Loading...
1/*
2 * processor_idle - idle state submodule to the ACPI processor driver
3 *
4 * Copyright (C) 2001, 2002 Andy Grover <andrew.grover@intel.com>
5 * Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
6 * Copyright (C) 2004, 2005 Dominik Brodowski <linux@brodo.de>
7 * Copyright (C) 2004 Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
8 * - Added processor hotplug support
9 * Copyright (C) 2005 Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
10 * - Added support for C3 on SMP
11 *
12 * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
13 *
14 * This program is free software; you can redistribute it and/or modify
15 * it under the terms of the GNU General Public License as published by
16 * the Free Software Foundation; either version 2 of the License, or (at
17 * your option) any later version.
18 *
19 * This program is distributed in the hope that it will be useful, but
20 * WITHOUT ANY WARRANTY; without even the implied warranty of
21 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
22 * General Public License for more details.
23 *
24 * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
25 */
26#define pr_fmt(fmt) "ACPI: " fmt
27
28#include <linux/module.h>
29#include <linux/acpi.h>
30#include <linux/dmi.h>
31#include <linux/sched.h> /* need_resched() */
32#include <linux/tick.h>
33#include <linux/cpuidle.h>
34#include <acpi/processor.h>
35
36/*
37 * Include the apic definitions for x86 to have the APIC timer related defines
38 * available also for UP (on SMP it gets magically included via linux/smp.h).
39 * asm/acpi.h is not an option, as it would require more include magic. Also
40 * creating an empty asm-ia64/apic.h would just trade pest vs. cholera.
41 */
42#ifdef CONFIG_X86
43#include <asm/apic.h>
44#endif
45
46#define ACPI_PROCESSOR_CLASS "processor"
47#define _COMPONENT ACPI_PROCESSOR_COMPONENT
48ACPI_MODULE_NAME("processor_idle");
49
50static unsigned int max_cstate __read_mostly = ACPI_PROCESSOR_MAX_POWER;
51module_param(max_cstate, uint, 0000);
52static unsigned int nocst __read_mostly;
53module_param(nocst, uint, 0000);
54static int bm_check_disable __read_mostly;
55module_param(bm_check_disable, uint, 0000);
56
57static unsigned int latency_factor __read_mostly = 2;
58module_param(latency_factor, uint, 0644);
59
60static DEFINE_PER_CPU(struct cpuidle_device *, acpi_cpuidle_device);
61
62static
63DEFINE_PER_CPU(struct acpi_processor_cx * [CPUIDLE_STATE_MAX], acpi_cstate);
64
65static int disabled_by_idle_boot_param(void)
66{
67 return boot_option_idle_override == IDLE_POLL ||
68 boot_option_idle_override == IDLE_HALT;
69}
70
71/*
72 * IBM ThinkPad R40e crashes mysteriously when going into C2 or C3.
73 * For now disable this. Probably a bug somewhere else.
74 *
75 * To skip this limit, boot/load with a large max_cstate limit.
76 */
77static int set_max_cstate(const struct dmi_system_id *id)
78{
79 if (max_cstate > ACPI_PROCESSOR_MAX_POWER)
80 return 0;
81
82 pr_notice("%s detected - limiting to C%ld max_cstate."
83 " Override with \"processor.max_cstate=%d\"\n", id->ident,
84 (long)id->driver_data, ACPI_PROCESSOR_MAX_POWER + 1);
85
86 max_cstate = (long)id->driver_data;
87
88 return 0;
89}
90
91static const struct dmi_system_id processor_power_dmi_table[] = {
92 { set_max_cstate, "Clevo 5600D", {
93 DMI_MATCH(DMI_BIOS_VENDOR,"Phoenix Technologies LTD"),
94 DMI_MATCH(DMI_BIOS_VERSION,"SHE845M0.86C.0013.D.0302131307")},
95 (void *)2},
96 { set_max_cstate, "Pavilion zv5000", {
97 DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
98 DMI_MATCH(DMI_PRODUCT_NAME,"Pavilion zv5000 (DS502A#ABA)")},
99 (void *)1},
100 { set_max_cstate, "Asus L8400B", {
101 DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK Computer Inc."),
102 DMI_MATCH(DMI_PRODUCT_NAME,"L8400B series Notebook PC")},
103 (void *)1},
104 {},
105};
106
107
108/*
109 * Callers should disable interrupts before the call and enable
110 * interrupts after return.
111 */
112static void acpi_safe_halt(void)
113{
114 if (!tif_need_resched()) {
115 safe_halt();
116 local_irq_disable();
117 }
118}
119
120#ifdef ARCH_APICTIMER_STOPS_ON_C3
121
122/*
123 * Some BIOS implementations switch to C3 in the published C2 state.
124 * This seems to be a common problem on AMD boxen, but other vendors
125 * are affected too. We pick the most conservative approach: we assume
126 * that the local APIC stops in both C2 and C3.
127 */
128static void lapic_timer_check_state(int state, struct acpi_processor *pr,
129 struct acpi_processor_cx *cx)
130{
131 struct acpi_processor_power *pwr = &pr->power;
132 u8 type = local_apic_timer_c2_ok ? ACPI_STATE_C3 : ACPI_STATE_C2;
133
134 if (cpu_has(&cpu_data(pr->id), X86_FEATURE_ARAT))
135 return;
136
137 if (amd_e400_c1e_detected)
138 type = ACPI_STATE_C1;
139
140 /*
141 * Check, if one of the previous states already marked the lapic
142 * unstable
143 */
144 if (pwr->timer_broadcast_on_state < state)
145 return;
146
147 if (cx->type >= type)
148 pr->power.timer_broadcast_on_state = state;
149}
150
151static void __lapic_timer_propagate_broadcast(void *arg)
152{
153 struct acpi_processor *pr = (struct acpi_processor *) arg;
154
155 if (pr->power.timer_broadcast_on_state < INT_MAX)
156 tick_broadcast_enable();
157 else
158 tick_broadcast_disable();
159}
160
161static void lapic_timer_propagate_broadcast(struct acpi_processor *pr)
162{
163 smp_call_function_single(pr->id, __lapic_timer_propagate_broadcast,
164 (void *)pr, 1);
165}
166
167/* Power(C) State timer broadcast control */
168static void lapic_timer_state_broadcast(struct acpi_processor *pr,
169 struct acpi_processor_cx *cx,
170 int broadcast)
171{
172 int state = cx - pr->power.states;
173
174 if (state >= pr->power.timer_broadcast_on_state) {
175 if (broadcast)
176 tick_broadcast_enter();
177 else
178 tick_broadcast_exit();
179 }
180}
181
182#else
183
184static void lapic_timer_check_state(int state, struct acpi_processor *pr,
185 struct acpi_processor_cx *cstate) { }
186static void lapic_timer_propagate_broadcast(struct acpi_processor *pr) { }
187static void lapic_timer_state_broadcast(struct acpi_processor *pr,
188 struct acpi_processor_cx *cx,
189 int broadcast)
190{
191}
192
193#endif
194
195#if defined(CONFIG_X86)
196static void tsc_check_state(int state)
197{
198 switch (boot_cpu_data.x86_vendor) {
199 case X86_VENDOR_AMD:
200 case X86_VENDOR_INTEL:
201 /*
202 * AMD Fam10h TSC will tick in all
203 * C/P/S0/S1 states when this bit is set.
204 */
205 if (boot_cpu_has(X86_FEATURE_NONSTOP_TSC))
206 return;
207
208 /*FALL THROUGH*/
209 default:
210 /* TSC could halt in idle, so notify users */
211 if (state > ACPI_STATE_C1)
212 mark_tsc_unstable("TSC halts in idle");
213 }
214}
215#else
216static void tsc_check_state(int state) { return; }
217#endif
218
219static int acpi_processor_get_power_info_fadt(struct acpi_processor *pr)
220{
221
222 if (!pr->pblk)
223 return -ENODEV;
224
225 /* if info is obtained from pblk/fadt, type equals state */
226 pr->power.states[ACPI_STATE_C2].type = ACPI_STATE_C2;
227 pr->power.states[ACPI_STATE_C3].type = ACPI_STATE_C3;
228
229#ifndef CONFIG_HOTPLUG_CPU
230 /*
231 * Check for P_LVL2_UP flag before entering C2 and above on
232 * an SMP system.
233 */
234 if ((num_online_cpus() > 1) &&
235 !(acpi_gbl_FADT.flags & ACPI_FADT_C2_MP_SUPPORTED))
236 return -ENODEV;
237#endif
238
239 /* determine C2 and C3 address from pblk */
240 pr->power.states[ACPI_STATE_C2].address = pr->pblk + 4;
241 pr->power.states[ACPI_STATE_C3].address = pr->pblk + 5;
242
243 /* determine latencies from FADT */
244 pr->power.states[ACPI_STATE_C2].latency = acpi_gbl_FADT.c2_latency;
245 pr->power.states[ACPI_STATE_C3].latency = acpi_gbl_FADT.c3_latency;
246
247 /*
248 * FADT specified C2 latency must be less than or equal to
249 * 100 microseconds.
250 */
251 if (acpi_gbl_FADT.c2_latency > ACPI_PROCESSOR_MAX_C2_LATENCY) {
252 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
253 "C2 latency too large [%d]\n", acpi_gbl_FADT.c2_latency));
254 /* invalidate C2 */
255 pr->power.states[ACPI_STATE_C2].address = 0;
256 }
257
258 /*
259 * FADT supplied C3 latency must be less than or equal to
260 * 1000 microseconds.
