sched_ule.c revision 230083
1/*- 2 * Copyright (c) 2002-2007, Jeffrey Roberson <jeff@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 unmodified, this list of conditions, and the following 10 * disclaimer. 11 * 2. Redistributions in binary form must reproduce the above copyright 12 * notice, this list of conditions and the following disclaimer in the 13 * documentation and/or other materials provided with the distribution. 14 * 15 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR 16 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES 17 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 18 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, 19 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT 20 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 21 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 22 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 23 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF 24 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 25 */ 26 27/* 28 * This file implements the ULE scheduler. ULE supports independent CPU 29 * run queues and fine grain locking. It has superior interactive 30 * performance under load even on uni-processor systems. 31 * 32 * etymology: 33 * ULE is the last three letters in schedule. It owes its name to a 34 * generic user created for a scheduling system by Paul Mikesell at 35 * Isilon Systems and a general lack of creativity on the part of the author. 36 */ 37 38#include <sys/cdefs.h> 39__FBSDID("$FreeBSD: stable/9/sys/kern/sched_ule.c 230083 2012-01-13 20:25:38Z jhb $"); 40 41#include "opt_hwpmc_hooks.h" 42#include "opt_kdtrace.h" 43#include "opt_sched.h" 44 45#include <sys/param.h> 46#include <sys/systm.h> 47#include <sys/kdb.h> 48#include <sys/kernel.h> 49#include <sys/ktr.h> 50#include <sys/lock.h> 51#include <sys/mutex.h> 52#include <sys/proc.h> 53#include <sys/resource.h> 54#include <sys/resourcevar.h> 55#include <sys/sched.h> 56#include <sys/smp.h> 57#include <sys/sx.h> 58#include <sys/sysctl.h> 59#include <sys/sysproto.h> 60#include <sys/turnstile.h> 61#include <sys/umtx.h> 62#include <sys/vmmeter.h> 63#include <sys/cpuset.h> 64#include <sys/sbuf.h> 65 66#ifdef HWPMC_HOOKS 67#include <sys/pmckern.h> 68#endif 69 70#ifdef KDTRACE_HOOKS 71#include <sys/dtrace_bsd.h> 72int dtrace_vtime_active; 73dtrace_vtime_switch_func_t dtrace_vtime_switch_func; 74#endif 75 76#include <machine/cpu.h> 77#include <machine/smp.h> 78 79#if defined(__sparc64__) 80#error "This architecture is not currently compatible with ULE" 81#endif 82 83#define KTR_ULE 0 84 85#define TS_NAME_LEN (MAXCOMLEN + sizeof(" td ") + sizeof(__XSTRING(UINT_MAX))) 86#define TDQ_NAME_LEN (sizeof("sched lock ") + sizeof(__XSTRING(MAXCPU))) 87#define TDQ_LOADNAME_LEN (sizeof("CPU ") + sizeof(__XSTRING(MAXCPU)) - 1 + sizeof(" load")) 88 89/* 90 * Thread scheduler specific section. All fields are protected 91 * by the thread lock. 92 */ 93struct td_sched { 94 struct runq *ts_runq; /* Run-queue we're queued on. */ 95 short ts_flags; /* TSF_* flags. */ 96 u_char ts_cpu; /* CPU that we have affinity for. */ 97 int ts_rltick; /* Real last tick, for affinity. */ 98 int ts_slice; /* Ticks of slice remaining. */ 99 u_int ts_slptime; /* Number of ticks we vol. slept */ 100 u_int ts_runtime; /* Number of ticks we were running */ 101 int ts_ltick; /* Last tick that we were running on */ 102 int ts_incrtick; /* Last tick that we incremented on */ 103 int ts_ftick; /* First tick that we were running on */ 104 int ts_ticks; /* Tick count */ 105#ifdef KTR 106 char ts_name[TS_NAME_LEN]; 107#endif 108}; 109/* flags kept in ts_flags */ 110#define TSF_BOUND 0x0001 /* Thread can not migrate. */ 111#define TSF_XFERABLE 0x0002 /* Thread was added as transferable. */ 112 113static struct td_sched td_sched0; 114 115#define THREAD_CAN_MIGRATE(td) ((td)->td_pinned == 0) 116#define THREAD_CAN_SCHED(td, cpu) \ 117 CPU_ISSET((cpu), &(td)->td_cpuset->cs_mask) 118 119/* 120 * Priority ranges used for interactive and non-interactive timeshare 121 * threads. The timeshare priorities are split up into four ranges. 122 * The first range handles interactive threads. The last three ranges 123 * (NHALF, x, and NHALF) handle non-interactive threads with the outer 124 * ranges supporting nice values. 125 */ 126#define PRI_TIMESHARE_RANGE (PRI_MAX_TIMESHARE - PRI_MIN_TIMESHARE + 1) 127#define PRI_INTERACT_RANGE ((PRI_TIMESHARE_RANGE - SCHED_PRI_NRESV) / 2) 128 129#define PRI_MIN_INTERACT PRI_MIN_TIMESHARE 130#define PRI_MAX_INTERACT (PRI_MIN_TIMESHARE + PRI_INTERACT_RANGE - 1) 131#define PRI_MIN_BATCH (PRI_MIN_TIMESHARE + PRI_INTERACT_RANGE) 132#define PRI_MAX_BATCH PRI_MAX_TIMESHARE 133 134/* 135 * Cpu percentage computation macros and defines. 136 * 137 * SCHED_TICK_SECS: Number of seconds to average the cpu usage across. 138 * SCHED_TICK_TARG: Number of hz ticks to average the cpu usage across. 139 * SCHED_TICK_MAX: Maximum number of ticks before scaling back. 140 * SCHED_TICK_SHIFT: Shift factor to avoid rounding away results. 141 * SCHED_TICK_HZ: Compute the number of hz ticks for a given ticks count. 142 * SCHED_TICK_TOTAL: Gives the amount of time we've been recording ticks. 143 */ 144#define SCHED_TICK_SECS 10 145#define SCHED_TICK_TARG (hz * SCHED_TICK_SECS) 146#define SCHED_TICK_MAX (SCHED_TICK_TARG + hz) 147#define SCHED_TICK_SHIFT 10 148#define SCHED_TICK_HZ(ts) ((ts)->ts_ticks >> SCHED_TICK_SHIFT) 149#define SCHED_TICK_TOTAL(ts) (max((ts)->ts_ltick - (ts)->ts_ftick, hz)) 150 151/* 152 * These macros determine priorities for non-interactive threads. They are 153 * assigned a priority based on their recent cpu utilization as expressed 154 * by the ratio of ticks to the tick total. NHALF priorities at the start 155 * and end of the MIN to MAX timeshare range are only reachable with negative 156 * or positive nice respectively. 157 * 158 * PRI_RANGE: Priority range for utilization dependent priorities. 159 * PRI_NRESV: Number of nice values. 160 * PRI_TICKS: Compute a priority in PRI_RANGE from the ticks count and total. 161 * PRI_NICE: Determines the part of the priority inherited from nice. 162 */ 163#define SCHED_PRI_NRESV (PRIO_MAX - PRIO_MIN) 164#define SCHED_PRI_NHALF (SCHED_PRI_NRESV / 2) 165#define SCHED_PRI_MIN (PRI_MIN_BATCH + SCHED_PRI_NHALF) 166#define SCHED_PRI_MAX (PRI_MAX_BATCH - SCHED_PRI_NHALF) 167#define SCHED_PRI_RANGE (SCHED_PRI_MAX - SCHED_PRI_MIN + 1) 168#define SCHED_PRI_TICKS(ts) \ 169 (SCHED_TICK_HZ((ts)) / \ 170 (roundup(SCHED_TICK_TOTAL((ts)), SCHED_PRI_RANGE) / SCHED_PRI_RANGE)) 171#define SCHED_PRI_NICE(nice) (nice) 172 173/* 174 * These determine the interactivity of a process. Interactivity differs from 175 * cpu utilization in that it expresses the voluntary time slept vs time ran 176 * while cpu utilization includes all time not running. This more accurately 177 * models the intent of the thread. 178 * 179 * SLP_RUN_MAX: Maximum amount of sleep time + run time we'll accumulate 180 * before throttling back. 181 * SLP_RUN_FORK: Maximum slp+run time to inherit at fork time. 182 * INTERACT_MAX: Maximum interactivity value. Smaller is better. 183 * INTERACT_THRESH: Threshold for placement on the current runq. 184 */ 185#define SCHED_SLP_RUN_MAX ((hz * 5) << SCHED_TICK_SHIFT) 186#define SCHED_SLP_RUN_FORK ((hz / 2) << SCHED_TICK_SHIFT) 187#define SCHED_INTERACT_MAX (100) 188#define SCHED_INTERACT_HALF (SCHED_INTERACT_MAX / 2) 189#define SCHED_INTERACT_THRESH (30) 190 191/* 192 * tickincr: Converts a stathz tick into a hz domain scaled by 193 * the shift factor. Without the shift the error rate 194 * due to rounding would be unacceptably high. 195 * realstathz: stathz is sometimes 0 and run off of hz. 196 * sched_slice: Runtime of each thread before rescheduling. 197 * preempt_thresh: Priority threshold for preemption and remote IPIs. 198 */ 199static int sched_interact = SCHED_INTERACT_THRESH; 200static int realstathz; 201static int tickincr; 202static int sched_slice = 1; 203#ifdef PREEMPTION 204#ifdef FULL_PREEMPTION 205static int preempt_thresh = PRI_MAX_IDLE; 206#else 207static int preempt_thresh = PRI_MIN_KERN; 208#endif 209#else 210static int preempt_thresh = 0; 211#endif 212static int static_boost = PRI_MIN_BATCH; 213static int sched_idlespins = 10000; 214static int sched_idlespinthresh = 16; 215 216/* 217 * tdq - per processor runqs and statistics. All fields are protected by the 218 * tdq_lock. The load and lowpri may be accessed without to avoid excess 219 * locking in sched_pickcpu(); 220 */ 221struct tdq { 222 /* Ordered to improve efficiency of cpu_search() and switch(). */ 223 struct mtx tdq_lock; /* run queue lock. */ 224 struct cpu_group *tdq_cg; /* Pointer to cpu topology. */ 225 volatile int tdq_load; /* Aggregate load. */ 226 volatile int tdq_cpu_idle; /* cpu_idle() is active. */ 227 int tdq_sysload; /* For loadavg, !ITHD load. */ 228 int tdq_transferable; /* Transferable thread count. */ 229 short tdq_switchcnt; /* Switches this tick. */ 230 short tdq_oldswitchcnt; /* Switches last tick. */ 231 u_char tdq_lowpri; /* Lowest priority thread. */ 232 u_char tdq_ipipending; /* IPI pending. */ 233 u_char tdq_idx; /* Current insert index. */ 234 u_char tdq_ridx; /* Current removal index. */ 235 struct runq tdq_realtime; /* real-time run queue. */ 236 struct runq tdq_timeshare; /* timeshare run queue. */ 237 struct runq tdq_idle; /* Queue of IDLE threads. */ 238 char tdq_name[TDQ_NAME_LEN]; 239#ifdef KTR 240 char tdq_loadname[TDQ_LOADNAME_LEN]; 241#endif 242} __aligned(64); 243 244/* Idle thread states and config. */ 245#define TDQ_RUNNING 1 246#define TDQ_IDLE 2 247 248#ifdef SMP 249struct cpu_group *cpu_top; /* CPU topology */ 250 251#define SCHED_AFFINITY_DEFAULT (max(1, hz / 1000)) 252#define SCHED_AFFINITY(ts, t) ((ts)->ts_rltick > ticks - ((t) * affinity)) 253 254/* 255 * Run-time tunables. 256 */ 257static int rebalance = 1; 258static int balance_interval = 128; /* Default set in sched_initticks(). */ 259static int affinity; 260static int steal_htt = 1; 261static int steal_idle = 1; 262static int steal_thresh = 2; 263 264/* 265 * One thread queue per processor. 