[linux.git] / init / calibrate.c

/* calibrate.c: default delay calibration
 *
 * Excised from init/main.c
 *  Copyright (C) 1991, 1992  Linus Torvalds
 */

#include <linux/jiffies.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <linux/timex.h>
#include <linux/smp.h>

unsigned long lpj_fine;
unsigned long preset_lpj;
static int __init lpj_setup(char *str)
{
	preset_lpj = simple_strtoul(str,NULL,0);
	return 1;
}

__setup("lpj=", lpj_setup);

#ifdef ARCH_HAS_READ_CURRENT_TIMER

/* This routine uses the read_current_timer() routine and gets the
 * loops per jiffy directly, instead of guessing it using delay().
 * Also, this code tries to handle non-maskable asynchronous events
 * (like SMIs)
 */
#define DELAY_CALIBRATION_TICKS			((HZ < 100) ? 1 : (HZ/100))
#define MAX_DIRECT_CALIBRATION_RETRIES		5

static unsigned long __cpuinit calibrate_delay_direct(void)
{
	unsigned long pre_start, start, post_start;
	unsigned long pre_end, end, post_end;
	unsigned long start_jiffies;
	unsigned long timer_rate_min, timer_rate_max;
	unsigned long good_timer_sum = 0;
	unsigned long good_timer_count = 0;
	unsigned long measured_times[MAX_DIRECT_CALIBRATION_RETRIES];
	int max = -1; /* index of measured_times with max/min values or not set */
	int min = -1;
	int i;

	if (read_current_timer(&pre_start) < 0 )
		return 0;

	/*
	 * A simple loop like
	 *	while ( jiffies < start_jiffies+1)
	 *		start = read_current_timer();
	 * will not do. As we don't really know whether jiffy switch
	 * happened first or timer_value was read first. And some asynchronous
	 * event can happen between these two events introducing errors in lpj.
	 *
	 * So, we do
	 * 1. pre_start <- When we are sure that jiffy switch hasn't happened
	 * 2. check jiffy switch
	 * 3. start <- timer value before or after jiffy switch
	 * 4. post_start <- When we are sure that jiffy switch has happened
	 *
	 * Note, we don't know anything about order of 2 and 3.
	 * Now, by looking at post_start and pre_start difference, we can
	 * check whether any asynchronous event happened or not
	 */

	for (i = 0; i < MAX_DIRECT_CALIBRATION_RETRIES; i++) {
		pre_start = 0;
		read_current_timer(&start);
		start_jiffies = jiffies;
		while (time_before_eq(jiffies, start_jiffies + 1)) {
			pre_start = start;
			read_current_timer(&start);
		}
		read_current_timer(&post_start);

		pre_end = 0;
		end = post_start;
		while (time_before_eq(jiffies, start_jiffies + 1 +
					       DELAY_CALIBRATION_TICKS)) {
			pre_end = end;
			read_current_timer(&end);
		}
		read_current_timer(&post_end);

		timer_rate_max = (post_end - pre_start) /
					DELAY_CALIBRATION_TICKS;
		timer_rate_min = (pre_end - post_start) /
					DELAY_CALIBRATION_TICKS;

		/*
		 * If the upper limit and lower limit of the timer_rate is
		 * >= 12.5% apart, redo calibration.
		 */
		if (start >= post_end)
			printk(KERN_NOTICE "calibrate_delay_direct() ignoring "
					"timer_rate as we had a TSC wrap around"
					" start=%lu >=post_end=%lu\n",
				start, post_end);
		if (start < post_end && pre_start != 0 && pre_end != 0 &&
		    (timer_rate_max - timer_rate_min) < (timer_rate_max >> 3)) {
			good_timer_count++;
			good_timer_sum += timer_rate_max;
			measured_times[i] = timer_rate_max;
			if (max < 0 || timer_rate_max > measured_times[max])
				max = i;
			if (min < 0 || timer_rate_max < measured_times[min])
				min = i;
		} else
			measured_times[i] = 0;

	}

	/*
	 * Find the maximum & minimum - if they differ too much throw out the
	 * one with the largest difference from the mean and try again...
	 */
	while (good_timer_count > 1) {
		unsigned long estimate;
		unsigned long maxdiff;

		/* compute the estimate */
		estimate = (good_timer_sum/good_timer_count);
		maxdiff = estimate >> 3;

