1// SPDX-License-Identifier: GPL-2.0-only
2/*
3 * CPPC (Collaborative Processor Performance Control) methods used by CPUfreq drivers.
4 *
5 * (C) Copyright 2014, 2015 Linaro Ltd.
6 * Author: Ashwin Chaugule <ashwin.chaugule@linaro.org>
7 *
8 * CPPC describes a few methods for controlling CPU performance using
9 * information from a per CPU table called CPC. This table is described in
10 * the ACPI v5.0+ specification. The table consists of a list of
11 * registers which may be memory mapped or hardware registers and also may
12 * include some static integer values.
13 *
14 * CPU performance is on an abstract continuous scale as against a discretized
15 * P-state scale which is tied to CPU frequency only. In brief, the basic
16 * operation involves:
17 *
18 * - OS makes a CPU performance request. (Can provide min and max bounds)
19 *
20 * - Platform (such as BMC) is free to optimize request within requested bounds
21 *   depending on power/thermal budgets etc.
22 *
23 * - Platform conveys its decision back to OS
24 *
25 * The communication between OS and platform occurs through another medium
26 * called (PCC) Platform Communication Channel. This is a generic mailbox like
27 * mechanism which includes doorbell semantics to indicate register updates.
28 * See drivers/mailbox/pcc.c for details on PCC.
29 *
30 * Finer details about the PCC and CPPC spec are available in the ACPI v5.1 and
31 * above specifications.
32 */
33
34#define pr_fmt(fmt)	"ACPI CPPC: " fmt
35
36#include <linux/cpufreq.h>
37#include <linux/delay.h>
38#include <linux/iopoll.h>
39#include <linux/ktime.h>
40#include <linux/rwsem.h>
41#include <linux/wait.h>
42
43#include <acpi/cppc_acpi.h>
44
45struct cppc_pcc_data {
46	struct mbox_chan *pcc_channel;
47	void __iomem *pcc_comm_addr;
48	bool pcc_channel_acquired;
49	unsigned int deadline_us;
50	unsigned int pcc_mpar, pcc_mrtt, pcc_nominal;
51
52	bool pending_pcc_write_cmd;	/* Any pending/batched PCC write cmds? */
53	bool platform_owns_pcc;		/* Ownership of PCC subspace */
54	unsigned int pcc_write_cnt;	/* Running count of PCC write commands */
55
56	/*
57	 * Lock to provide controlled access to the PCC channel.
58	 *
59	 * For performance critical usecases(currently cppc_set_perf)
60	 *	We need to take read_lock and check if channel belongs to OSPM
61	 * before reading or writing to PCC subspace
62	 *	We need to take write_lock before transferring the channel
63	 * ownership to the platform via a Doorbell
64	 *	This allows us to batch a number of CPPC requests if they happen
65	 * to originate in about the same time
66	 *
67	 * For non-performance critical usecases(init)
68	 *	Take write_lock for all purposes which gives exclusive access
69	 */
70	struct rw_semaphore pcc_lock;
71
72	/* Wait queue for CPUs whose requests were batched */
73	wait_queue_head_t pcc_write_wait_q;
74	ktime_t last_cmd_cmpl_time;
75	ktime_t last_mpar_reset;
76	int mpar_count;
77	int refcount;
78};
79
80/* Array to represent the PCC channel per subspace ID */
81static struct cppc_pcc_data *pcc_data[MAX_PCC_SUBSPACES];
82/* The cpu_pcc_subspace_idx contains per CPU subspace ID */
83static DEFINE_PER_CPU(int, cpu_pcc_subspace_idx);
84
85/*
86 * The cpc_desc structure contains the ACPI register details
87 * as described in the per CPU _CPC tables. The details
88 * include the type of register (e.g. PCC, System IO, FFH etc.)
89 * and destination addresses which lets us READ/WRITE CPU performance
90 * information using the appropriate I/O methods.
91 */
92static DEFINE_PER_CPU(struct cpc_desc *, cpc_desc_ptr);
93
94/* pcc mapped address + header size + offset within PCC subspace */
95#define GET_PCC_VADDR(offs, pcc_ss_id) (pcc_data[pcc_ss_id]->pcc_comm_addr + \
96						0x8 + (offs))
97
98/* Check if a CPC register is in PCC */
99#define CPC_IN_PCC(cpc) ((cpc)->type == ACPI_TYPE_BUFFER &&		\
100				(cpc)->cpc_entry.reg.space_id ==	\
101				ACPI_ADR_SPACE_PLATFORM_COMM)
102
103/* Evalutes to True if reg is a NULL register descriptor */
104#define IS_NULL_REG(reg) ((reg)->space_id ==  ACPI_ADR_SPACE_SYSTEM_MEMORY && \
105				(reg)->address == 0 &&			\
106				(reg)->bit_width == 0 &&		\
107				(reg)->bit_offset == 0 &&		\
108				(reg)->access_width == 0)
109
110/* Evalutes to True if an optional cpc field is supported */
111#define CPC_SUPPORTED(cpc) ((cpc)->type == ACPI_TYPE_INTEGER ?		\
112				!!(cpc)->cpc_entry.int_value :		\
113				!IS_NULL_REG(&(cpc)->cpc_entry.reg))
114/*
115 * Arbitrary Retries in case the remote processor is slow to respond
116 * to PCC commands. Keeping it high enough to cover emulators where
117 * the processors run painfully slow.