266 */ 267static struct tdq tdq_cpu[MAXCPU]; 268static struct tdq *balance_tdq; 269static int balance_ticks; 270 271#define TDQ_SELF() (&tdq_cpu[PCPU_GET(cpuid)]) 272#define TDQ_CPU(x) (&tdq_cpu[(x)]) 273#define TDQ_ID(x) ((int)((x) - tdq_cpu)) 274#else /* !SMP */ 275static struct tdq tdq_cpu; 276 277#define TDQ_ID(x) (0) 278#define TDQ_SELF() (&tdq_cpu) 279#define TDQ_CPU(x) (&tdq_cpu) 280#endif 281 282#define TDQ_LOCK_ASSERT(t, type) mtx_assert(TDQ_LOCKPTR((t)), (type)) 283#define TDQ_LOCK(t) mtx_lock_spin(TDQ_LOCKPTR((t))) 284#define TDQ_LOCK_FLAGS(t, f) mtx_lock_spin_flags(TDQ_LOCKPTR((t)), (f)) 285#define TDQ_UNLOCK(t) mtx_unlock_spin(TDQ_LOCKPTR((t))) 286#define TDQ_LOCKPTR(t) (&(t)->tdq_lock) 287 288static void sched_priority(struct thread *); 289static void sched_thread_priority(struct thread *, u_char); 290static int sched_interact_score(struct thread *); 291static void sched_interact_update(struct thread *); 292static void sched_interact_fork(struct thread *); 293static void sched_pctcpu_update(struct td_sched *); 294 295/* Operations on per processor queues */ 296static struct thread *tdq_choose(struct tdq *); 297static void tdq_setup(struct tdq *); 298static void tdq_load_add(struct tdq *, struct thread *); 299static void tdq_load_rem(struct tdq *, struct thread *); 300static __inline void tdq_runq_add(struct tdq *, struct thread *, int); 301static __inline void tdq_runq_rem(struct tdq *, struct thread *); 302static inline int sched_shouldpreempt(int, int, int); 303void tdq_print(int cpu); 304static void runq_print(struct runq *rq); 305static void tdq_add(struct tdq *, struct thread *, int); 306#ifdef SMP 307static int tdq_move(struct tdq *, struct tdq *); 308static int tdq_idled(struct tdq *); 309static void tdq_notify(struct tdq *, struct thread *); 310static struct thread *tdq_steal(struct tdq *, int); 311static struct thread *runq_steal(struct runq *, int); 312static int sched_pickcpu(struct thread *, int); 313static void sched_balance(void); 314static int sched_balance_pair(struct tdq *, struct tdq *); 315static inline struct tdq *sched_setcpu(struct thread *, int, int); 316static inline void thread_unblock_switch(struct thread *, struct mtx *); 317static struct mtx *sched_switch_migrate(struct tdq *, struct thread *, int); 318static int sysctl_kern_sched_topology_spec(SYSCTL_HANDLER_ARGS); 319static int sysctl_kern_sched_topology_spec_internal(struct sbuf *sb, 320 struct cpu_group *cg, int indent); 321#endif 322 323static void sched_setup(void *dummy); 324SYSINIT(sched_setup, SI_SUB_RUN_QUEUE, SI_ORDER_FIRST, sched_setup, NULL); 325 326static void sched_initticks(void *dummy); 327SYSINIT(sched_initticks, SI_SUB_CLOCKS, SI_ORDER_THIRD, sched_initticks, 328 NULL); 329 330/* 331 * Print the threads waiting on a run-queue. 332 */ 333static void 334runq_print(struct runq *rq) 335{ 336 struct rqhead *rqh; 337 struct thread *td; 338 int pri; 339 int j; 340 int i; 341 342 for (i = 0; i < RQB_LEN; i++) { 343 printf("\t\trunq bits %d 0x%zx\n", 344 i, rq->rq_status.rqb_bits[i]); 345 for (j = 0; j < RQB_BPW; j++) 346 if (rq->rq_status.rqb_bits[i] & (1ul << j)) { 347 pri = j + (i << RQB_L2BPW); 348 rqh = &rq->rq_queues[pri]; 349 TAILQ_FOREACH(td, rqh, td_runq) { 350 printf("\t\t\ttd %p(%s) priority %d rqindex %d pri %d\n", 351 td, td->td_name, td->td_priority, 352 td->td_rqindex, pri); 353 } 354 } 355 } 356} 357 358/* 359 * Print the status of a per-cpu thread queue. Should be a ddb show cmd. 360 */ 361void 362tdq_print(int cpu) 363{ 364 struct tdq *tdq; 365 366 tdq = TDQ_CPU(cpu); 367 368 printf("tdq %d:\n", TDQ_ID(tdq)); 369 printf("\tlock %p\n", TDQ_LOCKPTR(tdq)); 370 printf("\tLock name: %s\n", tdq->tdq_name); 371 printf("\tload: %d\n", tdq->tdq_load); 372 printf("\tswitch cnt: %d\n", tdq->tdq_switchcnt); 373 printf("\told switch cnt: %d\n", tdq->tdq_oldswitchcnt); 374 printf("\ttimeshare idx: %d\n", tdq->tdq_idx); 375 printf("\ttimeshare ridx: %d\n", tdq->tdq_ridx); 376 printf("\tload transferable: %d\n", tdq->tdq_transferable); 377 printf("\tlowest priority: %d\n", tdq->tdq_lowpri); 378 printf("\trealtime runq:\n"); 379 runq_print(&tdq->tdq_realtime); 380 printf("\ttimeshare runq:\n"); 381 runq_print(&tdq->tdq_timeshare); 382 printf("\tidle runq:\n"); 383 runq_print(&tdq->tdq_idle); 384} 385 386static inline int 387sched_shouldpreempt(int pri, int cpri, int remote) 388{ 389 /* 390 * If the new priority is not better than the current priority there is 391 * nothing to do. 392 */ 393 if (pri >= cpri) 394 return (0); 395 /* 396 * Always preempt idle. 397 */ 398 if (cpri >= PRI_MIN_IDLE) 399 return (1); 400 /* 401 * If preemption is disabled don't preempt others. 402 */ 403 if (preempt_thresh == 0) 404 return (0); 405 /* 406 * Preempt if we exceed the threshold. 407 */ 408 if (pri <= preempt_thresh) 409 return (1); 410 /* 411 * If we're interactive or better and there is non-interactive 412 * or worse running preempt only remote processors. 413 */ 414 if (remote && pri <= PRI_MAX_INTERACT && cpri > PRI_MAX_INTERACT) 415 return (1); 416 return (0); 417} 418 419#define TS_RQ_PPQ (((PRI_MAX_BATCH - PRI_MIN_BATCH) + 1) / RQ_NQS) 420/* 421 * Add a thread to the actual run-queue. Keeps transferable counts up to 422 * date with what is actually on the run-queue. Selects the correct 423 * queue position for timeshare threads. 424 */ 425static __inline void 426tdq_runq_add(struct tdq *tdq, struct thread *td, int flags) 427{ 428 struct td_sched *ts; 429 u_char pri; 430 431 TDQ_LOCK_ASSERT(tdq, MA_OWNED); 432 THREAD_LOCK_ASSERT(td, MA_OWNED); 433 434 pri = td->td_priority; 435 ts = td->td_sched; 436 TD_SET_RUNQ(td); 437 if (THREAD_CAN_MIGRATE(td)) { 438 tdq->tdq_transferable++; 439 ts->ts_flags |= TSF_XFERABLE; 440 } 441 if (pri < PRI_MIN_BATCH) { 442 ts->ts_runq = &tdq->tdq_realtime; 443 } else if (pri <= PRI_MAX_BATCH) { 444 ts->ts_runq = &tdq->tdq_timeshare; 445 KASSERT(pri <= PRI_MAX_BATCH && pri >= PRI_MIN_BATCH, 446 ("Invalid priority %d on timeshare runq", pri)); 447 /* 448 * This queue contains only priorities between MIN and MAX 449 * realtime. Use the whole queue to represent these values. 450 */ 451 if ((flags & (SRQ_BORROWING|SRQ_PREEMPTED)) == 0) { 452 pri = (pri - PRI_MIN_BATCH) / TS_RQ_PPQ; 453 pri = (pri + tdq->tdq_idx) % RQ_NQS; 454 /* 455 * This effectively shortens the queue by one so we 456 * can have a one slot difference between idx and 457 * ridx while we wait for threads to drain. 458 */ 459 if (tdq->tdq_ridx != tdq->tdq_idx && 460 pri == tdq->tdq_ridx) 461 pri = (unsigned char)(pri - 1) % RQ_NQS; 462 } else 463 pri = tdq->tdq_ridx; 464 runq_add_pri(ts->ts_runq, td, pri, flags); 465 return; 466 } else 467 ts->ts_runq = &tdq->tdq_idle; 468 runq_add(ts->ts_runq, td, flags); 469} 470 471/* 472 * Remove a thread from a run-queue. This typically happens when a thread 473 * is selected to run. Running threads are not on the queue and the 474 * transferable count does not reflect them. 475 */ 476static __inline void 477tdq_runq_rem(struct tdq *tdq, struct thread *td) 478{ 479 struct td_sched *ts; 480 481 ts = td->td_sched; 482 TDQ_LOCK_ASSERT(tdq, MA_OWNED); 483 KASSERT(ts->ts_runq != NULL, 484 ("tdq_runq_remove: thread %p null ts_runq", td)); 485 if (ts->ts_flags & TSF_XFERABLE) { 486 tdq->tdq_transferable--; 487 ts->ts_flags &= ~TSF_XFERABLE; 488 } 489 if (ts->ts_runq == &tdq->tdq_timeshare) { 490 if (tdq->tdq_idx != tdq->tdq_ridx) 491 runq_remove_idx(ts->ts_runq, td, &tdq->tdq_ridx); 492 else 493 runq_remove_idx(ts->ts_runq, td, NULL); 494 } else 495 runq_remove(ts->ts_runq, td); 496} 497 498/* 499 * Load is maintained for all threads RUNNING and ON_RUNQ. Add the load 500 * for this thread to the referenced thread queue. 501 */ 502static void 503tdq_load_add(struct tdq *tdq, struct thread *td) 504{ 505 506 TDQ_LOCK_ASSERT(tdq, MA_OWNED); 507 THREAD_LOCK_ASSERT(td, MA_OWNED); 508 509 tdq->tdq_load++; 510 if ((td->td_flags & TDF_NOLOAD) == 0) 511 tdq->tdq_sysload++; 512 KTR_COUNTER0(KTR_SCHED, "load", tdq->tdq_loadname, tdq->tdq_load); 513} 514 515/* 516 * Remove the load from a thread that is transitioning to a sleep state or 517 * exiting. 518 */ 519static void 520tdq_load_rem(struct tdq *tdq, struct thread *td) 521{ 522 523 THREAD_LOCK_ASSERT(td, MA_OWNED); 524 TDQ_LOCK_ASSERT(tdq, MA_OWNED); 525 KASSERT(tdq->tdq_load != 0, 526 ("tdq_load_rem: Removing with 0 load on queue %d", TDQ_ID(tdq))); 527 528 tdq->tdq_load--; 529 if ((td->td_flags & TDF_NOLOAD) == 0) 530 tdq->tdq_sysload--; 531 KTR_COUNTER0(KTR_SCHED, "load", tdq->tdq_loadname, tdq->tdq_load); 532} 533 534/* 535 * Set lowpri to its exact value by searching the run-queue and 536 * evaluating curthread. curthread may be passed as an optimization. 537 */ 538static void 539tdq_setlowpri(struct tdq *tdq, struct thread *ctd) 540{ 541 struct thread *td; 542 543 TDQ_LOCK_ASSERT(tdq, MA_OWNED); 544 if (ctd == NULL) 545 ctd = pcpu_find(TDQ_ID(tdq))->pc_curthread; 546 td = tdq_choose(tdq); 547 if (td == NULL || td->td_priority > ctd->td_priority) 548 tdq->tdq_lowpri = ctd->td_priority; 549 else 550 tdq->tdq_lowpri = td->td_priority; 551} 552 553#ifdef SMP 554struct cpu_search { 555 cpuset_t cs_mask; 556 u_int cs_load; 557 u_int cs_cpu; 558 int cs_limit; /* Min priority for low min load for high. */ 559}; 560 561#define CPU_SEARCH_LOWEST 0x1 562#define CPU_SEARCH_HIGHEST 0x2 563#define CPU_SEARCH_BOTH (CPU_SEARCH_LOWEST|CPU_SEARCH_HIGHEST) 564 565#define CPUSET_FOREACH(cpu, mask) \ 566 for ((cpu) = 0; (cpu) <= mp_maxid; (cpu)++) \ 567 if (CPU_ISSET(cpu, &mask)) 568 569static __inline int cpu_search(struct cpu_group *cg, struct cpu_search *low, 570 struct cpu_search *high, const int match); 571int cpu_search_lowest(struct cpu_group *cg, struct cpu_search *low); 572int cpu_search_highest(struct cpu_group *cg, struct cpu_search *high); 573int cpu_search_both(struct cpu_group *cg, struct cpu_search *low, 574 struct cpu_search *high); 575 576/* 577 * This routine compares according to the match argument and should be 578 * reduced in actual instantiations via constant propagation and dead code 579 * elimination. 