261 */
262 if (acpi_gbl_FADT.c3_latency > ACPI_PROCESSOR_MAX_C3_LATENCY) {
263 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
264 "C3 latency too large [%d]\n", acpi_gbl_FADT.c3_latency));
265 /* invalidate C3 */
266 pr->power.states[ACPI_STATE_C3].address = 0;
267 }
268
269 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
270 "lvl2[0x%08x] lvl3[0x%08x]\n",
271 pr->power.states[ACPI_STATE_C2].address,
272 pr->power.states[ACPI_STATE_C3].address));
273
274 return 0;
275}
276
277static int acpi_processor_get_power_info_default(struct acpi_processor *pr)
278{
279 if (!pr->power.states[ACPI_STATE_C1].valid) {
280 /* set the first C-State to C1 */
281 /* all processors need to support C1 */
282 pr->power.states[ACPI_STATE_C1].type = ACPI_STATE_C1;
283 pr->power.states[ACPI_STATE_C1].valid = 1;
284 pr->power.states[ACPI_STATE_C1].entry_method = ACPI_CSTATE_HALT;
285 }
286 /* the C0 state only exists as a filler in our array */
287 pr->power.states[ACPI_STATE_C0].valid = 1;
288 return 0;
289}
290
291static int acpi_processor_get_power_info_cst(struct acpi_processor *pr)
292{
293 acpi_status status;
294 u64 count;
295 int current_count;
296 int i, ret = 0;
297 struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
298 union acpi_object *cst;
299
300
301 if (nocst)
302 return -ENODEV;
303
304 current_count = 0;
305
306 status = acpi_evaluate_object(pr->handle, "_CST", NULL, &buffer);
307 if (ACPI_FAILURE(status)) {
308 ACPI_DEBUG_PRINT((ACPI_DB_INFO, "No _CST, giving up\n"));
309 return -ENODEV;
310 }
311
312 cst = buffer.pointer;
313
314 /* There must be at least 2 elements */
315 if (!cst || (cst->type != ACPI_TYPE_PACKAGE) || cst->package.count < 2) {
316 pr_err("not enough elements in _CST\n");
317 ret = -EFAULT;
318 goto end;
319 }
320
321 count = cst->package.elements[0].integer.value;
322
323 /* Validate number of power states. */
324 if (count < 1 || count != cst->package.count - 1) {
325 pr_err("count given by _CST is not valid\n");
326 ret = -EFAULT;
327 goto end;
328 }
329
330 /* Tell driver that at least _CST is supported. */
331 pr->flags.has_cst = 1;
332
333 for (i = 1; i <= count; i++) {
334 union acpi_object *element;
335 union acpi_object *obj;
336 struct acpi_power_register *reg;
337 struct acpi_processor_cx cx;
338
339 memset(&cx, 0, sizeof(cx));
340
341 element = &(cst->package.elements[i]);
342 if (element->type != ACPI_TYPE_PACKAGE)
343 continue;
344
345 if (element->package.count != 4)
346 continue;
347
348 obj = &(element->package.elements[0]);
349
350 if (obj->type != ACPI_TYPE_BUFFER)
351 continue;
352
353 reg = (struct acpi_power_register *)obj->buffer.pointer;
354
355 if (reg->space_id != ACPI_ADR_SPACE_SYSTEM_IO &&
356 (reg->space_id != ACPI_ADR_SPACE_FIXED_HARDWARE))
357 continue;
358
359 /* There should be an easy way to extract an integer... */
360 obj = &(element->package.elements[1]);
361 if (obj->type != ACPI_TYPE_INTEGER)
362 continue;
363
364 cx.type = obj->integer.value;
365 /*
366 * Some buggy BIOSes won't list C1 in _CST -
367 * Let acpi_processor_get_power_info_default() handle them later
368 */
369 if (i == 1 && cx.type != ACPI_STATE_C1)
370 current_count++;
371
372 cx.address = reg->address;
373 cx.index = current_count + 1;
374
375 cx.entry_method = ACPI_CSTATE_SYSTEMIO;
376 if (reg->space_id == ACPI_ADR_SPACE_FIXED_HARDWARE) {
377 if (acpi_processor_ffh_cstate_probe
378 (pr->id, &cx, reg) == 0) {
379 cx.entry_method = ACPI_CSTATE_FFH;
380 } else if (cx.type == ACPI_STATE_C1) {
381 /*
382 * C1 is a special case where FIXED_HARDWARE
383 * can be handled in non-MWAIT way as well.
384 * In that case, save this _CST entry info.
385 * Otherwise, ignore this info and continue.
386 */
387 cx.entry_method = ACPI_CSTATE_HALT;
388 snprintf(cx.desc, ACPI_CX_DESC_LEN, "ACPI HLT");
389 } else {
390 continue;
391 }
392 if (cx.type == ACPI_STATE_C1 &&
393 (boot_option_idle_override == IDLE_NOMWAIT)) {
394 /*
395 * In most cases the C1 space_id obtained from
396 * _CST object is FIXED_HARDWARE access mode.
397 * But when the option of idle=halt is added,
398 * the entry_method type should be changed from
399 * CSTATE_FFH to CSTATE_HALT.
400 * When the option of idle=nomwait is added,
401 * the C1 entry_method type should be
402 * CSTATE_HALT.
403 */
404 cx.entry_method = ACPI_CSTATE_HALT;
405 snprintf(cx.desc, ACPI_CX_DESC_LEN, "ACPI HLT");
406 }
407 } else {
408 snprintf(cx.desc, ACPI_CX_DESC_LEN, "ACPI IOPORT 0x%x",
409 cx.address);
410 }
411
412 if (cx.type == ACPI_STATE_C1) {
413 cx.valid = 1;
414 }
415
416 obj = &(element->package.elements[2]);
417 if (obj->type != ACPI_TYPE_INTEGER)
418 continue;
419
420 cx.latency = obj->integer.value;
421
422 obj = &(element->package.elements[3]);
423 if (obj->type != ACPI_TYPE_INTEGER)
424 continue;
425
426 current_count++;
427 memcpy(&(pr->power.states[current_count]), &cx, sizeof(cx));
428
429 /*
430 * We support total ACPI_PROCESSOR_MAX_POWER - 1
431 * (From 1 through ACPI_PROCESSOR_MAX_POWER - 1)
432 */
433 if (current_count >= (ACPI_PROCESSOR_MAX_POWER - 1)) {
434 pr_warn("Limiting number of power states to max (%d)\n",
435 ACPI_PROCESSOR_MAX_POWER);
436 pr_warn("Please increase ACPI_PROCESSOR_MAX_POWER if needed.\n");
437 break;
438 }
439 }
440
441 ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Found %d power states\n",
442 current_count));
443
444 /* Validate number of power states discovered */
445 if (current_count < 2)
446 ret = -EFAULT;
447
448 end:
449 kfree(buffer.pointer);
450
451 return ret;
452}
453
454static void acpi_processor_power_verify_c3(struct acpi_processor *pr,
455 struct acpi_processor_cx *cx)
456{
457 static int bm_check_flag = -1;
458 static int bm_control_flag = -1;
459
460
461 if (!cx->address)
462 return;
463
464 /*
465 * PIIX4 Erratum #18: We don't support C3 when Type-F (fast)
466 * DMA transfers are used by any ISA device to avoid livelock.
467 * Note that we could disable Type-F DMA (as recommended by
468 * the erratum), but this is known to disrupt certain ISA
469 * devices thus we take the conservative approach.
470 */
471 else if (errata.piix4.fdma) {
472 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
473 "C3 not supported on PIIX4 with Type-F DMA\n"));
474 return;
475 }
476
477 /* All the logic here assumes flags.bm_check is same across all CPUs */
478 if (bm_check_flag == -1) {
479 /* Determine whether bm_check is needed based on CPU */
480 acpi_processor_power_init_bm_check(&(pr->flags), pr->id);
481 bm_check_flag = pr->flags.bm_check;
482 bm_control_flag = pr->flags.bm_control;
483 } else {
484 pr->flags.bm_check = bm_check_flag;
485 pr->flags.bm_control = bm_control_flag;
486 }
487
488 if (pr->flags.bm_check) {
489 if (!pr->flags.bm_control) {
490 if (pr->flags.has_cst != 1) {
491 /* bus mastering control is necessary */
492 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
493 "C3 support requires BM control\n"));
494 return;
495 } else {
496 /* Here we enter C3 without bus mastering */
497 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
498 "C3 support without BM control\n"));
499 }
500 }
501 } else {
502 /*
503 * WBINVD should be set in fadt, for C3 state to be
504 * supported on when bm_check is not required.
505 */
506 if (!(acpi_gbl_FADT.flags & ACPI_FADT_WBINVD)) {
507 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
508 "Cache invalidation should work properly"
509 " for C3 to be enabled on SMP systems\n"));
510 return;
511 }
512 }
513
514 /*
515 * Otherwise we've met all of our C3 requirements.
516 * Normalize the C3 latency to expidite policy. Enable
517 * checking of bus mastering status (bm_check) so we can
518 * use this in our C3 policy
519 */
520 cx->valid = 1;
521
522 /*
523 * On older chipsets, BM_RLD needs to be set
524 * in order for Bus Master activity to wake the
525 * system from C3. Newer chipsets handle DMA
526 * during C3 automatically and BM_RLD is a NOP.
527 * In either case, the proper way to
528 * handle BM_RLD is to set it and leave it set.
529 */
530 acpi_write_bit_register(ACPI_BITREG_BUS_MASTER_RLD, 1);
531
532 return;
533}
534
535static int acpi_processor_power_verify(struct acpi_processor *pr)
536{
537 unsigned int i;
538 unsigned int working = 0;
539
540 pr->power.timer_broadcast_on_state = INT_MAX;
541
542 for (i = 1; i < ACPI_PROCESSOR_MAX_POWER && i <= max_cstate; i++) {
543 struct acpi_processor_cx *cx = &pr->power.states[i];
544
545 switch (cx->type) {
546 case ACPI_STATE_C1:
547 cx->valid = 1;
548 break;
549
550 case ACPI_STATE_C2:
551 if (!cx->address)
552 break;
553 cx->valid = 1;
554 break;
555
556 case ACPI_STATE_C3:
557 acpi_processor_power_verify_c3(pr, cx);
558 break;
559 }
560 if (!cx->valid)
561 continue;
562
563 lapic_timer_check_state(i, pr, cx);
564 tsc_check_state(cx->type);
565 working++;
566 }
567
568 lapic_timer_propagate_broadcast(pr);
569
570 return (working);
571}
572
573static int acpi_processor_get_power_info(struct acpi_processor *pr)
574{
575 unsigned int i;
576 int result;
577
578
579 /* NOTE: the idle thread may not be running while calling
580 * this function */
581
582 /* Zero initialize all the C-states info. */
583 memset(pr->power.states, 0, sizeof(pr->power.states));
584
585 result = acpi_processor_get_power_info_cst(pr);
586 if (result == -ENODEV)
587 result = acpi_processor_get_power_info_fadt(pr);
588
589 if (result)
590 return result;
591
592 acpi_processor_get_power_info_default(pr);
593
594 pr->power.count = acpi_processor_power_verify(pr);
595
596 /*
597 * if one state of type C2 or C3 is available, mark this
598 * CPU as being "idle manageable"
599 */
600 for (i = 1; i < ACPI_PROCESSOR_MAX_POWER; i++) {
601 if (pr->power.states[i].valid) {
602 pr->power.count = i;
603 if (pr->power.states[i].type >= ACPI_STATE_C2)
604 pr->flags.power = 1;
605 }
606 }
607
608 return 0;
609}
610
611/**
612 * acpi_idle_bm_check - checks if bus master activity was detected
613 */
614static int acpi_idle_bm_check(void)
615{
616 u32 bm_status = 0;
617
618 if (bm_check_disable)
619 return 0;
620
621 acpi_read_bit_register(ACPI_BITREG_BUS_MASTER_STATUS, &bm_status);
622 if (bm_status)
623 acpi_write_bit_register(ACPI_BITREG_BUS_MASTER_STATUS, 1);
624 /*
625 * PIIX4 Erratum #18: Note that BM_STS doesn't always reflect
626 * the true state of bus mastering activity; forcing us to
627 * manually check the BMIDEA bit of each IDE channel.
628 */
629 else if (errata.piix4.bmisx) {
630 if ((inb_p(errata.piix4.bmisx + 0x02) & 0x01)
631 || (inb_p(errata.piix4.bmisx + 0x0A) & 0x01))
632 bm_status = 1;
633 }
634 return bm_status;
635}
636
637/**
638 * acpi_idle_do_entry - enter idle state using the appropriate method
639 * @cx: cstate data
640 *
641 * Caller disables interrupt before call and enables interrupt after return.