118 */
119#define NUM_RETRIES 500ULL
120
121#define define_one_cppc_ro(_name)		\
122static struct kobj_attribute _name =		\
123__ATTR(_name, 0444, show_##_name, NULL)
124
125#define to_cpc_desc(a) container_of(a, struct cpc_desc, kobj)
126
127#define show_cppc_data(access_fn, struct_name, member_name)		\
128	static ssize_t show_##member_name(struct kobject *kobj,		\
129				struct kobj_attribute *attr, char *buf)	\
130	{								\
131		struct cpc_desc *cpc_ptr = to_cpc_desc(kobj);		\
132		struct struct_name st_name = {0};			\
133		int ret;						\
134									\
135		ret = access_fn(cpc_ptr->cpu_id, &st_name);		\
136		if (ret)						\
137			return ret;					\
138									\
139		return scnprintf(buf, PAGE_SIZE, "%llu\n",		\
140				(u64)st_name.member_name);		\
141	}								\
142	define_one_cppc_ro(member_name)
143
144show_cppc_data(cppc_get_perf_caps, cppc_perf_caps, highest_perf);
145show_cppc_data(cppc_get_perf_caps, cppc_perf_caps, lowest_perf);
146show_cppc_data(cppc_get_perf_caps, cppc_perf_caps, nominal_perf);
147show_cppc_data(cppc_get_perf_caps, cppc_perf_caps, lowest_nonlinear_perf);
148show_cppc_data(cppc_get_perf_caps, cppc_perf_caps, lowest_freq);
149show_cppc_data(cppc_get_perf_caps, cppc_perf_caps, nominal_freq);
150
151show_cppc_data(cppc_get_perf_ctrs, cppc_perf_fb_ctrs, reference_perf);
152show_cppc_data(cppc_get_perf_ctrs, cppc_perf_fb_ctrs, wraparound_time);
153
154static ssize_t show_feedback_ctrs(struct kobject *kobj,
155		struct kobj_attribute *attr, char *buf)
156{
157	struct cpc_desc *cpc_ptr = to_cpc_desc(kobj);
158	struct cppc_perf_fb_ctrs fb_ctrs = {0};
159	int ret;
160
161	ret = cppc_get_perf_ctrs(cpc_ptr->cpu_id, &fb_ctrs);
162	if (ret)
163		return ret;
164
165	return scnprintf(buf, PAGE_SIZE, "ref:%llu del:%llu\n",
166			fb_ctrs.reference, fb_ctrs.delivered);
167}
168define_one_cppc_ro(feedback_ctrs);
169
170static struct attribute *cppc_attrs[] = {
171	&feedback_ctrs.attr,
172	&reference_perf.attr,
173	&wraparound_time.attr,
174	&highest_perf.attr,
175	&lowest_perf.attr,
176	&lowest_nonlinear_perf.attr,
177	&nominal_perf.attr,
178	&nominal_freq.attr,
179	&lowest_freq.attr,
180	NULL
181};
182
183static struct kobj_type cppc_ktype = {
184	.sysfs_ops = &kobj_sysfs_ops,
185	.default_attrs = cppc_attrs,
186};
187
188static int check_pcc_chan(int pcc_ss_id, bool chk_err_bit)
189{
190	int ret, status;
191	struct cppc_pcc_data *pcc_ss_data = pcc_data[pcc_ss_id];
192	struct acpi_pcct_shared_memory __iomem *generic_comm_base =
193		pcc_ss_data->pcc_comm_addr;
194
195	if (!pcc_ss_data->platform_owns_pcc)
196		return 0;
197
198	/*
199	 * Poll PCC status register every 3us(delay_us) for maximum of
200	 * deadline_us(timeout_us) until PCC command complete bit is set(cond)
201	 */
202	ret = readw_relaxed_poll_timeout(&generic_comm_base->status, status,
203					status & PCC_CMD_COMPLETE_MASK, 3,
204					pcc_ss_data->deadline_us);
205
206	if (likely(!ret)) {
207		pcc_ss_data->platform_owns_pcc = false;
208		if (chk_err_bit && (status & PCC_ERROR_MASK))
209			ret = -EIO;
210	}
211
212	if (unlikely(ret))
213		pr_err("PCC check channel failed for ss: %d. ret=%d\n",
214		       pcc_ss_id, ret);
215
216	return ret;
217}
218
219/*
220 * This function transfers the ownership of the PCC to the platform
221 * So it must be called while holding write_lock(pcc_lock)
222 */
223static int send_pcc_cmd(int pcc_ss_id, u16 cmd)
224{
225	int ret = -EIO, i;
226	struct cppc_pcc_data *pcc_ss_data = pcc_data[pcc_ss_id];
227	struct acpi_pcct_shared_memory *generic_comm_base =
228		(struct acpi_pcct_shared_memory *)pcc_ss_data->pcc_comm_addr;
229	unsigned int time_delta;
230
231	/*
232	 * For CMD_WRITE we know for a fact the caller should have checked
233	 * the channel before writing to PCC space
234	 */
235	if (cmd == CMD_READ) {
236		/*
237		 * If there are pending cpc_writes, then we stole the channel
238		 * before write completion, so first send a WRITE command to
239		 * platform
240		 */
241		if (pcc_ss_data->pending_pcc_write_cmd)
242			send_pcc_cmd(pcc_ss_id, CMD_WRITE);
243
244		ret = check_pcc_chan(pcc_ss_id, false);
245		if (ret)
246			goto end;
247	} else /* CMD_WRITE */
248		pcc_ss_data->pending_pcc_write_cmd = FALSE;
249
250	/*
251	 * Handle the Minimum Request Turnaround Time(MRTT)
252	 * "The minimum amount of time that OSPM must wait after the completion
253	 * of a command before issuing the next command, in microseconds"
254	 */
255	if (pcc_ss_data->pcc_mrtt) {
256		time_delta = ktime_us_delta(ktime_get(),
257					    pcc_ss_data->last_cmd_cmpl_time);
258		if (pcc_ss_data->pcc_mrtt > time_delta)
259			udelay(pcc_ss_data->pcc_mrtt - time_delta);
260	}
261
262	/*
263	 * Handle the non-zero Maximum Periodic Access Rate(MPAR)
264	 * "The maximum number of periodic requests that the subspace channel can
265	 * support, reported in commands per minute. 0 indicates no limitation."
266	 *
267	 * This parameter should be ideally zero or large enough so that it can
268	 * handle maximum number of requests that all the cores in the system can
269	 * collectively generate. If it is not, we will follow the spec and just
270	 * not send the request to the platform after hitting the MPAR limit in
271	 * any 60s window
272	 */
273	if (pcc_ss_data->pcc_mpar) {
274		if (pcc_ss_data->mpar_count == 0) {
275			time_delta = ktime_ms_delta(ktime_get(),
276						    pcc_ss_data->last_mpar_reset);
277			if ((time_delta < 60 * MSEC_PER_SEC) && pcc_ss_data->last_mpar_reset) {
278				pr_debug("PCC cmd for subspace %d not sent due to MPAR limit",
279					 pcc_ss_id);
280				ret = -EIO;
281				goto end;
282			}
283			pcc_ss_data->last_mpar_reset = ktime_get();
284			pcc_ss_data->mpar_count = pcc_ss_data->pcc_mpar;
285		}
286		pcc_ss_data->mpar_count--;
287	}
288
289	/* Write to the shared comm region. */
290	writew_relaxed(cmd, &generic_comm_base->command);
291
292	/* Flip CMD COMPLETE bit */
293	writew_relaxed(0, &generic_comm_base->status);
294
295	pcc_ss_data->platform_owns_pcc = true;
296
297	/* Ring doorbell */
298	ret = mbox_send_message(pcc_ss_data->pcc_channel, &cmd);
299	if (ret < 0) {
300		pr_err("Err sending PCC mbox message. ss: %d cmd:%d, ret:%d\n",
301		       pcc_ss_id, cmd, ret);
302		goto end;
303	}
304
305	/* wait for completion and check for PCC errro bit */
306	ret = check_pcc_chan(pcc_ss_id, true);
307
308	if (pcc_ss_data->pcc_mrtt)
309		pcc_ss_data->last_cmd_cmpl_time = ktime_get();
310
311	if (pcc_ss_data->pcc_channel->mbox->txdone_irq)
312		mbox_chan_txdone(pcc_ss_data->pcc_channel, ret);
313	else
314		mbox_client_txdone(pcc_ss_data->pcc_channel, ret);
315
316end:
317	if (cmd == CMD_WRITE) {
318		if (unlikely(ret)) {
319			for_each_possible_cpu(i) {
320				struct cpc_desc *desc = per_cpu(cpc_desc_ptr, i);
321				if (!desc)
322					continue;
323
324				if (desc->write_cmd_id == pcc_ss_data->pcc_write_cnt)
325					desc->write_cmd_status = ret;
326			}
327		}
328		pcc_ss_data->pcc_write_cnt++;
329		wake_up_all(&pcc_ss_data->pcc_write_wait_q);
330	}
331
332	return ret;
333}
334
335static void cppc_chan_tx_done(struct mbox_client *cl, void *msg, int ret)
336{
337	if (ret < 0)
338		pr_debug("TX did not complete: CMD sent:%x, ret:%d\n",
339				*(u16 *)msg, ret);
340	else
341		pr_debug("TX completed. CMD sent:%x, ret:%d\n",
342				*(u16 *)msg, ret);
343}
344
345static struct mbox_client cppc_mbox_cl = {
346	.tx_done = cppc_chan_tx_done,
347	.knows_txdone = true,
348};
349
350static int acpi_get_psd(struct cpc_desc *cpc_ptr, acpi_handle handle)
351{
352	int result = -EFAULT;
353	acpi_status status = AE_OK;
354	struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL};
355	struct acpi_buffer format = {sizeof("NNNNN"), "NNNNN"};
356	struct acpi_buffer state = {0, NULL};
357	union acpi_object  *psd = NULL;
358	struct acpi_psd_package *pdomain;
359
360	status = acpi_evaluate_object_typed(handle, "_PSD", NULL,
361					    &buffer, ACPI_TYPE_PACKAGE);
362	if (status == AE_NOT_FOUND)	/* _PSD is optional */
363		return 0;
364	if (ACPI_FAILURE(status))
365		return -ENODEV;
366
367	psd = buffer.pointer;
368	if (!psd || psd->package.count != 1) {
369		pr_debug("Invalid _PSD data\n");
370		goto end;
371	}
372
373	pdomain = &(cpc_ptr->domain_info);
374
375	state.length = sizeof(struct acpi_psd_package);
376	state.pointer = pdomain;
377
378	status = acpi_extract_package(&(psd->package.elements[0]),
379		&format, &state);
380	if (ACPI_FAILURE(status)) {
381		pr_debug("Invalid _PSD data for CPU:%d\n", cpc_ptr->cpu_id);
382		goto end;
383	}
384
385	if (pdomain->num_entries != ACPI_PSD_REV0_ENTRIES) {
386		pr_debug("Unknown _PSD:num_entries for CPU:%d\n", cpc_ptr->cpu_id);
387		goto end;
388	}
389
390	if (pdomain->revision != ACPI_PSD_REV0_REVISION) {
391		pr_debug("Unknown _PSD:revision for CPU: %d\n", cpc_ptr->cpu_id);
392		goto end;
393	}
394
395	if (pdomain->coord_type != DOMAIN_COORD_TYPE_SW_ALL &&
396	    pdomain->coord_type != DOMAIN_COORD_TYPE_SW_ANY &&
397	    pdomain->coord_type != DOMAIN_COORD_TYPE_HW_ALL) {
398		pr_debug("Invalid _PSD:coord_type for CPU:%d\n", cpc_ptr->cpu_id);
399		goto end;
400	}
401
402	result = 0;
403end:
404	kfree(buffer.pointer);
405	return result;
406}
407
408/**
409 * acpi_get_psd_map - Map the CPUs in a common freq domain.