580 */ 581static __inline int 582cpu_compare(int cpu, struct cpu_search *low, struct cpu_search *high, 583 const int match) 584{ 585 struct tdq *tdq; 586 587 tdq = TDQ_CPU(cpu); 588 if (match & CPU_SEARCH_LOWEST) 589 if (CPU_ISSET(cpu, &low->cs_mask) && 590 tdq->tdq_load < low->cs_load && 591 tdq->tdq_lowpri > low->cs_limit) { 592 low->cs_cpu = cpu; 593 low->cs_load = tdq->tdq_load; 594 } 595 if (match & CPU_SEARCH_HIGHEST) 596 if (CPU_ISSET(cpu, &high->cs_mask) && 597 tdq->tdq_load >= high->cs_limit && 598 tdq->tdq_load > high->cs_load && 599 tdq->tdq_transferable) { 600 high->cs_cpu = cpu; 601 high->cs_load = tdq->tdq_load; 602 } 603 return (tdq->tdq_load); 604} 605 606/* 607 * Search the tree of cpu_groups for the lowest or highest loaded cpu 608 * according to the match argument. This routine actually compares the 609 * load on all paths through the tree and finds the least loaded cpu on 610 * the least loaded path, which may differ from the least loaded cpu in 611 * the system. This balances work among caches and busses. 612 * 613 * This inline is instantiated in three forms below using constants for the 614 * match argument. It is reduced to the minimum set for each case. It is 615 * also recursive to the depth of the tree. 616 */ 617static __inline int 618cpu_search(struct cpu_group *cg, struct cpu_search *low, 619 struct cpu_search *high, const int match) 620{ 621 int total; 622 623 total = 0; 624 if (cg->cg_children) { 625 struct cpu_search lgroup; 626 struct cpu_search hgroup; 627 struct cpu_group *child; 628 u_int lload; 629 int hload; 630 int load; 631 int i; 632 633 lload = -1; 634 hload = -1; 635 for (i = 0; i < cg->cg_children; i++) { 636 child = &cg->cg_child[i]; 637 if (match & CPU_SEARCH_LOWEST) { 638 lgroup = *low; 639 lgroup.cs_load = -1; 640 } 641 if (match & CPU_SEARCH_HIGHEST) { 642 hgroup = *high; 643 lgroup.cs_load = 0; 644 } 645 switch (match) { 646 case CPU_SEARCH_LOWEST: 647 load = cpu_search_lowest(child, &lgroup); 648 break; 649 case CPU_SEARCH_HIGHEST: 650 load = cpu_search_highest(child, &hgroup); 651 break; 652 case CPU_SEARCH_BOTH: 653 load = cpu_search_both(child, &lgroup, &hgroup); 654 break; 655 } 656 total += load; 657 if (match & CPU_SEARCH_LOWEST) 658 if (load < lload || low->cs_cpu == -1) { 659 *low = lgroup; 660 lload = load; 661 } 662 if (match & CPU_SEARCH_HIGHEST) 663 if (load > hload || high->cs_cpu == -1) { 664 hload = load; 665 *high = hgroup; 666 } 667 } 668 } else { 669 int cpu; 670 671 CPUSET_FOREACH(cpu, cg->cg_mask) 672 total += cpu_compare(cpu, low, high, match); 673 } 674 return (total); 675} 676 677/* 678 * cpu_search instantiations must pass constants to maintain the inline 679 * optimization. 680 */ 681int 682cpu_search_lowest(struct cpu_group *cg, struct cpu_search *low) 683{ 684 return cpu_search(cg, low, NULL, CPU_SEARCH_LOWEST); 685} 686 687int 688cpu_search_highest(struct cpu_group *cg, struct cpu_search *high) 689{ 690 return cpu_search(cg, NULL, high, CPU_SEARCH_HIGHEST); 691} 692 693int 694cpu_search_both(struct cpu_group *cg, struct cpu_search *low, 695 struct cpu_search *high) 696{ 697 return cpu_search(cg, low, high, CPU_SEARCH_BOTH); 698} 699 700/* 701 * Find the cpu with the least load via the least loaded path that has a 702 * lowpri greater than pri pri. A pri of -1 indicates any priority is 703 * acceptable. 704 */ 705static inline int 706sched_lowest(struct cpu_group *cg, cpuset_t mask, int pri) 707{ 708 struct cpu_search low; 709 710 low.cs_cpu = -1; 711 low.cs_load = -1; 712 low.cs_mask = mask; 713 low.cs_limit = pri; 714 cpu_search_lowest(cg, &low); 715 return low.cs_cpu; 716} 717 718/* 719 * Find the cpu with the highest load via the highest loaded path. 720 */ 721static inline int 722sched_highest(struct cpu_group *cg, cpuset_t mask, int minload) 723{ 724 struct cpu_search high; 725 726 high.cs_cpu = -1; 727 high.cs_load = 0; 728 high.cs_mask = mask; 729 high.cs_limit = minload; 730 cpu_search_highest(cg, &high); 731 return high.cs_cpu; 732} 733 734/* 735 * Simultaneously find the highest and lowest loaded cpu reachable via 736 * cg. 737 */ 738static inline void 739sched_both(struct cpu_group *cg, cpuset_t mask, int *lowcpu, int *highcpu) 740{ 741 struct cpu_search high; 742 struct cpu_search low; 743 744 low.cs_cpu = -1; 745 low.cs_limit = -1; 746 low.cs_load = -1; 747 low.cs_mask = mask; 748 high.cs_load = 0; 749 high.cs_cpu = -1; 750 high.cs_limit = -1; 751 high.cs_mask = mask; 752 cpu_search_both(cg, &low, &high); 753 *lowcpu = low.cs_cpu; 754 *highcpu = high.cs_cpu; 755 return; 756} 757 758static void 759sched_balance_group(struct cpu_group *cg) 760{ 761 cpuset_t mask; 762 int high; 763 int low; 764 int i; 765 766 CPU_FILL(&mask); 767 for (;;) { 768 sched_both(cg, mask, &low, &high); 769 if (low == high || low == -1 || high == -1) 770 break; 771 if (sched_balance_pair(TDQ_CPU(high), TDQ_CPU(low))) 772 break; 773 /* 774 * If we failed to move any threads determine which cpu 775 * to kick out of the set and try again. 776 */ 777 if (TDQ_CPU(high)->tdq_transferable == 0) 778 CPU_CLR(high, &mask); 779 else 780 CPU_CLR(low, &mask); 781 } 782 783 for (i = 0; i < cg->cg_children; i++) 784 sched_balance_group(&cg->cg_child[i]); 785} 786 787static void 788sched_balance(void) 789{ 790 struct tdq *tdq; 791 792 /* 793 * Select a random time between .5 * balance_interval and 794 * 1.5 * balance_interval. 795 */ 796 balance_ticks = max(balance_interval / 2, 1); 797 balance_ticks += random() % balance_interval; 798 if (smp_started == 0 || rebalance == 0) 799 return; 800 tdq = TDQ_SELF(); 801 TDQ_UNLOCK(tdq); 802 sched_balance_group(cpu_top); 803 TDQ_LOCK(tdq); 804} 805 806/* 807 * Lock two thread queues using their address to maintain lock order. 808 */ 809static void 810tdq_lock_pair(struct tdq *one, struct tdq *two) 811{ 812 if (one < two) { 813 TDQ_LOCK(one); 814 TDQ_LOCK_FLAGS(two, MTX_DUPOK); 815 } else { 816 TDQ_LOCK(two); 817 TDQ_LOCK_FLAGS(one, MTX_DUPOK); 818 } 819} 820 821/* 822 * Unlock two thread queues. Order is not important here. 823 */ 824static void 825tdq_unlock_pair(struct tdq *one, struct tdq *two) 826{ 827 TDQ_UNLOCK(one); 828 TDQ_UNLOCK(two); 829} 830 831/* 832 * Transfer load between two imbalanced thread queues. 833 */ 834static int 835sched_balance_pair(struct tdq *high, struct tdq *low) 836{ 837 int transferable; 838 int high_load; 839 int low_load; 840 int moved; 841 int move; 842 int diff; 843 int i; 844 845 tdq_lock_pair(high, low); 846 transferable = high->tdq_transferable; 847 high_load = high->tdq_load; 848 low_load = low->tdq_load; 849 moved = 0; 850 /* 851 * Determine what the imbalance is and then adjust that to how many 852 * threads we actually have to give up (transferable). 853 */ 854 if (transferable != 0) { 855 diff = high_load - low_load; 856 move = diff / 2; 857 if (diff & 0x1) 858 move++; 859 move = min(move, transferable); 860 for (i = 0; i < move; i++) 861 moved += tdq_move(high, low); 862 /* 863 * IPI the target cpu to force it to reschedule with the new 864 * workload. 865 */ 866 ipi_cpu(TDQ_ID(low), IPI_PREEMPT); 867 } 868 tdq_unlock_pair(high, low); 869 return (moved); 870} 871 872/* 873 * Move a thread from one thread queue to another. 874 */ 875static int 876tdq_move(struct tdq *from, struct tdq *to) 877{ 878 struct td_sched *ts; 879 struct thread *td; 880 struct tdq *tdq; 881 int cpu; 882 883 TDQ_LOCK_ASSERT(from, MA_OWNED); 884 TDQ_LOCK_ASSERT(to, MA_OWNED); 885 886 tdq = from; 887 cpu = TDQ_ID(to); 888 td = tdq_steal(tdq, cpu); 889 if (td == NULL) 890 return (0); 891 ts = td->td_sched; 892 /* 893 * Although the run queue is locked the thread may be blocked. Lock 894 * it to clear this and acquire the run-queue lock. 895 */ 896 thread_lock(td); 897 /* Drop recursive lock on from acquired via thread_lock(). */ 898 TDQ_UNLOCK(from); 899 sched_rem(td); 900 ts->ts_cpu = cpu; 901 td->td_lock = TDQ_LOCKPTR(to); 902 tdq_add(to, td, SRQ_YIELDING); 903 return (1); 904} 905 906/* 907 * This tdq has idled. Try to steal a thread from another cpu and switch 908 * to it. 909 */ 910static int 911tdq_idled(struct tdq *tdq) 912{ 913 struct cpu_group *cg; 914 struct tdq *steal; 915 cpuset_t mask; 916 int thresh; 917 int cpu; 918 919 if (smp_started == 0 || steal_idle == 0) 920 return (1); 921 CPU_FILL(&mask); 922 CPU_CLR(PCPU_GET(cpuid), &mask); 923 /* We don't want to be preempted while we're iterating. */ 924 spinlock_enter(); 925 for (cg = tdq->tdq_cg; cg != NULL; ) { 926 if ((cg->cg_flags & CG_FLAG_THREAD) == 0) 927 thresh = steal_thresh; 928 else 929 thresh = 1; 930 cpu = sched_highest(cg, mask, thresh); 931 if (cpu == -1) { 932 cg = cg->cg_parent; 933 continue; 934 } 935 steal = TDQ_CPU(cpu); 936 CPU_CLR(cpu, &mask); 937 tdq_lock_pair(tdq, steal); 938 if (steal->tdq_load < thresh || steal->tdq_transferable == 0) { 939 tdq_unlock_pair(tdq, steal); 940 continue; 941 } 942 /* 943 * If a thread was added while interrupts were disabled don't 944 * steal one here. If we fail to acquire one due to affinity 945 * restrictions loop again with this cpu removed from the 946 * set. 947 */ 948 if (tdq->tdq_load == 0 && tdq_move(steal, tdq) == 0) { 949 tdq_unlock_pair(tdq, steal); 950 continue; 951 } 952 spinlock_exit(); 953 TDQ_UNLOCK(steal); 954 mi_switch(SW_VOL | SWT_IDLE, NULL); 955 thread_unlock(curthread); 956 957 return (0); 958 } 959 spinlock_exit(); 960 return (1); 961} 962 963/* 964 * Notify a remote cpu of new work. Sends an IPI if criteria are met. 965 */ 966static void 967tdq_notify(struct tdq *tdq, struct thread *td) 968{ 969 struct thread *ctd; 970 int pri; 971 int cpu; 972 973 if (tdq->tdq_ipipending) 974 return; 975 cpu = td->td_sched->ts_cpu; 976 pri = td->td_priority; 977 ctd = pcpu_find(cpu)->pc_curthread; 978 if (!sched_shouldpreempt(pri, ctd->td_priority, 1)) 979 return; 980 if (TD_IS_IDLETHREAD(ctd)) { 981 /* 982 * If the MD code has an idle wakeup routine try that before 983 * falling back to IPI. 984 */ 985 if (!tdq->tdq_cpu_idle || cpu_idle_wakeup(cpu)) 986 return; 987 } 988 tdq->tdq_ipipending = 1; 989 ipi_cpu(cpu, IPI_PREEMPT); 990} 991 992/* 993 * Steals load from a timeshare queue. Honors the rotating queue head 994 * index. 