642 */
643static void acpi_idle_do_entry(struct acpi_processor_cx *cx)
644{
645 if (cx->entry_method == ACPI_CSTATE_FFH) {
646 /* Call into architectural FFH based C-state */
647 acpi_processor_ffh_cstate_enter(cx);
648 } else if (cx->entry_method == ACPI_CSTATE_HALT) {
649 acpi_safe_halt();
650 } else {
651 /* IO port based C-state */
652 inb(cx->address);
653 /* Dummy wait op - must do something useless after P_LVL2 read
654 because chipsets cannot guarantee that STPCLK# signal
655 gets asserted in time to freeze execution properly. */
656 inl(acpi_gbl_FADT.xpm_timer_block.address);
657 }
658}
659
660/**
661 * acpi_idle_play_dead - enters an ACPI state for long-term idle (i.e. off-lining)
662 * @dev: the target CPU
663 * @index: the index of suggested state
664 */
665static int acpi_idle_play_dead(struct cpuidle_device *dev, int index)
666{
667 struct acpi_processor_cx *cx = per_cpu(acpi_cstate[index], dev->cpu);
668
669 ACPI_FLUSH_CPU_CACHE();
670
671 while (1) {
672
673 if (cx->entry_method == ACPI_CSTATE_HALT)
674 safe_halt();
675 else if (cx->entry_method == ACPI_CSTATE_SYSTEMIO) {
676 inb(cx->address);
677 /* See comment in acpi_idle_do_entry() */
678 inl(acpi_gbl_FADT.xpm_timer_block.address);
679 } else
680 return -ENODEV;
681 }
682
683 /* Never reached */
684 return 0;
685}
686
687static bool acpi_idle_fallback_to_c1(struct acpi_processor *pr)
688{
689 return IS_ENABLED(CONFIG_HOTPLUG_CPU) && !pr->flags.has_cst &&
690 !(acpi_gbl_FADT.flags & ACPI_FADT_C2_MP_SUPPORTED);
691}
692
693static int c3_cpu_count;
694static DEFINE_RAW_SPINLOCK(c3_lock);
695
696/**
697 * acpi_idle_enter_bm - enters C3 with proper BM handling
698 * @pr: Target processor
699 * @cx: Target state context
700 * @timer_bc: Whether or not to change timer mode to broadcast
701 */
702static void acpi_idle_enter_bm(struct acpi_processor *pr,
703 struct acpi_processor_cx *cx, bool timer_bc)
704{
705 acpi_unlazy_tlb(smp_processor_id());
706
707 /*
708 * Must be done before busmaster disable as we might need to
709 * access HPET !
710 */
711 if (timer_bc)
712 lapic_timer_state_broadcast(pr, cx, 1);
713
714 /*
715 * disable bus master
716 * bm_check implies we need ARB_DIS
717 * bm_control implies whether we can do ARB_DIS
718 *
719 * That leaves a case where bm_check is set and bm_control is
720 * not set. In that case we cannot do much, we enter C3
721 * without doing anything.
722 */
723 if (pr->flags.bm_control) {
724 raw_spin_lock(&c3_lock);
725 c3_cpu_count++;
726 /* Disable bus master arbitration when all CPUs are in C3 */
727 if (c3_cpu_count == num_online_cpus())
728 acpi_write_bit_register(ACPI_BITREG_ARB_DISABLE, 1);
729 raw_spin_unlock(&c3_lock);
730 }
731
732 acpi_idle_do_entry(cx);
733
734 /* Re-enable bus master arbitration */
735 if (pr->flags.bm_control) {
736 raw_spin_lock(&c3_lock);
737 acpi_write_bit_register(ACPI_BITREG_ARB_DISABLE, 0);
738 c3_cpu_count--;
739 raw_spin_unlock(&c3_lock);
740 }
741
742 if (timer_bc)
743 lapic_timer_state_broadcast(pr, cx, 0);
744}
745
746static int acpi_idle_enter(struct cpuidle_device *dev,
747 struct cpuidle_driver *drv, int index)
748{
749 struct acpi_processor_cx *cx = per_cpu(acpi_cstate[index], dev->cpu);
750 struct acpi_processor *pr;
751
752 pr = __this_cpu_read(processors);
753 if (unlikely(!pr))
754 return -EINVAL;
755
756 if (cx->type != ACPI_STATE_C1) {
757 if (acpi_idle_fallback_to_c1(pr) && num_online_cpus() > 1) {
758 index = CPUIDLE_DRIVER_STATE_START;
759 cx = per_cpu(acpi_cstate[index], dev->cpu);
760 } else if (cx->type == ACPI_STATE_C3 && pr->flags.bm_check) {
761 if (cx->bm_sts_skip || !acpi_idle_bm_check()) {
762 acpi_idle_enter_bm(pr, cx, true);
763 return index;
764 } else if (drv->safe_state_index >= 0) {
765 index = drv->safe_state_index;
766 cx = per_cpu(acpi_cstate[index], dev->cpu);
767 } else {
768 acpi_safe_halt();
769 return -EBUSY;
770 }
771 }
772 }
773
774 lapic_timer_state_broadcast(pr, cx, 1);
775
776 if (cx->type == ACPI_STATE_C3)
777 ACPI_FLUSH_CPU_CACHE();
778
779 acpi_idle_do_entry(cx);
780
781 lapic_timer_state_broadcast(pr, cx, 0);
782
783 return index;
784}
785
786static void acpi_idle_enter_freeze(struct cpuidle_device *dev,
787 struct cpuidle_driver *drv, int index)
788{
789 struct acpi_processor_cx *cx = per_cpu(acpi_cstate[index], dev->cpu);
790
791 if (cx->type == ACPI_STATE_C3) {
792 struct acpi_processor *pr = __this_cpu_read(processors);
793
794 if (unlikely(!pr))
795 return;
796
797 if (pr->flags.bm_check) {
798 acpi_idle_enter_bm(pr, cx, false);
799 return;
800 } else {
801 ACPI_FLUSH_CPU_CACHE();
802 }
803 }
804 acpi_idle_do_entry(cx);
805}
806
807struct cpuidle_driver acpi_idle_driver = {
808 .name = "acpi_idle",
809 .owner = THIS_MODULE,
810};
811
812/**
813 * acpi_processor_setup_cpuidle_cx - prepares and configures CPUIDLE
814 * device i.e. per-cpu data
815 *
816 * @pr: the ACPI processor
817 * @dev : the cpuidle device
818 */
819static int acpi_processor_setup_cpuidle_cx(struct acpi_processor *pr,
820 struct cpuidle_device *dev)
821{
822 int i, count = CPUIDLE_DRIVER_STATE_START;
823 struct acpi_processor_cx *cx;
824
825 if (!pr->flags.power_setup_done)
826 return -EINVAL;
827
828 if (pr->flags.power == 0) {
829 return -EINVAL;
830 }
831
832 if (!dev)
833 return -EINVAL;
834
835 dev->cpu = pr->id;
836
837 if (max_cstate == 0)
838 max_cstate = 1;
839
840 for (i = 1; i < ACPI_PROCESSOR_MAX_POWER && i <= max_cstate; i++) {
841 cx = &pr->power.states[i];
842
843 if (!cx->valid)
844 continue;
845
846 per_cpu(acpi_cstate[count], dev->cpu) = cx;
847
848 count++;
849 if (count == CPUIDLE_STATE_MAX)
850 break;
851 }
852
853 if (!count)
854 return -EINVAL;
855
856 return 0;
857}
858
859/**
860 * acpi_processor_setup_cpuidle states- prepares and configures cpuidle
861 * global state data i.e. idle routines
862 *
863 * @pr: the ACPI processor
864 */
865static int acpi_processor_setup_cpuidle_states(struct acpi_processor *pr)
866{
867 int i, count = CPUIDLE_DRIVER_STATE_START;
868 struct acpi_processor_cx *cx;
869 struct cpuidle_state *state;
870 struct cpuidle_driver *drv = &acpi_idle_driver;
871
872 if (!pr->flags.power_setup_done)
873 return -EINVAL;
874
875 if (pr->flags.power == 0)
876 return -EINVAL;
877
878 drv->safe_state_index = -1;
879 for (i = CPUIDLE_DRIVER_STATE_START; i < CPUIDLE_STATE_MAX; i++) {
880 drv->states[i].name[0] = '\0';
881 drv->states[i].desc[0] = '\0';
882 }
883
884 if (max_cstate == 0)
885 max_cstate = 1;
886
887 for (i = 1; i < ACPI_PROCESSOR_MAX_POWER && i <= max_cstate; i++) {
888 cx = &pr->power.states[i];
889
890 if (!cx->valid)
891 continue;
892
893 state = &drv->states[count];
894 snprintf(state->name, CPUIDLE_NAME_LEN, "C%d", i);
895 strncpy(state->desc, cx->desc, CPUIDLE_DESC_LEN);
896 state->exit_latency = cx->latency;
897 state->target_residency = cx->latency * latency_factor;
898 state->enter = acpi_idle_enter;
899
900 state->flags = 0;
901 if (cx->type == ACPI_STATE_C1 || cx->type == ACPI_STATE_C2) {
902 state->enter_dead = acpi_idle_play_dead;
903 drv->safe_state_index = count;
904 }
905 /*
906 * Halt-induced C1 is not good for ->enter_freeze, because it
907 * re-enables interrupts on exit. Moreover, C1 is generally not
908 * particularly interesting from the suspend-to-idle angle, so
909 * avoid C1 and the situations in which we may need to fall back
910 * to it altogether.
911 */
912 if (cx->type != ACPI_STATE_C1 && !acpi_idle_fallback_to_c1(pr))
913 state->enter_freeze = acpi_idle_enter_freeze;
914
915 count++;
916 if (count == CPUIDLE_STATE_MAX)
917 break;
918 }
919
920 drv->state_count = count;
921
922 if (!count)
923 return -EINVAL;
924
925 return 0;
926}
927
928int acpi_processor_hotplug(struct acpi_processor *pr)
929{
930 int ret = 0;
931 struct cpuidle_device *dev;
932
933 if (disabled_by_idle_boot_param())
934 return 0;
935
936 if (nocst)
937 return -ENODEV;
938
939 if (!pr->flags.power_setup_done)
940 return -ENODEV;
941
942 dev = per_cpu(acpi_cpuidle_device, pr->id);
943 cpuidle_pause_and_lock();
944 cpuidle_disable_device(dev);
945 acpi_processor_get_power_info(pr);
946 if (pr->flags.power) {
947 acpi_processor_setup_cpuidle_cx(pr, dev);
948 ret = cpuidle_enable_device(dev);
949 }
950 cpuidle_resume_and_unlock();
951
952 return ret;
953}
954
955int acpi_processor_cst_has_changed(struct acpi_processor *pr)
956{
957 int cpu;
958 struct acpi_processor *_pr;
959 struct cpuidle_device *dev;
960
961 if (disabled_by_idle_boot_param())
962 return 0;
963
964 if (nocst)
965 return -ENODEV;
966
967 if (!pr->flags.power_setup_done)
968 return -ENODEV;
969
970 /*
971 * FIXME: Design the ACPI notification to make it once per
972 * system instead of once per-cpu. This condition is a hack
973 * to make the code that updates C-States be called once.
