time.c

一个2.4.21版本的嵌入式linux内核

/*
 * 
 * Common time routines among all ppc machines.
 *
 * Written by Cort Dougan ([e-mail]) to merge
 * Paul Mackerras' version and mine for PReP and Pmac.
 * MPC8xx/MBX changes by Dan Malek ([e-mail]).
 * Converted for 64-bit by Mike Corrigan ([e-mail])
 *
 * First round of bugfixes by Gabriel Paubert ([e-mail])
 * to make clock more stable (2.4.0-test5). The only thing
 * that this code assumes is that the timebases have been synchronized
 * by firmware on SMP and are never stopped (never do sleep
 * on SMP then, nap and doze are OK).
 * 
 * Speeded up do_gettimeofday by getting rid of references to
 * xtime (which required locks for consistency). ([e-mail])
 *
 * TODO (not necessarily in this file):
 * - improve precision and reproducibility of timebase frequency
 * measurement at boot time. (for iSeries, we calibrate the timebase
 * against the Titan chip's clock.)
 * - for astronomical applications: add a new function to get
 * non ambiguous timestamps even around leap seconds. This needs
 * a new timestamp format and a good name.
 *
 * 1997-09-10  Updated NTP code according to technical memorandum Jan '96
 *             "A Kernel Model for Precision Timekeeping" by Dave Mills
 *
 *      This program is free software; you can redistribute it and/or
 *      modify it under the terms of the GNU General Public License
 *      as published by the Free Software Foundation; either version
 *      2 of the License, or (at your option) any later version.
 */

#include <linux/config.h>
#include <linux/errno.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/param.h>
#include <linux/string.h>
#include <linux/mm.h>
#include <linux/interrupt.h>
#include <linux/timex.h>
#include <linux/kernel_stat.h>
#include <linux/mc146818rtc.h>
#include <linux/time.h>
#include <linux/init.h>

#include <asm/naca.h>
#include <asm/segment.h>
#include <asm/io.h>
#include <asm/processor.h>
#include <asm/nvram.h>
#include <asm/cache.h>
#include <asm/machdep.h>
#include <asm/init.h>
#ifdef CONFIG_PPC_ISERIES
#include <asm/iSeries/HvCallXm.h>
#endif
#include <asm/uaccess.h>

#include <asm/time.h>
#include <asm/ppcdebug.h>

void smp_local_timer_interrupt(struct pt_regs *);

extern void setup_before_console_init();

/* keep track of when we need to update the rtc */
time_t last_rtc_update;
extern rwlock_t xtime_lock;
extern int piranha_simulator;
#ifdef CONFIG_PPC_ISERIES
unsigned long iSeries_recal_titan = 0;
unsigned long iSeries_recal_tb = 0; 
static unsigned long first_settimeofday = 1;
#endif

#define XSEC_PER_SEC (1024*1024)
#define USEC_PER_SEC (1000000)

unsigned long tb_ticks_per_jiffy;
unsigned long tb_ticks_per_usec;
unsigned long tb_ticks_per_sec;
unsigned long next_xtime_sync_tb;
unsigned long xtime_sync_interval;
unsigned long tb_to_xs;
unsigned long processor_freq;
spinlock_t rtc_lock = SPIN_LOCK_UNLOCKED;

extern unsigned long wall_jiffies;
extern unsigned long lpEvent_count;
extern int smp_tb_synchronized;

extern unsigned long prof_cpu_mask;
extern unsigned int * prof_buffer;
extern unsigned long prof_len;
extern unsigned long prof_shift;
extern char _stext;

extern struct timezone sys_tz;

void ppc_adjtimex(void);

static unsigned adjusting_time = 0;

static void ppc_do_profile (unsigned long nip)
{
	/*
	 * Only measure the CPUs specified by /proc/irq/prof_cpu_mask.
	 * (default is all CPUs.)
	 */
	if (!((1<<smp_processor_id()) & prof_cpu_mask))
		return;

	nip -= (unsigned long) &_stext;
	nip >>= prof_shift;
	/*
	 * Don't ignore out-of-bounds EIP values silently,
	 * put them into the last histogram slot, so if
	 * present, they will show up as a sharp peak.
	 */
	if (nip > prof_len-1)
		nip = prof_len-1;
	atomic_inc((atomic_t *)&prof_buffer[nip]);
}


static __inline__ void timer_check_rtc(void)
{
        /*
         * update the rtc when needed, this should be performed on the
         * right fraction of a second. Half or full second ?
         * Full second works on mk48t59 clocks, others need testing.
         * Note that this update is basically only used through 
         * the adjtimex system calls. Setting the HW clock in
         * any other way is a /dev/rtc and userland business.
         * This is still wrong by -0.5/+1.5 jiffies because of the
         * timer interrupt resolution and possible delay, but here we 
         * hit a quantization limit which can only be solved by higher
         * resolution timers and decoupling time management from timer
         * interrupts. This is also wrong on the clocks
         * which require being written at the half second boundary.
         * We should have an rtc call that only sets the minutes and
         * seconds like on Intel to avoid problems with non UTC clocks.
         */
        if ( (time_status & STA_UNSYNC) == 0 &&
             xtime.tv_sec - last_rtc_update >= 659 &&
             abs(xtime.tv_usec - (1000000-1000000/HZ)) < 500000/HZ &&
             jiffies - wall_jiffies == 1) {
	    struct rtc_time tm;
	    to_tm(xtime.tv_sec+1, &tm);
	    tm.tm_year -= 1900;
	    tm.tm_mon -= 1;
            if (ppc_md.set_rtc_time(&tm) == 0)
                last_rtc_update = xtime.tv_sec+1;
            else
                /* Try again one minute later */
                last_rtc_update += 60;
        }
}

