Loading...
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 | /* * acpi_pad.c ACPI Processor Aggregator Driver * * Copyright (c) 2009, Intel Corporation. * * This program is free software; you can redistribute it and/or modify it * under the terms and conditions of the GNU General Public License, * version 2, as published by the Free Software Foundation. * * This program is distributed in the hope it will be useful, but WITHOUT * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for * more details. * */ #include <linux/kernel.h> #include <linux/cpumask.h> #include <linux/module.h> #include <linux/init.h> #include <linux/types.h> #include <linux/kthread.h> #include <linux/freezer.h> #include <linux/cpu.h> #include <linux/tick.h> #include <linux/slab.h> #include <linux/acpi.h> #include <asm/mwait.h> #define ACPI_PROCESSOR_AGGREGATOR_CLASS "acpi_pad" #define ACPI_PROCESSOR_AGGREGATOR_DEVICE_NAME "Processor Aggregator" #define ACPI_PROCESSOR_AGGREGATOR_NOTIFY 0x80 static DEFINE_MUTEX(isolated_cpus_lock); static DEFINE_MUTEX(round_robin_lock); static unsigned long power_saving_mwait_eax; static unsigned char tsc_detected_unstable; static unsigned char tsc_marked_unstable; static void power_saving_mwait_init(void) { unsigned int eax, ebx, ecx, edx; unsigned int highest_cstate = 0; unsigned int highest_subcstate = 0; int i; if (!boot_cpu_has(X86_FEATURE_MWAIT)) return; if (boot_cpu_data.cpuid_level < CPUID_MWAIT_LEAF) return; cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &edx); if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED) || !(ecx & CPUID5_ECX_INTERRUPT_BREAK)) return; edx >>= MWAIT_SUBSTATE_SIZE; for (i = 0; i < 7 && edx; i++, edx >>= MWAIT_SUBSTATE_SIZE) { if (edx & MWAIT_SUBSTATE_MASK) { highest_cstate = i; highest_subcstate = edx & MWAIT_SUBSTATE_MASK; } } power_saving_mwait_eax = (highest_cstate << MWAIT_SUBSTATE_SIZE) | (highest_subcstate - 1); #if defined(CONFIG_X86) switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: case X86_VENDOR_INTEL: /* * AMD Fam10h TSC will tick in all * C/P/S0/S1 states when this bit is set. */ if (!boot_cpu_has(X86_FEATURE_NONSTOP_TSC)) tsc_detected_unstable = 1; break; default: /* TSC could halt in idle */ tsc_detected_unstable = 1; } #endif } static unsigned long cpu_weight[NR_CPUS]; static int tsk_in_cpu[NR_CPUS] = {[0 ... NR_CPUS-1] = -1}; static DECLARE_BITMAP(pad_busy_cpus_bits, NR_CPUS); static void round_robin_cpu(unsigned int tsk_index) { struct cpumask *pad_busy_cpus = to_cpumask(pad_busy_cpus_bits); cpumask_var_t tmp; int cpu; unsigned long min_weight = -1; unsigned long uninitialized_var(preferred_cpu); if (!alloc_cpumask_var(&tmp, GFP_KERNEL)) return; mutex_lock(&round_robin_lock); cpumask_clear(tmp); for_each_cpu(cpu, pad_busy_cpus) cpumask_or(tmp, tmp, topology_sibling_cpumask(cpu)); cpumask_andnot(tmp, cpu_online_mask, tmp); /* avoid HT sibilings if possible */ if (cpumask_empty(tmp)) cpumask_andnot(tmp, cpu_online_mask, pad_busy_cpus); if (cpumask_empty(tmp)) { mutex_unlock(&round_robin_lock); return; } for_each_cpu(cpu, tmp) { if (cpu_weight[cpu] < min_weight) { min_weight = cpu_weight[cpu]; preferred_cpu = cpu; } } if (tsk_in_cpu[tsk_index] != -1) cpumask_clear_cpu(tsk_in_cpu[tsk_index], pad_busy_cpus); tsk_in_cpu[tsk_index] = preferred_cpu; cpumask_set_cpu(preferred_cpu, pad_busy_cpus); cpu_weight[preferred_cpu]++; mutex_unlock(&round_robin_lock); set_cpus_allowed_ptr(current, cpumask_of(preferred_cpu)); } static void exit_round_robin(unsigned int tsk_index) { struct cpumask *pad_busy_cpus = to_cpumask(pad_busy_cpus_bits); cpumask_clear_cpu(tsk_in_cpu[tsk_index], pad_busy_cpus); tsk_in_cpu[tsk_index] = -1; } static unsigned int idle_pct = 5; /* percentage */ static unsigned int round_robin_time = 1; /* second */ static int power_saving_thread(void *data) { struct sched_param param = {.sched_priority = 1}; int do_sleep; unsigned int tsk_index = (unsigned long)data; u64 last_jiffies = 