974 */
975
976 if (pr->id == 0 && cpuidle_get_driver() == &acpi_idle_driver) {
977
978 /* Protect against cpu-hotplug */
979 get_online_cpus();
980 cpuidle_pause_and_lock();
981
982 /* Disable all cpuidle devices */
983 for_each_online_cpu(cpu) {
984 _pr = per_cpu(processors, cpu);
985 if (!_pr || !_pr->flags.power_setup_done)
986 continue;
987 dev = per_cpu(acpi_cpuidle_device, cpu);
988 cpuidle_disable_device(dev);
989 }
990
991 /* Populate Updated C-state information */
992 acpi_processor_get_power_info(pr);
993 acpi_processor_setup_cpuidle_states(pr);
994
995 /* Enable all cpuidle devices */
996 for_each_online_cpu(cpu) {
997 _pr = per_cpu(processors, cpu);
998 if (!_pr || !_pr->flags.power_setup_done)
999 continue;
1000 acpi_processor_get_power_info(_pr);
1001 if (_pr->flags.power) {
1002 dev = per_cpu(acpi_cpuidle_device, cpu);
1003 acpi_processor_setup_cpuidle_cx(_pr, dev);
1004 cpuidle_enable_device(dev);
1005 }
1006 }
1007 cpuidle_resume_and_unlock();
1008 put_online_cpus();
1009 }
1010
1011 return 0;
1012}
1013
1014static int acpi_processor_registered;
1015
1016int acpi_processor_power_init(struct acpi_processor *pr)
1017{
1018 acpi_status status;
1019 int retval;
1020 struct cpuidle_device *dev;
1021 static int first_run;
1022
1023 if (disabled_by_idle_boot_param())
1024 return 0;
1025
1026 if (!first_run) {
1027 dmi_check_system(processor_power_dmi_table);
1028 max_cstate = acpi_processor_cstate_check(max_cstate);
1029 if (max_cstate < ACPI_C_STATES_MAX)
1030 printk(KERN_NOTICE
1031 "ACPI: processor limited to max C-state %d\n",
1032 max_cstate);
1033 first_run++;
1034 }
1035
1036 if (acpi_gbl_FADT.cst_control && !nocst) {
1037 status =
1038 acpi_os_write_port(acpi_gbl_FADT.smi_command, acpi_gbl_FADT.cst_control, 8);
1039 if (ACPI_FAILURE(status)) {
1040 ACPI_EXCEPTION((AE_INFO, status,
1041 "Notifying BIOS of _CST ability failed"));
1042 }
1043 }
1044
1045 acpi_processor_get_power_info(pr);
1046 pr->flags.power_setup_done = 1;
1047
1048 /*
1049 * Install the idle handler if processor power management is supported.
1050 * Note that we use previously set idle handler will be used on
1051 * platforms that only support C1.
1052 */
1053 if (pr->flags.power) {
1054 /* Register acpi_idle_driver if not already registered */
1055 if (!acpi_processor_registered) {
1056 acpi_processor_setup_cpuidle_states(pr);
1057 retval = cpuidle_register_driver(&acpi_idle_driver);
1058 if (retval)
1059 return retval;
1060 pr_debug("%s registered with cpuidle\n",
1061 acpi_idle_driver.name);
1062 }
1063
1064 dev = kzalloc(sizeof(*dev), GFP_KERNEL);
1065 if (!dev)
1066 return -ENOMEM;
1067 per_cpu(acpi_cpuidle_device, pr->id) = dev;
1068
1069 acpi_processor_setup_cpuidle_cx(pr, dev);
1070
1071 /* Register per-cpu cpuidle_device. Cpuidle driver
1072 * must already be registered before registering device
1073 */
1074 retval = cpuidle_register_device(dev);
1075 if (retval) {
1076 if (acpi_processor_registered == 0)
1077 cpuidle_unregister_driver(&acpi_idle_driver);
1078 return retval;
1079 }
1080 acpi_processor_registered++;
1081 }
1082 return 0;
1083}
1084
1085int acpi_processor_power_exit(struct acpi_processor *pr)
1086{
1087 struct cpuidle_device *dev = per_cpu(acpi_cpuidle_device, pr->id);
1088
1089 if (disabled_by_idle_boot_param())
1090 return 0;
1091
1092 if (pr->flags.power) {
1093 cpuidle_unregister_device(dev);
1094 acpi_processor_registered--;
1095 if (acpi_processor_registered == 0)
1096 cpuidle_unregister_driver(&acpi_idle_driver);
1097 }
1098
1099 pr->flags.power_setup_done = 0;
1100 return 0;
1101}
1/*
2 * processor_idle - idle state submodule to the ACPI processor driver
3 *
4 * Copyright (C) 2001, 2002 Andy Grover <andrew.grover@intel.com>
5 * Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
6 * Copyright (C) 2004, 2005 Dominik Brodowski <linux@brodo.de>
7 * Copyright (C) 2004 Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
8 * - Added processor hotplug support
9 * Copyright (C) 2005 Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
10 * - Added support for C3 on SMP
11 *
12 * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
13 *
14 * This program is free software; you can redistribute it and/or modify
15 * it under the terms of the GNU General Public License as published by
16 * the Free Software Foundation; either version 2 of the License, or (at
17 * your option) any later version.
18 *
19 * This program is distributed in the hope that it will be useful, but
20 * WITHOUT ANY WARRANTY; without even the implied warranty of
21 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
22 * General Public License for more details.
23 *
24 * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
25 */
26#define pr_fmt(fmt) "ACPI: " fmt
27
28#include <linux/module.h>
29#include <linux/acpi.h>
30#include <linux/dmi.h>
31#include <linux/sched.h> /* need_resched() */
32#include <linux/tick.h>
33#include <linux/cpuidle.h>
34#include <linux/cpu.h>
35#include <acpi/processor.h>
36
37/*
38 * Include the apic definitions for x86 to have the APIC timer related defines
39 * available also for UP (on SMP it gets magically included via linux/smp.h).
40 * asm/acpi.h is not an option, as it would require more include magic. Also
41 * creating an empty asm-ia64/apic.h would just trade pest vs. cholera.
42 */
43#ifdef CONFIG_X86
44#include <asm/apic.h>
45#endif
46
47#define ACPI_PROCESSOR_CLASS "processor"
48#define _COMPONENT ACPI_PROCESSOR_COMPONENT
49ACPI_MODULE_NAME("processor_idle");
50
51static unsigned int max_cstate __read_mostly = ACPI_PROCESSOR_MAX_POWER;
52module_param(max_cstate, uint, 0000);
53static unsigned int nocst __read_mostly;
54module_param(nocst, uint, 0000);
55static int bm_check_disable __read_mostly;
56module_param(bm_check_disable, uint, 0000);
57
58static unsigned int latency_factor __read_mostly = 2;
59module_param(latency_factor, uint, 0644);
60
61static DEFINE_PER_CPU(struct cpuidle_device *, acpi_cpuidle_device);
62
63struct cpuidle_driver acpi_idle_driver = {
64 .name = "acpi_idle",
65 .owner = THIS_MODULE,
66};
67
68#ifdef CONFIG_ACPI_PROCESSOR_CSTATE
69static
70DEFINE_PER_CPU(struct acpi_processor_cx * [CPUIDLE_STATE_MAX], acpi_cstate);
71
72static int disabled_by_idle_boot_param(void)
73{
74 return boot_option_idle_override == IDLE_POLL ||
75 boot_option_idle_override == IDLE_HALT;
76}
77
78/*
79 * IBM ThinkPad R40e crashes mysteriously when going into C2 or C3.
80 * For now disable this. Probably a bug somewhere else.
81 *
82 * To skip this limit, boot/load with a large max_cstate limit.
83 */
84static int set_max_cstate(const struct dmi_system_id *id)
85{
86 if (max_cstate > ACPI_PROCESSOR_MAX_POWER)
87 return 0;
88
89 pr_notice("%s detected - limiting to C%ld max_cstate."
90 " Override with \"processor.max_cstate=%d\"\n", id->ident,
91 (long)id->driver_data, ACPI_PROCESSOR_MAX_POWER + 1);
92
93 max_cstate = (long)id->driver_data;
94
95 return 0;
96}
97
98static const struct dmi_system_id processor_power_dmi_table[] = {
99 { set_max_cstate, "Clevo 5600D", {
100 DMI_MATCH(DMI_BIOS_VENDOR,"Phoenix Technologies LTD"),
101 DMI_MATCH(DMI_BIOS_VERSION,"SHE845M0.86C.0013.D.0302131307")},
102 (void *)2},
103 { set_max_cstate, "Pavilion zv5000", {
104 DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
105 DMI_MATCH(DMI_PRODUCT_NAME,"Pavilion zv5000 (DS502A#ABA)")},
106 (void *)1},
107 { set_max_cstate, "Asus L8400B", {
108 DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK Computer Inc."),
109 DMI_MATCH(DMI_PRODUCT_NAME,"L8400B series Notebook PC")},
110 (void *)1},
111 {},
112};
113
114
115/*
116 * Callers should disable interrupts before the call and enable
117 * interrupts after return.
118 */
119static void __cpuidle acpi_safe_halt(void)
120{
121 if (!tif_need_resched()) {
122 safe_halt();
123 local_irq_disable();
124 }
125}
126
127#ifdef ARCH_APICTIMER_STOPS_ON_C3
128
129/*
130 * Some BIOS implementations switch to C3 in the published C2 state.
131 * This seems to be a common problem on AMD boxen, but other vendors
132 * are affected too. We pick the most conservative approach: we assume
133 * that the local APIC stops in both C2 and C3.
134 */
135static void lapic_timer_check_state(int state, struct acpi_processor *pr,
136 struct acpi_processor_cx *cx)
137{
138 struct acpi_processor_power *pwr = &pr->power;
139 u8 type = local_apic_timer_c2_ok ? ACPI_STATE_C3 : ACPI_STATE_C2;
140
141 if (cpu_has(&cpu_data(pr->id), X86_FEATURE_ARAT))
142 return;
143
144 if (boot_cpu_has_bug(X86_BUG_AMD_APIC_C1E))
145 type = ACPI_STATE_C1;
146
147 /*
148 * Check, if one of the previous states already marked the lapic
149 * unstable
150 */
151 if (pwr->timer_broadcast_on_state < state)
152 return;
153
154 if (cx->type >= type)
155 pr->power.timer_broadcast_on_state = state;
156}
157
158static void __lapic_timer_propagate_broadcast(void *arg)
159{
160 struct acpi_processor *pr = (struct acpi_processor *) arg;
161
162 if (pr->power.timer_broadcast_on_state < INT_MAX)
163 tick_broadcast_enable();
164 else
165 tick_broadcast_disable();
166}
167
168static void lapic_timer_propagate_broadcast(struct acpi_processor *pr)
169{
170 smp_call_function_single(pr->id, __lapic_timer_propagate_broadcast,
171 (void *)pr, 1);
172}
173
174/* Power(C) State timer broadcast control */
175static void lapic_timer_state_broadcast(struct acpi_processor *pr,
176 struct acpi_processor_cx *cx,
177 int broadcast)
178{
179 int state = cx - pr->power.states;
180
181 if (state >= pr->power.timer_broadcast_on_state) {
182 if (broadcast)
183 tick_broadcast_enter();
184 else
185 tick_broadcast_exit();
186 }
187}
188
189#else
190
191static void lapic_timer_check_state(int state, struct acpi_processor *pr,
192 struct acpi_processor_cx *cstate) { }
193static void lapic_timer_propagate_broadcast(struct acpi_processor *pr) { }
194static void lapic_timer_state_broadcast(struct acpi_processor *pr,
195 struct acpi_processor_cx *cx,
196 int broadcast)
197{
198}
199
200#endif
201
202#if defined(CONFIG_X86)
203static void tsc_check_state(int state)
204{
205 switch (boot_cpu_data.x86_vendor) {
206 case X86_VENDOR_AMD:
207 case X86_VENDOR_INTEL:
208 /*
209 * AMD Fam10h TSC will tick in all
210 * C/P/S0/S1 states when this bit is set.
