kern_clock.c revision 1.124
1 /* $NetBSD: kern_clock.c,v 1.124 2008/06/01 21:24:15 ad Exp $ */ 2 3 /*- 4 * Copyright (c) 2000, 2004, 2006, 2007, 2008 The NetBSD Foundation, Inc. 5 * All rights reserved. 6 * 7 * This code is derived from software contributed to The NetBSD Foundation 8 * by Jason R. Thorpe of the Numerical Aerospace Simulation Facility, 9 * NASA Ames Research Center. 10 * This code is derived from software contributed to The NetBSD Foundation 11 * by Charles M. Hannum. 12 * 13 * Redistribution and use in source and binary forms, with or without 14 * modification, are permitted provided that the following conditions 15 * are met: 16 * 1. Redistributions of source code must retain the above copyright 17 * notice, this list of conditions and the following disclaimer. 18 * 2. Redistributions in binary form must reproduce the above copyright 19 * notice, this list of conditions and the following disclaimer in the 20 * documentation and/or other materials provided with the distribution. 21 * 22 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 23 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 24 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 25 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 26 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 27 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 28 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 29 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 30 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 31 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 32 * POSSIBILITY OF SUCH DAMAGE. 33 */ 34 35 /*- 36 * Copyright (c) 1982, 1986, 1991, 1993 37 * The Regents of the University of California. All rights reserved. 38 * (c) UNIX System Laboratories, Inc. 39 * All or some portions of this file are derived from material licensed 40 * to the University of California by American Telephone and Telegraph 41 * Co. or Unix System Laboratories, Inc. and are reproduced herein with 42 * the permission of UNIX System Laboratories, Inc. 43 * 44 * Redistribution and use in source and binary forms, with or without 45 * modification, are permitted provided that the following conditions 46 * are met: 47 * 1. Redistributions of source code must retain the above copyright 48 * notice, this list of conditions and the following disclaimer. 49 * 2. Redistributions in binary form must reproduce the above copyright 50 * notice, this list of conditions and the following disclaimer in the 51 * documentation and/or other materials provided with the distribution. 52 * 3. Neither the name of the University nor the names of its contributors 53 * may be used to endorse or promote products derived from this software 54 * without specific prior written permission. 55 * 56 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 57 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 58 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 59 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 60 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 61 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 62 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 63 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 64 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 65 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 66 * SUCH DAMAGE. 67 * 68 * @(#)kern_clock.c 8.5 (Berkeley) 1/21/94 69 */ 70 71 #include <sys/cdefs.h> 72 __KERNEL_RCSID(0, "$NetBSD: kern_clock.c,v 1.124 2008/06/01 21:24:15 ad Exp $"); 73 74 #include "opt_ntp.h" 75 #include "opt_perfctrs.h" 76 77 #include <sys/param.h> 78 #include <sys/systm.h> 79 #include <sys/callout.h> 80 #include <sys/kernel.h> 81 #include <sys/proc.h> 82 #include <sys/resourcevar.h> 83 #include <sys/signalvar.h> 84 #include <sys/sysctl.h> 85 #include <sys/timex.h> 86 #include <sys/sched.h> 87 #include <sys/time.h> 88 #include <sys/timetc.h> 89 #include <sys/cpu.h> 90 #include <sys/atomic.h> 91 92 #include <uvm/uvm_extern.h> 93 94 #ifdef GPROF 95 #include <sys/gmon.h> 96 #endif 97 98 /* 99 * Clock handling routines. 100 * 101 * This code is written to operate with two timers that run independently of 102 * each other. The main clock, running hz times per second, is used to keep 103 * track of real time. The second timer handles kernel and user profiling, 104 * and does resource use estimation. If the second timer is programmable, 105 * it is randomized to avoid aliasing between the two clocks. For example, 106 * the randomization prevents an adversary from always giving up the CPU 107 * just before its quantum expires. Otherwise, it would never accumulate 108 * CPU ticks. The mean frequency of the second timer is stathz. 