linux_old1/arch/ia64/kernel/time.c

305 lines
8.2 KiB
C
Raw Normal View History

/*
* linux/arch/ia64/kernel/time.c
*
* Copyright (C) 1998-2003 Hewlett-Packard Co
* Stephane Eranian <eranian@hpl.hp.com>
* David Mosberger <davidm@hpl.hp.com>
* Copyright (C) 1999 Don Dugger <don.dugger@intel.com>
* Copyright (C) 1999-2000 VA Linux Systems
* Copyright (C) 1999-2000 Walt Drummond <drummond@valinux.com>
*/
#include <linux/cpu.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/profile.h>
#include <linux/sched.h>
#include <linux/time.h>
#include <linux/interrupt.h>
#include <linux/efi.h>
#include <linux/profile.h>
#include <linux/timex.h>
#include <asm/machvec.h>
#include <asm/delay.h>
#include <asm/hw_irq.h>
#include <asm/ptrace.h>
#include <asm/sal.h>
#include <asm/sections.h>
#include <asm/system.h>
volatile int time_keeper_id = 0; /* smp_processor_id() of time-keeper */
#ifdef CONFIG_IA64_DEBUG_IRQ
unsigned long last_cli_ip;
EXPORT_SYMBOL(last_cli_ip);
#endif
static struct time_interpolator itc_interpolator = {
.shift = 16,
.mask = 0xffffffffffffffffLL,
.source = TIME_SOURCE_CPU
};
static irqreturn_t
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 21:55:46 +08:00
timer_interrupt (int irq, void *dev_id)
{
unsigned long new_itm;
if (unlikely(cpu_is_offline(smp_processor_id()))) {
return IRQ_HANDLED;
}
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 21:55:46 +08:00
platform_timer_interrupt(irq, dev_id);
new_itm = local_cpu_data->itm_next;
if (!time_after(ia64_get_itc(), new_itm))
printk(KERN_ERR "Oops: timer tick before it's due (itc=%lx,itm=%lx)\n",
ia64_get_itc(), new_itm);
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 21:55:46 +08:00
profile_tick(CPU_PROFILING);
while (1) {
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 21:55:46 +08:00
update_process_times(user_mode(get_irq_regs()));
new_itm += local_cpu_data->itm_delta;
if (smp_processor_id() == time_keeper_id) {
/*
* Here we are in the timer irq handler. We have irqs locally
* disabled, but we don't know if the timer_bh is running on
* another CPU. We need to avoid to SMP race by acquiring the
* xtime_lock.
*/
write_seqlock(&xtime_lock);
[PATCH] simplify update_times (avoid jiffies/jiffies_64 aliasing problem) Pass ticks to do_timer() and update_times(), and adjust x86_64 and s390 timer interrupt handler with this change. Currently update_times() calculates ticks by "jiffies - wall_jiffies", but callers of do_timer() should know how many ticks to update. Passing ticks get rid of this redundant calculation. Also there are another redundancy pointed out by Martin Schwidefsky. This cleanup make a barrier added by 5aee405c662ca644980c184774277fc6d0769a84 needless. So this patch removes it. As a bonus, this cleanup make wall_jiffies can be removed easily, since now wall_jiffies is always synced with jiffies. (This patch does not really remove wall_jiffies. It would be another cleanup patch) Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@elte.hu> Cc: john stultz <johnstul@us.ibm.com> Cc: Andi Kleen <ak@muc.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Acked-by: Russell King <rmk@arm.linux.org.uk> Cc: Ian Molton <spyro@f2s.com> Cc: Mikael Starvik <starvik@axis.com> Acked-by: David Howells <dhowells@redhat.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Hirokazu Takata <takata.hirokazu@renesas.com> Acked-by: Ralf Baechle <ralf@linux-mips.org> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Kazumoto Kojima <kkojima@rr.iij4u.or.jp> Cc: Richard Curnow <rc@rc0.org.uk> Cc: William Lee Irwin III <wli@holomorphy.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Jeff Dike <jdike@addtoit.com> Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Cc: Miles Bader <uclinux-v850@lsi.nec.co.jp> Cc: Chris Zankel <chris@zankel.net> Acked-by: "Luck, Tony" <tony.luck@intel.com> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Roman Zippel <zippel@linux-m68k.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-09-29 17:00:32 +08:00
do_timer(1);
local_cpu_data->itm_next = new_itm;
write_sequnlock(&xtime_lock);
} else
local_cpu_data->itm_next = new_itm;
if (time_after(new_itm, ia64_get_itc()))
break;
}
do {
/*
* If we're too close to the next clock tick for
* comfort, we increase the safety margin by
* intentionally dropping the next tick(s). We do NOT
* update itm.next because that would force us to call
* do_timer() which in turn would let our clock run
* too fast (with the potentially devastating effect
* of losing monotony of time).