211 */
212 if (boot_cpu_has(X86_FEATURE_NONSTOP_TSC))
213 return;
214
215 /*FALL THROUGH*/
216 default:
217 /* TSC could halt in idle, so notify users */
218 if (state > ACPI_STATE_C1)
219 mark_tsc_unstable("TSC halts in idle");
220 }
221}
222#else
223static void tsc_check_state(int state) { return; }
224#endif
225
226static int acpi_processor_get_power_info_fadt(struct acpi_processor *pr)
227{
228
229 if (!pr->pblk)
230 return -ENODEV;
231
232 /* if info is obtained from pblk/fadt, type equals state */
233 pr->power.states[ACPI_STATE_C2].type = ACPI_STATE_C2;
234 pr->power.states[ACPI_STATE_C3].type = ACPI_STATE_C3;
235
236#ifndef CONFIG_HOTPLUG_CPU
237 /*
238 * Check for P_LVL2_UP flag before entering C2 and above on
239 * an SMP system.
240 */
241 if ((num_online_cpus() > 1) &&
242 !(acpi_gbl_FADT.flags & ACPI_FADT_C2_MP_SUPPORTED))
243 return -ENODEV;
244#endif
245
246 /* determine C2 and C3 address from pblk */
247 pr->power.states[ACPI_STATE_C2].address = pr->pblk + 4;
248 pr->power.states[ACPI_STATE_C3].address = pr->pblk + 5;
249
250 /* determine latencies from FADT */
251 pr->power.states[ACPI_STATE_C2].latency = acpi_gbl_FADT.c2_latency;
252 pr->power.states[ACPI_STATE_C3].latency = acpi_gbl_FADT.c3_latency;
253
254 /*
255 * FADT specified C2 latency must be less than or equal to
256 * 100 microseconds.
257 */
258 if (acpi_gbl_FADT.c2_latency > ACPI_PROCESSOR_MAX_C2_LATENCY) {
259 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
260 "C2 latency too large [%d]\n", acpi_gbl_FADT.c2_latency));
261 /* invalidate C2 */
262 pr->power.states[ACPI_STATE_C2].address = 0;
263 }
264
265 /*
266 * FADT supplied C3 latency must be less than or equal to
267 * 1000 microseconds.
268 */
269 if (acpi_gbl_FADT.c3_latency > ACPI_PROCESSOR_MAX_C3_LATENCY) {
270 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
271 "C3 latency too large [%d]\n", acpi_gbl_FADT.c3_latency));
272 /* invalidate C3 */
273 pr->power.states[ACPI_STATE_C3].address = 0;
274 }
275
276 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
277 "lvl2[0x%08x] lvl3[0x%08x]\n",
278 pr->power.states[ACPI_STATE_C2].address,
279 pr->power.states[ACPI_STATE_C3].address));
280
281 return 0;
282}
283
284static int acpi_processor_get_power_info_default(struct acpi_processor *pr)
285{
286 if (!pr->power.states[ACPI_STATE_C1].valid) {
287 /* set the first C-State to C1 */
288 /* all processors need to support C1 */
289 pr->power.states[ACPI_STATE_C1].type = ACPI_STATE_C1;
290 pr->power.states[ACPI_STATE_C1].valid = 1;
291 pr->power.states[ACPI_STATE_C1].entry_method = ACPI_CSTATE_HALT;
292 }
293 /* the C0 state only exists as a filler in our array */
294 pr->power.states[ACPI_STATE_C0].valid = 1;
295 return 0;
296}
297
298static int acpi_processor_get_power_info_cst(struct acpi_processor *pr)
299{
300 acpi_status status;
301 u64 count;
302 int current_count;
303 int i, ret = 0;
304 struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
305 union acpi_object *cst;
306
307 if (nocst)
308 return -ENODEV;
309
310 current_count = 0;
311
312 status = acpi_evaluate_object(pr->handle, "_CST", NULL, &buffer);
313 if (ACPI_FAILURE(status)) {
314 ACPI_DEBUG_PRINT((ACPI_DB_INFO, "No _CST, giving up\n"));
315 return -ENODEV;
316 }
317
318 cst = buffer.pointer;
319
320 /* There must be at least 2 elements */
321 if (!cst || (cst->type != ACPI_TYPE_PACKAGE) || cst->package.count < 2) {
322 pr_err("not enough elements in _CST\n");
323 ret = -EFAULT;
324 goto end;
325 }
326
327 count = cst->package.elements[0].integer.value;
328
329 /* Validate number of power states. */
330 if (count < 1 || count != cst->package.count - 1) {
331 pr_err("count given by _CST is not valid\n");
332 ret = -EFAULT;
333 goto end;
334 }
335
336 /* Tell driver that at least _CST is supported. */
337 pr->flags.has_cst = 1;
338
339 for (i = 1; i <= count; i++) {
340 union acpi_object *element;
341 union acpi_object *obj;
342 struct acpi_power_register *reg;
343 struct acpi_processor_cx cx;
344
345 memset(&cx, 0, sizeof(cx));
346
347 element = &(cst->package.elements[i]);
348 if (element->type != ACPI_TYPE_PACKAGE)
349 continue;
350
351 if (element->package.count != 4)
352 continue;
353
354 obj = &(element->package.elements[0]);
355
356 if (obj->type != ACPI_TYPE_BUFFER)
357 continue;
358
359 reg = (struct acpi_power_register *)obj->buffer.pointer;
360
361 if (reg->space_id != ACPI_ADR_SPACE_SYSTEM_IO &&
362 (reg->space_id != ACPI_ADR_SPACE_FIXED_HARDWARE))
363 continue;
364
365 /* There should be an easy way to extract an integer... */
366 obj = &(element->package.elements[1]);
367 if (obj->type != ACPI_TYPE_INTEGER)
368 continue;
369
370 cx.type = obj->integer.value;
371 /*
372 * Some buggy BIOSes won't list C1 in _CST -
373 * Let acpi_processor_get_power_info_default() handle them later
374 */
375 if (i == 1 && cx.type != ACPI_STATE_C1)
376 current_count++;
377
378 cx.address = reg->address;
379 cx.index = current_count + 1;
380
381 cx.entry_method = ACPI_CSTATE_SYSTEMIO;
382 if (reg->space_id == ACPI_ADR_SPACE_FIXED_HARDWARE) {
383 if (acpi_processor_ffh_cstate_probe
384 (pr->id, &cx, reg) == 0) {
385 cx.entry_method = ACPI_CSTATE_FFH;
386 } else if (cx.type == ACPI_STATE_C1) {
387 /*
388 * C1 is a special case where FIXED_HARDWARE
389 * can be handled in non-MWAIT way as well.
390 * In that case, save this _CST entry info.
391 * Otherwise, ignore this info and continue.
392 */
393 cx.entry_method = ACPI_CSTATE_HALT;
394 snprintf(cx.desc, ACPI_CX_DESC_LEN, "ACPI HLT");
395 } else {
396 continue;
397 }
398 if (cx.type == ACPI_STATE_C1 &&
399 (boot_option_idle_override == IDLE_NOMWAIT)) {
400 /*
401 * In most cases the C1 space_id obtained from
402 * _CST object is FIXED_HARDWARE access mode.
403 * But when the option of idle=halt is added,
404 * the entry_method type should be changed from
405 * CSTATE_FFH to CSTATE_HALT.
406 * When the option of idle=nomwait is added,
407 * the C1 entry_method type should be
408 * CSTATE_HALT.
409 */
410 cx.entry_method = ACPI_CSTATE_HALT;
411 snprintf(cx.desc, ACPI_CX_DESC_LEN, "ACPI HLT");
412 }
413 } else {
414 snprintf(cx.desc, ACPI_CX_DESC_LEN, "ACPI IOPORT 0x%x",
415 cx.address);
416 }
417
418 if (cx.type == ACPI_STATE_C1) {
419 cx.valid = 1;
420 }
421
422 obj = &(element->package.elements[2]);
423 if (obj->type != ACPI_TYPE_INTEGER)
424 continue;
425
426 cx.latency = obj->integer.value;
427
428 obj = &(element->package.elements[3]);
429 if (obj->type != ACPI_TYPE_INTEGER)
430 continue;
431
432 current_count++;
433 memcpy(&(pr->power.states[current_count]), &cx, sizeof(cx));
434
435 /*
436 * We support total ACPI_PROCESSOR_MAX_POWER - 1
437 * (From 1 through ACPI_PROCESSOR_MAX_POWER - 1)
438 */
439 if (current_count >= (ACPI_PROCESSOR_MAX_POWER - 1)) {
440 pr_warn("Limiting number of power states to max (%d)\n",
441 ACPI_PROCESSOR_MAX_POWER);
442 pr_warn("Please increase ACPI_PROCESSOR_MAX_POWER if needed.\n");
443 break;
444 }
445 }
446
447 ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Found %d power states\n",
448 current_count));
449
450 /* Validate number of power states discovered */
451 if (current_count < 2)
452 ret = -EFAULT;
453
454 end:
455 kfree(buffer.pointer);
456
457 return ret;
458}
459
460static void acpi_processor_power_verify_c3(struct acpi_processor *pr,
461 struct acpi_processor_cx *cx)
462{
463 static int bm_check_flag = -1;
464 static int bm_control_flag = -1;
465
466
467 if (!cx->address)
468 return;
469
470 /*
471 * PIIX4 Erratum #18: We don't support C3 when Type-F (fast)
472 * DMA transfers are used by any ISA device to avoid livelock.
473 * Note that we could disable Type-F DMA (as recommended by
474 * the erratum), but this is known to disrupt certain ISA
475 * devices thus we take the conservative approach.
476 */
477 else if (errata.piix4.fdma) {
478 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
479 "C3 not supported on PIIX4 with Type-F DMA\n"));
480 return;
481 }
482
483 /* All the logic here assumes flags.bm_check is same across all CPUs */
484 if (bm_check_flag == -1) {
485 /* Determine whether bm_check is needed based on CPU */
486 acpi_processor_power_init_bm_check(&(pr->flags), pr->id);
487 bm_check_flag = pr->flags.bm_check;
488 bm_control_flag = pr->flags.bm_control;
489 } else {
490 pr->flags.bm_check = bm_check_flag;
491 pr->flags.bm_control = bm_control_flag;
492 }
493
494 if (pr->flags.bm_check) {
495 if (!pr->flags.bm_control) {
496 if (pr->flags.has_cst != 1) {
497 /* bus mastering control is necessary */
498 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
499 "C3 support requires BM control\n"));
500 return;
501 } else {
502 /* Here we enter C3 without bus mastering */
503 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
504 "C3 support without BM control\n"));
505 }
506 }
507 } else {
508 /*
509 * WBINVD should be set in fadt, for C3 state to be
510 * supported on when bm_check is not required.
511 */
512 if (!(acpi_gbl_FADT.flags & ACPI_FADT_WBINVD)) {
513 ACPI_DEBUG_PRINT((ACPI_DB_INFO,
514 "Cache invalidation should work properly"
515 " for C3 to be enabled on SMP systems\n"));
516 return;
517 }
518 }
519
520 /*
521 * Otherwise we've met all of our C3 requirements.