		/* if range is within 12% let's take it */
		if ((measured_times[max] - measured_times[min]) < maxdiff)
			return estimate;

		/* ok - drop the worse value and try again... */
		good_timer_sum = 0;
		good_timer_count = 0;
		if ((measured_times[max] - estimate) <
				(estimate - measured_times[min])) {
			printk(KERN_NOTICE "calibrate_delay_direct() dropping "
					"min bogoMips estimate %d = %lu\n",
				min, measured_times[min]);
			measured_times[min] = 0;
			min = max;
		} else {
			printk(KERN_NOTICE "calibrate_delay_direct() dropping "
					"max bogoMips estimate %d = %lu\n",
				max, measured_times[max]);
			measured_times[max] = 0;
			max = min;
		}

		for (i = 0; i < MAX_DIRECT_CALIBRATION_RETRIES; i++) {
			if (measured_times[i] == 0)
				continue;
			good_timer_count++;
			good_timer_sum += measured_times[i];
			if (measured_times[i] < measured_times[min])
				min = i;
			if (measured_times[i] > measured_times[max])
				max = i;
		}

	}

	printk(KERN_NOTICE "calibrate_delay_direct() failed to get a good "
	       "estimate for loops_per_jiffy.\nProbably due to long platform "
		"interrupts. Consider using \"lpj=\" boot option.\n");
	return 0;
}
#else
static unsigned long __cpuinit calibrate_delay_direct(void) {return 0;}
#endif

/*
 * This is the number of bits of precision for the loops_per_jiffy.  Each
 * time we refine our estimate after the first takes 1.5/HZ seconds, so try
 * to start with a good estimate.
 * For the boot cpu we can skip the delay calibration and assign it a value
 * calculated based on the timer frequency.
 * For the rest of the CPUs we cannot assume that the timer frequency is same as
 * the cpu frequency, hence do the calibration for those.
 */
#define LPS_PREC 8

static unsigned long __cpuinit calibrate_delay_converge(void)
{
	/* First stage - slowly accelerate to find initial bounds */
	unsigned long lpj, lpj_base, ticks, loopadd, loopadd_base, chop_limit;
	int trials = 0, band = 0, trial_in_band = 0;

	lpj = (1<<12);

	/* wait for "start of" clock tick */
	ticks = jiffies;
	while (ticks == jiffies)
		; /* nothing */
	/* Go .. */
	ticks = jiffies;
	do {
		if (++trial_in_band == (1<<band)) {
			++band;
			trial_in_band = 0;
		}
		__delay(lpj * band);
		trials += band;
	} while (ticks == jiffies);
	/*
	 * We overshot, so retreat to a clear underestimate. Then estimate
	 * the largest likely undershoot. This defines our chop bounds.
	 */
	trials -= band;
	loopadd_base = lpj * band;
	lpj_base = lpj * trials;

recalibrate:
	lpj = lpj_base;
	loopadd = loopadd_base;

	/*
	 * Do a binary approximation to get lpj set to
	 * equal one clock (up to LPS_PREC bits)
	 */
	chop_limit = lpj >> LPS_PREC;
	while (loopadd > chop_limit) {
		lpj += loopadd;
		ticks = jiffies;
		while (ticks == jiffies)
			; /* nothing */
		ticks = jiffies;
		__delay(lpj);
		if (jiffies != ticks)	/* longer than 1 tick */
			lpj -= loopadd;
		loopadd >>= 1;
	}
	/*
	 * If we incremented every single time possible, presume we've
	 * massively underestimated initially, and retry with a higher
	 * start, and larger range. (Only seen on x86_64, due to SMIs)
	 */
	if (lpj + loopadd * 2 == lpj_base + loopadd_base * 2) {
		lpj_base = lpj;
		loopadd_base <<= 2;
		goto recalibrate;
	}

	return lpj;
}

void __cpuinit calibrate_delay(void)
{
	static bool printed;