410 * @all_cpu_data: Ptrs to CPU specific CPPC data including PSD info.
411 *
412 *	Return: 0 for success or negative value for err.
413 */
414int acpi_get_psd_map(struct cppc_cpudata **all_cpu_data)
415{
416	int count_target;
417	int retval = 0;
418	unsigned int i, j;
419	cpumask_var_t covered_cpus;
420	struct cppc_cpudata *pr, *match_pr;
421	struct acpi_psd_package *pdomain;
422	struct acpi_psd_package *match_pdomain;
423	struct cpc_desc *cpc_ptr, *match_cpc_ptr;
424
425	if (!zalloc_cpumask_var(&covered_cpus, GFP_KERNEL))
426		return -ENOMEM;
427
428	/*
429	 * Now that we have _PSD data from all CPUs, let's setup P-state
430	 * domain info.
431	 */
432	for_each_possible_cpu(i) {
433		if (cpumask_test_cpu(i, covered_cpus))
434			continue;
435
436		pr = all_cpu_data[i];
437		cpc_ptr = per_cpu(cpc_desc_ptr, i);
438		if (!cpc_ptr) {
439			retval = -EFAULT;
440			goto err_ret;
441		}
442
443		pdomain = &(cpc_ptr->domain_info);
444		cpumask_set_cpu(i, pr->shared_cpu_map);
445		cpumask_set_cpu(i, covered_cpus);
446		if (pdomain->num_processors <= 1)
447			continue;
448
449		/* Validate the Domain info */
450		count_target = pdomain->num_processors;
451		if (pdomain->coord_type == DOMAIN_COORD_TYPE_SW_ALL)
452			pr->shared_type = CPUFREQ_SHARED_TYPE_ALL;
453		else if (pdomain->coord_type == DOMAIN_COORD_TYPE_HW_ALL)
454			pr->shared_type = CPUFREQ_SHARED_TYPE_HW;
455		else if (pdomain->coord_type == DOMAIN_COORD_TYPE_SW_ANY)
456			pr->shared_type = CPUFREQ_SHARED_TYPE_ANY;
457
458		for_each_possible_cpu(j) {
459			if (i == j)
460				continue;
461
462			match_cpc_ptr = per_cpu(cpc_desc_ptr, j);
463			if (!match_cpc_ptr) {
464				retval = -EFAULT;
465				goto err_ret;
466			}
467
468			match_pdomain = &(match_cpc_ptr->domain_info);
469			if (match_pdomain->domain != pdomain->domain)
470				continue;
471
472			/* Here i and j are in the same domain */
473			if (match_pdomain->num_processors != count_target) {
474				retval = -EFAULT;
475				goto err_ret;
476			}
477
478			if (pdomain->coord_type != match_pdomain->coord_type) {
479				retval = -EFAULT;
480				goto err_ret;
481			}
482
483			cpumask_set_cpu(j, covered_cpus);
484			cpumask_set_cpu(j, pr->shared_cpu_map);
485		}
486
487		for_each_cpu(j, pr->shared_cpu_map) {
488			if (i == j)
489				continue;
490
491			match_pr = all_cpu_data[j];
492			match_pr->shared_type = pr->shared_type;
493			cpumask_copy(match_pr->shared_cpu_map,
494				     pr->shared_cpu_map);
495		}
496	}
497	goto out;
498
499err_ret:
500	for_each_possible_cpu(i) {
501		pr = all_cpu_data[i];
502
503		/* Assume no coordination on any error parsing domain info */
504		cpumask_clear(pr->shared_cpu_map);
505		cpumask_set_cpu(i, pr->shared_cpu_map);
506		pr->shared_type = CPUFREQ_SHARED_TYPE_ALL;
507	}
508out:
509	free_cpumask_var(covered_cpus);
510	return retval;
511}
512EXPORT_SYMBOL_GPL(acpi_get_psd_map);
513
514static int register_pcc_channel(int pcc_ss_idx)
515{
516	struct acpi_pcct_hw_reduced *cppc_ss;
517	u64 usecs_lat;
518
519	if (pcc_ss_idx >= 0) {
520		pcc_data[pcc_ss_idx]->pcc_channel =
521			pcc_mbox_request_channel(&cppc_mbox_cl,	pcc_ss_idx);
522
523		if (IS_ERR(pcc_data[pcc_ss_idx]->pcc_channel)) {
524			pr_err("Failed to find PCC channel for subspace %d\n",
525			       pcc_ss_idx);
526			return -ENODEV;
527		}
528
529		/*
530		 * The PCC mailbox controller driver should
531		 * have parsed the PCCT (global table of all
532		 * PCC channels) and stored pointers to the
533		 * subspace communication region in con_priv.
534		 */
535		cppc_ss = (pcc_data[pcc_ss_idx]->pcc_channel)->con_priv;
536
537		if (!cppc_ss) {
538			pr_err("No PCC subspace found for %d CPPC\n",
539			       pcc_ss_idx);
540			return -ENODEV;
541		}
542
543		/*
544		 * cppc_ss->latency is just a Nominal value. In reality
545		 * the remote processor could be much slower to reply.
546		 * So add an arbitrary amount of wait on top of Nominal.