/* Synchronize xtime with do_gettimeofday */ 

static __inline__ void timer_sync_xtime( unsigned long cur_tb )
{
	struct timeval my_tv;

	if ( cur_tb > next_xtime_sync_tb ) {
		next_xtime_sync_tb = cur_tb + xtime_sync_interval;
		do_gettimeofday( &my_tv );
		if ( xtime.tv_sec <= my_tv.tv_sec ) {
			xtime.tv_sec = my_tv.tv_sec;
			xtime.tv_usec = my_tv.tv_usec;
		}
	}
}

#ifdef CONFIG_PPC_ISERIES

/* 
 * This function recalibrates the timebase based on the 49-bit time-of-day
 * value in the Titan chip.  The Titan is much more accurate than the value
 * returned by the service processor for the timebase frequency.  
 */

static void iSeries_tb_recal(void)
{
	struct div_result divres;
	unsigned long titan, tb;
	tb = get_tb();
	titan = HvCallXm_loadTod();
	if ( iSeries_recal_titan ) {
		unsigned long tb_ticks = tb - iSeries_recal_tb;
		unsigned long titan_usec = (titan - iSeries_recal_titan) >> 12;
		unsigned long new_tb_ticks_per_sec   = (tb_ticks * USEC_PER_SEC)/titan_usec;
		unsigned long new_tb_ticks_per_jiffy = (new_tb_ticks_per_sec+(HZ/2))/HZ;
		long tick_diff = new_tb_ticks_per_jiffy - tb_ticks_per_jiffy;
		char sign = '+';		
		/* make sure tb_ticks_per_sec and tb_ticks_per_jiffy are consistent */
		new_tb_ticks_per_sec = new_tb_ticks_per_jiffy * HZ;

		if ( tick_diff < 0 ) {
			tick_diff = -tick_diff;
			sign = '-';
		}
		if ( tick_diff ) {
			if ( tick_diff < tb_ticks_per_jiffy/25 ) {
				printk( "Titan recalibrate: new tb_ticks_per_jiffy = %lu (%c%ld)\n",
						new_tb_ticks_per_jiffy, sign, tick_diff );
				tb_ticks_per_jiffy = new_tb_ticks_per_jiffy;
				tb_ticks_per_sec   = new_tb_ticks_per_sec;
				div128_by_32( XSEC_PER_SEC, 0, tb_ticks_per_sec, &divres );
				systemcfg->tb_ticks_per_sec = tb_ticks_per_sec;
				tb_to_xs = divres.result_low;
				systemcfg->tb_to_xs = tb_to_xs;
			}
			else {
				printk( "Titan recalibrate: FAILED (difference > 4 percent)\n"
					"                   new tb_ticks_per_jiffy = %lu\n"
					"                   old tb_ticks_per_jiffy = %lu\n",
					new_tb_ticks_per_jiffy, tb_ticks_per_jiffy );
			}
		}
	}
	iSeries_recal_titan = titan;
	iSeries_recal_tb = tb;
}
#endif

/*
 * For iSeries shared processors, we have to let the hypervisor
 * set the hardware decrementer.  We set a virtual decrementer
 * in the ItLpPaca and call the hypervisor if the virtual
 * decrementer is less than the current value in the hardware
 * decrementer. (almost always the new decrementer value will
 * be greater than the current hardware decementer so the hypervisor
 * call will not be needed)
 */

unsigned long tb_last_stamp=0;

/*
 * timer_interrupt - gets called when the decrementer overflows,
 * with interrupts disabled.
 */
int timer_interrupt(struct pt_regs * regs)
{
	int next_dec;
	unsigned long cur_tb;
	struct paca_struct *lpaca = get_paca();
	unsigned long cpu = lpaca->xPacaIndex;
	struct ItLpQueue * lpq;

	irq_enter(cpu);

	if ((!user_mode(regs)) && (prof_buffer))
		ppc_do_profile(instruction_pointer(regs));

	pmc_timeslice_tick(); /* Hack this in for now */

	lpaca->xLpPaca.xIntDword.xFields.xDecrInt = 0;