109 * 110 * If no second timer exists, stathz will be zero; in this case we drive 111 * profiling and statistics off the main clock. This WILL NOT be accurate; 112 * do not do it unless absolutely necessary. 113 * 114 * The statistics clock may (or may not) be run at a higher rate while 115 * profiling. This profile clock runs at profhz. We require that profhz 116 * be an integral multiple of stathz. 117 * 118 * If the statistics clock is running fast, it must be divided by the ratio 119 * profhz/stathz for statistics. (For profiling, every tick counts.) 120 */ 121 122 int stathz; 123 int profhz; 124 int profsrc; 125 int schedhz; 126 int profprocs; 127 int hardclock_ticks; 128 static int hardscheddiv; /* hard => sched divider (used if schedhz == 0) */ 129 static int psdiv; /* prof => stat divider */ 130 int psratio; /* ratio: prof / stat */ 131 132 static u_int get_intr_timecount(struct timecounter *); 133 134 static struct timecounter intr_timecounter = { 135 get_intr_timecount, /* get_timecount */ 136 0, /* no poll_pps */ 137 ~0u, /* counter_mask */ 138 0, /* frequency */ 139 "clockinterrupt", /* name */ 140 0, /* quality - minimum implementation level for a clock */ 141 NULL, /* prev */ 142 NULL, /* next */ 143 }; 144 145 static u_int 146 get_intr_timecount(struct timecounter *tc) 147 { 148 149 return (u_int)hardclock_ticks; 150 } 151 152 /* 153 * Initialize clock frequencies and start both clocks running. 154 */ 155 void 156 initclocks(void) 157 { 158 int i; 159 160 /* 161 * Set divisors to 1 (normal case) and let the machine-specific 162 * code do its bit. 163 */ 164 psdiv = 1; 165 /* 166 * provide minimum default time counter 167 * will only run at interrupt resolution 168 */ 169 intr_timecounter.tc_frequency = hz; 170 tc_init(&intr_timecounter); 171 cpu_initclocks(); 172 173 /* 174 * Compute profhz and stathz, fix profhz if needed. 175 */ 176 i = stathz ? stathz : hz; 177 if (profhz == 0) 178 profhz = i; 179 psratio = profhz / i; 180 if (schedhz == 0) { 181 /* 16Hz is best */ 182 hardscheddiv = hz / 16; 183 if (hardscheddiv <= 0) 184 panic("hardscheddiv"); 185 } 186 187 } 188 189 /* 190 * The real-time timer, interrupting hz times per second. 191 */ 192 void 193 hardclock(struct clockframe *frame) 194 { 195 struct lwp *l; 196 struct cpu_info *ci; 197 198 ci = curcpu(); 199 l = ci->ci_data.cpu_onproc; 200 201 timer_tick(l, CLKF_USERMODE(frame)); 202 203 /* 204 * If no separate statistics clock is available, run it from here. 205 */ 206 if (stathz == 0) 207 statclock(frame); 208 /* 209 * If no separate schedclock is provided, call it here 210 * at about 16 Hz. 211 */ 212 if (schedhz == 0) { 213 if ((int)(--ci->ci_schedstate.spc_schedticks) <= 0) { 214 schedclock(l); 215 ci->ci_schedstate.spc_schedticks = hardscheddiv; 216 } 217 } 218 if ((--ci->ci_schedstate.spc_ticks) <= 0) 219 sched_tick(ci); 220 221 if (CPU_IS_PRIMARY(ci)) { 222 hardclock_ticks++; 223 tc_ticktock(); 224 } 225 226 /* 227 * Update real-time timeout queue. Callouts are processed at a 228 * very low CPU priority, so we don't keep the relatively high 229 * clock interrupt priority any longer than necessary. 230 */ 231 callout_hardclock(); 232 } 233 234 /* 235 * Start profiling on a process. 236 * 237 * Kernel profiling passes proc0 which never exits and hence 238 * keeps the profile clock running constantly. 239 */ 240 void 241 startprofclock(struct proc *p) 242 { 243 244 KASSERT(mutex_owned(&p->p_stmutex)); 245 246 if ((p->p_stflag & PST_PROFIL) == 0) { 247 p->p_stflag |= PST_PROFIL; 248 /* 249 * This is only necessary if using the clock as the 250 * profiling source. 251 */ 252 if (++profprocs == 1 && stathz != 0) 253 psdiv = psratio; 254 } 255 } 256 257 /* 258 * Stop profiling on a process. 259 */ 260 void 261 stopprofclock(struct proc *p) 262 { 263 264 KASSERT(mutex_owned(&p->p_stmutex)); 265 266 if (p->p_stflag & PST_PROFIL) { 267 p->p_stflag &= ~PST_PROFIL; 268 /* 269 * This is only necessary if using the clock as the 270 * profiling source. 