995 */ 996static struct thread * 997runq_steal_from(struct runq *rq, int cpu, u_char start) 998{ 999 struct rqbits *rqb; 1000 struct rqhead *rqh; 1001 struct thread *td; 1002 int first; 1003 int bit; 1004 int pri; 1005 int i; 1006 1007 rqb = &rq->rq_status; 1008 bit = start & (RQB_BPW -1); 1009 pri = 0; 1010 first = 0; 1011again: 1012 for (i = RQB_WORD(start); i < RQB_LEN; bit = 0, i++) { 1013 if (rqb->rqb_bits[i] == 0) 1014 continue; 1015 if (bit != 0) { 1016 for (pri = bit; pri < RQB_BPW; pri++) 1017 if (rqb->rqb_bits[i] & (1ul << pri)) 1018 break; 1019 if (pri >= RQB_BPW) 1020 continue; 1021 } else 1022 pri = RQB_FFS(rqb->rqb_bits[i]); 1023 pri += (i << RQB_L2BPW); 1024 rqh = &rq->rq_queues[pri]; 1025 TAILQ_FOREACH(td, rqh, td_runq) { 1026 if (first && THREAD_CAN_MIGRATE(td) && 1027 THREAD_CAN_SCHED(td, cpu)) 1028 return (td); 1029 first = 1; 1030 } 1031 } 1032 if (start != 0) { 1033 start = 0; 1034 goto again; 1035 } 1036 1037 return (NULL); 1038} 1039 1040/* 1041 * Steals load from a standard linear queue. 1042 */ 1043static struct thread * 1044runq_steal(struct runq *rq, int cpu) 1045{ 1046 struct rqhead *rqh; 1047 struct rqbits *rqb; 1048 struct thread *td; 1049 int word; 1050 int bit; 1051 1052 rqb = &rq->rq_status; 1053 for (word = 0; word < RQB_LEN; word++) { 1054 if (rqb->rqb_bits[word] == 0) 1055 continue; 1056 for (bit = 0; bit < RQB_BPW; bit++) { 1057 if ((rqb->rqb_bits[word] & (1ul << bit)) == 0) 1058 continue; 1059 rqh = &rq->rq_queues[bit + (word << RQB_L2BPW)]; 1060 TAILQ_FOREACH(td, rqh, td_runq) 1061 if (THREAD_CAN_MIGRATE(td) && 1062 THREAD_CAN_SCHED(td, cpu)) 1063 return (td); 1064 } 1065 } 1066 return (NULL); 1067} 1068 1069/* 1070 * Attempt to steal a thread in priority order from a thread queue. 1071 */ 1072static struct thread * 1073tdq_steal(struct tdq *tdq, int cpu) 1074{ 1075 struct thread *td; 1076 1077 TDQ_LOCK_ASSERT(tdq, MA_OWNED); 1078 if ((td = runq_steal(&tdq->tdq_realtime, cpu)) != NULL) 1079 return (td); 1080 if ((td = runq_steal_from(&tdq->tdq_timeshare, 1081 cpu, tdq->tdq_ridx)) != NULL) 1082 return (td); 1083 return (runq_steal(&tdq->tdq_idle, cpu)); 1084} 1085 1086/* 1087 * Sets the thread lock and ts_cpu to match the requested cpu. Unlocks the 1088 * current lock and returns with the assigned queue locked. 1089 */ 1090static inline struct tdq * 1091sched_setcpu(struct thread *td, int cpu, int flags) 1092{ 1093 1094 struct tdq *tdq; 1095 1096 THREAD_LOCK_ASSERT(td, MA_OWNED); 1097 tdq = TDQ_CPU(cpu); 1098 td->td_sched->ts_cpu = cpu; 1099 /* 1100 * If the lock matches just return the queue. 1101 */ 1102 if (td->td_lock == TDQ_LOCKPTR(tdq)) 1103 return (tdq); 1104#ifdef notyet 1105 /* 1106 * If the thread isn't running its lockptr is a 1107 * turnstile or a sleepqueue. We can just lock_set without 1108 * blocking. 1109 */ 1110 if (TD_CAN_RUN(td)) { 1111 TDQ_LOCK(tdq); 1112 thread_lock_set(td, TDQ_LOCKPTR(tdq)); 1113 return (tdq); 1114 } 1115#endif 1116 /* 1117 * The hard case, migration, we need to block the thread first to 1118 * prevent order reversals with other cpus locks. 1119 */ 1120 spinlock_enter(); 1121 thread_lock_block(td); 1122 TDQ_LOCK(tdq); 1123 thread_lock_unblock(td, TDQ_LOCKPTR(tdq)); 1124 spinlock_exit(); 1125 return (tdq); 1126} 1127 1128SCHED_STAT_DEFINE(pickcpu_intrbind, "Soft interrupt binding"); 1129SCHED_STAT_DEFINE(pickcpu_idle_affinity, "Picked idle cpu based on affinity"); 1130SCHED_STAT_DEFINE(pickcpu_affinity, "Picked cpu based on affinity"); 1131SCHED_STAT_DEFINE(pickcpu_lowest, "Selected lowest load"); 1132SCHED_STAT_DEFINE(pickcpu_local, "Migrated to current cpu"); 1133SCHED_STAT_DEFINE(pickcpu_migration, "Selection may have caused migration"); 1134 1135static int 1136sched_pickcpu(struct thread *td, int flags) 1137{ 1138 struct cpu_group *cg; 1139 struct td_sched *ts; 1140 struct tdq *tdq; 1141 cpuset_t mask; 1142 int self; 1143 int pri; 1144 int cpu; 1145 1146 self = PCPU_GET(cpuid); 1147 ts = td->td_sched; 1148 if (smp_started == 0) 1149 return (self); 1150 /* 1151 * Don't migrate a running thread from sched_switch(). 1152 */ 1153 if ((flags & SRQ_OURSELF) || !THREAD_CAN_MIGRATE(td)) 1154 return (ts->ts_cpu); 1155 /* 1156 * Prefer to run interrupt threads on the processors that generate 1157 * the interrupt. 1158 */ 1159 if (td->td_priority <= PRI_MAX_ITHD && THREAD_CAN_SCHED(td, self) && 1160 curthread->td_intr_nesting_level && ts->ts_cpu != self) { 1161 SCHED_STAT_INC(pickcpu_intrbind); 1162 ts->ts_cpu = self; 1163 } 1164 /* 1165 * If the thread can run on the last cpu and the affinity has not 1166 * expired or it is idle run it there. 1167 */ 1168 pri = td->td_priority; 1169 tdq = TDQ_CPU(ts->ts_cpu); 1170 if (THREAD_CAN_SCHED(td, ts->ts_cpu)) { 1171 if (tdq->tdq_lowpri > PRI_MIN_IDLE) { 1172 SCHED_STAT_INC(pickcpu_idle_affinity); 1173 return (ts->ts_cpu); 1174 } 1175 if (SCHED_AFFINITY(ts, CG_SHARE_L2) && tdq->tdq_lowpri > pri) { 1176 SCHED_STAT_INC(pickcpu_affinity); 1177 return (ts->ts_cpu); 1178 } 1179 } 1180 /* 1181 * Search for the highest level in the tree that still has affinity. 1182 */ 1183 cg = NULL; 1184 for (cg = tdq->tdq_cg; cg != NULL; cg = cg->cg_parent) 1185 if (SCHED_AFFINITY(ts, cg->cg_level)) 1186 break; 1187 cpu = -1; 1188 mask = td->td_cpuset->cs_mask; 1189 if (cg) 1190 cpu = sched_lowest(cg, mask, pri); 1191 if (cpu == -1) 1192 cpu = sched_lowest(cpu_top, mask, -1); 1193 /* 1194 * Compare the lowest loaded cpu to current cpu. 1195 */ 1196 if (THREAD_CAN_SCHED(td, self) && TDQ_CPU(self)->tdq_lowpri > pri && 1197 TDQ_CPU(cpu)->tdq_lowpri < PRI_MIN_IDLE) { 1198 SCHED_STAT_INC(pickcpu_local); 1199 cpu = self; 1200 } else 1201 SCHED_STAT_INC(pickcpu_lowest); 1202 if (cpu != ts->ts_cpu) 1203 SCHED_STAT_INC(pickcpu_migration); 1204 KASSERT(cpu != -1, ("sched_pickcpu: Failed to find a cpu.")); 1205 return (cpu); 1206} 1207#endif 1208 1209/* 1210 * Pick the highest priority task we have and return it. 1211 */ 1212static struct thread * 1213tdq_choose(struct tdq *tdq) 1214{ 1215 struct thread *td; 1216 1217 TDQ_LOCK_ASSERT(tdq, MA_OWNED); 1218 td = runq_choose(&tdq->tdq_realtime); 1219 if (td != NULL) 1220 return (td); 1221 td = runq_choose_from(&tdq->tdq_timeshare, tdq->tdq_ridx); 1222 if (td != NULL) { 1223 KASSERT(td->td_priority >= PRI_MIN_BATCH, 1224 ("tdq_choose: Invalid priority on timeshare queue %d", 1225 td->td_priority)); 1226 return (td); 1227 } 1228 td = runq_choose(&tdq->tdq_idle); 1229 if (td != NULL) { 1230 KASSERT(td->td_priority >= PRI_MIN_IDLE, 1231 ("tdq_choose: Invalid priority on idle queue %d", 1232 td->td_priority)); 1233 return (td); 1234 } 1235 1236 return (NULL); 1237} 1238 1239/* 1240 * Initialize a thread queue. 1241 */ 1242static void 1243tdq_setup(struct tdq *tdq) 1244{ 1245 1246 if (bootverbose) 1247 printf("ULE: setup cpu %d\n", TDQ_ID(tdq)); 1248 runq_init(&tdq->tdq_realtime); 1249 runq_init(&tdq->tdq_timeshare); 1250 runq_init(&tdq->tdq_idle); 1251 snprintf(tdq->tdq_name, sizeof(tdq->tdq_name), 1252 "sched lock %d", (int)TDQ_ID(tdq)); 1253 mtx_init(&tdq->tdq_lock, tdq->tdq_name, "sched lock", 1254 MTX_SPIN | MTX_RECURSE); 1255#ifdef KTR 1256 snprintf(tdq->tdq_loadname, sizeof(tdq->tdq_loadname), 1257 "CPU %d load", (int)TDQ_ID(tdq)); 1258#endif 1259} 1260 1261#ifdef SMP 1262static void 1263sched_setup_smp(void) 1264{ 1265 struct tdq *tdq; 1266 int i; 1267 1268 cpu_top = smp_topo(); 1269 CPU_FOREACH(i) { 1270 tdq = TDQ_CPU(i); 1271 tdq_setup(tdq); 1272 tdq->tdq_cg = smp_topo_find(cpu_top, i); 1273 if (tdq->tdq_cg == NULL) 1274 panic("Can't find cpu group for %d\n", i); 1275 } 1276 balance_tdq = TDQ_SELF(); 1277 sched_balance(); 1278} 1279#endif 1280 1281/* 1282 * Setup the thread queues and initialize the topology based on MD 1283 * information. 1284 */ 1285static void 1286sched_setup(void *dummy) 1287{ 1288 struct tdq *tdq; 1289 1290 tdq = TDQ_SELF(); 1291#ifdef SMP 1292 sched_setup_smp(); 1293#else 1294 tdq_setup(tdq); 1295#endif 1296 /* 1297 * To avoid divide-by-zero, we set realstathz a dummy value 1298 * in case which sched_clock() called before sched_initticks(). 1299 */ 1300 realstathz = hz; 1301 sched_slice = (realstathz/10); /* ~100ms */ 1302 tickincr = 1 << SCHED_TICK_SHIFT; 1303 1304 /* Add thread0's load since it's running. */ 1305 TDQ_LOCK(tdq); 1306 thread0.td_lock = TDQ_LOCKPTR(TDQ_SELF()); 1307 tdq_load_add(tdq, &thread0); 1308 tdq->tdq_lowpri = thread0.td_priority; 1309 TDQ_UNLOCK(tdq); 1310} 1311 1312/* 1313 * This routine determines the tickincr after stathz and hz are setup. 1314 */ 1315/* ARGSUSED */ 1316static void 1317sched_initticks(void *dummy) 1318{ 1319 int incr; 1320 1321 realstathz = stathz ? stathz : hz; 1322 sched_slice = (realstathz/10); /* ~100ms */ 1323 1324 /* 1325 * tickincr is shifted out by 10 to avoid rounding errors due to 1326 * hz not being evenly divisible by stathz on all platforms. 1327 */ 1328 incr = (hz << SCHED_TICK_SHIFT) / realstathz; 1329 /* 1330 * This does not work for values of stathz that are more than 1331 * 1 << SCHED_TICK_SHIFT * hz. In practice this does not happen. 1332 */ 1333 if (incr == 0) 1334 incr = 1; 1335 tickincr = incr; 1336#ifdef SMP 1337 /* 1338 * Set the default balance interval now that we know 1339 * what realstathz is. 1340 */ 1341 balance_interval = realstathz; 1342 /* 1343 * Set steal thresh to roughly log2(mp_ncpu) but no greater than 4. 1344 * This prevents excess thrashing on large machines and excess idle 1345 * on smaller machines. 1346 */ 1347 steal_thresh = min(fls(mp_ncpus) - 1, 3); 1348 affinity = SCHED_AFFINITY_DEFAULT; 1349#endif 1350} 1351 1352 1353/* 1354 * This is the core of the interactivity algorithm. Determines a score based 1355 * on past behavior. It is the ratio of sleep time to run time scaled to 1356 * a [0, 100] integer. This is the voluntary sleep time of a process, which 1357 * differs from the cpu usage because it does not account for time spent 1358 * waiting on a run-queue. Would be prettier if we had floating point. 1359 */ 1360static int 1361sched_interact_score(struct thread *td) 1362{ 1363 struct td_sched *ts; 1364 int div; 1365 1366 ts = td->td_sched; 1367 /* 1368 * The score is only needed if this is likely to be an interactive 1369 * task. Don't go through the expense of computing it if there's 1370 * no chance. 