	if (preset_lpj) {
		loops_per_jiffy = preset_lpj;
		if (!printed)
			pr_info("Calibrating delay loop (skipped) "
				"preset value.. ");
	} else if ((!printed) && lpj_fine) {
		loops_per_jiffy = lpj_fine;
		pr_info("Calibrating delay loop (skipped), "
			"value calculated using timer frequency.. ");
	} else if ((loops_per_jiffy = calibrate_delay_direct()) != 0) {
		if (!printed)
			pr_info("Calibrating delay using timer "
				"specific routine.. ");
	} else {
		if (!printed)
			pr_info("Calibrating delay loop... ");
		loops_per_jiffy = calibrate_delay_converge();
	}
	if (!printed)
		pr_cont("%lu.%02lu BogoMIPS (lpj=%lu)\n",
			loops_per_jiffy/(500000/HZ),
			(loops_per_jiffy/(5000/HZ)) % 100, loops_per_jiffy);

	printed = true;
}
Commit	Line	Data
1da177e4 LT	1	/* calibrate.c: default delay calibration
	2	*
	3	* Excised from init/main.c
	4	* Copyright (C) 1991, 1992 Linus Torvalds
	5	*/
	6
cd354f1a	7	#include <linux/jiffies.h>
1da177e4 LT	8	#include <linux/delay.h>
1da177e4 LT	9	#include <linux/init.h>
941e492b	10	#include <linux/timex.h>
3da757da	11	#include <linux/smp.h>
8a9e1b0f	12
f3f3149f	13	unsigned long lpj_fine;
bfe8df3d	14	unsigned long preset_lpj;
1da177e4 LT	15	static int __init lpj_setup(char *str)
	16	{
	17	preset_lpj = simple_strtoul(str,NULL,0);
	18	return 1;
	19	}
	20
	21	__setup("lpj=", lpj_setup);
	22
8a9e1b0f VP	23	#ifdef ARCH_HAS_READ_CURRENT_TIMER
	24
	25	/* This routine uses the read_current_timer() routine and gets the
	26	* loops per jiffy directly, instead of guessing it using delay().
	27	* Also, this code tries to handle non-maskable asynchronous events
	28	* (like SMIs)
	29	*/
	30	#define DELAY_CALIBRATION_TICKS ((HZ < 100) ? 1 : (HZ/100))
	31	#define MAX_DIRECT_CALIBRATION_RETRIES 5
	32
6c81c32f	33	static unsigned long __cpuinit calibrate_delay_direct(void)
8a9e1b0f VP	34	{
	35	unsigned long pre_start, start, post_start;
	36	unsigned long pre_end, end, post_end;
	37	unsigned long start_jiffies;
f3f3149f AK	38	unsigned long timer_rate_min, timer_rate_max;
	39	unsigned long good_timer_sum = 0;
	40	unsigned long good_timer_count = 0;
d2b46313 AW	41	unsigned long measured_times[MAX_DIRECT_CALIBRATION_RETRIES];
	42	int max = -1; /* index of measured_times with max/min values or not set */
	43	int min = -1;
8a9e1b0f VP	44	int i;
	45
	46	if (read_current_timer(&pre_start) < 0 )
	47	return 0;
	48
	49	/*
	50	* A simple loop like
	51	* while ( jiffies < start_jiffies+1)
	52	* start = read_current_timer();
	53	* will not do. As we don't really know whether jiffy switch
	54	* happened first or timer_value was read first. And some asynchronous
	55	* event can happen between these two events introducing errors in lpj.
	56	*
	57	* So, we do
	58	* 1. pre_start <- When we are sure that jiffy switch hasn't happened
	59	* 2. check jiffy switch
	60	* 3. start <- timer value before or after jiffy switch
	61	* 4. post_start <- When we are sure that jiffy switch has happened
	62	*
	63	* Note, we don't know anything about order of 2 and 3.
	64	* Now, by looking at post_start and pre_start difference, we can
	65	* check whether any asynchronous event happened or not
	66	*/
	67
	68	for (i = 0; i < MAX_DIRECT_CALIBRATION_RETRIES; i++) {
	69	pre_start = 0;
	70	read_current_timer(&start);
	71	start_jiffies = jiffies;
70a06228	72	while (time_before_eq(jiffies, start_jiffies + 1)) {
8a9e1b0f VP	73	pre_start = start;
	74	read_current_timer(&start);