0; sched_setscheduler(current, SCHED_RR, ¶m); while (!kthread_should_stop()) { unsigned long expire_time; /* round robin to cpus */ expire_time = last_jiffies + round_robin_time * HZ; if (time_before(expire_time, jiffies)) { last_jiffies = jiffies; round_robin_cpu(tsk_index); } do_sleep = 0; expire_time = jiffies + HZ * (100 - idle_pct) / 100; while (!need_resched()) { if (tsc_detected_unstable && !tsc_marked_unstable) { /* TSC could halt in idle, so notify users */ mark_tsc_unstable("TSC halts in idle"); tsc_marked_unstable = 1; } local_irq_disable(); tick_broadcast_enable(); tick_broadcast_enter(); stop_critical_timings(); mwait_idle_with_hints(power_saving_mwait_eax, 1); start_critical_timings(); tick_broadcast_exit(); local_irq_enable(); if (time_before(expire_time, jiffies)) { do_sleep = 1; break; } } /* * current sched_rt has threshold for rt task running time. * When a rt task uses 95% CPU time, the rt thread will be * scheduled out for 5% CPU time to not starve other tasks. But * the mechanism only works when all CPUs have RT task running, * as if one CPU hasn't RT task, RT task from other CPUs will * borrow CPU time from this CPU and cause RT task use > 95% * CPU time. To make 'avoid starvation' work, takes a nap here. */ if (unlikely(do_sleep)) schedule_timeout_killable(HZ * idle_pct / 100); /* If an external event has set the need_resched flag, then * we need to deal with it, or this loop will continue to * spin without calling __mwait(). */ if (unlikely(need_resched())) schedule(); } exit_round_robin(tsk_index); return 0; } static struct task_struct *ps_tsks[NR_CPUS]; static unsigned int ps_tsk_num; static int create_power_saving_task(void) { int rc; ps_tsks[ps_tsk_num] = kthread_run(power_saving_thread, (void *)(unsigned long)ps_tsk_num, "acpi_pad/%d", ps_tsk_num); if (IS_ERR(ps_tsks[ps_tsk_num])) { rc = PTR_ERR(ps_tsks[ps_tsk_num]); ps_tsks[ps_tsk_num] = NULL; } else { rc = 0; ps_tsk_num++; } return rc; } static void destroy_power_saving_task(void) { if (ps_tsk_num > 0) { ps_tsk_num--; kthread_stop(ps_tsks[ps_tsk_num]); ps_tsks[ps_tsk_num] = NULL; } } static void set_power_saving_task_num(unsigned int num) { if (num > ps_tsk_num) { while (ps_tsk_num < num) { if (create_power_saving_task()) return; } } else if (num < ps_tsk_num) { while (ps_tsk_num > num) destroy_power_saving_task(); } } static void acpi_pad_idle_cpus(unsigned int num_cpus) { get_online_cpus(); num_cpus = min_t(unsigned int, num_cpus, num_online_cpus()); set_power_saving_task_num(num_cpus); put_online_cpus(); } static uint32_t acpi_pad_idle_cpus_num(void) { return ps_tsk_num; } static ssize_t acpi_pad_rrtime_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { unsigned long num; if (kstrtoul(buf, 0, &num)) return -EINVAL; if (num < 1 || num >= 100) return -EINVAL; mutex_lock(&isolated_cpus_lock); round_robin_time = num; mutex_unlock(&isolated_cpus_lock); return count; } static ssize_t acpi_pad_rrtime_show(struct device *dev, struct device_attribute *attr, char *buf) { return scnprintf(buf, PAGE_SIZE, "%d\n", round_robin_time); } static DEVICE_ATTR(rrtime, S_IRUGO|S_IWUSR, acpi_pad_rrtime_show, acpi_pad_rrtime_store); static ssize_t acpi_pad_idlepct_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { unsigned long num; if (kstrtoul(buf, 0, &num)) return -EINVAL; if (num < 1 || num >= 100) return -EINVAL; mutex_lock(&isolated_cpus_lock); idle_pct = num; mutex_unlock(&isolated_cpus_lock); return count; } static ssize_t acpi_pad_idlepct_show(struct device *dev, struct device_attribute *attr, char *buf) { return scnprintf(buf, PAGE_SIZE, "%d\n", idle_pct); } static DEVICE_ATTR(idlepct, S_IRUGO|S_IWUSR, acpi_pad_idlepct_show, acpi_pad_idlepct_store); static ssize_t acpi_pad_idlecpus_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { unsigned long num; if (kstrtoul(buf, 0, &num)) return -EINVAL; mutex_lock(&isolated_cpus_lock); acpi_pad_idle_cpus(num); mutex_unlock(&isolated_cpus_lock); return count; } static ssize_t