271 */ 272 if (--profprocs == 0 && stathz != 0) 273 psdiv = 1; 274 } 275 } 276 277 #if defined(PERFCTRS) 278 /* 279 * Independent profiling "tick" in case we're using a separate 280 * clock or profiling event source. Currently, that's just 281 * performance counters--hence the wrapper. 282 */ 283 void 284 proftick(struct clockframe *frame) 285 { 286 #ifdef GPROF 287 struct gmonparam *g; 288 intptr_t i; 289 #endif 290 struct lwp *l; 291 struct proc *p; 292 293 l = curcpu()->ci_data.cpu_onproc; 294 p = (l ? l->l_proc : NULL); 295 if (CLKF_USERMODE(frame)) { 296 mutex_spin_enter(&p->p_stmutex); 297 if (p->p_stflag & PST_PROFIL) 298 addupc_intr(l, CLKF_PC(frame)); 299 mutex_spin_exit(&p->p_stmutex); 300 } else { 301 #ifdef GPROF 302 g = &_gmonparam; 303 if (g->state == GMON_PROF_ON) { 304 i = CLKF_PC(frame) - g->lowpc; 305 if (i < g->textsize) { 306 i /= HISTFRACTION * sizeof(*g->kcount); 307 g->kcount[i]++; 308 } 309 } 310 #endif 311 #ifdef LWP_PC 312 if (p != NULL && (p->p_stflag & PST_PROFIL) != 0) 313 addupc_intr(l, LWP_PC(l)); 314 #endif 315 } 316 } 317 #endif 318 319 void 320 schedclock(struct lwp *l) 321 { 322 struct cpu_info *ci; 323 324 ci = l->l_cpu; 325 326 /* Accumulate syscall and context switch counts. */ 327 atomic_add_int((unsigned *)&uvmexp.swtch, ci->ci_data.cpu_nswtch); 328 ci->ci_data.cpu_nswtch = 0; 329 atomic_add_int((unsigned *)&uvmexp.syscalls, ci->ci_data.cpu_nsyscall); 330 ci->ci_data.cpu_nsyscall = 0; 331 atomic_add_int((unsigned *)&uvmexp.traps, ci->ci_data.cpu_ntrap); 332 ci->ci_data.cpu_ntrap = 0; 333 334 if ((l->l_flag & LW_IDLE) != 0) 335 return; 336 337 sched_schedclock(l); 338 } 339 340 /* 341 * Statistics clock. Grab profile sample, and if divider reaches 0, 342 * do process and kernel statistics. 343 */ 344 void 345 statclock(struct clockframe *frame) 346 { 347 #ifdef GPROF 348 struct gmonparam *g; 349 intptr_t i; 350 #endif 351 struct cpu_info *ci = curcpu(); 352 struct schedstate_percpu *spc = &ci->ci_schedstate; 353 struct proc *p; 354 struct lwp *l; 355 356 /* 357 * Notice changes in divisor frequency, and adjust clock 358 * frequency accordingly. 359 */ 360 if (spc->spc_psdiv != psdiv) { 361 spc->spc_psdiv = psdiv; 362 spc->spc_pscnt = psdiv; 363 if (psdiv == 1) { 364 setstatclockrate(stathz); 365 } else { 366 setstatclockrate(profhz); 367 } 368 } 369 l = ci->ci_data.cpu_onproc; 370 if ((l->l_flag & LW_IDLE) != 0) { 371 /* 372 * don't account idle lwps as swapper. 373 */ 374 p = NULL; 375 } else { 376 p = l->l_proc; 377 mutex_spin_enter(&p->p_stmutex); 378 } 379 380 if (CLKF_USERMODE(frame)) { 381 if ((p->p_stflag & PST_PROFIL) && profsrc == PROFSRC_CLOCK) 382 addupc_intr(l, CLKF_PC(frame)); 383 if (--spc->spc_pscnt > 0) { 384 mutex_spin_exit(&p->p_stmutex); 385 return; 386 } 387 388 /* 389 * Came from user mode; CPU was in user state. 390 * If this process is being profiled record the tick. 391 */ 392 p->p_uticks++; 393 if (p->p_nice > NZERO) 394 spc->spc_cp_time[CP_NICE]++; 395 else 396 spc->spc_cp_time[CP_USER]++; 397 } else { 398 #ifdef GPROF 399 /* 400 * Kernel statistics are just like addupc_intr, only easier. 401 */ 402 g = &_gmonparam; 403 if (profsrc == PROFSRC_CLOCK && g->state == GMON_PROF_ON) { 404 i = CLKF_PC(frame) - g->lowpc; 405 if (i < g->textsize) { 406 i /= HISTFRACTION * sizeof(*g->kcount); 407 g->kcount[i]++; 408 } 409 } 410 #endif 411 #ifdef LWP_PC 412 if (p != NULL && profsrc == PROFSRC_CLOCK && 413 (p->p_stflag & PST_PROFIL)) { 414 addupc_intr(l, LWP_PC(l)); 415 } 416 #endif 417 if (--spc->spc_pscnt > 0) { 418 if (p != NULL) 419 mutex_spin_exit(&p->p_stmutex); 420 return; 421 } 422 /* 423 * Came from kernel mode, so we were: 424 * - handling an interrupt, 425 * - doing syscall or trap work on behalf of the current 426 * user process, or 427 * - spinning in the idle loop. 428 * Whichever it is, charge the time as appropriate. 429 * Note that we charge interrupts to the current process, 430 * regardless of whether they are ``for'' that process, 431 * so that we know how much of its real time was spent 432 * in ``non-process'' (i.e., interrupt) work. 433 */ 434 if (CLKF_INTR(frame) || (curlwp->l_pflag & LP_INTR) != 0) { 435 if (p != NULL) { 436 p->p_iticks++; 437 } 438 spc->spc_cp_time[CP_INTR]++; 439 } else if (p != NULL) { 440 p->p_sticks++; 441 spc->spc_cp_time[CP_SYS]++; 442 } else { 443 spc->spc_cp_time[CP_IDLE]++; 444 } 445 } 446 spc->spc_pscnt = psdiv; 447 448 if (p != NULL) { 449 ++l->l_cpticks; 450 mutex_spin_exit(&p->p_stmutex); 451 } 452 } 453
Indexes created Thu Sep 25 08:09:54 GMT 2025