547		 */
548		usecs_lat = NUM_RETRIES * cppc_ss->latency;
549		pcc_data[pcc_ss_idx]->deadline_us = usecs_lat;
550		pcc_data[pcc_ss_idx]->pcc_mrtt = cppc_ss->min_turnaround_time;
551		pcc_data[pcc_ss_idx]->pcc_mpar = cppc_ss->max_access_rate;
552		pcc_data[pcc_ss_idx]->pcc_nominal = cppc_ss->latency;
553
554		pcc_data[pcc_ss_idx]->pcc_comm_addr =
555			acpi_os_ioremap(cppc_ss->base_address, cppc_ss->length);
556		if (!pcc_data[pcc_ss_idx]->pcc_comm_addr) {
557			pr_err("Failed to ioremap PCC comm region mem for %d\n",
558			       pcc_ss_idx);
559			return -ENOMEM;
560		}
561
562		/* Set flag so that we don't come here for each CPU. */
563		pcc_data[pcc_ss_idx]->pcc_channel_acquired = true;
564	}
565
566	return 0;
567}
568
569/**
570 * cpc_ffh_supported() - check if FFH reading supported
571 *
572 * Check if the architecture has support for functional fixed hardware
573 * read/write capability.
574 *
575 * Return: true for supported, false for not supported
576 */
577bool __weak cpc_ffh_supported(void)
578{
579	return false;
580}
581
582/**
583 * pcc_data_alloc() - Allocate the pcc_data memory for pcc subspace
584 *
585 * Check and allocate the cppc_pcc_data memory.
586 * In some processor configurations it is possible that same subspace
587 * is shared between multiple CPUs. This is seen especially in CPUs
588 * with hardware multi-threading support.
589 *
590 * Return: 0 for success, errno for failure
591 */
592static int pcc_data_alloc(int pcc_ss_id)
593{
594	if (pcc_ss_id < 0 || pcc_ss_id >= MAX_PCC_SUBSPACES)
595		return -EINVAL;
596
597	if (pcc_data[pcc_ss_id]) {
598		pcc_data[pcc_ss_id]->refcount++;
599	} else {
600		pcc_data[pcc_ss_id] = kzalloc(sizeof(struct cppc_pcc_data),
601					      GFP_KERNEL);
602		if (!pcc_data[pcc_ss_id])
603			return -ENOMEM;
604		pcc_data[pcc_ss_id]->refcount++;
605	}
606
607	return 0;
608}
609
610/*
611 * An example CPC table looks like the following.
612 *
613 *	Name(_CPC, Package()
614 *			{
615 *			17,
616 *			NumEntries
617 *			1,
618 *			// Revision
619 *			ResourceTemplate(){Register(PCC, 32, 0, 0x120, 2)},
620 *			// Highest Performance
621 *			ResourceTemplate(){Register(PCC, 32, 0, 0x124, 2)},
622 *			// Nominal Performance
623 *			ResourceTemplate(){Register(PCC, 32, 0, 0x128, 2)},
624 *			// Lowest Nonlinear Performance
625 *			ResourceTemplate(){Register(PCC, 32, 0, 0x12C, 2)},
626 *			// Lowest Performance
627 *			ResourceTemplate(){Register(PCC, 32, 0, 0x130, 2)},
628 *			// Guaranteed Performance Register
629 *			ResourceTemplate(){Register(PCC, 32, 0, 0x110, 2)},
630 *			// Desired Performance Register
631 *			ResourceTemplate(){Register(SystemMemory, 0, 0, 0, 0)},
632 *			..
633 *			..
634 *			..
635 *
636 *		}
637 * Each Register() encodes how to access that specific register.
638 * e.g. a sample PCC entry has the following encoding:
639 *
640 *	Register (
641 *		PCC,
642 *		AddressSpaceKeyword
643 *		8,
644 *		//RegisterBitWidth
645 *		8,
646 *		//RegisterBitOffset
647 *		0x30,
648 *		//RegisterAddress
649 *		9
650 *		//AccessSize (subspace ID)
651 *		0
652 *		)
653 *	}
654 */
655
656/**
657 * acpi_cppc_processor_probe - Search for per CPU _CPC objects.
658 * @pr: Ptr to acpi_processor containing this CPU's logical ID.
659 *
660 *	Return: 0 for success or negative value for err.
661 */
662int acpi_cppc_processor_probe(struct acpi_processor *pr)
663{
664	struct acpi_buffer output = {ACPI_ALLOCATE_BUFFER, NULL};
665	union acpi_object *out_obj, *cpc_obj;
666	struct cpc_desc *cpc_ptr;
667	struct cpc_reg *gas_t;
668	struct device *cpu_dev;
669	acpi_handle handle = pr->handle;
670	unsigned int num_ent, i, cpc_rev;
671	int pcc_subspace_id = -1;
672	acpi_status status;
673	int ret = -EFAULT;
674
675	/* Parse the ACPI _CPC table for this CPU. */
676	status = acpi_evaluate_object_typed(handle, "_CPC", NULL, &output,
677			ACPI_TYPE_PACKAGE);
678	if (ACPI_FAILURE(status)) {
679		ret = -ENODEV;
680		goto out_buf_free;
681	}
682
683	out_obj = (union acpi_object *) output.pointer;
684
685	cpc_ptr = kzalloc(sizeof(struct cpc_desc), GFP_KERNEL);
686	if (!cpc_ptr) {
687		ret = -ENOMEM;
688		goto out_buf_free;
689	}
690
691	/* First entry is NumEntries. */
692	cpc_obj = &out_obj->package.elements[0];
693	if (cpc_obj->type == ACPI_TYPE_INTEGER)	{
694		num_ent = cpc_obj->integer.value;
695		if (num_ent <= 1) {
696			pr_debug("Unexpected _CPC NumEntries value (%d) for CPU:%d\n",
697				 num_ent, pr->id);
698			goto out_free;
699		}
700	} else {
701		pr_debug("Unexpected entry type(%d) for NumEntries\n",
702				cpc_obj->type);
703		goto out_free;
704	}
705
706	/* Second entry should be revision. */
707	cpc_obj = &out_obj->package.elements[1];
708	if (cpc_obj->type == ACPI_TYPE_INTEGER)	{
709		cpc_rev = cpc_obj->integer.value;
710	} else {
711		pr_debug("Unexpected entry type(%d) for Revision\n",
712				cpc_obj->type);
713		goto out_free;
714	}
715
716	if (cpc_rev < CPPC_V2_REV) {
717		pr_debug("Unsupported _CPC Revision (%d) for CPU:%d\n", cpc_rev,
718			 pr->id);
719		goto out_free;
720	}
721
722	/*
723	 * Disregard _CPC if the number of entries in the return pachage is not
724	 * as expected, but support future revisions being proper supersets of
725	 * the v3 and only causing more entries to be returned by _CPC.
726	 */
727	if ((cpc_rev == CPPC_V2_REV && num_ent != CPPC_V2_NUM_ENT) ||
728	    (cpc_rev == CPPC_V3_REV && num_ent != CPPC_V3_NUM_ENT) ||
729	    (cpc_rev > CPPC_V3_REV && num_ent <= CPPC_V3_NUM_ENT)) {
730		pr_debug("Unexpected number of _CPC return package entries (%d) for CPU:%d\n",
731			 num_ent, pr->id);
732		goto out_free;
733	}
734	if (cpc_rev > CPPC_V3_REV) {
735		num_ent = CPPC_V3_NUM_ENT;
736		cpc_rev = CPPC_V3_REV;
737	}
738
739	cpc_ptr->num_entries = num_ent;
740	cpc_ptr->version = cpc_rev;
741
742	/* Iterate through remaining entries in _CPC */
743	for (i = 2; i < num_ent; i++) {
744		cpc_obj = &out_obj->package.elements[i];
745
746		if (cpc_obj->type == ACPI_TYPE_INTEGER)	{
747			cpc_ptr->cpc_regs[i-2].type = ACPI_TYPE_INTEGER;
748			cpc_ptr->cpc_regs[i-2].cpc_entry.int_value = cpc_obj->integer.value;
749		} else if (cpc_obj->type == ACPI_TYPE_BUFFER) {
750			gas_t = (struct cpc_reg *)
751				cpc_obj->buffer.pointer;
752
753			/*
754			 * The PCC Subspace index is encoded inside
755			 * the CPC table entries. The same PCC index
756			 * will be used for all the PCC entries,
757			 * so extract it only once.