522 * Normalize the C3 latency to expidite policy. Enable
523 * checking of bus mastering status (bm_check) so we can
524 * use this in our C3 policy
525 */
526 cx->valid = 1;
527
528 /*
529 * On older chipsets, BM_RLD needs to be set
530 * in order for Bus Master activity to wake the
531 * system from C3. Newer chipsets handle DMA
532 * during C3 automatically and BM_RLD is a NOP.
533 * In either case, the proper way to
534 * handle BM_RLD is to set it and leave it set.
535 */
536 acpi_write_bit_register(ACPI_BITREG_BUS_MASTER_RLD, 1);
537
538 return;
539}
540
541static int acpi_processor_power_verify(struct acpi_processor *pr)
542{
543 unsigned int i;
544 unsigned int working = 0;
545
546 pr->power.timer_broadcast_on_state = INT_MAX;
547
548 for (i = 1; i < ACPI_PROCESSOR_MAX_POWER && i <= max_cstate; i++) {
549 struct acpi_processor_cx *cx = &pr->power.states[i];
550
551 switch (cx->type) {
552 case ACPI_STATE_C1:
553 cx->valid = 1;
554 break;
555
556 case ACPI_STATE_C2:
557 if (!cx->address)
558 break;
559 cx->valid = 1;
560 break;
561
562 case ACPI_STATE_C3:
563 acpi_processor_power_verify_c3(pr, cx);
564 break;
565 }
566 if (!cx->valid)
567 continue;
568
569 lapic_timer_check_state(i, pr, cx);
570 tsc_check_state(cx->type);
571 working++;
572 }
573
574 lapic_timer_propagate_broadcast(pr);
575
576 return (working);
577}
578
579static int acpi_processor_get_cstate_info(struct acpi_processor *pr)
580{
581 unsigned int i;
582 int result;
583
584
585 /* NOTE: the idle thread may not be running while calling
586 * this function */
587
588 /* Zero initialize all the C-states info. */
589 memset(pr->power.states, 0, sizeof(pr->power.states));
590
591 result = acpi_processor_get_power_info_cst(pr);
592 if (result == -ENODEV)
593 result = acpi_processor_get_power_info_fadt(pr);
594
595 if (result)
596 return result;
597
598 acpi_processor_get_power_info_default(pr);
599
600 pr->power.count = acpi_processor_power_verify(pr);
601
602 /*
603 * if one state of type C2 or C3 is available, mark this
604 * CPU as being "idle manageable"
605 */
606 for (i = 1; i < ACPI_PROCESSOR_MAX_POWER; i++) {
607 if (pr->power.states[i].valid) {
608 pr->power.count = i;
609 if (pr->power.states[i].type >= ACPI_STATE_C2)
610 pr->flags.power = 1;
611 }
612 }
613
614 return 0;
615}
616
617/**
618 * acpi_idle_bm_check - checks if bus master activity was detected
619 */
620static int acpi_idle_bm_check(void)
621{
622 u32 bm_status = 0;
623
624 if (bm_check_disable)
625 return 0;
626
627 acpi_read_bit_register(ACPI_BITREG_BUS_MASTER_STATUS, &bm_status);
628 if (bm_status)
629 acpi_write_bit_register(ACPI_BITREG_BUS_MASTER_STATUS, 1);
630 /*
631 * PIIX4 Erratum #18: Note that BM_STS doesn't always reflect
632 * the true state of bus mastering activity; forcing us to
633 * manually check the BMIDEA bit of each IDE channel.
634 */
635 else if (errata.piix4.bmisx) {
636 if ((inb_p(errata.piix4.bmisx + 0x02) & 0x01)
637 || (inb_p(errata.piix4.bmisx + 0x0A) & 0x01))
638 bm_status = 1;
639 }
640 return bm_status;
641}
642
643/**
644 * acpi_idle_do_entry - enter idle state using the appropriate method
645 * @cx: cstate data
646 *
647 * Caller disables interrupt before call and enables interrupt after return.
648 */
649static void __cpuidle acpi_idle_do_entry(struct acpi_processor_cx *cx)
650{
651 if (cx->entry_method == ACPI_CSTATE_FFH) {
652 /* Call into architectural FFH based C-state */
653 acpi_processor_ffh_cstate_enter(cx);
654 } else if (cx->entry_method == ACPI_CSTATE_HALT) {
655 acpi_safe_halt();
656 } else {
657 /* IO port based C-state */
658 inb(cx->address);
659 /* Dummy wait op - must do something useless after P_LVL2 read
660 because chipsets cannot guarantee that STPCLK# signal
661 gets asserted in time to freeze execution properly. */
662 inl(acpi_gbl_FADT.xpm_timer_block.address);
663 }
664}
665
666/**
667 * acpi_idle_play_dead - enters an ACPI state for long-term idle (i.e. off-lining)
668 * @dev: the target CPU
669 * @index: the index of suggested state
670 */
671static int acpi_idle_play_dead(struct cpuidle_device *dev, int index)
672{
673 struct acpi_processor_cx *cx = per_cpu(acpi_cstate[index], dev->cpu);
674
675 ACPI_FLUSH_CPU_CACHE();
676
677 while (1) {
678
679 if (cx->entry_method == ACPI_CSTATE_HALT)
680 safe_halt();
681 else if (cx->entry_method == ACPI_CSTATE_SYSTEMIO) {
682 inb(cx->address);
683 /* See comment in acpi_idle_do_entry() */
684 inl(acpi_gbl_FADT.xpm_timer_block.address);
685 } else
686 return -ENODEV;
687 }
688
689 /* Never reached */
690 return 0;
691}
692
693static bool acpi_idle_fallback_to_c1(struct acpi_processor *pr)
694{
695 return IS_ENABLED(CONFIG_HOTPLUG_CPU) && !pr->flags.has_cst &&
696 !(acpi_gbl_FADT.flags & ACPI_FADT_C2_MP_SUPPORTED);
697}
698
699static int c3_cpu_count;
700static DEFINE_RAW_SPINLOCK(c3_lock);
701
702/**
703 * acpi_idle_enter_bm - enters C3 with proper BM handling
704 * @pr: Target processor
705 * @cx: Target state context
706 * @timer_bc: Whether or not to change timer mode to broadcast
707 */
708static void acpi_idle_enter_bm(struct acpi_processor *pr,
709 struct acpi_processor_cx *cx, bool timer_bc)
710{
711 acpi_unlazy_tlb(smp_processor_id());
712
713 /*
714 * Must be done before busmaster disable as we might need to
715 * access HPET !
716 */
717 if (timer_bc)
718 lapic_timer_state_broadcast(pr, cx, 1);
719
720 /*
721 * disable bus master
722 * bm_check implies we need ARB_DIS
723 * bm_control implies whether we can do ARB_DIS
724 *
725 * That leaves a case where bm_check is set and bm_control is
726 * not set. In that case we cannot do much, we enter C3
727 * without doing anything.
728 */
729 if (pr->flags.bm_control) {
730 raw_spin_lock(&c3_lock);
731 c3_cpu_count++;
732 /* Disable bus master arbitration when all CPUs are in C3 */
733 if (c3_cpu_count == num_online_cpus())
734 acpi_write_bit_register(ACPI_BITREG_ARB_DISABLE, 1);
735 raw_spin_unlock(&c3_lock);
736 }
737
738 acpi_idle_do_entry(cx);
739
740 /* Re-enable bus master arbitration */
741 if (pr->flags.bm_control) {
742 raw_spin_lock(&c3_lock);
743 acpi_write_bit_register(ACPI_BITREG_ARB_DISABLE, 0);
744 c3_cpu_count--;
745 raw_spin_unlock(&c3_lock);
746 }
747
748 if (timer_bc)
749 lapic_timer_state_broadcast(pr, cx, 0);
750}
751
752static int acpi_idle_enter(struct cpuidle_device *dev,
753 struct cpuidle_driver *drv, int index)
754{
755 struct acpi_processor_cx *cx = per_cpu(acpi_cstate[index], dev->cpu);
756 struct acpi_processor *pr;
757
758 pr = __this_cpu_read(processors);
759 if (unlikely(!pr))
760 return -EINVAL;
761
762 if (cx->type != ACPI_STATE_C1) {
763 if (acpi_idle_fallback_to_c1(pr) && num_online_cpus() > 1) {
764 index = CPUIDLE_DRIVER_STATE_START;
765 cx = per_cpu(acpi_cstate[index], dev->cpu);
766 } else if (cx->type == ACPI_STATE_C3 && pr->flags.bm_check) {
767 if (cx->bm_sts_skip || !acpi_idle_bm_check()) {
768 acpi_idle_enter_bm(pr, cx, true);
769 return index;
770 } else if (drv->safe_state_index >= 0) {
771 index = drv->safe_state_index;
772 cx = per_cpu(acpi_cstate[index], dev->cpu);
773 } else {
774 acpi_safe_halt();
775 return -EBUSY;
776 }
777 }
778 }
779
780 lapic_timer_state_broadcast(pr, cx, 1);
781
782 if (cx->type == ACPI_STATE_C3)
783 ACPI_FLUSH_CPU_CACHE();
784
785 acpi_idle_do_entry(cx);
786
787 lapic_timer_state_broadcast(pr, cx, 0);
788
789 return index;
790}
791
792static void acpi_idle_enter_freeze(struct cpuidle_device *dev,
793 struct cpuidle_driver *drv, int index)
794{
795 struct acpi_processor_cx *cx = per_cpu(acpi_cstate[index], dev->cpu);
796
797 if (cx->type == ACPI_STATE_C3) {
798 struct acpi_processor *pr = __this_cpu_read(processors);
799
800 if (unlikely(!pr))
801 return;
802
803 if (pr->flags.bm_check) {
804 acpi_idle_enter_bm(pr, cx, false);
805 return;
806 } else {
807 ACPI_FLUSH_CPU_CACHE();
808 }
809 }
810 acpi_idle_do_entry(cx);
811}
812
813static int acpi_processor_setup_cpuidle_cx(struct acpi_processor *pr,
814 struct cpuidle_device *dev)
815{
816 int i, count = CPUIDLE_DRIVER_STATE_START;
817 struct acpi_processor_cx *cx;
818
819 if (max_cstate == 0)
820 max_cstate = 1;
821
822 for (i = 1; i < ACPI_PROCESSOR_MAX_POWER && i <= max_cstate; i++) {
823 cx = &pr->power.states[i];
824
825 if (!cx->valid)
826 continue;
827
828 per_cpu(acpi_cstate[count], dev->cpu) = cx;
829
830 count++;
831 if (count == CPUIDLE_STATE_MAX)
832 break;
833 }
834
835 if (!count)
836 return -EINVAL;
837
838 return 0;
839}
840
841static int acpi_processor_setup_cstates(struct acpi_processor *pr)
842{
843 int i, count = CPUIDLE_DRIVER_STATE_START;
844 struct acpi_processor_cx *cx;
845 struct cpuidle_state *state;
846 struct cpuidle_driver *drv = &acpi_idle_driver;
847
848 if (max_cstate == 0)
849 max_cstate = 1;
850
851 for (i = 1; i < ACPI_PROCESSOR_MAX_POWER && i <= max_cstate; i++) {
852 cx = &pr->power.states[i];
853
854 if (!cx->valid)
855 continue;
856
857 state = &drv->states[count];
858 snprintf(state->name, CPUIDLE_NAME_LEN, "C%d", i);
859 strlcpy(state->desc, cx->desc, CPUIDLE_DESC_LEN);
860 state->exit_latency = cx->latency;
861 state->target_residency = cx->latency * latency_factor;
862 state->enter = acpi_idle_enter;
863
864 state->flags = 0;
865 if (cx->type == ACPI_STATE_C1 || cx->type == ACPI_STATE_C2) {
866 state->enter_dead = acpi_idle_play_dead;
867 drv->safe_state_index = count;
868 }
869 /*
870 * Halt-induced C1 is not good for ->enter_freeze, because it
871 * re-enables interrupts on exit. Moreover, C1 is generally not
872 * particularly interesting from the suspend-to-idle angle, so
873 * avoid C1 and the situations in which we may need to fall back
874 * to it altogether.