acpi_pad_idlecpus_show(struct device *dev, struct device_attribute *attr, char *buf) { return cpumap_print_to_pagebuf(false, buf, to_cpumask(pad_busy_cpus_bits)); } static DEVICE_ATTR(idlecpus, S_IRUGO|S_IWUSR, acpi_pad_idlecpus_show, acpi_pad_idlecpus_store); static int acpi_pad_add_sysfs(struct acpi_device *device) { int result; result = device_create_file(&device->dev, &dev_attr_idlecpus); if (result) return -ENODEV; result = device_create_file(&device->dev, &dev_attr_idlepct); if (result) { device_remove_file(&device->dev, &dev_attr_idlecpus); return -ENODEV; } result = device_create_file(&device->dev, &dev_attr_rrtime); if (result) { device_remove_file(&device->dev, &dev_attr_idlecpus); device_remove_file(&device->dev, &dev_attr_idlepct); return -ENODEV; } return 0; } static void acpi_pad_remove_sysfs(struct acpi_device *device) { device_remove_file(&device->dev, &dev_attr_idlecpus); device_remove_file(&device->dev, &dev_attr_idlepct); device_remove_file(&device->dev, &dev_attr_rrtime); } /* * Query firmware how many CPUs should be idle * return -1 on failure */ static int acpi_pad_pur(acpi_handle handle) { struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL}; union acpi_object *package; int num = -1; if (ACPI_FAILURE(acpi_evaluate_object(handle, "_PUR", NULL, &buffer))) return num; if (!buffer.length || !buffer.pointer) return num; package = buffer.pointer; if (package->type == ACPI_TYPE_PACKAGE && package->package.count == 2 && package->package.elements[0].integer.value == 1) /* rev 1 */ num = package->package.elements[1].integer.value; kfree(buffer.pointer); return num; } static void acpi_pad_handle_notify(acpi_handle handle) { int num_cpus; uint32_t idle_cpus; struct acpi_buffer param = { .length = 4, .pointer = (void *)&idle_cpus, }; mutex_lock(&isolated_cpus_lock); num_cpus = acpi_pad_pur(handle); if (num_cpus < 0) { mutex_unlock(&isolated_cpus_lock); return; } acpi_pad_idle_cpus(num_cpus); idle_cpus = acpi_pad_idle_cpus_num(); acpi_evaluate_ost(handle, ACPI_PROCESSOR_AGGREGATOR_NOTIFY, 0, ¶m); mutex_unlock(&isolated_cpus_lock); } static void acpi_pad_notify(acpi_handle handle, u32 event, void *data) { struct acpi_device *device = data; switch (event) { case ACPI_PROCESSOR_AGGREGATOR_NOTIFY: acpi_pad_handle_notify(handle); acpi_bus_generate_netlink_event(device->pnp.device_class, dev_name(&device->dev), event, 0); break; default: pr_warn("Unsupported event [0x%x]\n", event); break; } } static int acpi_pad_add(struct acpi_device *device) { acpi_status status; strcpy(acpi_device_name(device), ACPI_PROCESSOR_AGGREGATOR_DEVICE_NAME); strcpy(acpi_device_class(device), ACPI_PROCESSOR_AGGREGATOR_CLASS); if (acpi_pad_add_sysfs(device)) return -ENODEV; status = acpi_install_notify_handler(device->handle, ACPI_DEVICE_NOTIFY, acpi_pad_notify, device); if (ACPI_FAILURE(status)) { acpi_pad_remove_sysfs(device); return -ENODEV; } return 0; } static int acpi_pad_remove(struct acpi_device *device) { mutex_lock(&isolated_cpus_lock); acpi_pad_idle_cpus(0); mutex_unlock(&isolated_cpus_lock); acpi_remove_notify_handler(device->handle, ACPI_DEVICE_NOTIFY, acpi_pad_notify); acpi_pad_remove_sysfs(device); return 0; } static const struct acpi_device_id pad_device_ids[] = { {"ACPI000C", 0}, {"", 0}, }; MODULE_DEVICE_TABLE(acpi, pad_device_ids); static struct acpi_driver acpi_pad_driver = { .name = "processor_aggregator", .class = ACPI_PROCESSOR_AGGREGATOR_CLASS, .ids = pad_device_ids, .ops = { .add = acpi_pad_add, .remove = acpi_pad_remove, }, }; static int __init acpi_pad_init(void) { power_saving_mwait_init(); if (power_saving_mwait_eax == 0) return -EINVAL; return acpi_bus_register_driver(&acpi_pad_driver); } static void __exit acpi_pad_exit(void) { acpi_bus_unregister_driver(&acpi_pad_driver); } module_init(acpi_pad_init); module_exit(acpi_pad_exit); MODULE_AUTHOR("Shaohua Li<shaohua.li@intel.com>"); MODULE_DESCRIPTION("ACPI Processor Aggregator Driver"); MODULE_LICENSE("GPL"); |