758			 */
759			if (gas_t->space_id == ACPI_ADR_SPACE_PLATFORM_COMM) {
760				if (pcc_subspace_id < 0) {
761					pcc_subspace_id = gas_t->access_width;
762					if (pcc_data_alloc(pcc_subspace_id))
763						goto out_free;
764				} else if (pcc_subspace_id != gas_t->access_width) {
765					pr_debug("Mismatched PCC ids.\n");
766					goto out_free;
767				}
768			} else if (gas_t->space_id == ACPI_ADR_SPACE_SYSTEM_MEMORY) {
769				if (gas_t->address) {
770					void __iomem *addr;
771
772					addr = ioremap(gas_t->address, gas_t->bit_width/8);
773					if (!addr)
774						goto out_free;
775					cpc_ptr->cpc_regs[i-2].sys_mem_vaddr = addr;
776				}
777			} else {
778				if (gas_t->space_id != ACPI_ADR_SPACE_FIXED_HARDWARE || !cpc_ffh_supported()) {
779					/* Support only PCC ,SYS MEM and FFH type regs */
780					pr_debug("Unsupported register type: %d\n", gas_t->space_id);
781					goto out_free;
782				}
783			}
784
785			cpc_ptr->cpc_regs[i-2].type = ACPI_TYPE_BUFFER;
786			memcpy(&cpc_ptr->cpc_regs[i-2].cpc_entry.reg, gas_t, sizeof(*gas_t));
787		} else {
788			pr_debug("Err in entry:%d in CPC table of CPU:%d \n", i, pr->id);
789			goto out_free;
790		}
791	}
792	per_cpu(cpu_pcc_subspace_idx, pr->id) = pcc_subspace_id;
793
794	/*
795	 * Initialize the remaining cpc_regs as unsupported.
796	 * Example: In case FW exposes CPPC v2, the below loop will initialize
797	 * LOWEST_FREQ and NOMINAL_FREQ regs as unsupported
798	 */
799	for (i = num_ent - 2; i < MAX_CPC_REG_ENT; i++) {
800		cpc_ptr->cpc_regs[i].type = ACPI_TYPE_INTEGER;
801		cpc_ptr->cpc_regs[i].cpc_entry.int_value = 0;
802	}
803
804
805	/* Store CPU Logical ID */
806	cpc_ptr->cpu_id = pr->id;
807
808	/* Parse PSD data for this CPU */
809	ret = acpi_get_psd(cpc_ptr, handle);
810	if (ret)
811		goto out_free;
812
813	/* Register PCC channel once for all PCC subspace ID. */
814	if (pcc_subspace_id >= 0 && !pcc_data[pcc_subspace_id]->pcc_channel_acquired) {
815		ret = register_pcc_channel(pcc_subspace_id);
816		if (ret)
817			goto out_free;
818
819		init_rwsem(&pcc_data[pcc_subspace_id]->pcc_lock);
820		init_waitqueue_head(&pcc_data[pcc_subspace_id]->pcc_write_wait_q);
821	}
822
823	/* Everything looks okay */
824	pr_debug("Parsed CPC struct for CPU: %d\n", pr->id);
825
826	/* Add per logical CPU nodes for reading its feedback counters. */
827	cpu_dev = get_cpu_device(pr->id);
828	if (!cpu_dev) {
829		ret = -EINVAL;
830		goto out_free;
831	}
832
833	/* Plug PSD data into this CPU's CPC descriptor. */
834	per_cpu(cpc_desc_ptr, pr->id) = cpc_ptr;
835
836	ret = kobject_init_and_add(&cpc_ptr->kobj, &cppc_ktype, &cpu_dev->kobj,
837			"acpi_cppc");
838	if (ret) {
839		per_cpu(cpc_desc_ptr, pr->id) = NULL;
840		kobject_put(&cpc_ptr->kobj);
841		goto out_free;
842	}
843
844	kfree(output.pointer);
845	return 0;
846
847out_free:
848	/* Free all the mapped sys mem areas for this CPU */
849	for (i = 2; i < cpc_ptr->num_entries; i++) {
850		void __iomem *addr = cpc_ptr->cpc_regs[i-2].sys_mem_vaddr;
851
852		if (addr)
853			iounmap(addr);
854	}
855	kfree(cpc_ptr);
856
857out_buf_free:
858	kfree(output.pointer);
859	return ret;
860}
861EXPORT_SYMBOL_GPL(acpi_cppc_processor_probe);
862
863/**
864 * acpi_cppc_processor_exit - Cleanup CPC structs.
865 * @pr: Ptr to acpi_processor containing this CPU's logical ID.
866 *
867 * Return: Void
868 */
869void acpi_cppc_processor_exit(struct acpi_processor *pr)
870{
871	struct cpc_desc *cpc_ptr;
872	unsigned int i;
873	void __iomem *addr;
874	int pcc_ss_id = per_cpu(cpu_pcc_subspace_idx, pr->id);
875
876	if (pcc_ss_id >=0 && pcc_data[pcc_ss_id]) {
877		if (pcc_data[pcc_ss_id]->pcc_channel_acquired) {
878			pcc_data[pcc_ss_id]->refcount--;
879			if (!pcc_data[pcc_ss_id]->refcount) {
880				pcc_mbox_free_channel(pcc_data[pcc_ss_id]->pcc_channel);
881				kfree(pcc_data[pcc_ss_id]);
882				pcc_data[pcc_ss_id] = NULL;
883			}
884		}
885	}
886
887	cpc_ptr = per_cpu(cpc_desc_ptr, pr->id);
888	if (!cpc_ptr)
889		return;
890
891	/* Free all the mapped sys mem areas for this CPU */
892	for (i = 2; i < cpc_ptr->num_entries; i++) {
893		addr = cpc_ptr->cpc_regs[i-2].sys_mem_vaddr;
894		if (addr)
895			iounmap(addr);
896	}
897
898	kobject_put(&cpc_ptr->kobj);
899	kfree(cpc_ptr);
900}
901EXPORT_SYMBOL_GPL(acpi_cppc_processor_exit);
902
903/**
904 * cpc_read_ffh() - Read FFH register
905 * @cpunum:	CPU number to read
906 * @reg:	cppc register information
907 * @val:	place holder for return value
908 *
909 * Read bit_width bits from a specified address and bit_offset
910 *
911 * Return: 0 for success and error code
912 */
913int __weak cpc_read_ffh(int cpunum, struct cpc_reg *reg, u64 *val)
914{
915	return -ENOTSUPP;
916}
917
918/**
919 * cpc_write_ffh() - Write FFH register
920 * @cpunum:	CPU number to write
921 * @reg:	cppc register information
922 * @val:	value to write
923 *
924 * Write value of bit_width bits to a specified address and bit_offset
925 *
926 * Return: 0 for success and error code
927 */
928int __weak cpc_write_ffh(int cpunum, struct cpc_reg *reg, u64 val)
929{
930	return -ENOTSUPP;
931}
932
933/*
934 * Since cpc_read and cpc_write are called while holding pcc_lock, it should be
935 * as fast as possible. We have already mapped the PCC subspace during init, so
936 * we can directly write to it.