	75	}
	76	read_current_timer(&post_start);
	77
	78	pre_end = 0;
	79	end = post_start;
70a06228 TD	80	while (time_before_eq(jiffies, start_jiffies + 1 +
70a06228 TD	81	DELAY_CALIBRATION_TICKS)) {
8a9e1b0f VP	82	pre_end = end;
	83	read_current_timer(&end);
	84	}
	85	read_current_timer(&post_end);
	86
f3f3149f AK	87	timer_rate_max = (post_end - pre_start) /
	88	DELAY_CALIBRATION_TICKS;
	89	timer_rate_min = (pre_end - post_start) /
	90	DELAY_CALIBRATION_TICKS;
8a9e1b0f VP	91
8a9e1b0f VP	92	/*
f3f3149f	93	* If the upper limit and lower limit of the timer_rate is
8a9e1b0f VP	94	* >= 12.5% apart, redo calibration.
8a9e1b0f VP	95	*/
d2b46313 AW	96	if (start >= post_end)
	97	printk(KERN_NOTICE "calibrate_delay_direct() ignoring "
	98	"timer_rate as we had a TSC wrap around"
	99	" start=%lu >=post_end=%lu\n",
	100	start, post_end);
	101	if (start < post_end && pre_start != 0 && pre_end != 0 &&
f3f3149f AK	102	(timer_rate_max - timer_rate_min) < (timer_rate_max >> 3)) {
	103	good_timer_count++;
	104	good_timer_sum += timer_rate_max;
d2b46313 AW	105	measured_times[i] = timer_rate_max;
	106	if (max < 0 \|\| timer_rate_max > measured_times[max])
	107	max = i;
	108	if (min < 0 \|\| timer_rate_max < measured_times[min])
	109	min = i;
	110	} else
	111	measured_times[i] = 0;
	112
8a9e1b0f VP	113	}
8a9e1b0f VP	114
d2b46313 AW	115	/*
	116	* Find the maximum & minimum - if they differ too much throw out the
	117	* one with the largest difference from the mean and try again...
	118	*/
	119	while (good_timer_count > 1) {
	120	unsigned long estimate;
	121	unsigned long maxdiff;
	122
	123	/* compute the estimate */
	124	estimate = (good_timer_sum/good_timer_count);
	125	maxdiff = estimate >> 3;
	126
	127	/* if range is within 12% let's take it */
	128	if ((measured_times[max] - measured_times[min]) < maxdiff)
	129	return estimate;
	130
	131	/* ok - drop the worse value and try again... */
	132	good_timer_sum = 0;
	133	good_timer_count = 0;
	134	if ((measured_times[max] - estimate) <
	135	(estimate - measured_times[min])) {
	136	printk(KERN_NOTICE "calibrate_delay_direct() dropping "
	137	"min bogoMips estimate %d = %lu\n",
	138	min, measured_times[min]);
	139	measured_times[min] = 0;
	140	min = max;
	141	} else {
	142	printk(KERN_NOTICE "calibrate_delay_direct() dropping "
	143	"max bogoMips estimate %d = %lu\n",
	144	max, measured_times[max]);
	145	measured_times[max] = 0;
	146	max = min;
	147	}
	148
	149	for (i = 0; i < MAX_DIRECT_CALIBRATION_RETRIES; i++) {
	150	if (measured_times[i] == 0)
	151	continue;
	152	good_timer_count++;
	153	good_timer_sum += measured_times[i];
	154	if (measured_times[i] < measured_times[min])
	155	min = i;
	156	if (measured_times[i] > measured_times[max])
	157	max = i;
	158	}
	159
	160	}
8a9e1b0f	161
d2b46313 AW	162	printk(KERN_NOTICE "calibrate_delay_direct() failed to get a good "
	163	"estimate for loops_per_jiffy.\nProbably due to long platform "
	164	"interrupts. Consider using \"lpj=\" boot option.\n");
8a9e1b0f VP	165	return 0;
	166	}
	167	#else
6c81c32f	168	static unsigned long __cpuinit calibrate_delay_direct(void) {return 0;}
8a9e1b0f VP	169	#endif
8a9e1b0f VP	170
1da177e4 LT	171	/*
1da177e4 LT	172	* This is the number of bits of precision for the loops_per_jiffy. Each
191e5688 PC	173	* time we refine our estimate after the first takes 1.5/HZ seconds, so try
191e5688 PC	174	* to start with a good estimate.