1371 */ 1372 if (sched_interact <= SCHED_INTERACT_HALF && 1373 ts->ts_runtime >= ts->ts_slptime) 1374 return (SCHED_INTERACT_HALF); 1375 1376 if (ts->ts_runtime > ts->ts_slptime) { 1377 div = max(1, ts->ts_runtime / SCHED_INTERACT_HALF); 1378 return (SCHED_INTERACT_HALF + 1379 (SCHED_INTERACT_HALF - (ts->ts_slptime / div))); 1380 } 1381 if (ts->ts_slptime > ts->ts_runtime) { 1382 div = max(1, ts->ts_slptime / SCHED_INTERACT_HALF); 1383 return (ts->ts_runtime / div); 1384 } 1385 /* runtime == slptime */ 1386 if (ts->ts_runtime) 1387 return (SCHED_INTERACT_HALF); 1388 1389 /* 1390 * This can happen if slptime and runtime are 0. 1391 */ 1392 return (0); 1393 1394} 1395 1396/* 1397 * Scale the scheduling priority according to the "interactivity" of this 1398 * process. 1399 */ 1400static void 1401sched_priority(struct thread *td) 1402{ 1403 int score; 1404 int pri; 1405 1406 if (PRI_BASE(td->td_pri_class) != PRI_TIMESHARE) 1407 return; 1408 /* 1409 * If the score is interactive we place the thread in the realtime 1410 * queue with a priority that is less than kernel and interrupt 1411 * priorities. These threads are not subject to nice restrictions. 1412 * 1413 * Scores greater than this are placed on the normal timeshare queue 1414 * where the priority is partially decided by the most recent cpu 1415 * utilization and the rest is decided by nice value. 1416 * 1417 * The nice value of the process has a linear effect on the calculated 1418 * score. Negative nice values make it easier for a thread to be 1419 * considered interactive. 1420 */ 1421 score = imax(0, sched_interact_score(td) + td->td_proc->p_nice); 1422 if (score < sched_interact) { 1423 pri = PRI_MIN_INTERACT; 1424 pri += ((PRI_MAX_INTERACT - PRI_MIN_INTERACT + 1) / 1425 sched_interact) * score; 1426 KASSERT(pri >= PRI_MIN_INTERACT && pri <= PRI_MAX_INTERACT, 1427 ("sched_priority: invalid interactive priority %d score %d", 1428 pri, score)); 1429 } else { 1430 pri = SCHED_PRI_MIN; 1431 if (td->td_sched->ts_ticks) 1432 pri += min(SCHED_PRI_TICKS(td->td_sched), 1433 SCHED_PRI_RANGE); 1434 pri += SCHED_PRI_NICE(td->td_proc->p_nice); 1435 KASSERT(pri >= PRI_MIN_BATCH && pri <= PRI_MAX_BATCH, 1436 ("sched_priority: invalid priority %d: nice %d, " 1437 "ticks %d ftick %d ltick %d tick pri %d", 1438 pri, td->td_proc->p_nice, td->td_sched->ts_ticks, 1439 td->td_sched->ts_ftick, td->td_sched->ts_ltick, 1440 SCHED_PRI_TICKS(td->td_sched))); 1441 } 1442 sched_user_prio(td, pri); 1443 1444 return; 1445} 1446 1447/* 1448 * This routine enforces a maximum limit on the amount of scheduling history 1449 * kept. It is called after either the slptime or runtime is adjusted. This 1450 * function is ugly due to integer math. 1451 */ 1452static void 1453sched_interact_update(struct thread *td) 1454{ 1455 struct td_sched *ts; 1456 u_int sum; 1457 1458 ts = td->td_sched; 1459 sum = ts->ts_runtime + ts->ts_slptime; 1460 if (sum < SCHED_SLP_RUN_MAX) 1461 return; 1462 /* 1463 * This only happens from two places: 1464 * 1) We have added an unusual amount of run time from fork_exit. 1465 * 2) We have added an unusual amount of sleep time from sched_sleep(). 1466 */ 1467 if (sum > SCHED_SLP_RUN_MAX * 2) { 1468 if (ts->ts_runtime > ts->ts_slptime) { 1469 ts->ts_runtime = SCHED_SLP_RUN_MAX; 1470 ts->ts_slptime = 1; 1471 } else { 1472 ts->ts_slptime = SCHED_SLP_RUN_MAX; 1473 ts->ts_runtime = 1; 1474 } 1475 return; 1476 } 1477 /* 1478 * If we have exceeded by more than 1/5th then the algorithm below 1479 * will not bring us back into range. Dividing by two here forces 1480 * us into the range of [4/5 * SCHED_INTERACT_MAX, SCHED_INTERACT_MAX] 1481 */ 1482 if (sum > (SCHED_SLP_RUN_MAX / 5) * 6) { 1483 ts->ts_runtime /= 2; 1484 ts->ts_slptime /= 2; 1485 return; 1486 } 1487 ts->ts_runtime = (ts->ts_runtime / 5) * 4; 1488 ts->ts_slptime = (ts->ts_slptime / 5) * 4; 1489} 1490 1491/* 1492 * Scale back the interactivity history when a child thread is created. The 1493 * history is inherited from the parent but the thread may behave totally 1494 * differently. For example, a shell spawning a compiler process. We want 1495 * to learn that the compiler is behaving badly very quickly. 1496 */ 1497static void 1498sched_interact_fork(struct thread *td) 1499{ 1500 int ratio; 1501 int sum; 1502 1503 sum = td->td_sched->ts_runtime + td->td_sched->ts_slptime; 1504 if (sum > SCHED_SLP_RUN_FORK) { 1505 ratio = sum / SCHED_SLP_RUN_FORK; 1506 td->td_sched->ts_runtime /= ratio; 1507 td->td_sched->ts_slptime /= ratio; 1508 } 1509} 1510 1511/* 1512 * Called from proc0_init() to setup the scheduler fields. 1513 */ 1514void 1515schedinit(void) 1516{ 1517 1518 /* 1519 * Set up the scheduler specific parts of proc0. 1520 */ 1521 proc0.p_sched = NULL; /* XXX */ 1522 thread0.td_sched = &td_sched0; 1523 td_sched0.ts_ltick = ticks; 1524 td_sched0.ts_ftick = ticks; 1525 td_sched0.ts_slice = sched_slice; 1526} 1527 1528/* 1529 * This is only somewhat accurate since given many processes of the same 1530 * priority they will switch when their slices run out, which will be 1531 * at most sched_slice stathz ticks. 1532 */ 1533int 1534sched_rr_interval(void) 1535{ 1536 1537 /* Convert sched_slice to hz */ 1538 return (hz/(realstathz/sched_slice)); 1539} 1540 1541/* 1542 * Update the percent cpu tracking information when it is requested or 1543 * the total history exceeds the maximum. We keep a sliding history of 1544 * tick counts that slowly decays. This is less precise than the 4BSD 1545 * mechanism since it happens with less regular and frequent events. 1546 */ 1547static void 1548sched_pctcpu_update(struct td_sched *ts) 1549{ 1550 1551 if (ts->ts_ticks == 0) 1552 return; 1553 if (ticks - (hz / 10) < ts->ts_ltick && 1554 SCHED_TICK_TOTAL(ts) < SCHED_TICK_MAX) 1555 return; 1556 /* 1557 * Adjust counters and watermark for pctcpu calc. 1558 */ 1559 if (ts->ts_ltick > ticks - SCHED_TICK_TARG) 1560 ts->ts_ticks = (ts->ts_ticks / (ticks - ts->ts_ftick)) * 1561 SCHED_TICK_TARG; 1562 else 1563 ts->ts_ticks = 0; 1564 ts->ts_ltick = ticks; 1565 ts->ts_ftick = ts->ts_ltick - SCHED_TICK_TARG; 1566} 1567 1568/* 1569 * Adjust the priority of a thread. Move it to the appropriate run-queue 1570 * if necessary. This is the back-end for several priority related 1571 * functions. 1572 */ 1573static void 1574sched_thread_priority(struct thread *td, u_char prio) 1575{ 1576 struct td_sched *ts; 1577 struct tdq *tdq; 1578 int oldpri; 1579 1580 KTR_POINT3(KTR_SCHED, "thread", sched_tdname(td), "prio", 1581 "prio:%d", td->td_priority, "new prio:%d", prio, 1582 KTR_ATTR_LINKED, sched_tdname(curthread)); 1583 if (td != curthread && prio > td->td_priority) { 1584 KTR_POINT3(KTR_SCHED, "thread", sched_tdname(curthread), 1585 "lend prio", "prio:%d", td->td_priority, "new prio:%d", 1586 prio, KTR_ATTR_LINKED, sched_tdname(td)); 1587 } 1588 ts = td->td_sched; 1589 THREAD_LOCK_ASSERT(td, MA_OWNED); 1590 if (td->td_priority == prio) 1591 return; 1592 /* 1593 * If the priority has been elevated due to priority 1594 * propagation, we may have to move ourselves to a new 1595 * queue. This could be optimized to not re-add in some 1596 * cases. 1597 */ 1598 if (TD_ON_RUNQ(td) && prio < td->td_priority) { 1599 sched_rem(td); 1600 td->td_priority = prio; 1601 sched_add(td, SRQ_BORROWING); 1602 return; 1603 } 1604 /* 1605 * If the thread is currently running we may have to adjust the lowpri 1606 * information so other cpus are aware of our current priority. 1607 */ 1608 if (TD_IS_RUNNING(td)) { 1609 tdq = TDQ_CPU(ts->ts_cpu); 1610 oldpri = td->td_priority; 1611 td->td_priority = prio; 1612 if (prio < tdq->tdq_lowpri) 1613 tdq->tdq_lowpri = prio; 1614 else if (tdq->tdq_lowpri == oldpri) 1615 tdq_setlowpri(tdq, td); 1616 return; 1617 } 1618 td->td_priority = prio; 1619} 1620 1621/* 1622 * Update a thread's priority when it is lent another thread's 1623 * priority. 1624 */ 1625void 1626sched_lend_prio(struct thread *td, u_char prio) 1627{ 1628 1629 td->td_flags |= TDF_BORROWING; 1630 sched_thread_priority(td, prio); 1631} 1632 1633/* 1634 * Restore a thread's priority when priority propagation is 1635 * over. The prio argument is the minimum priority the thread 1636 * needs to have to satisfy other possible priority lending 1637 * requests. If the thread's regular priority is less 1638 * important than prio, the thread will keep a priority boost 1639 * of prio. 1640 */ 1641void 1642sched_unlend_prio(struct thread *td, u_char prio) 1643{ 1644 u_char base_pri; 1645 1646 if (td->td_base_pri >= PRI_MIN_TIMESHARE && 1647 td->td_base_pri <= PRI_MAX_TIMESHARE) 1648 base_pri = td->td_user_pri; 1649 else 1650 base_pri = td->td_base_pri; 1651 if (prio >= base_pri) { 1652 td->td_flags &= ~TDF_BORROWING; 1653 sched_thread_priority(td, base_pri); 1654 } else 1655 sched_lend_prio(td, prio); 1656} 1657 1658/* 1659 * Standard entry for setting the priority to an absolute value. 1660 */ 1661void 1662sched_prio(struct thread *td, u_char prio) 1663{ 1664 u_char oldprio; 1665 1666 /* First, update the base priority. */ 1667 td->td_base_pri = prio; 1668 1669 /* 1670 * If the thread is borrowing another thread's priority, don't 1671 * ever lower the priority. 1672 */ 1673 if (td->td_flags & TDF_BORROWING && td->td_priority < prio) 1674 return; 1675 1676 /* Change the real priority. */ 1677 oldprio = td->td_priority; 1678 sched_thread_priority(td, prio); 1679 1680 /* 1681 * If the thread is on a turnstile, then let the turnstile update 1682 * its state. 1683 */ 1684 if (TD_ON_LOCK(td) && oldprio != prio) 1685 turnstile_adjust(td, oldprio); 1686} 1687 1688/* 1689 * Set the base user priority, does not effect current running priority. 