875 */
876 if (cx->type != ACPI_STATE_C1 && !acpi_idle_fallback_to_c1(pr))
877 state->enter_freeze = acpi_idle_enter_freeze;
878
879 count++;
880 if (count == CPUIDLE_STATE_MAX)
881 break;
882 }
883
884 drv->state_count = count;
885
886 if (!count)
887 return -EINVAL;
888
889 return 0;
890}
891
892static inline void acpi_processor_cstate_first_run_checks(void)
893{
894 acpi_status status;
895 static int first_run;
896
897 if (first_run)
898 return;
899 dmi_check_system(processor_power_dmi_table);
900 max_cstate = acpi_processor_cstate_check(max_cstate);
901 if (max_cstate < ACPI_C_STATES_MAX)
902 pr_notice("ACPI: processor limited to max C-state %d\n",
903 max_cstate);
904 first_run++;
905
906 if (acpi_gbl_FADT.cst_control && !nocst) {
907 status = acpi_os_write_port(acpi_gbl_FADT.smi_command,
908 acpi_gbl_FADT.cst_control, 8);
909 if (ACPI_FAILURE(status))
910 ACPI_EXCEPTION((AE_INFO, status,
911 "Notifying BIOS of _CST ability failed"));
912 }
913}
914#else
915
916static inline int disabled_by_idle_boot_param(void) { return 0; }
917static inline void acpi_processor_cstate_first_run_checks(void) { }
918static int acpi_processor_get_cstate_info(struct acpi_processor *pr)
919{
920 return -ENODEV;
921}
922
923static int acpi_processor_setup_cpuidle_cx(struct acpi_processor *pr,
924 struct cpuidle_device *dev)
925{
926 return -EINVAL;
927}
928
929static int acpi_processor_setup_cstates(struct acpi_processor *pr)
930{
931 return -EINVAL;
932}
933
934#endif /* CONFIG_ACPI_PROCESSOR_CSTATE */
935
936struct acpi_lpi_states_array {
937 unsigned int size;
938 unsigned int composite_states_size;
939 struct acpi_lpi_state *entries;
940 struct acpi_lpi_state *composite_states[ACPI_PROCESSOR_MAX_POWER];
941};
942
943static int obj_get_integer(union acpi_object *obj, u32 *value)
944{
945 if (obj->type != ACPI_TYPE_INTEGER)
946 return -EINVAL;
947
948 *value = obj->integer.value;
949 return 0;
950}
951
952static int acpi_processor_evaluate_lpi(acpi_handle handle,
953 struct acpi_lpi_states_array *info)
954{
955 acpi_status status;
956 int ret = 0;
957 int pkg_count, state_idx = 1, loop;
958 struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
959 union acpi_object *lpi_data;
960 struct acpi_lpi_state *lpi_state;
961
962 status = acpi_evaluate_object(handle, "_LPI", NULL, &buffer);
963 if (ACPI_FAILURE(status)) {
964 ACPI_DEBUG_PRINT((ACPI_DB_INFO, "No _LPI, giving up\n"));
965 return -ENODEV;
966 }
967
968 lpi_data = buffer.pointer;
969
970 /* There must be at least 4 elements = 3 elements + 1 package */
971 if (!lpi_data || lpi_data->type != ACPI_TYPE_PACKAGE ||
972 lpi_data->package.count < 4) {
973 pr_debug("not enough elements in _LPI\n");
974 ret = -ENODATA;
975 goto end;
976 }
977
978 pkg_count = lpi_data->package.elements[2].integer.value;
979
980 /* Validate number of power states. */
981 if (pkg_count < 1 || pkg_count != lpi_data->package.count - 3) {
982 pr_debug("count given by _LPI is not valid\n");
983 ret = -ENODATA;
984 goto end;
985 }
986
987 lpi_state = kcalloc(pkg_count, sizeof(*lpi_state), GFP_KERNEL);
988 if (!lpi_state) {
989 ret = -ENOMEM;
990 goto end;
991 }
992
993 info->size = pkg_count;
994 info->entries = lpi_state;
995
996 /* LPI States start at index 3 */
997 for (loop = 3; state_idx <= pkg_count; loop++, state_idx++, lpi_state++) {
998 union acpi_object *element, *pkg_elem, *obj;
999
1000 element = &lpi_data->package.elements[loop];
1001 if (element->type != ACPI_TYPE_PACKAGE || element->package.count < 7)
1002 continue;
1003
1004 pkg_elem = element->package.elements;
1005
1006 obj = pkg_elem + 6;
1007 if (obj->type == ACPI_TYPE_BUFFER) {
1008 struct acpi_power_register *reg;
1009
1010 reg = (struct acpi_power_register *)obj->buffer.pointer;
1011 if (reg->space_id != ACPI_ADR_SPACE_SYSTEM_IO &&
1012 reg->space_id != ACPI_ADR_SPACE_FIXED_HARDWARE)
1013 continue;
1014
1015 lpi_state->address = reg->address;
1016 lpi_state->entry_method =
1017 reg->space_id == ACPI_ADR_SPACE_FIXED_HARDWARE ?
1018 ACPI_CSTATE_FFH : ACPI_CSTATE_SYSTEMIO;
1019 } else if (obj->type == ACPI_TYPE_INTEGER) {
1020 lpi_state->entry_method = ACPI_CSTATE_INTEGER;
1021 lpi_state->address = obj->integer.value;
1022 } else {
1023 continue;
1024 }
1025
1026 /* elements[7,8] skipped for now i.e. Residency/Usage counter*/
1027
1028 obj = pkg_elem + 9;
1029 if (obj->type == ACPI_TYPE_STRING)
1030 strlcpy(lpi_state->desc, obj->string.pointer,
1031 ACPI_CX_DESC_LEN);
1032
1033 lpi_state->index = state_idx;
1034 if (obj_get_integer(pkg_elem + 0, &lpi_state->min_residency)) {
1035 pr_debug("No min. residency found, assuming 10 us\n");
1036 lpi_state->min_residency = 10;
1037 }
1038
1039 if (obj_get_integer(pkg_elem + 1, &lpi_state->wake_latency)) {
1040 pr_debug("No wakeup residency found, assuming 10 us\n");
1041 lpi_state->wake_latency = 10;
1042 }
1043
1044 if (obj_get_integer(pkg_elem + 2, &lpi_state->flags))
1045 lpi_state->flags = 0;
1046
1047 if (obj_get_integer(pkg_elem + 3, &lpi_state->arch_flags))
1048 lpi_state->arch_flags = 0;
1049
1050 if (obj_get_integer(pkg_elem + 4, &lpi_state->res_cnt_freq))
1051 lpi_state->res_cnt_freq = 1;
1052
1053 if (obj_get_integer(pkg_elem + 5, &lpi_state->enable_parent_state))
1054 lpi_state->enable_parent_state = 0;
1055 }
1056
1057 acpi_handle_debug(handle, "Found %d power states\n", state_idx);
1058end:
1059 kfree(buffer.pointer);
1060 return ret;
1061}
1062
1063/*
1064 * flat_state_cnt - the number of composite LPI states after the process of flattening
1065 */
1066static int flat_state_cnt;
1067
1068/**
1069 * combine_lpi_states - combine local and parent LPI states to form a composite LPI state
1070 *
1071 * @local: local LPI state
1072 * @parent: parent LPI state
1073 * @result: composite LPI state
1074 */
1075static bool combine_lpi_states(struct acpi_lpi_state *local,
1076 struct acpi_lpi_state *parent,
1077 struct acpi_lpi_state *result)
1078{
1079 if (parent->entry_method == ACPI_CSTATE_INTEGER) {
1080 if (!parent->address) /* 0 means autopromotable */
1081 return false;
1082 result->address = local->address + parent->address;
1083 } else {
1084 result->address = parent->address;
1085 }
1086
1087 result->min_residency = max(local->min_residency, parent->min_residency);
1088 result->wake_latency = local->wake_latency + parent->wake_latency;
1089 result->enable_parent_state = parent->enable_parent_state;
1090 result->entry_method = local->entry_method;
1091
1092 result->flags = parent->flags;
1093 result->arch_flags = parent->arch_flags;
1094 result->index = parent->index;
1095
1096 strlcpy(result->desc, local->desc, ACPI_CX_DESC_LEN);
1097 strlcat(result->desc, "+", ACPI_CX_DESC_LEN);
1098 strlcat(result->desc, parent->desc, ACPI_CX_DESC_LEN);
1099 return true;
1100}
1101
1102#define ACPI_LPI_STATE_FLAGS_ENABLED BIT(0)
1103
1104static void stash_composite_state(struct acpi_lpi_states_array *curr_level,
1105 struct acpi_lpi_state *t)
1106{
1107 curr_level->composite_states[curr_level->composite_states_size++] = t;
1108}
1109
1110static int flatten_lpi_states(struct acpi_processor *pr,
1111 struct acpi_lpi_states_array *curr_level,
1112 struct acpi_lpi_states_array *prev_level)
1113{
1114 int i, j, state_count = curr_level->size;
1115 struct acpi_lpi_state *p, *t = curr_level->entries;
1116
1117 curr_level->composite_states_size = 0;
1118 for (j = 0; j < state_count; j++, t++) {
1119 struct acpi_lpi_state *flpi;
1120
1121 if (!(t->flags & ACPI_LPI_STATE_FLAGS_ENABLED))
1122 continue;
1123
1124 if (flat_state_cnt >= ACPI_PROCESSOR_MAX_POWER) {
1125 pr_warn("Limiting number of LPI states to max (%d)\n",
1126 ACPI_PROCESSOR_MAX_POWER);
1127 pr_warn("Please increase ACPI_PROCESSOR_MAX_POWER if needed.\n");
1128 break;
1129 }
1130
1131 flpi = &pr->power.lpi_states[flat_state_cnt];
1132
1133 if (!prev_level) { /* leaf/processor node */
1134 memcpy(flpi, t, sizeof(*t));
1135 stash_composite_state(curr_level, flpi);
1136 flat_state_cnt++;
1137 continue;
1138 }
1139
1140 for (i = 0; i < prev_level->composite_states_size; i++) {
1141 p = prev_level->composite_states[i];
1142 if (t->index <= p->enable_parent_state &&
1143 combine_lpi_states(p, t, flpi)) {
1144 stash_composite_state(curr_level, flpi);
1145 flat_state_cnt++;
1146 flpi++;
1147 }
1148 }
1149 }
1150
1151 kfree(curr_level->entries);
1152 return 0;
1153}
1154
1155static int acpi_processor_get_lpi_info(struct acpi_processor *pr)
1156{
1157 int ret, i;
1158 acpi_status status;
1159 acpi_handle handle = pr->handle, pr_ahandle;
1160 struct acpi_device *d = NULL;
1161 struct acpi_lpi_states_array info[2], *tmp, *prev, *curr;
1162
1163 if (!osc_pc_lpi_support_confirmed)
1164 return -EOPNOTSUPP;
1165