3da757da	175	* For the boot cpu we can skip the delay calibration and assign it a value
f3f3149f AK	176	* calculated based on the timer frequency.
f3f3149f AK	177	* For the rest of the CPUs we cannot assume that the timer frequency is same as
3da757da	178	* the cpu frequency, hence do the calibration for those.
1da177e4 LT	179	*/
	180	#define LPS_PREC 8
	181
71c696b1	182	static unsigned long __cpuinit calibrate_delay_converge(void)
1da177e4	183	{
191e5688	184	/* First stage - slowly accelerate to find initial bounds */
b1b5f65e	185	unsigned long lpj, lpj_base, ticks, loopadd, loopadd_base, chop_limit;
191e5688	186	int trials = 0, band = 0, trial_in_band = 0;
71c696b1 PC	187
71c696b1 PC	188	lpj = (1<<12);
191e5688 PC	189
	190	/* wait for "start of" clock tick */
	191	ticks = jiffies;
	192	while (ticks == jiffies)
	193	; /* nothing */
	194	/* Go .. */
	195	ticks = jiffies;
	196	do {
	197	if (++trial_in_band == (1<<band)) {
	198	++band;
	199	trial_in_band = 0;
	200	}
	201	__delay(lpj * band);
	202	trials += band;
	203	} while (ticks == jiffies);
	204	/*
	205	* We overshot, so retreat to a clear underestimate. Then estimate
	206	* the largest likely undershoot. This defines our chop bounds.
	207	*/
	208	trials -= band;
b1b5f65e PC	209	loopadd_base = lpj * band;
	210	lpj_base = lpj * trials;
	211
	212	recalibrate:
	213	lpj = lpj_base;
	214	loopadd = loopadd_base;
71c696b1 PC	215
	216	/*
	217	* Do a binary approximation to get lpj set to
191e5688	218	* equal one clock (up to LPS_PREC bits)
71c696b1	219	*/
b1b5f65e	220	chop_limit = lpj >> LPS_PREC;
191e5688 PC	221	while (loopadd > chop_limit) {
191e5688 PC	222	lpj += loopadd;
71c696b1 PC	223	ticks = jiffies;
71c696b1 PC	224	while (ticks == jiffies)
191e5688	225	; /* nothing */
71c696b1 PC	226	ticks = jiffies;
	227	__delay(lpj);
	228	if (jiffies != ticks) /* longer than 1 tick */
191e5688 PC	229	lpj -= loopadd;
191e5688 PC	230	loopadd >>= 1;
71c696b1	231	}
b1b5f65e PC	232	/*
	233	* If we incremented every single time possible, presume we've
	234	* massively underestimated initially, and retry with a higher
	235	* start, and larger range. (Only seen on x86_64, due to SMIs)
	236	*/
	237	if (lpj + loopadd * 2 == lpj_base + loopadd_base * 2) {
	238	lpj_base = lpj;
	239	loopadd_base <<= 2;
	240	goto recalibrate;
	241	}
71c696b1 PC	242
	243	return lpj;
	244	}
	245
	246	void __cpuinit calibrate_delay(void)
	247	{
feae3203	248	static bool printed;
1da177e4 LT	249
	250	if (preset_lpj) {
	251	loops_per_jiffy = preset_lpj;
feae3203 MT	252	if (!printed)
	253	pr_info("Calibrating delay loop (skipped) "
	254	"preset value.. ");
	255	} else if ((!printed) && lpj_fine) {
f3f3149f	256	loops_per_jiffy = lpj_fine;
feae3203	257	pr_info("Calibrating delay loop (skipped), "
f3f3149f	258	"value calculated using timer frequency.. ");
8a9e1b0f	259	} else if ((loops_per_jiffy = calibrate_delay_direct()) != 0) {
feae3203 MT	260	if (!printed)
	261	pr_info("Calibrating delay using timer "
	262	"specific routine.. ");
1da177e4	263	} else {
feae3203 MT	264	if (!printed)
feae3203 MT	265	pr_info("Calibrating delay loop... ");
71c696b1	266	loops_per_jiffy = calibrate_delay_converge();
1da177e4	267	}
feae3203 MT	268	if (!printed)
feae3203 MT	269	pr_cont("%lu.%02lu BogoMIPS (lpj=%lu)\n",
3da757da AK	270	loops_per_jiffy/(500000/HZ),
3da757da AK	271	(loops_per_jiffy/(5000/HZ)) % 100, loops_per_jiffy);
feae3203 MT	272
feae3203 MT	273	printed = true;
1da177e4	274	}