937 */
938
939static int cpc_read(int cpu, struct cpc_register_resource *reg_res, u64 *val)
940{
941	int ret_val = 0;
942	void __iomem *vaddr = 0;
943	int pcc_ss_id = per_cpu(cpu_pcc_subspace_idx, cpu);
944	struct cpc_reg *reg = &reg_res->cpc_entry.reg;
945
946	if (reg_res->type == ACPI_TYPE_INTEGER) {
947		*val = reg_res->cpc_entry.int_value;
948		return ret_val;
949	}
950
951	*val = 0;
952	if (reg->space_id == ACPI_ADR_SPACE_PLATFORM_COMM && pcc_ss_id >= 0)
953		vaddr = GET_PCC_VADDR(reg->address, pcc_ss_id);
954	else if (reg->space_id == ACPI_ADR_SPACE_SYSTEM_MEMORY)
955		vaddr = reg_res->sys_mem_vaddr;
956	else if (reg->space_id == ACPI_ADR_SPACE_FIXED_HARDWARE)
957		return cpc_read_ffh(cpu, reg, val);
958	else
959		return acpi_os_read_memory((acpi_physical_address)reg->address,
960				val, reg->bit_width);
961
962	switch (reg->bit_width) {
963		case 8:
964			*val = readb_relaxed(vaddr);
965			break;
966		case 16:
967			*val = readw_relaxed(vaddr);
968			break;
969		case 32:
970			*val = readl_relaxed(vaddr);
971			break;
972		case 64:
973			*val = readq_relaxed(vaddr);
974			break;
975		default:
976			pr_debug("Error: Cannot read %u bit width from PCC for ss: %d\n",
977				 reg->bit_width, pcc_ss_id);
978			ret_val = -EFAULT;
979	}
980
981	return ret_val;
982}
983
984static int cpc_write(int cpu, struct cpc_register_resource *reg_res, u64 val)
985{
986	int ret_val = 0;
987	void __iomem *vaddr = 0;
988	int pcc_ss_id = per_cpu(cpu_pcc_subspace_idx, cpu);
989	struct cpc_reg *reg = &reg_res->cpc_entry.reg;
990
991	if (reg->space_id == ACPI_ADR_SPACE_PLATFORM_COMM && pcc_ss_id >= 0)
992		vaddr = GET_PCC_VADDR(reg->address, pcc_ss_id);
993	else if (reg->space_id == ACPI_ADR_SPACE_SYSTEM_MEMORY)
994		vaddr = reg_res->sys_mem_vaddr;
995	else if (reg->space_id == ACPI_ADR_SPACE_FIXED_HARDWARE)
996		return cpc_write_ffh(cpu, reg, val);
997	else
998		return acpi_os_write_memory((acpi_physical_address)reg->address,
999				val, reg->bit_width);
1000
1001	switch (reg->bit_width) {
1002		case 8:
1003			writeb_relaxed(val, vaddr);
1004			break;
1005		case 16:
1006			writew_relaxed(val, vaddr);
1007			break;
1008		case 32:
1009			writel_relaxed(val, vaddr);
1010			break;
1011		case 64:
1012			writeq_relaxed(val, vaddr);
1013			break;
1014		default:
1015			pr_debug("Error: Cannot write %u bit width to PCC for ss: %d\n",
1016				 reg->bit_width, pcc_ss_id);
1017			ret_val = -EFAULT;
1018			break;
1019	}
1020
1021	return ret_val;
1022}
1023
1024/**
1025 * cppc_get_desired_perf - Get the value of desired performance register.
1026 * @cpunum: CPU from which to get desired performance.
1027 * @desired_perf: address of a variable to store the returned desired performance
1028 *
1029 * Return: 0 for success, -EIO otherwise.
1030 */
1031int cppc_get_desired_perf(int cpunum, u64 *desired_perf)
1032{
1033	struct cpc_desc *cpc_desc = per_cpu(cpc_desc_ptr, cpunum);
1034	int pcc_ss_id = per_cpu(cpu_pcc_subspace_idx, cpunum);
1035	struct cpc_register_resource *desired_reg;
1036	struct cppc_pcc_data *pcc_ss_data = NULL;
1037
1038	desired_reg = &cpc_desc->cpc_regs[DESIRED_PERF];
1039
1040	if (CPC_IN_PCC(desired_reg)) {
1041		int ret = 0;
1042
1043		if (pcc_ss_id < 0)
1044			return -EIO;
1045
1046		pcc_ss_data = pcc_data[pcc_ss_id];
1047
1048		down_write(&pcc_ss_data->pcc_lock);
1049
1050		if (send_pcc_cmd(pcc_ss_id, CMD_READ) >= 0)
1051			cpc_read(cpunum, desired_reg, desired_perf);
1052		else
1053			ret = -EIO;
1054
1055		up_write(&pcc_ss_data->pcc_lock);
1056
1057		return ret;
1058	}
1059
1060	cpc_read(cpunum, desired_reg, desired_perf);
1061
1062	return 0;
1063}
1064EXPORT_SYMBOL_GPL(cppc_get_desired_perf);
1065
1066/**
1067 * cppc_get_perf_caps - Get a CPU's performance capabilities.
1068 * @cpunum: CPU from which to get capabilities info.
1069 * @perf_caps: ptr to cppc_perf_caps. See cppc_acpi.h
1070 *
1071 * Return: 0 for success with perf_caps populated else -ERRNO.
