstate.spc_sched_info; 441 ci_rq->r_ev_push.ev_count++; 442 443 return tci; 444 } 445 446 /* 447 * Tries to catch an LWP from the runqueue of other CPU. 448 */ 449 static struct lwp * 450 sched_catchlwp(struct cpu_info *ci) 451 { 452 struct cpu_info *curci = curcpu(); 453 struct schedstate_percpu *spc, *curspc; 454 TAILQ_HEAD(, lwp) *q_head; 455 runqueue_t *ci_rq; 456 struct lwp *l; 457 458 curspc = &curci->ci_schedstate; 459 spc = &ci->ci_schedstate; 460 KASSERT(curspc->spc_psid == spc->spc_psid); 461 462 ci_rq = spc->spc_sched_info; 463 if (ci_rq->r_mcount < min_catch) { 464 spc_unlock(ci); 465 return NULL; 466 } 467 468 /* Take the highest priority thread */ 469 q_head = sched_getrq(ci_rq, spc->spc_maxpriority); 470 l = TAILQ_FIRST(q_head); 471 472 for (;;) { 473 /* Check the first and next result from the queue */ 474 if (l == NULL) { 475 break; 476 } 477 KASSERTMSG(l->l_stat == LSRUN, "%s l %p (%s) l_stat %d", 478 ci->ci_data.cpu_name, 479 l, (l->l_name ? l->l_name : l->l_proc->p_comm), l->l_stat); 480 481 /* Look for threads, whose are allowed to migrate */ 482 if ((l->l_pflag & LP_BOUND) || lwp_cache_hot(l) || 483 !sched_migratable(l, curci)) { 484 l = TAILQ_NEXT(l, l_runq); 485 continue; 486 } 487 488 /* Grab the thread, and move to the local run queue */ 489 sched_dequeue(l); 490 491 /* 492 * If LWP is still context switching, we may need to 493 * spin-wait before changing its CPU. 494 */ 495 if (__predict_false(l->l_ctxswtch != 0)) { 496 u_int count; 497 count = SPINLOCK_BACKOFF_MIN; 498 while (l->l_ctxswtch) 499 SPINLOCK_BACKOFF(count); 500 } 501 l->l_cpu = curci; 502 ci_rq->r_ev_pull.ev_count++; 503 lwp_unlock_to(l, curspc->spc_mutex); 504 sched_enqueue(l, false); 505 return l; 506 } 507 spc_unlock(ci); 508 509 return l; 510 } 511 512 /* 513 * Periodical calculations for balancing. 514 */ 515 static void 516 sched_balance(void *nocallout) 517 { 518 struct cpu_info *ci, *hci; 519 runqueue_t *ci_rq; 520 CPU_INFO_ITERATOR cii; 521 u_int highest; 522 523 hci = curcpu(); 524 highest = 0; 525 526 /* Make lockless countings */ 527 for (CPU_INFO_FOREACH(cii, ci)) { 528 ci_rq = ci->ci_schedstate.spc_sched_info; 529 530 /* Average count of the threads */ 531 ci_rq->r_avgcount = (ci_rq->r_avgcount + ci_rq->r_mcount) >> 1; 532 533 /* Look for CPU with the highest average */ 534 if (ci_rq->r_avgcount > highest) { 535 hci = ci; 536 highest = ci_rq->r_avgcount; 537 } 538 } 539 540 /* Update the worker */ 541 worker_ci = hci; 542 543 if (nocallout == NULL) 544 callout_schedule(&balance_ch, balance_period); 545 } 546 547 /* 548 * Called from each CPU's idle loop. 549 */ 550 void 551 sched_idle(void) 552 { 553 struct cpu_info *ci = curcpu(), *tci = NULL; 554 struct schedstate_percpu *spc, *tspc; 555 runqueue_t *ci_rq; 556 bool dlock = false; 557 558 /* Check if there is a migrating LWP */ 559 spc = &ci->ci_schedstate; 560 if (spc->spc_migrating == NULL) 561 goto no_migration; 562 563 spc_lock(ci); 564 for (;;) { 565 struct lwp *l; 566 567 l = spc->spc_migrating; 568 if (l == NULL) 569 break; 570 571 /* 572 * If second attempt, and target CPU has changed, 573 * drop the old lock. 574 */ 575 if (dlock == true && tci != l->l_target_cpu) { 576 KASSERT(tci != NULL); 577 spc_unlock(tci); 578 dlock = false; 579 } 580 581 /* 582 * Nothing to do if destination has changed to the 583 * local CPU, or migration was done by other CPU. 584 */ 585 tci = l->l_target_cpu; 586 if (tci == NULL || tci == ci) { 587 spc->spc_migrating = NULL; 588 l->l_target_cpu = NULL; 589 break; 590 } 591 tspc = &tci->ci_schedstate; 592 593 /* 594 * Double-lock the runqueues. 595 * We do that only once. 596 */ 597 if (dlock == false) { 598 dlock = true; 599 if (ci < tci) { 600 spc_lock(tci); 601 } else if (!mutex_tryenter(tspc->spc_mutex)) { 602 spc_unlock(ci); 603 spc_lock(tci); 604 spc_lock(ci); 605 /* Check the situation again.. */ 606 continue; 607 } 608 } 609 610 /* Migrate the thread */ 611 KASSERT(l->l_stat == LSRUN); 612 spc->spc_migrating = NULL; 613 l->l_target_cpu = NULL; 614 sched_dequeue(l); 615 l->l_cpu = tci; 616 lwp_setlock(l, tspc->spc_mutex); 617 sched_enqueue(l, false); 618 break; 619 } 620 if (dlock == true) { 621 KASSERT(tci != NULL); 622 spc_unlock(tci); 623 } 624 spc_unlock(ci); 625 626 no_migration: 627 ci_rq = spc->spc_sched_info; 628 if ((spc->spc_flags & SPCF_OFFLINE) != 0 || ci_rq->r_count != 0) { 629 return; 630 } 631 632 /* Reset the counter, and call the balancer */ 633 ci_rq->r_avgcount = 0; 634 sched_balance(ci); 635 tci = worker_ci; 636 tspc = &tci->ci_schedstate; 637 if (ci == tci || spc->spc_psid != tspc->spc_psid) 638 return; 639 spc_dlock(ci, tci); 640 (void)sched_catchlwp(tci); 641 spc_unlock(ci); 642 } 643 644 #else 645 646 /* 647 * stubs for !MULTIPROCESSOR 648 */ 649 650 struct cpu_info * 651 sched_takecpu(struct lwp *l) 652 { 653 654 return l->l_cpu; 655 } 656 657 void 658 sched_idle(void) 659 { 660 661 } 662 #endif /* MULTIPROCESSOR */ 663 664 /* 665 * Scheduling statistics and balancing. 666 */ 667 void 668 sched_lwp_stats(struct lwp *l) 669 { 670 int batch; 671 672 KASSERT(lwp_locked(l, NULL)); 673 674 /* Update sleep time */ 675 if (l->l_stat == LSSLEEP || l->l_stat == LSSTOP || 676 l->l_stat == LSSUSPENDED) 677 l->l_slptime++; 678 679 /* 680 * Set that thread is more CPU-bound, if sum of run time exceeds the 681 * sum of sleep time. Check if thread is CPU-bound a first time. 682 */ 683 batch = (l->l_rticksum > l->l_slpticksum); 684 if (batch != 0) { 685 if ((l->l_flag & LW_BATCH) == 0) 686 batch = 0; 687 l->l_flag |= LW_BATCH; 688 } else 689 l->l_flag &= ~LW_BATCH; 690 691 /* 692 * If thread is CPU-bound and never sleeps, it would occupy the CPU. 693 * In such case reset the value of last sleep, and check it later, if 694 * it is still zero - perform the migration, unmark the batch flag. 695 */ 696 if (batch && (l->l_slptime + l->l_slpticksum) == 0) { 697 if (l->l_slpticks == 0) { 698 if (l->l_target_cpu == NULL && 699 (l->l_stat == LSRUN || l->l_stat == LSONPROC)) { 700 struct cpu_info *ci = sched_takecpu(l); 701 l->l_target_cpu = (ci != l->l_cpu) ? ci : NULL; 702 } 703 l->l_flag &= ~LW_BATCH; 704 } else { 705 l->l_slpticks = 0; 706 } 707 } 708 709 /* Reset the time sums */ 710 l->l_slpticksum = 0; 711 l->l_rticksum = 0; 712 713 /* Scheduler-specific hook */ 714 sched_pstats_hook(l, batch); 715 } 716 717 /* 718 * Scheduler mill. 719 */ 720 struct lwp * 721 sched_nextlwp(void) 722 { 723 struct cpu_info *ci = curcpu(); 724 struct schedstate_percpu *spc; 725 TAILQ_HEAD(, lwp) *q_head; 726 runqueue_t *ci_rq; 727 struct lwp *l; 728 729 /* Return to idle LWP if there is a migrating thread */ 730 spc = &ci->ci_schedstate; 731 if (__predict_false(spc->spc_migrating != NULL)) 732 return NULL; 733 ci_rq = spc->spc_sched_info; 734 735 #ifdef MULTIPROCESSOR 736 /* If runqueue is empty, try to catch some thread from other CPU */ 737 if (__predict_false(ci_rq->r_count == 0)) { 738 struct schedstate_percpu *cspc; 739 struct cpu_info *cci; 740 741 /* Offline CPUs should not perform this, however */ 742 if (__predict_false(spc->spc_flags & SPCF_OFFLINE)) 743 return NULL; 744 745 /* Reset the counter, and call the balancer */ 746 ci_rq->r_avgcount = 0; 747 sched_balance(ci); 748 cci = worker_ci; 749 cspc = &cci->ci_schedstate; 750 if (ci == cci || spc->spc_psid != cspc->spc_psid || 751 !mutex_tryenter(cci->ci_schedstate.spc_mutex)) 752 return NULL; 753 return sched_catchlwp(cci); 754 } 755 #else 756 if (__predict_false(ci_rq->r_count == 0)) 757 return NULL; 758 #endif 759 760 /* Take the highest priority thread */ 761 KASSERT(ci_rq->r_bitmap[spc->spc_maxpriority >> BITMAP_SHIFT]); 762 q_head = sched_getrq(ci_rq, spc->spc_maxpriority); 763 l = TAILQ_FIRST(q_head); 764 KASSERT(l != NULL); 765 766 sched_oncpu(l); 767 l->l_rticks = hardclock_ticks; 768 769 return l; 770 } 771 772 /* 773 * sched_curcpu_runnable_p: return if curcpu() should exit the idle loop. 774 */ 775 776 bool 777 sched_curcpu_runnable_p(void) 778 { 779 const struct cpu_info *ci; 780 const struct schedstate_percpu *spc; 781 const runqueue_t *ci_rq; 782 bool rv; 783 784 kpreempt_disable(); 785 ci = curcpu(); 786 spc = &ci->ci_schedstate; 787 ci_rq = spc->spc_sched_info; 788 789 #ifndef __HAVE_FAST_SOFTINTS 790 if (ci->ci_data.cpu_softints) { 791 kpreempt_enable(); 792 return true; 793 } 794 #endif 795 796 rv = (ci_rq->r_count != 0) ? true : false; 797 kpreempt_enable(); 798 799 return rv; 800 } 801 802 /* 803 * Sysctl nodes and initialization. 