	while (lpaca->next_jiffy_update_tb <= (cur_tb = get_tb())) {

#ifdef CONFIG_SMP
		smp_local_timer_interrupt(regs);
#endif
		if (cpu == 0) {
			write_lock(&xtime_lock);
			tb_last_stamp = lpaca->next_jiffy_update_tb;
			do_timer(regs);
			timer_sync_xtime( cur_tb );
			timer_check_rtc();
			write_unlock(&xtime_lock);
			if ( adjusting_time && (time_adjust == 0) )
				ppc_adjtimex();
		}
		lpaca->next_jiffy_update_tb += tb_ticks_per_jiffy;
	}
	
	next_dec = lpaca->next_jiffy_update_tb - cur_tb;
	if (next_dec > lpaca->default_decr)
        	next_dec = lpaca->default_decr;
	set_dec(next_dec);

	lpq = lpaca->lpQueuePtr;
	if (lpq && ItLpQueue_isLpIntPending(lpq))
		lpEvent_count += ItLpQueue_process(lpq, regs); 

	irq_exit(cpu);

	if (softirq_pending(cpu))
		do_softirq();
	
	return 1;
}


/*
 * This version of gettimeofday has microsecond resolution.
 */
void do_gettimeofday(struct timeval *tv)
{
        unsigned long sec, usec, tb_ticks;
	unsigned long xsec, tb_xsec;
	unsigned long temp_tb_to_xs, temp_stamp_xsec;
	unsigned long tb_count_1, tb_count_2;
	unsigned long always_zero;
	struct systemcfg *gtdp;
	
	gtdp = systemcfg;
	/* 
	 * The following loop guarantees that we see a consistent view of the
	 * tb_to_xs and stamp_xsec variables.  These two variables can change
	 * (eg. when xntpd adjusts the clock frequency) and an inconsistent
	 * view (one variable changed, the other not) could result in a wildly
	 * wrong result for do_gettimeofday. 
	 *
	 * The code which updates these variables (ppc_adjtimex below)
	 * increments tb_update_count, then updates the two variables and then
	 * increments tb_update_count again.  This code reads tb_update_count,
	 * reads the two variables and then reads tb_update_count again.  It
	 * loops doing this until the two reads of tb_update_count yield the
	 * same value and that value is even.  This ensures a consistent view
	 * of the two variables.
	 *
	 * The strange looking assembler code below causes the hardware to
	 * think that reading the two variables is dependent on the first read
	 * of tb_update_count and that the second reading of tb_update_count is
	 * dependent on reading the two variables.  This assures ordering
	 * without the need for a lwsync, which is much more expensive.
	 */
	do {
		tb_ticks = get_tb() - gtdp->tb_orig_stamp;

		tb_count_1 = gtdp->tb_update_count;

		__asm__ __volatile__ (
"		andc 	%0,%2,%2\n\
		add	%1,%3,%0\n\
"		: "=&r"(always_zero), "=r"(gtdp)
		: "r"(tb_count_1), "r"(gtdp) );

		temp_tb_to_xs = gtdp->tb_to_xs;
		temp_stamp_xsec = gtdp->stamp_xsec;

		__asm__ __volatile__ (
"		add	%0,%2,%3\n\
		andc	%0,%0,%0\n\
		add	%1,%4,%0\n\
"		: "=&r"(always_zero), "=r"(gtdp)
		: "r"(temp_stamp_xsec), "r"(temp_tb_to_xs), "r"(gtdp) );

		tb_count_2 = gtdp->tb_update_count;

	} while ( tb_count_2 - ( tb_count_1 & 0xfffffffffffffffe ) ); 

	/* These calculations are faster (gets rid of divides)
	 * if done in units of 1/2^20 rather than microseconds.
	 * The conversion to microseconds at the end is done
	 * without a divide (and in fact, without a multiply) */
	tb_xsec = mulhdu( tb_ticks, temp_tb_to_xs );
	xsec = temp_stamp_xsec + tb_xsec;
	sec = xsec / XSEC_PER_SEC;
	xsec -= sec * XSEC_PER_SEC;
	usec = (xsec * USEC_PER_SEC)/XSEC_PER_SEC;

        tv->tv_sec = sec;
        tv->tv_usec = usec;
}

void do_settimeofday(struct timeval *tv)
{
	unsigned long flags;
	unsigned long delta_xsec;
	long int tb_delta, new_usec, new_sec;
	unsigned long new_xsec;

	write_lock_irqsave(&xtime_lock, flags);
	/* Updating the RTC is not the job of this code. If the time is
	 * stepped under NTP, the RTC will be update after STA_UNSYNC
	 * is cleared. Tool like clock/hwclock either copy the RTC
	 * to the system time, in which case there is no point in writing
	 * to the RTC again, or write to the RTC but then they don't call
	 * settimeofday to perform this operation.
	 */
#ifdef CONFIG_PPC_ISERIES
	if ( first_settimeofday ) {
		iSeries_tb_recal();
		first_settimeofday = 0;
	}
#endif
	tb_delta = tb_ticks_since(tb_last_stamp);
	tb_delta += (jiffies - wall_jiffies) * tb_ticks_per_jiffy;

	new_sec = tv->tv_sec;
	new_usec = tv->tv_usec - tb_delta / tb_ticks_per_usec;
	while (new_usec <0) {
		new_sec--; 
		new_usec += USEC_PER_SEC;