1072 */
1073int cppc_get_perf_caps(int cpunum, struct cppc_perf_caps *perf_caps)
1074{
1075	struct cpc_desc *cpc_desc = per_cpu(cpc_desc_ptr, cpunum);
1076	struct cpc_register_resource *highest_reg, *lowest_reg,
1077		*lowest_non_linear_reg, *nominal_reg, *guaranteed_reg,
1078		*low_freq_reg = NULL, *nom_freq_reg = NULL;
1079	u64 high, low, guaranteed, nom, min_nonlinear, low_f = 0, nom_f = 0;
1080	int pcc_ss_id = per_cpu(cpu_pcc_subspace_idx, cpunum);
1081	struct cppc_pcc_data *pcc_ss_data = NULL;
1082	int ret = 0, regs_in_pcc = 0;
1083
1084	if (!cpc_desc) {
1085		pr_debug("No CPC descriptor for CPU:%d\n", cpunum);
1086		return -ENODEV;
1087	}
1088
1089	highest_reg = &cpc_desc->cpc_regs[HIGHEST_PERF];
1090	lowest_reg = &cpc_desc->cpc_regs[LOWEST_PERF];
1091	lowest_non_linear_reg = &cpc_desc->cpc_regs[LOW_NON_LINEAR_PERF];
1092	nominal_reg = &cpc_desc->cpc_regs[NOMINAL_PERF];
1093	low_freq_reg = &cpc_desc->cpc_regs[LOWEST_FREQ];
1094	nom_freq_reg = &cpc_desc->cpc_regs[NOMINAL_FREQ];
1095	guaranteed_reg = &cpc_desc->cpc_regs[GUARANTEED_PERF];
1096
1097	/* Are any of the regs PCC ?*/
1098	if (CPC_IN_PCC(highest_reg) || CPC_IN_PCC(lowest_reg) ||
1099		CPC_IN_PCC(lowest_non_linear_reg) || CPC_IN_PCC(nominal_reg) ||
1100		CPC_IN_PCC(low_freq_reg) || CPC_IN_PCC(nom_freq_reg)) {
1101		if (pcc_ss_id < 0) {
1102			pr_debug("Invalid pcc_ss_id\n");
1103			return -ENODEV;
1104		}
1105		pcc_ss_data = pcc_data[pcc_ss_id];
1106		regs_in_pcc = 1;
1107		down_write(&pcc_ss_data->pcc_lock);
1108		/* Ring doorbell once to update PCC subspace */
1109		if (send_pcc_cmd(pcc_ss_id, CMD_READ) < 0) {
1110			ret = -EIO;
1111			goto out_err;
1112		}
1113	}
1114
1115	cpc_read(cpunum, highest_reg, &high);
1116	perf_caps->highest_perf = high;
1117
1118	cpc_read(cpunum, lowest_reg, &low);
1119	perf_caps->lowest_perf = low;
1120
1121	cpc_read(cpunum, nominal_reg, &nom);
1122	perf_caps->nominal_perf = nom;
1123
1124	if (guaranteed_reg->type != ACPI_TYPE_BUFFER  ||
1125	    IS_NULL_REG(&guaranteed_reg->cpc_entry.reg)) {
1126		perf_caps->guaranteed_perf = 0;
1127	} else {
1128		cpc_read(cpunum, guaranteed_reg, &guaranteed);
1129		perf_caps->guaranteed_perf = guaranteed;
1130	}
1131
1132	cpc_read(cpunum, lowest_non_linear_reg, &min_nonlinear);
1133	perf_caps->lowest_nonlinear_perf = min_nonlinear;
1134
1135	if (!high || !low || !nom || !min_nonlinear)
1136		ret = -EFAULT;
1137
1138	/* Read optional lowest and nominal frequencies if present */
1139	if (CPC_SUPPORTED(low_freq_reg))
1140		cpc_read(cpunum, low_freq_reg, &low_f);
1141
1142	if (CPC_SUPPORTED(nom_freq_reg))
1143		cpc_read(cpunum, nom_freq_reg, &nom_f);
1144
1145	perf_caps->lowest_freq = low_f;
1146	perf_caps->nominal_freq = nom_f;
1147
1148
1149out_err:
1150	if (regs_in_pcc)
1151		up_write(&pcc_ss_data->pcc_lock);
1152	return ret;
1153}
1154EXPORT_SYMBOL_GPL(cppc_get_perf_caps);
1155
1156/**
1157 * cppc_get_perf_ctrs - Read a CPU's performance feedback counters.
1158 * @cpunum: CPU from which to read counters.
1159 * @perf_fb_ctrs: ptr to cppc_perf_fb_ctrs. See cppc_acpi.h
1160 *
1161 * Return: 0 for success with perf_fb_ctrs populated else -ERRNO.
1162 */
1163int cppc_get_perf_ctrs(int cpunum, struct cppc_perf_fb_ctrs *perf_fb_ctrs)
1164{
1165	struct cpc_desc *cpc_desc = per_cpu(cpc_desc_ptr, cpunum);
1166	struct cpc_register_resource *delivered_reg, *reference_reg,
1167		*ref_perf_reg, *ctr_wrap_reg;
1168	int pcc_ss_id = per_cpu(cpu_pcc_subspace_idx, cpunum);
1169	struct cppc_pcc_data *pcc_ss_data = NULL;
1170	u64 delivered, reference, ref_perf, ctr_wrap_time;
1171	int ret = 0, regs_in_pcc = 0;
1172
1173	if (!cpc_desc) {
1174		pr_debug("No CPC descriptor for CPU:%d\n", cpunum);
1175		return -ENODEV;
1176	}
1177
1178	delivered_reg = &cpc_desc->cpc_regs[DELIVERED_CTR];
1179	reference_reg = &cpc_desc->cpc_regs[REFERENCE_CTR];
1180	ref_perf_reg = &cpc_desc->cpc_regs[REFERENCE_PERF];
1181	ctr_wrap_reg = &cpc_desc->cpc_regs[CTR_WRAP_TIME];
1182
1183	/*
1184	 * If reference perf register is not supported then we should
1185	 * use the nominal perf value
1186	 */
1187	if (!CPC_SUPPORTED(ref_perf_reg))
1188		ref_perf_reg = &cpc_desc->cpc_regs[NOMINAL_PERF];
1189
1190	/* Are any of the regs PCC ?*/
1191	if (CPC_IN_PCC(delivered_reg) || CPC_IN_PCC(reference_reg) ||
1192		CPC_IN_PCC(ctr_wrap_reg) || CPC_IN_PCC(ref_perf_reg)) {
1193		if (pcc_ss_id < 0) {
1194			pr_debug("Invalid pcc_ss_id\n");
1195			return -ENODEV;
1196		}
1197		pcc_ss_data = pcc_data[pcc_ss_id];
1198		down_write(&pcc_ss_data->pcc_lock);
1199		regs_in_pcc = 1;
1200		/* Ring doorbell once to update PCC subspace */
1201		if (send_pcc_cmd(pcc_ss_id, CMD_READ) < 0) {
1202			ret = -EIO;
1203			goto out_err;
1204		}
1205	}
1206
1207	cpc_read(cpunum, delivered_reg, &delivered);
1208	cpc_read(cpunum, reference_reg, &reference);
1209	cpc_read(cpunum, ref_perf_reg, &ref_perf);
1210
1211	/*
1212	 * Per spec, if ctr_wrap_time optional register is unsupported, then the
1213	 * performance counters are assumed to never wrap during the lifetime of
1214	 * platform
1215	 */
1216	ctr_wrap_time = (u64)(~((u64)0));
1217	if (CPC_SUPPORTED(ctr_wrap_reg))
1218		cpc_read(cpunum, ctr_wrap_reg, &ctr_wrap_time);
1219
1220	if (!delivered || !reference ||	!ref_perf) {
1221		ret = -EFAULT;
1222		goto out_err;
1223	}
1224
1225	perf_fb_ctrs->delivered = delivered;
1226	perf_fb_ctrs->reference = reference;
1227	perf_fb_ctrs->reference_perf = ref_perf;
1228	perf_fb_ctrs->wraparound_time = ctr_wrap_time;
1229out_err:
1230	if (regs_in_pcc)
1231		up_write(&pcc_ss_data->pcc_lock);
1232	return ret;
1233}
1234EXPORT_SYMBOL_GPL(cppc_get_perf_ctrs);
1235
1236/**
1237 * cppc_set_perf - Set a CPU's performance controls.
1238 * @cpu: CPU for which to set performance controls.
1239 * @perf_ctrls: ptr to cppc_perf_ctrls. See cppc_acpi.h
1240 *
1241 * Return: 0 for success, -ERRNO otherwise.