1690 */ 1691void 1692sched_user_prio(struct thread *td, u_char prio) 1693{ 1694 1695 td->td_base_user_pri = prio; 1696 if (td->td_lend_user_pri <= prio) 1697 return; 1698 td->td_user_pri = prio; 1699} 1700 1701void 1702sched_lend_user_prio(struct thread *td, u_char prio) 1703{ 1704 1705 THREAD_LOCK_ASSERT(td, MA_OWNED); 1706 td->td_lend_user_pri = prio; 1707 td->td_user_pri = min(prio, td->td_base_user_pri); 1708 if (td->td_priority > td->td_user_pri) 1709 sched_prio(td, td->td_user_pri); 1710 else if (td->td_priority != td->td_user_pri) 1711 td->td_flags |= TDF_NEEDRESCHED; 1712} 1713 1714/* 1715 * Handle migration from sched_switch(). This happens only for 1716 * cpu binding. 1717 */ 1718static struct mtx * 1719sched_switch_migrate(struct tdq *tdq, struct thread *td, int flags) 1720{ 1721 struct tdq *tdn; 1722 1723 tdn = TDQ_CPU(td->td_sched->ts_cpu); 1724#ifdef SMP 1725 tdq_load_rem(tdq, td); 1726 /* 1727 * Do the lock dance required to avoid LOR. We grab an extra 1728 * spinlock nesting to prevent preemption while we're 1729 * not holding either run-queue lock. 1730 */ 1731 spinlock_enter(); 1732 thread_lock_block(td); /* This releases the lock on tdq. */ 1733 1734 /* 1735 * Acquire both run-queue locks before placing the thread on the new 1736 * run-queue to avoid deadlocks created by placing a thread with a 1737 * blocked lock on the run-queue of a remote processor. The deadlock 1738 * occurs when a third processor attempts to lock the two queues in 1739 * question while the target processor is spinning with its own 1740 * run-queue lock held while waiting for the blocked lock to clear. 1741 */ 1742 tdq_lock_pair(tdn, tdq); 1743 tdq_add(tdn, td, flags); 1744 tdq_notify(tdn, td); 1745 TDQ_UNLOCK(tdn); 1746 spinlock_exit(); 1747#endif 1748 return (TDQ_LOCKPTR(tdn)); 1749} 1750 1751/* 1752 * Variadic version of thread_lock_unblock() that does not assume td_lock 1753 * is blocked. 1754 */ 1755static inline void 1756thread_unblock_switch(struct thread *td, struct mtx *mtx) 1757{ 1758 atomic_store_rel_ptr((volatile uintptr_t *)&td->td_lock, 1759 (uintptr_t)mtx); 1760} 1761 1762/* 1763 * Switch threads. This function has to handle threads coming in while 1764 * blocked for some reason, running, or idle. It also must deal with 1765 * migrating a thread from one queue to another as running threads may 1766 * be assigned elsewhere via binding. 1767 */ 1768void 1769sched_switch(struct thread *td, struct thread *newtd, int flags) 1770{ 1771 struct tdq *tdq; 1772 struct td_sched *ts; 1773 struct mtx *mtx; 1774 int srqflag; 1775 int cpuid; 1776 1777 THREAD_LOCK_ASSERT(td, MA_OWNED); 1778 KASSERT(newtd == NULL, ("sched_switch: Unsupported newtd argument")); 1779 1780 cpuid = PCPU_GET(cpuid); 1781 tdq = TDQ_CPU(cpuid); 1782 ts = td->td_sched; 1783 mtx = td->td_lock; 1784 ts->ts_rltick = ticks; 1785 td->td_lastcpu = td->td_oncpu; 1786 td->td_oncpu = NOCPU; 1787 if (!(flags & SW_PREEMPT)) 1788 td->td_flags &= ~TDF_NEEDRESCHED; 1789 td->td_owepreempt = 0; 1790 tdq->tdq_switchcnt++; 1791 /* 1792 * The lock pointer in an idle thread should never change. Reset it 1793 * to CAN_RUN as well. 1794 */ 1795 if (TD_IS_IDLETHREAD(td)) { 1796 MPASS(td->td_lock == TDQ_LOCKPTR(tdq)); 1797 TD_SET_CAN_RUN(td); 1798 } else if (TD_IS_RUNNING(td)) { 1799 MPASS(td->td_lock == TDQ_LOCKPTR(tdq)); 1800 srqflag = (flags & SW_PREEMPT) ? 1801 SRQ_OURSELF|SRQ_YIELDING|SRQ_PREEMPTED : 1802 SRQ_OURSELF|SRQ_YIELDING; 1803#ifdef SMP 1804 if (THREAD_CAN_MIGRATE(td) && !THREAD_CAN_SCHED(td, ts->ts_cpu)) 1805 ts->ts_cpu = sched_pickcpu(td, 0); 1806#endif 1807 if (ts->ts_cpu == cpuid) 1808 tdq_runq_add(tdq, td, srqflag); 1809 else { 1810 KASSERT(THREAD_CAN_MIGRATE(td) || 1811 (ts->ts_flags & TSF_BOUND) != 0, 1812 ("Thread %p shouldn't migrate", td)); 1813 mtx = sched_switch_migrate(tdq, td, srqflag); 1814 } 1815 } else { 1816 /* This thread must be going to sleep. */ 1817 TDQ_LOCK(tdq); 1818 mtx = thread_lock_block(td); 1819 tdq_load_rem(tdq, td); 1820 } 1821 /* 1822 * We enter here with the thread blocked and assigned to the 1823 * appropriate cpu run-queue or sleep-queue and with the current 1824 * thread-queue locked. 1825 */ 1826 TDQ_LOCK_ASSERT(tdq, MA_OWNED | MA_NOTRECURSED); 1827 newtd = choosethread(); 1828 /* 1829 * Call the MD code to switch contexts if necessary. 1830 */ 1831 if (td != newtd) { 1832#ifdef HWPMC_HOOKS 1833 if (PMC_PROC_IS_USING_PMCS(td->td_proc)) 1834 PMC_SWITCH_CONTEXT(td, PMC_FN_CSW_OUT); 1835#endif 1836 lock_profile_release_lock(&TDQ_LOCKPTR(tdq)->lock_object); 1837 TDQ_LOCKPTR(tdq)->mtx_lock = (uintptr_t)newtd; 1838 1839#ifdef KDTRACE_HOOKS 1840 /* 1841 * If DTrace has set the active vtime enum to anything 1842 * other than INACTIVE (0), then it should have set the 1843 * function to call. 1844 */ 1845 if (dtrace_vtime_active) 1846 (*dtrace_vtime_switch_func)(newtd); 1847#endif 1848 1849 cpu_switch(td, newtd, mtx); 1850 /* 1851 * We may return from cpu_switch on a different cpu. However, 1852 * we always return with td_lock pointing to the current cpu's 1853 * run queue lock. 1854 */ 1855 cpuid = PCPU_GET(cpuid); 1856 tdq = TDQ_CPU(cpuid); 1857 lock_profile_obtain_lock_success( 1858 &TDQ_LOCKPTR(tdq)->lock_object, 0, 0, __FILE__, __LINE__); 1859#ifdef HWPMC_HOOKS 1860 if (PMC_PROC_IS_USING_PMCS(td->td_proc)) 1861 PMC_SWITCH_CONTEXT(td, PMC_FN_CSW_IN); 1862#endif 1863 } else 1864 thread_unblock_switch(td, mtx); 1865 /* 1866 * Assert that all went well and return. 1867 */ 1868 TDQ_LOCK_ASSERT(tdq, MA_OWNED|MA_NOTRECURSED); 1869 MPASS(td->td_lock == TDQ_LOCKPTR(tdq)); 1870 td->td_oncpu = cpuid; 1871} 1872 1873/* 1874 * Adjust thread priorities as a result of a nice request. 1875 */ 1876void 1877sched_nice(struct proc *p, int nice) 1878{ 1879 struct thread *td; 1880 1881 PROC_LOCK_ASSERT(p, MA_OWNED); 1882 1883 p->p_nice = nice; 1884 FOREACH_THREAD_IN_PROC(p, td) { 1885 thread_lock(td); 1886 sched_priority(td); 1887 sched_prio(td, td->td_base_user_pri); 1888 thread_unlock(td); 1889 } 1890} 1891 1892/* 1893 * Record the sleep time for the interactivity scorer. 1894 */ 1895void 1896sched_sleep(struct thread *td, int prio) 1897{ 1898 1899 THREAD_LOCK_ASSERT(td, MA_OWNED); 1900 1901 td->td_slptick = ticks; 1902 if (TD_IS_SUSPENDED(td) || prio >= PSOCK) 1903 td->td_flags |= TDF_CANSWAP; 1904 if (PRI_BASE(td->td_pri_class) != PRI_TIMESHARE) 1905 return; 1906 if (static_boost == 1 && prio) 1907 sched_prio(td, prio); 1908 else if (static_boost && td->td_priority > static_boost) 1909 sched_prio(td, static_boost); 1910} 1911 1912/* 1913 * Schedule a thread to resume execution and record how long it voluntarily 1914 * slept. We also update the pctcpu, interactivity, and priority. 1915 */ 1916void 1917sched_wakeup(struct thread *td) 1918{ 1919 struct td_sched *ts; 1920 int slptick; 1921 1922 THREAD_LOCK_ASSERT(td, MA_OWNED); 1923 ts = td->td_sched; 1924 td->td_flags &= ~TDF_CANSWAP; 1925 /* 1926 * If we slept for more than a tick update our interactivity and 1927 * priority. 1928 */ 1929 slptick = td->td_slptick; 1930 td->td_slptick = 0; 1931 if (slptick && slptick != ticks) { 1932 u_int hzticks; 1933 1934 hzticks = (ticks - slptick) << SCHED_TICK_SHIFT; 1935 ts->ts_slptime += hzticks; 1936 sched_interact_update(td); 1937 sched_pctcpu_update(ts); 1938 } 1939 /* Reset the slice value after we sleep. */ 1940 ts->ts_slice = sched_slice; 1941 sched_add(td, SRQ_BORING); 1942} 1943 1944/* 1945 * Penalize the parent for creating a new child and initialize the child's 1946 * priority. 1947 */ 1948void 1949sched_fork(struct thread *td, struct thread *child) 1950{ 1951 THREAD_LOCK_ASSERT(td, MA_OWNED); 1952 sched_fork_thread(td, child); 1953 /* 1954 * Penalize the parent and child for forking. 1955 */ 1956 sched_interact_fork(child); 1957 sched_priority(child); 1958 td->td_sched->ts_runtime += tickincr; 1959 sched_interact_update(td); 1960 sched_priority(td); 1961} 1962 1963/* 1964 * Fork a new thread, may be within the same process. 1965 */ 1966void 1967sched_fork_thread(struct thread *td, struct thread *child) 1968{ 1969 struct td_sched *ts; 1970 struct td_sched *ts2; 1971 1972 THREAD_LOCK_ASSERT(td, MA_OWNED); 1973 /* 1974 * Initialize child. 1975 */ 1976 ts = td->td_sched; 1977 ts2 = child->td_sched; 1978 child->td_lock = TDQ_LOCKPTR(TDQ_SELF()); 1979 child->td_cpuset = cpuset_ref(td->td_cpuset); 1980 ts2->ts_cpu = ts->ts_cpu; 1981 ts2->ts_flags = 0; 1982 /* 1983 * Grab our parents cpu estimation information. 1984 */ 1985 ts2->ts_ticks = ts->ts_ticks; 1986 ts2->ts_ltick = ts->ts_ltick; 1987 ts2->ts_incrtick = ts->ts_incrtick; 1988 ts2->ts_ftick = ts->ts_ftick; 1989 /* 1990 * Do not inherit any borrowed priority from the parent. 1991 */ 1992 child->td_priority = child->td_base_pri; 1993 /* 1994 * And update interactivity score. 1995 */ 1996 ts2->ts_slptime = ts->ts_slptime; 1997 ts2->ts_runtime = ts->ts_runtime; 1998 ts2->ts_slice = 1; /* Attempt to quickly learn interactivity. */ 1999#ifdef KTR 2000 bzero(ts2->ts_name, sizeof(ts2->ts_name)); 2001#endif 2002} 2003 2004/* 2005 * Adjust the priority class of a thread. 2006 */ 2007void 2008sched_class(struct thread *td, int class) 2009{ 2010 2011 THREAD_LOCK_ASSERT(td, MA_OWNED); 2012 if (td->td_pri_class == class) 2013 return; 2014 td->td_pri_class = class; 2015} 2016 2017/* 2018 * Return some of the child's priority and interactivity to the parent. 2019 */ 2020void 2021sched_exit(struct proc *p, struct thread *child) 2022{ 2023 struct thread *td; 2024 2025 KTR_STATE1(KTR_SCHED, "thread", sched_tdname(child), "proc exit", 2026 "prio:%d", child->td_priority); 2027 PROC_LOCK_ASSERT(p, MA_OWNED); 2028 td = FIRST_THREAD_IN_PROC(p); 2029 sched_exit_thread(td, child); 2030} 2031 2032/* 2033 * Penalize another thread for the time spent on this one. This helps to 2034 * worsen the priority and interactivity of processes which schedule batch 2035 * jobs such as make. This has little effect on the make process itself but 2036 * causes new processes spawned by it to receive worse scores immediately. 2037 */ 2038void 2039sched_exit_thread(struct thread *td, struct thread *child) 2040{ 2041 2042 KTR_STATE1(KTR_SCHED, "thread", sched_tdname(child), "thread exit", 2043 "prio:%d", child->td_priority); 2044 /* 2045 * Give the child's runtime to the parent without returning the 2046 * sleep time as a penalty to the parent. This causes shells that 2047 * launch expensive things to mark their children as expensive. 