804 */ 805 806 SYSCTL_SETUP(sysctl_sched_setup, "sysctl sched setup") 807 { 808 const struct sysctlnode *node = NULL; 809 810 sysctl_createv(clog, 0, NULL, &node, 811 CTLFLAG_PERMANENT, 812 CTLTYPE_NODE, "sched", 813 SYSCTL_DESCR("Scheduler options"), 814 NULL, 0, NULL, 0, 815 CTL_KERN, CTL_CREATE, CTL_EOL); 816 817 if (node == NULL) 818 return; 819 820 sysctl_createv(clog, 0, &node, NULL, 821 CTLFLAG_PERMANENT | CTLFLAG_READWRITE, 822 CTLTYPE_INT, "cacheht_time", 823 SYSCTL_DESCR("Cache hotness time (in ticks)"), 824 NULL, 0, &cacheht_time, 0, 825 CTL_CREATE, CTL_EOL); 826 sysctl_createv(clog, 0, &node, NULL, 827 CTLFLAG_PERMANENT | CTLFLAG_READWRITE, 828 CTLTYPE_INT, "balance_period", 829 SYSCTL_DESCR("Balance period (in ticks)"), 830 NULL, 0, &balance_period, 0, 831 CTL_CREATE, CTL_EOL); 832 sysctl_createv(clog, 0, &node, NULL, 833 CTLFLAG_PERMANENT | CTLFLAG_READWRITE, 834 CTLTYPE_INT, "min_catch", 835 SYSCTL_DESCR("Minimal count of threads for catching"), 836 NULL, 0, &min_catch, 0, 837 CTL_CREATE, CTL_EOL); 838 sysctl_createv(clog, 0, &node, NULL, 839 CTLFLAG_PERMANENT | CTLFLAG_READWRITE, 840 CTLTYPE_INT, "timesoftints", 841 SYSCTL_DESCR("Track CPU time for soft interrupts"), 842 NULL, 0, &softint_timing, 0, 843 CTL_CREATE, CTL_EOL); 844 sysctl_createv(clog, 0, &node, NULL, 845 CTLFLAG_PERMANENT | CTLFLAG_READWRITE, 846 CTLTYPE_INT, "kpreempt_pri", 847 SYSCTL_DESCR("Minimum priority to trigger kernel preemption"), 848 NULL, 0, &sched_kpreempt_pri, 0, 849 CTL_CREATE, CTL_EOL); 850 sysctl_createv(clog, 0, &node, NULL, 851 CTLFLAG_PERMANENT | CTLFLAG_READWRITE, 852 CTLTYPE_INT, "upreempt_pri", 853 SYSCTL_DESCR("Minimum priority to trigger user preemption"), 854 NULL, 0, &sched_upreempt_pri, 0, 855 CTL_CREATE, CTL_EOL); 856 } 857 858 /* 859 * Debugging. 860 */ 861 862 #ifdef DDB 863 864 void 865 sched_print_runqueue(void (*pr)(const char *, ...)) 866 { 867 runqueue_t *ci_rq; 868 struct cpu_info *ci, *tci; 869 struct schedstate_percpu *spc; 870 struct lwp *l; 871 struct proc *p; 872 CPU_INFO_ITERATOR cii; 873 874 for (CPU_INFO_FOREACH(cii, ci)) { 875 int i; 876 877 spc = &ci->ci_schedstate; 878 ci_rq = spc->spc_sched_info; 879 880 (*pr)("Run-queue (CPU = %u):\n", ci->ci_index); 881 (*pr)(" pid.lid = %d.%d, r_count = %u, r_avgcount = %u, " 882 "maxpri = %d, mlwp = %p\n", 883 #ifdef MULTIPROCESSOR 884 ci->ci_curlwp->l_proc->p_pid, ci->ci_curlwp->l_lid, 885 #else 886 curlwp->l_proc->p_pid, curlwp->l_lid, 887 #endif 888 ci_rq->r_count, ci_rq->r_avgcount, spc->spc_maxpriority, 889 spc->spc_migrating); 890 i = (PRI_COUNT >> BITMAP_SHIFT) - 1; 891 do { 892 uint32_t q; 893 q = ci_rq->r_bitmap[i]; 894 (*pr)(" bitmap[%d] => [ %d (0x%x) ]\n", i, ffs(q), q); 895 } while (i--); 896 } 897 898 (*pr)(" %5s %4s %4s %10s %3s %18s %4s %4s %s\n", 899 "LID", "PRI", "EPRI", "FL", "ST", "LWP", "CPU", "TCI", "LRTICKS"); 900 901 PROCLIST_FOREACH(p, &allproc) { 902 (*pr)(" /- %d (%s)\n", (int)p->p_pid, p->p_comm); 903 LIST_FOREACH(l, &p->p_lwps, l_sibling) { 904 ci = l->l_cpu; 905 tci = l->l_target_cpu; 906 (*pr)(" | %5d %4u %4u 0x%8.8x %3s %18p %4u %4d %u\n", 907 (int)l->l_lid, l->l_priority, lwp_eprio(l), 908 l->l_flag, l->l_stat == LSRUN ? "RQ" : 909 (l->l_stat == LSSLEEP ? "SQ" : "-"), 910 l, ci->ci_index, (tci ? tci->ci_index : -1), 911 (u_int)(hardclock_ticks - l->l_rticks)); 912 } 913 } 914 } 915 916 #endif 917