1242 */
1243int cppc_set_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls)
1244{
1245	struct cpc_desc *cpc_desc = per_cpu(cpc_desc_ptr, cpu);
1246	struct cpc_register_resource *desired_reg;
1247	int pcc_ss_id = per_cpu(cpu_pcc_subspace_idx, cpu);
1248	struct cppc_pcc_data *pcc_ss_data = NULL;
1249	int ret = 0;
1250
1251	if (!cpc_desc) {
1252		pr_debug("No CPC descriptor for CPU:%d\n", cpu);
1253		return -ENODEV;
1254	}
1255
1256	desired_reg = &cpc_desc->cpc_regs[DESIRED_PERF];
1257
1258	/*
1259	 * This is Phase-I where we want to write to CPC registers
1260	 * -> We want all CPUs to be able to execute this phase in parallel
1261	 *
1262	 * Since read_lock can be acquired by multiple CPUs simultaneously we
1263	 * achieve that goal here
1264	 */
1265	if (CPC_IN_PCC(desired_reg)) {
1266		if (pcc_ss_id < 0) {
1267			pr_debug("Invalid pcc_ss_id\n");
1268			return -ENODEV;
1269		}
1270		pcc_ss_data = pcc_data[pcc_ss_id];
1271		down_read(&pcc_ss_data->pcc_lock); /* BEGIN Phase-I */
1272		if (pcc_ss_data->platform_owns_pcc) {
1273			ret = check_pcc_chan(pcc_ss_id, false);
1274			if (ret) {
1275				up_read(&pcc_ss_data->pcc_lock);
1276				return ret;
1277			}
1278		}
1279		/*
1280		 * Update the pending_write to make sure a PCC CMD_READ will not
1281		 * arrive and steal the channel during the switch to write lock
1282		 */
1283		pcc_ss_data->pending_pcc_write_cmd = true;
1284		cpc_desc->write_cmd_id = pcc_ss_data->pcc_write_cnt;
1285		cpc_desc->write_cmd_status = 0;
1286	}
1287
1288	/*
1289	 * Skip writing MIN/MAX until Linux knows how to come up with
1290	 * useful values.
1291	 */
1292	cpc_write(cpu, desired_reg, perf_ctrls->desired_perf);
1293
1294	if (CPC_IN_PCC(desired_reg))
1295		up_read(&pcc_ss_data->pcc_lock);	/* END Phase-I */
1296	/*
1297	 * This is Phase-II where we transfer the ownership of PCC to Platform
1298	 *
1299	 * Short Summary: Basically if we think of a group of cppc_set_perf
1300	 * requests that happened in short overlapping interval. The last CPU to
1301	 * come out of Phase-I will enter Phase-II and ring the doorbell.
1302	 *
1303	 * We have the following requirements for Phase-II:
1304	 *     1. We want to execute Phase-II only when there are no CPUs
1305	 * currently executing in Phase-I
1306	 *     2. Once we start Phase-II we want to avoid all other CPUs from
1307	 * entering Phase-I.
1308	 *     3. We want only one CPU among all those who went through Phase-I
1309	 * to run phase-II
1310	 *
1311	 * If write_trylock fails to get the lock and doesn't transfer the
1312	 * PCC ownership to the platform, then one of the following will be TRUE
1313	 *     1. There is at-least one CPU in Phase-I which will later execute
1314	 * write_trylock, so the CPUs in Phase-I will be responsible for
1315	 * executing the Phase-II.
1316	 *     2. Some other CPU has beaten this CPU to successfully execute the
1317	 * write_trylock and has already acquired the write_lock. We know for a
1318	 * fact it (other CPU acquiring the write_lock) couldn't have happened
1319	 * before this CPU's Phase-I as we held the read_lock.
1320	 *     3. Some other CPU executing pcc CMD_READ has stolen the
1321	 * down_write, in which case, send_pcc_cmd will check for pending
1322	 * CMD_WRITE commands by checking the pending_pcc_write_cmd.
1323	 * So this CPU can be certain that its request will be delivered
1324	 *    So in all cases, this CPU knows that its request will be delivered
1325	 * by another CPU and can return
1326	 *
1327	 * After getting the down_write we still need to check for
1328	 * pending_pcc_write_cmd to take care of the following scenario
1329	 *    The thread running this code could be scheduled out between
1330	 * Phase-I and Phase-II. Before it is scheduled back on, another CPU
1331	 * could have delivered the request to Platform by triggering the
1332	 * doorbell and transferred the ownership of PCC to platform. So this
1333	 * avoids triggering an unnecessary doorbell and more importantly before
1334	 * triggering the doorbell it makes sure that the PCC channel ownership
1335	 * is still with OSPM.
1336	 *   pending_pcc_write_cmd can also be cleared by a different CPU, if
1337	 * there was a pcc CMD_READ waiting on down_write and it steals the lock
1338	 * before the pcc CMD_WRITE is completed. pcc_send_cmd checks for this
1339	 * case during a CMD_READ and if there are pending writes it delivers
1340	 * the write command before servicing the read command
1341	 */
1342	if (CPC_IN_PCC(desired_reg)) {
1343		if (down_write_trylock(&pcc_ss_data->pcc_lock)) {/* BEGIN Phase-II */
1344			/* Update only if there are pending write commands */
1345			if (pcc_ss_data->pending_pcc_write_cmd)
1346				send_pcc_cmd(pcc_ss_id, CMD_WRITE);
1347			up_write(&pcc_ss_data->pcc_lock);	/* END Phase-II */
1348		} else
1349			/* Wait until pcc_write_cnt is updated by send_pcc_cmd */
1350			wait_event(pcc_ss_data->pcc_write_wait_q,
1351				   cpc_desc->write_cmd_id != pcc_ss_data->pcc_write_cnt);
1352
1353		/* send_pcc_cmd updates the status in case of failure */
1354		ret = cpc_desc->write_cmd_status;
1355	}
1356	return ret;
1357}
1358EXPORT_SYMBOL_GPL(cppc_set_perf);
1359
1360/**
1361 * cppc_get_transition_latency - returns frequency transition latency in ns
1362 *
1363 * ACPI CPPC does not explicitly specifiy how a platform can specify the
1364 * transition latency for perfromance change requests. The closest we have
1365 * is the timing information from the PCCT tables which provides the info
1366 * on the number and frequency of PCC commands the platform can handle.
1367 */
1368unsigned int cppc_get_transition_latency(int cpu_num)
1369{
1370	/*
1371	 * Expected transition latency is based on the PCCT timing values
1372	 * Below are definition from ACPI spec:
1373	 * pcc_nominal- Expected latency to process a command, in microseconds
1374	 * pcc_mpar   - The maximum number of periodic requests that the subspace
1375	 *              channel can support, reported in commands per minute. 0
1376	 *              indicates no limitation.
1377	 * pcc_mrtt   - The minimum amount of time that OSPM must wait after the
1378	 *              completion of a command before issuing the next command,
1379	 *              in microseconds.
1380	 */
1381	unsigned int latency_ns = 0;
1382	struct cpc_desc *cpc_desc;
1383	struct cpc_register_resource *desired_reg;
1384	int pcc_ss_id = per_cpu(cpu_pcc_subspace_idx, cpu_num);
1385	struct cppc_pcc_data *pcc_ss_data;
1386
1387	cpc_desc = per_cpu(cpc_desc_ptr, cpu_num);
1388	if (!cpc_desc)
1389		return CPUFREQ_ETERNAL;
1390
1391	desired_reg = &cpc_desc->cpc_regs[DESIRED_PERF];
1392	if (!CPC_IN_PCC(desired_reg))
1393		return CPUFREQ_ETERNAL;
1394
1395	if (pcc_ss_id < 0)
1396		return CPUFREQ_ETERNAL;
1397
1398	pcc_ss_data = pcc_data[pcc_ss_id];
1399	if (pcc_ss_data->pcc_mpar)
1400		latency_ns = 60 * (1000 * 1000 * 1000 / pcc_ss_data->pcc_mpar);
1401
1402	latency_ns = max(latency_ns, pcc_ss_data->pcc_nominal * 1000);
1403	latency_ns = max(latency_ns, pcc_ss_data->pcc_mrtt * 1000);
1404
1405	return latency_ns;
1406}
1407EXPORT_SYMBOL_GPL(cppc_get_transition_latency);
1408