2048 */ 2049 thread_lock(td); 2050 td->td_sched->ts_runtime += child->td_sched->ts_runtime; 2051 sched_interact_update(td); 2052 sched_priority(td); 2053 thread_unlock(td); 2054} 2055 2056void 2057sched_preempt(struct thread *td) 2058{ 2059 struct tdq *tdq; 2060 2061 thread_lock(td); 2062 tdq = TDQ_SELF(); 2063 TDQ_LOCK_ASSERT(tdq, MA_OWNED); 2064 tdq->tdq_ipipending = 0; 2065 if (td->td_priority > tdq->tdq_lowpri) { 2066 int flags; 2067 2068 flags = SW_INVOL | SW_PREEMPT; 2069 if (td->td_critnest > 1) 2070 td->td_owepreempt = 1; 2071 else if (TD_IS_IDLETHREAD(td)) 2072 mi_switch(flags | SWT_REMOTEWAKEIDLE, NULL); 2073 else 2074 mi_switch(flags | SWT_REMOTEPREEMPT, NULL); 2075 } 2076 thread_unlock(td); 2077} 2078 2079/* 2080 * Fix priorities on return to user-space. Priorities may be elevated due 2081 * to static priorities in msleep() or similar. 2082 */ 2083void 2084sched_userret(struct thread *td) 2085{ 2086 /* 2087 * XXX we cheat slightly on the locking here to avoid locking in 2088 * the usual case. Setting td_priority here is essentially an 2089 * incomplete workaround for not setting it properly elsewhere. 2090 * Now that some interrupt handlers are threads, not setting it 2091 * properly elsewhere can clobber it in the window between setting 2092 * it here and returning to user mode, so don't waste time setting 2093 * it perfectly here. 2094 */ 2095 KASSERT((td->td_flags & TDF_BORROWING) == 0, 2096 ("thread with borrowed priority returning to userland")); 2097 if (td->td_priority != td->td_user_pri) { 2098 thread_lock(td); 2099 td->td_priority = td->td_user_pri; 2100 td->td_base_pri = td->td_user_pri; 2101 tdq_setlowpri(TDQ_SELF(), td); 2102 thread_unlock(td); 2103 } 2104} 2105 2106/* 2107 * Handle a stathz tick. This is really only relevant for timeshare 2108 * threads. 2109 */ 2110void 2111sched_clock(struct thread *td) 2112{ 2113 struct tdq *tdq; 2114 struct td_sched *ts; 2115 2116 THREAD_LOCK_ASSERT(td, MA_OWNED); 2117 tdq = TDQ_SELF(); 2118#ifdef SMP 2119 /* 2120 * We run the long term load balancer infrequently on the first cpu. 2121 */ 2122 if (balance_tdq == tdq) { 2123 if (balance_ticks && --balance_ticks == 0) 2124 sched_balance(); 2125 } 2126#endif 2127 /* 2128 * Save the old switch count so we have a record of the last ticks 2129 * activity. Initialize the new switch count based on our load. 2130 * If there is some activity seed it to reflect that. 2131 */ 2132 tdq->tdq_oldswitchcnt = tdq->tdq_switchcnt; 2133 tdq->tdq_switchcnt = tdq->tdq_load; 2134 /* 2135 * Advance the insert index once for each tick to ensure that all 2136 * threads get a chance to run. 2137 */ 2138 if (tdq->tdq_idx == tdq->tdq_ridx) { 2139 tdq->tdq_idx = (tdq->tdq_idx + 1) % RQ_NQS; 2140 if (TAILQ_EMPTY(&tdq->tdq_timeshare.rq_queues[tdq->tdq_ridx])) 2141 tdq->tdq_ridx = tdq->tdq_idx; 2142 } 2143 ts = td->td_sched; 2144 if (td->td_pri_class & PRI_FIFO_BIT) 2145 return; 2146 if (PRI_BASE(td->td_pri_class) == PRI_TIMESHARE) { 2147 /* 2148 * We used a tick; charge it to the thread so 2149 * that we can compute our interactivity. 2150 */ 2151 td->td_sched->ts_runtime += tickincr; 2152 sched_interact_update(td); 2153 sched_priority(td); 2154 } 2155 /* 2156 * We used up one time slice. 2157 */ 2158 if (--ts->ts_slice > 0) 2159 return; 2160 /* 2161 * We're out of time, force a requeue at userret(). 2162 */ 2163 ts->ts_slice = sched_slice; 2164 td->td_flags |= TDF_NEEDRESCHED; 2165} 2166 2167/* 2168 * Called once per hz tick. Used for cpu utilization information. This 2169 * is easier than trying to scale based on stathz. 2170 */ 2171void 2172sched_tick(int cnt) 2173{ 2174 struct td_sched *ts; 2175 2176 ts = curthread->td_sched; 2177 /* 2178 * Ticks is updated asynchronously on a single cpu. Check here to 2179 * avoid incrementing ts_ticks multiple times in a single tick. 2180 */ 2181 if (ts->ts_incrtick == ticks) 2182 return; 2183 /* Adjust ticks for pctcpu */ 2184 ts->ts_ticks += cnt << SCHED_TICK_SHIFT; 2185 ts->ts_ltick = ticks; 2186 ts->ts_incrtick = ticks; 2187 /* 2188 * Update if we've exceeded our desired tick threshold by over one 2189 * second. 2190 */ 2191 if (ts->ts_ftick + SCHED_TICK_MAX < ts->ts_ltick) 2192 sched_pctcpu_update(ts); 2193} 2194 2195/* 2196 * Return whether the current CPU has runnable tasks. Used for in-kernel 2197 * cooperative idle threads. 2198 */ 2199int 2200sched_runnable(void) 2201{ 2202 struct tdq *tdq; 2203 int load; 2204 2205 load = 1; 2206 2207 tdq = TDQ_SELF(); 2208 if ((curthread->td_flags & TDF_IDLETD) != 0) { 2209 if (tdq->tdq_load > 0) 2210 goto out; 2211 } else 2212 if (tdq->tdq_load - 1 > 0) 2213 goto out; 2214 load = 0; 2215out: 2216 return (load); 2217} 2218 2219/* 2220 * Choose the highest priority thread to run. The thread is removed from 2221 * the run-queue while running however the load remains. For SMP we set 2222 * the tdq in the global idle bitmask if it idles here. 2223 */ 2224struct thread * 2225sched_choose(void) 2226{ 2227 struct thread *td; 2228 struct tdq *tdq; 2229 2230 tdq = TDQ_SELF(); 2231 TDQ_LOCK_ASSERT(tdq, MA_OWNED); 2232 td = tdq_choose(tdq); 2233 if (td) { 2234 td->td_sched->ts_ltick = ticks; 2235 tdq_runq_rem(tdq, td); 2236 tdq->tdq_lowpri = td->td_priority; 2237 return (td); 2238 } 2239 tdq->tdq_lowpri = PRI_MAX_IDLE; 2240 return (PCPU_GET(idlethread)); 2241} 2242 2243/* 2244 * Set owepreempt if necessary. Preemption never happens directly in ULE, 2245 * we always request it once we exit a critical section. 2246 */ 2247static inline void 2248sched_setpreempt(struct thread *td) 2249{ 2250 struct thread *ctd; 2251 int cpri; 2252 int pri; 2253 2254 THREAD_LOCK_ASSERT(curthread, MA_OWNED); 2255 2256 ctd = curthread; 2257 pri = td->td_priority; 2258 cpri = ctd->td_priority; 2259 if (pri < cpri) 2260 ctd->td_flags |= TDF_NEEDRESCHED; 2261 if (panicstr != NULL || pri >= cpri || cold || TD_IS_INHIBITED(ctd)) 2262 return; 2263 if (!sched_shouldpreempt(pri, cpri, 0)) 2264 return; 2265 ctd->td_owepreempt = 1; 2266} 2267 2268/* 2269 * Add a thread to a thread queue. Select the appropriate runq and add the 2270 * thread to it. This is the internal function called when the tdq is 2271 * predetermined. 2272 */ 2273void 2274tdq_add(struct tdq *tdq, struct thread *td, int flags) 2275{ 2276 2277 TDQ_LOCK_ASSERT(tdq, MA_OWNED); 2278 KASSERT((td->td_inhibitors == 0), 2279 ("sched_add: trying to run inhibited thread")); 2280 KASSERT((TD_CAN_RUN(td) || TD_IS_RUNNING(td)), 2281 ("sched_add: bad thread state")); 2282 KASSERT(td->td_flags & TDF_INMEM, 2283 ("sched_add: thread swapped out")); 2284 2285 if (td->td_priority < tdq->tdq_lowpri) 2286 tdq->tdq_lowpri = td->td_priority; 2287 tdq_runq_add(tdq, td, flags); 2288 tdq_load_add(tdq, td); 2289} 2290 2291/* 2292 * Select the target thread queue and add a thread to it. Request 2293 * preemption or IPI a remote processor if required. 2294 */ 2295void 2296sched_add(struct thread *td, int flags) 2297{ 2298 struct tdq *tdq; 2299#ifdef SMP 2300 int cpu; 2301#endif 2302 2303 KTR_STATE2(KTR_SCHED, "thread", sched_tdname(td), "runq add", 2304 "prio:%d", td->td_priority, KTR_ATTR_LINKED, 2305 sched_tdname(curthread)); 2306 KTR_POINT1(KTR_SCHED, "thread", sched_tdname(curthread), "wokeup", 2307 KTR_ATTR_LINKED, sched_tdname(td)); 2308 THREAD_LOCK_ASSERT(td, MA_OWNED); 2309 /* 2310 * Recalculate the priority before we select the target cpu or 2311 * run-queue. 2312 */ 2313 if (PRI_BASE(td->td_pri_class) == PRI_TIMESHARE) 2314 sched_priority(td); 2315#ifdef SMP 2316 /* 2317 * Pick the destination cpu and if it isn't ours transfer to the 2318 * target cpu. 2319 */ 2320 cpu = sched_pickcpu(td, flags); 2321 tdq = sched_setcpu(td, cpu, flags); 2322 tdq_add(tdq, td, flags); 2323 if (cpu != PCPU_GET(cpuid)) { 2324 tdq_notify(tdq, td); 2325 return; 2326 } 2327#else 2328 tdq = TDQ_SELF(); 2329 TDQ_LOCK(tdq); 2330 /* 2331 * Now that the thread is moving to the run-queue, set the lock 2332 * to the scheduler's lock. 2333 */ 2334 thread_lock_set(td, TDQ_LOCKPTR(tdq)); 2335 tdq_add(tdq, td, flags); 2336#endif 2337 if (!(flags & SRQ_YIELDING)) 2338 sched_setpreempt(td); 2339} 2340 2341/* 2342 * Remove a thread from a run-queue without running it. This is used 2343 * when we're stealing a thread from a remote queue. Otherwise all threads 2344 * exit by calling sched_exit_thread() and sched_throw() themselves. 2345 */ 2346void 2347sched_rem(struct thread *td) 2348{ 2349 struct tdq *tdq; 2350 2351 KTR_STATE1(KTR_SCHED, "thread", sched_tdname(td), "runq rem", 2352 "prio:%d", td->td_priority); 2353 tdq = TDQ_CPU(td->td_sched->ts_cpu); 2354 TDQ_LOCK_ASSERT(tdq, MA_OWNED); 2355 MPASS(td->td_lock == TDQ_LOCKPTR(tdq)); 2356 KASSERT(TD_ON_RUNQ(td), 2357 ("sched_rem: thread not on run queue")); 2358 tdq_runq_rem(tdq, td); 2359 tdq_load_rem(tdq, td); 2360 TD_SET_CAN_RUN(td); 2361 if (td->td_priority == tdq->tdq_lowpri) 2362 tdq_setlowpri(tdq, NULL); 2363} 2364 2365/* 2366 * Fetch cpu utilization information. Updates on demand. 2367 */ 2368fixpt_t 2369sched_pctcpu(struct thread *td) 2370{ 2371 fixpt_t pctcpu; 2372 struct td_sched *ts; 2373 2374 pctcpu = 0; 2375 ts = td->td_sched; 2376 if (ts == NULL) 2377 return (0); 2378 2379 THREAD_LOCK_ASSERT(td, MA_OWNED); 2380 if (ts->ts_ticks) { 2381 int rtick; 2382 2383 sched_pctcpu_update(ts); 2384 /* How many rtick per second ? */ 2385 rtick = min(SCHED_TICK_HZ(ts) / SCHED_TICK_SECS, hz); 2386 pctcpu = (FSCALE * ((FSCALE * rtick)/hz)) >> FSHIFT; 2387 } 2388 2389 return (pctcpu); 2390} 2391 2392/* 2393 * Enforce affinity settings for a thread. Called after adjustments to 2394 * cpumask. 2395 */ 2396void 2397sched_affinity(struct thread *td) 2398{ 2399#ifdef SMP 2400 struct td_sched *ts; 2401 2402 THREAD_LOCK_ASSERT(td, MA_OWNED); 2403 ts = td->td_sched; 2404 if (THREAD_CAN_SCHED(td, ts->ts_cpu)) 2405 return; 2406 if (TD_ON_RUNQ(td)) { 2407 sched_rem(td); 2408 sched_add(td, SRQ_BORING); 2409 return; 2410 } 2411 if (!TD_IS_RUNNING(td)) 2412 return; 2413 /* 2414 * Force a switch before returning to userspace. If the 2415 * target thread is not running locally send an ipi to force 2416 * the issue. 