1166 if (!acpi_has_method(handle, "_LPI"))
1167 return -EINVAL;
1168
1169 flat_state_cnt = 0;
1170 prev = &info[0];
1171 curr = &info[1];
1172 handle = pr->handle;
1173 ret = acpi_processor_evaluate_lpi(handle, prev);
1174 if (ret)
1175 return ret;
1176 flatten_lpi_states(pr, prev, NULL);
1177
1178 status = acpi_get_parent(handle, &pr_ahandle);
1179 while (ACPI_SUCCESS(status)) {
1180 acpi_bus_get_device(pr_ahandle, &d);
1181 handle = pr_ahandle;
1182
1183 if (strcmp(acpi_device_hid(d), ACPI_PROCESSOR_CONTAINER_HID))
1184 break;
1185
1186 /* can be optional ? */
1187 if (!acpi_has_method(handle, "_LPI"))
1188 break;
1189
1190 ret = acpi_processor_evaluate_lpi(handle, curr);
1191 if (ret)
1192 break;
1193
1194 /* flatten all the LPI states in this level of hierarchy */
1195 flatten_lpi_states(pr, curr, prev);
1196
1197 tmp = prev, prev = curr, curr = tmp;
1198
1199 status = acpi_get_parent(handle, &pr_ahandle);
1200 }
1201
1202 pr->power.count = flat_state_cnt;
1203 /* reset the index after flattening */
1204 for (i = 0; i < pr->power.count; i++)
1205 pr->power.lpi_states[i].index = i;
1206
1207 /* Tell driver that _LPI is supported. */
1208 pr->flags.has_lpi = 1;
1209 pr->flags.power = 1;
1210
1211 return 0;
1212}
1213
1214int __weak acpi_processor_ffh_lpi_probe(unsigned int cpu)
1215{
1216 return -ENODEV;
1217}
1218
1219int __weak acpi_processor_ffh_lpi_enter(struct acpi_lpi_state *lpi)
1220{
1221 return -ENODEV;
1222}
1223
1224/**
1225 * acpi_idle_lpi_enter - enters an ACPI any LPI state
1226 * @dev: the target CPU
1227 * @drv: cpuidle driver containing cpuidle state info
1228 * @index: index of target state
1229 *
1230 * Return: 0 for success or negative value for error
1231 */
1232static int acpi_idle_lpi_enter(struct cpuidle_device *dev,
1233 struct cpuidle_driver *drv, int index)
1234{
1235 struct acpi_processor *pr;
1236 struct acpi_lpi_state *lpi;
1237
1238 pr = __this_cpu_read(processors);
1239
1240 if (unlikely(!pr))
1241 return -EINVAL;
1242
1243 lpi = &pr->power.lpi_states[index];
1244 if (lpi->entry_method == ACPI_CSTATE_FFH)
1245 return acpi_processor_ffh_lpi_enter(lpi);
1246
1247 return -EINVAL;
1248}
1249
1250static int acpi_processor_setup_lpi_states(struct acpi_processor *pr)
1251{
1252 int i;
1253 struct acpi_lpi_state *lpi;
1254 struct cpuidle_state *state;
1255 struct cpuidle_driver *drv = &acpi_idle_driver;
1256
1257 if (!pr->flags.has_lpi)
1258 return -EOPNOTSUPP;
1259
1260 for (i = 0; i < pr->power.count && i < CPUIDLE_STATE_MAX; i++) {
1261 lpi = &pr->power.lpi_states[i];
1262
1263 state = &drv->states[i];
1264 snprintf(state->name, CPUIDLE_NAME_LEN, "LPI-%d", i);
1265 strlcpy(state->desc, lpi->desc, CPUIDLE_DESC_LEN);
1266 state->exit_latency = lpi->wake_latency;
1267 state->target_residency = lpi->min_residency;
1268 if (lpi->arch_flags)
1269 state->flags |= CPUIDLE_FLAG_TIMER_STOP;
1270 state->enter = acpi_idle_lpi_enter;
1271 drv->safe_state_index = i;
1272 }
1273
1274 drv->state_count = i;
1275
1276 return 0;
1277}
1278
1279/**
1280 * acpi_processor_setup_cpuidle_states- prepares and configures cpuidle
1281 * global state data i.e. idle routines
1282 *
1283 * @pr: the ACPI processor
1284 */
1285static int acpi_processor_setup_cpuidle_states(struct acpi_processor *pr)
1286{
1287 int i;
1288 struct cpuidle_driver *drv = &acpi_idle_driver;
1289
1290 if (!pr->flags.power_setup_done || !pr->flags.power)
1291 return -EINVAL;
1292
1293 drv->safe_state_index = -1;
1294 for (i = CPUIDLE_DRIVER_STATE_START; i < CPUIDLE_STATE_MAX; i++) {
1295 drv->states[i].name[0] = '\0';
1296 drv->states[i].desc[0] = '\0';
1297 }
1298
1299 if (pr->flags.has_lpi)
1300 return acpi_processor_setup_lpi_states(pr);
1301
1302 return acpi_processor_setup_cstates(pr);
1303}
1304
1305/**
1306 * acpi_processor_setup_cpuidle_dev - prepares and configures CPUIDLE
1307 * device i.e. per-cpu data
1308 *
1309 * @pr: the ACPI processor
1310 * @dev : the cpuidle device
1311 */
1312static int acpi_processor_setup_cpuidle_dev(struct acpi_processor *pr,
1313 struct cpuidle_device *dev)
1314{
1315 if (!pr->flags.power_setup_done || !pr->flags.power || !dev)
1316 return -EINVAL;
1317
1318 dev->cpu = pr->id;
1319 if (pr->flags.has_lpi)
1320 return acpi_processor_ffh_lpi_probe(pr->id);
1321
1322 return acpi_processor_setup_cpuidle_cx(pr, dev);
1323}
1324
1325static int acpi_processor_get_power_info(struct acpi_processor *pr)
1326{
1327 int ret;
1328
1329 ret = acpi_processor_get_lpi_info(pr);
1330 if (ret)
1331 ret = acpi_processor_get_cstate_info(pr);
1332
1333 return ret;
1334}
1335
1336int acpi_processor_hotplug(struct acpi_processor *pr)
1337{
1338 int ret = 0;
1339 struct cpuidle_device *dev;
1340
1341 if (disabled_by_idle_boot_param())
1342 return 0;
1343
1344 if (!pr->flags.power_setup_done)
1345 return -ENODEV;
1346
1347 dev = per_cpu(acpi_cpuidle_device, pr->id);
1348 cpuidle_pause_and_lock();
1349 cpuidle_disable_device(dev);
1350 ret = acpi_processor_get_power_info(pr);
1351 if (!ret && pr->flags.power) {
1352 acpi_processor_setup_cpuidle_dev(pr, dev);
1353 ret = cpuidle_enable_device(dev);
1354 }
1355 cpuidle_resume_and_unlock();
1356
1357 return ret;
1358}
1359
1360int acpi_processor_power_state_has_changed(struct acpi_processor *pr)
1361{
1362 int cpu;
1363 struct acpi_processor *_pr;
1364 struct cpuidle_device *dev;
1365
1366 if (disabled_by_idle_boot_param())
1367 return 0;
1368
1369 if (!pr->flags.power_setup_done)
1370 return -ENODEV;
1371
1372 /*
1373 * FIXME: Design the ACPI notification to make it once per
1374 * system instead of once per-cpu. This condition is a hack
1375 * to make the code that updates C-States be called once.
1376 */
1377
1378 if (pr->id == 0 && cpuidle_get_driver() == &acpi_idle_driver) {
1379
1380 /* Protect against cpu-hotplug */
1381 get_online_cpus();
1382 cpuidle_pause_and_lock();
1383
1384 /* Disable all cpuidle devices */
1385 for_each_online_cpu(cpu) {
1386 _pr = per_cpu(processors, cpu);
1387 if (!_pr || !_pr->flags.power_setup_done)
1388 continue;
1389 dev = per_cpu(acpi_cpuidle_device, cpu);
1390 cpuidle_disable_device(dev);
1391 }
1392
1393 /* Populate Updated C-state information */
1394 acpi_processor_get_power_info(pr);
1395 acpi_processor_setup_cpuidle_states(pr);
1396
1397 /* Enable all cpuidle devices */
1398 for_each_online_cpu(cpu) {
1399 _pr = per_cpu(processors, cpu);
1400 if (!_pr || !_pr->flags.power_setup_done)
1401 continue;
1402 acpi_processor_get_power_info(_pr);
1403 if (_pr->flags.power) {
1404 dev = per_cpu(acpi_cpuidle_device, cpu);
1405 acpi_processor_setup_cpuidle_dev(_pr, dev);
1406 cpuidle_enable_device(dev);
1407 }
1408 }
1409 cpuidle_resume_and_unlock();
1410 put_online_cpus();
1411 }
1412
1413 return 0;
1414}
1415
1416static int acpi_processor_registered;
1417
1418int acpi_processor_power_init(struct acpi_processor *pr)
1419{
1420 int retval;
1421 struct cpuidle_device *dev;
1422
1423 if (disabled_by_idle_boot_param())
1424 return 0;
1425
1426 acpi_processor_cstate_first_run_checks();
1427
1428 if (!acpi_processor_get_power_info(pr))
1429 pr->flags.power_setup_done = 1;
1430
1431 /*
1432 * Install the idle handler if processor power management is supported.
1433 * Note that we use previously set idle handler will be used on
1434 * platforms that only support C1.
1435 */
1436 if (pr->flags.power) {
1437 /* Register acpi_idle_driver if not already registered */
1438 if (!acpi_processor_registered) {
1439 acpi_processor_setup_cpuidle_states(pr);
1440 retval = cpuidle_register_driver(&acpi_idle_driver);
1441 if (retval)
1442 return retval;
1443 pr_debug("%s registered with cpuidle\n",
1444 acpi_idle_driver.name);
1445 }
1446
1447 dev = kzalloc(sizeof(*dev), GFP_KERNEL);
1448 if (!dev)
1449 return -ENOMEM;
1450 per_cpu(acpi_cpuidle_device, pr->id) = dev;
1451
1452 acpi_processor_setup_cpuidle_dev(pr, dev);
1453
1454 /* Register per-cpu cpuidle_device. Cpuidle driver
1455 * must already be registered before registering device
1456 */
1457 retval = cpuidle_register_device(dev);
1458 if (retval) {
1459 if (acpi_processor_registered == 0)
1460 cpuidle_unregister_driver(&acpi_idle_driver);
1461 return retval;
1462 }
1463 acpi_processor_registered++;
1464 }
1465 return 0;
1466}
1467
1468int acpi_processor_power_exit(struct acpi_processor *pr)
1469{
1470 struct cpuidle_device *dev = per_cpu(acpi_cpuidle_device, pr->id);
1471
1472 if (disabled_by_idle_boot_param())
1473 return 0;
1474
1475 if (pr->flags.power) {
1476 cpuidle_unregister_device(dev);
1477 acpi_processor_registered--;
1478 if (acpi_processor_registered == 0)
1479 cpuidle_unregister_driver(&acpi_idle_driver);
1480 }
1481
1482 pr->flags.power_setup_done = 0;
1483 return 0;
1484}