2417 */ 2418 td->td_flags |= TDF_NEEDRESCHED; 2419 if (td != curthread) 2420 ipi_cpu(ts->ts_cpu, IPI_PREEMPT); 2421#endif 2422} 2423 2424/* 2425 * Bind a thread to a target cpu. 2426 */ 2427void 2428sched_bind(struct thread *td, int cpu) 2429{ 2430 struct td_sched *ts; 2431 2432 THREAD_LOCK_ASSERT(td, MA_OWNED|MA_NOTRECURSED); 2433 KASSERT(td == curthread, ("sched_bind: can only bind curthread")); 2434 ts = td->td_sched; 2435 if (ts->ts_flags & TSF_BOUND) 2436 sched_unbind(td); 2437 KASSERT(THREAD_CAN_MIGRATE(td), ("%p must be migratable", td)); 2438 ts->ts_flags |= TSF_BOUND; 2439 sched_pin(); 2440 if (PCPU_GET(cpuid) == cpu) 2441 return; 2442 ts->ts_cpu = cpu; 2443 /* When we return from mi_switch we'll be on the correct cpu. */ 2444 mi_switch(SW_VOL, NULL); 2445} 2446 2447/* 2448 * Release a bound thread. 2449 */ 2450void 2451sched_unbind(struct thread *td) 2452{ 2453 struct td_sched *ts; 2454 2455 THREAD_LOCK_ASSERT(td, MA_OWNED); 2456 KASSERT(td == curthread, ("sched_unbind: can only bind curthread")); 2457 ts = td->td_sched; 2458 if ((ts->ts_flags & TSF_BOUND) == 0) 2459 return; 2460 ts->ts_flags &= ~TSF_BOUND; 2461 sched_unpin(); 2462} 2463 2464int 2465sched_is_bound(struct thread *td) 2466{ 2467 THREAD_LOCK_ASSERT(td, MA_OWNED); 2468 return (td->td_sched->ts_flags & TSF_BOUND); 2469} 2470 2471/* 2472 * Basic yield call. 2473 */ 2474void 2475sched_relinquish(struct thread *td) 2476{ 2477 thread_lock(td); 2478 mi_switch(SW_VOL | SWT_RELINQUISH, NULL); 2479 thread_unlock(td); 2480} 2481 2482/* 2483 * Return the total system load. 2484 */ 2485int 2486sched_load(void) 2487{ 2488#ifdef SMP 2489 int total; 2490 int i; 2491 2492 total = 0; 2493 CPU_FOREACH(i) 2494 total += TDQ_CPU(i)->tdq_sysload; 2495 return (total); 2496#else 2497 return (TDQ_SELF()->tdq_sysload); 2498#endif 2499} 2500 2501int 2502sched_sizeof_proc(void) 2503{ 2504 return (sizeof(struct proc)); 2505} 2506 2507int 2508sched_sizeof_thread(void) 2509{ 2510 return (sizeof(struct thread) + sizeof(struct td_sched)); 2511} 2512 2513#ifdef SMP 2514#define TDQ_IDLESPIN(tdq) \ 2515 ((tdq)->tdq_cg != NULL && ((tdq)->tdq_cg->cg_flags & CG_FLAG_THREAD) == 0) 2516#else 2517#define TDQ_IDLESPIN(tdq) 1 2518#endif 2519 2520/* 2521 * The actual idle process. 2522 */ 2523void 2524sched_idletd(void *dummy) 2525{ 2526 struct thread *td; 2527 struct tdq *tdq; 2528 int switchcnt; 2529 int i; 2530 2531 mtx_assert(&Giant, MA_NOTOWNED); 2532 td = curthread; 2533 tdq = TDQ_SELF(); 2534 for (;;) { 2535#ifdef SMP 2536 if (tdq_idled(tdq) == 0) 2537 continue; 2538#endif 2539 switchcnt = tdq->tdq_switchcnt + tdq->tdq_oldswitchcnt; 2540 /* 2541 * If we're switching very frequently, spin while checking 2542 * for load rather than entering a low power state that 2543 * may require an IPI. However, don't do any busy 2544 * loops while on SMT machines as this simply steals 2545 * cycles from cores doing useful work. 2546 */ 2547 if (TDQ_IDLESPIN(tdq) && switchcnt > sched_idlespinthresh) { 2548 for (i = 0; i < sched_idlespins; i++) { 2549 if (tdq->tdq_load) 2550 break; 2551 cpu_spinwait(); 2552 } 2553 } 2554 switchcnt = tdq->tdq_switchcnt + tdq->tdq_oldswitchcnt; 2555 if (tdq->tdq_load == 0) { 2556 tdq->tdq_cpu_idle = 1; 2557 if (tdq->tdq_load == 0) { 2558 cpu_idle(switchcnt > sched_idlespinthresh * 4); 2559 tdq->tdq_switchcnt++; 2560 } 2561 tdq->tdq_cpu_idle = 0; 2562 } 2563 if (tdq->tdq_load) { 2564 thread_lock(td); 2565 mi_switch(SW_VOL | SWT_IDLE, NULL); 2566 thread_unlock(td); 2567 } 2568 } 2569} 2570 2571/* 2572 * A CPU is entering for the first time or a thread is exiting. 2573 */ 2574void 2575sched_throw(struct thread *td) 2576{ 2577 struct thread *newtd; 2578 struct tdq *tdq; 2579 2580 tdq = TDQ_SELF(); 2581 if (td == NULL) { 2582 /* Correct spinlock nesting and acquire the correct lock. */ 2583 TDQ_LOCK(tdq); 2584 spinlock_exit(); 2585 PCPU_SET(switchtime, cpu_ticks()); 2586 PCPU_SET(switchticks, ticks); 2587 } else { 2588 MPASS(td->td_lock == TDQ_LOCKPTR(tdq)); 2589 tdq_load_rem(tdq, td); 2590 lock_profile_release_lock(&TDQ_LOCKPTR(tdq)->lock_object); 2591 } 2592 KASSERT(curthread->td_md.md_spinlock_count == 1, ("invalid count")); 2593 newtd = choosethread(); 2594 TDQ_LOCKPTR(tdq)->mtx_lock = (uintptr_t)newtd; 2595 cpu_throw(td, newtd); /* doesn't return */ 2596} 2597 2598/* 2599 * This is called from fork_exit(). Just acquire the correct locks and 2600 * let fork do the rest of the work. 2601 */ 2602void 2603sched_fork_exit(struct thread *td) 2604{ 2605 struct td_sched *ts; 2606 struct tdq *tdq; 2607 int cpuid; 2608 2609 /* 2610 * Finish setting up thread glue so that it begins execution in a 2611 * non-nested critical section with the scheduler lock held. 2612 */ 2613 cpuid = PCPU_GET(cpuid); 2614 tdq = TDQ_CPU(cpuid); 2615 ts = td->td_sched; 2616 if (TD_IS_IDLETHREAD(td)) 2617 td->td_lock = TDQ_LOCKPTR(tdq); 2618 MPASS(td->td_lock == TDQ_LOCKPTR(tdq)); 2619 td->td_oncpu = cpuid; 2620 TDQ_LOCK_ASSERT(tdq, MA_OWNED | MA_NOTRECURSED); 2621 lock_profile_obtain_lock_success( 2622 &TDQ_LOCKPTR(tdq)->lock_object, 0, 0, __FILE__, __LINE__); 2623} 2624 2625/* 2626 * Create on first use to catch odd startup conditons. 2627 */ 2628char * 2629sched_tdname(struct thread *td) 2630{ 2631#ifdef KTR 2632 struct td_sched *ts; 2633 2634 ts = td->td_sched; 2635 if (ts->ts_name[0] == '\0') 2636 snprintf(ts->ts_name, sizeof(ts->ts_name), 2637 "%s tid %d", td->td_name, td->td_tid); 2638 return (ts->ts_name); 2639#else 2640 return (td->td_name); 2641#endif 2642} 2643 2644#ifdef SMP 2645 2646/* 2647 * Build the CPU topology dump string. Is recursively called to collect 2648 * the topology tree. 2649 */ 2650static int 2651sysctl_kern_sched_topology_spec_internal(struct sbuf *sb, struct cpu_group *cg, 2652 int indent) 2653{ 2654 char cpusetbuf[CPUSETBUFSIZ]; 2655 int i, first; 2656 2657 sbuf_printf(sb, "%*s<group level=\"%d\" cache-level=\"%d\">\n", indent, 2658 "", 1 + indent / 2, cg->cg_level); 2659 sbuf_printf(sb, "%*s <cpu count=\"%d\" mask=\"%s\">", indent, "", 2660 cg->cg_count, cpusetobj_strprint(cpusetbuf, &cg->cg_mask)); 2661 first = TRUE; 2662 for (i = 0; i < MAXCPU; i++) { 2663 if (CPU_ISSET(i, &cg->cg_mask)) { 2664 if (!first) 2665 sbuf_printf(sb, ", "); 2666 else 2667 first = FALSE; 2668 sbuf_printf(sb, "%d", i); 2669 } 2670 } 2671 sbuf_printf(sb, "</cpu>\n"); 2672 2673 if (cg->cg_flags != 0) { 2674 sbuf_printf(sb, "%*s <flags>", indent, ""); 2675 if ((cg->cg_flags & CG_FLAG_HTT) != 0) 2676 sbuf_printf(sb, "<flag name=\"HTT\">HTT group</flag>"); 2677 if ((cg->cg_flags & CG_FLAG_THREAD) != 0) 2678 sbuf_printf(sb, "<flag name=\"THREAD\">THREAD group</flag>"); 2679 if ((cg->cg_flags & CG_FLAG_SMT) != 0) 2680 sbuf_printf(sb, "<flag name=\"SMT\">SMT group</flag>"); 2681 sbuf_printf(sb, "</flags>\n"); 2682 } 2683 2684 if (cg->cg_children > 0) { 2685 sbuf_printf(sb, "%*s <children>\n", indent, ""); 2686 for (i = 0; i < cg->cg_children; i++) 2687 sysctl_kern_sched_topology_spec_internal(sb, 2688 &cg->cg_child[i], indent+2); 2689 sbuf_printf(sb, "%*s </children>\n", indent, ""); 2690 } 2691 sbuf_printf(sb, "%*s</group>\n", indent, ""); 2692 return (0); 2693} 2694 2695/* 2696 * Sysctl handler for retrieving topology dump. It's a wrapper for 2697 * the recursive sysctl_kern_smp_topology_spec_internal(). 2698 */ 2699static int 2700sysctl_kern_sched_topology_spec(SYSCTL_HANDLER_ARGS) 2701{ 2702 struct sbuf *topo; 2703 int err; 2704 2705 KASSERT(cpu_top != NULL, ("cpu_top isn't initialized")); 2706 2707 topo = sbuf_new(NULL, NULL, 500, SBUF_AUTOEXTEND); 2708 if (topo == NULL) 2709 return (ENOMEM); 2710 2711 sbuf_printf(topo, "<groups>\n"); 2712 err = sysctl_kern_sched_topology_spec_internal(topo, cpu_top, 1); 2713 sbuf_printf(topo, "</groups>\n"); 2714 2715 if (err == 0) { 2716 sbuf_finish(topo); 2717 err = SYSCTL_OUT(req, sbuf_data(topo), sbuf_len(topo)); 2718 } 2719 sbuf_delete(topo); 2720 return (err); 2721} 2722 2723#endif 2724 2725SYSCTL_NODE(_kern, OID_AUTO, sched, CTLFLAG_RW, 0, "Scheduler"); 2726SYSCTL_STRING(_kern_sched, OID_AUTO, name, CTLFLAG_RD, "ULE", 0, 2727 "Scheduler name"); 2728SYSCTL_INT(_kern_sched, OID_AUTO, slice, CTLFLAG_RW, &sched_slice, 0, 2729 "Slice size for timeshare threads"); 2730SYSCTL_INT(_kern_sched, OID_AUTO, interact, CTLFLAG_RW, &sched_interact, 0, 2731 "Interactivity score threshold"); 2732SYSCTL_INT(_kern_sched, OID_AUTO, preempt_thresh, CTLFLAG_RW, &preempt_thresh, 2733 0,"Min priority for preemption, lower priorities have greater precedence"); 2734SYSCTL_INT(_kern_sched, OID_AUTO, static_boost, CTLFLAG_RW, &static_boost, 2735 0,"Controls whether static kernel priorities are assigned to sleeping threads."); 2736SYSCTL_INT(_kern_sched, OID_AUTO, idlespins, CTLFLAG_RW, &sched_idlespins, 2737 0,"Number of times idle will spin waiting for new work."); 2738SYSCTL_INT(_kern_sched, OID_AUTO, idlespinthresh, CTLFLAG_RW, &sched_idlespinthresh, 2739 0,"Threshold before we will permit idle spinning."); 2740#ifdef SMP 2741SYSCTL_INT(_kern_sched, OID_AUTO, affinity, CTLFLAG_RW, &affinity, 0, 2742 "Number of hz ticks to keep thread affinity for"); 2743SYSCTL_INT(_kern_sched, OID_AUTO, balance, CTLFLAG_RW, &rebalance, 0, 2744 "Enables the long-term load balancer"); 2745SYSCTL_INT(_kern_sched, OID_AUTO, balance_interval, CTLFLAG_RW, 2746 &balance_interval, 0, 2747 "Average frequency in stathz ticks to run the long-term balancer"); 2748SYSCTL_INT(_kern_sched, OID_AUTO, steal_htt, CTLFLAG_RW, &steal_htt, 0, 2749 "Steals work from another hyper-threaded core on idle"); 2750SYSCTL_INT(_kern_sched, OID_AUTO, steal_idle, CTLFLAG_RW, &steal_idle, 0, 2751 "Attempts to steal work from other cores before idling"); 2752SYSCTL_INT(_kern_sched, OID_AUTO, steal_thresh, CTLFLAG_RW, &steal_thresh, 0, 2753 "Minimum load on remote cpu before we'll steal"); 2754 2755/* Retrieve SMP topology */ 2756SYSCTL_PROC(_kern_sched, OID_AUTO, topology_spec, CTLTYPE_STRING | 2757 CTLFLAG_RD, NULL, 0, sysctl_kern_sched_topology_spec, "A", 2758 "XML dump of detected CPU topology"); 2759 2760#endif 2761 2762/* ps compat. All cpu percentages from ULE are weighted. */ 2763static int ccpu = 0; 2764SYSCTL_INT(_kern, OID_AUTO, ccpu, CTLFLAG_RD, &ccpu, 0, ""); 2765