*/
while (!time_after(new_itm, ia64_get_itc() + local_cpu_data->itm_delta/2))
new_itm += local_cpu_data->itm_delta;
ia64_set_itm(new_itm);
/* double check, in case we got hit by a (slow) PMI: */
} while (time_after_eq(ia64_get_itc(), new_itm));
return IRQ_HANDLED;
}
/*
* Encapsulate access to the itm structure for SMP.
*/
void
ia64_cpu_local_tick (void)
{
int cpu = smp_processor_id();
unsigned long shift = 0, delta;
/* arrange for the cycle counter to generate a timer interrupt: */
ia64_set_itv(IA64_TIMER_VECTOR);
delta = local_cpu_data->itm_delta;
/*
* Stagger the timer tick for each CPU so they don't occur all at (almost) the
* same time:
*/
if (cpu) {
unsigned long hi = 1UL << ia64_fls(cpu);
shift = (2*(cpu - hi) + 1) * delta/hi/2;
}
local_cpu_data->itm_next = ia64_get_itc() + delta + shift;
ia64_set_itm(local_cpu_data->itm_next);
}
static int nojitter;
static int __init nojitter_setup(char *str)
{
nojitter = 1;
printk("Jitter checking for ITC timers disabled\n");
return 1;
}
__setup("nojitter", nojitter_setup);
void __devinit
ia64_init_itm (void)
{
unsigned long platform_base_freq, itc_freq;
struct pal_freq_ratio itc_ratio, proc_ratio;
long status, platform_base_drift, itc_drift;
/*
* According to SAL v2.6, we need to use a SAL call to determine the platform base
* frequency and then a PAL call to determine the frequency ratio between the ITC
* and the base frequency.
*/
status = ia64_sal_freq_base(SAL_FREQ_BASE_PLATFORM,
&platform_base_freq, &platform_base_drift);
if (status != 0) {
printk(KERN_ERR "SAL_FREQ_BASE_PLATFORM failed: %s\n", ia64_sal_strerror(status));
} else {
status = ia64_pal_freq_ratios(&proc_ratio, NULL, &itc_ratio);
if (status != 0)
printk(KERN_ERR "PAL_FREQ_RATIOS failed with status=%ld\n", status);
}
if (status != 0) {
/* invent "random" values */
printk(KERN_ERR
"SAL/PAL failed to obtain frequency info---inventing reasonable values\n");
platform_base_freq = 100000000;
platform_base_drift = -1; /* no drift info */
itc_ratio.num = 3;
itc_ratio.den = 1;
}
if (platform_base_freq < 40000000) {
printk(KERN_ERR "Platform base frequency %lu bogus---resetting to 75MHz!\n",
platform_base_freq);
platform_base_freq = 75000000;
platform_base_drift = -1;
}
if (!proc_ratio.den)
proc_ratio.den = 1; /* avoid division by zero */
if (!itc_ratio.den)
itc_ratio.den = 1; /* avoid division by zero */
itc_freq = (platform_base_freq*itc_ratio.num)/itc_ratio.den;
local_cpu_data->itm_delta = (itc_freq + HZ/2) / HZ;
printk(KERN_DEBUG "CPU %d: base freq=%lu.%03luMHz, ITC ratio=%u/%u, "
"ITC freq=%lu.%03luMHz", smp_processor_id(),
platform_base_freq / 1000000, (platform_base_freq / 1000) % 1000,
itc_ratio.num, itc_ratio.den, itc_freq / 1000000, (itc_freq / 1000) % 1000);
if (platform_base_drift != -1) {
itc_drift = platform_base_drift*itc_ratio.num/itc_ratio.den;
printk("+/-%ldppm\n", itc_drift);
} else {
itc_drift = -1;
printk("\n");
}
local_cpu_data->proc_freq = (platform_base_freq*proc_ratio.num)/proc_ratio.den;
local_cpu_data->itc_freq = itc_freq;
local_cpu_data->cyc_per_usec = (itc_freq + USEC_PER_SEC/2) / USEC_PER_SEC;
local_cpu_data->nsec_per_cyc = ((NSEC_PER_SEC<<IA64_NSEC_PER_CYC_SHIFT)
+ itc_freq/2)/itc_freq;
if (!(sal_platform_features & IA64_SAL_PLATFORM_FEATURE_ITC_DRIFT)) {
itc_interpolator.frequency = local_cpu_data->itc_freq;
itc_interpolator.drift = itc_drift;
#ifdef CONFIG_SMP
/* On IA64 in an SMP configuration ITCs are never accurately synchronized.
* Jitter compensation requires a cmpxchg which may limit
* the scalability of the syscalls for retrieving time.
* The ITC synchronization is usually successful to within a few
* ITC ticks but this is not a sure thing. If you need to improve
* timer performance in SMP situations then boot the kernel with the
* "nojitter" option. However, doing so may result in time fluctuating (maybe
* even going backward) if the ITC offsets between the individual CPUs
* are too large.
*/
if (!nojitter) itc_interpolator.jitter = 1;
#endif
register_time_interpolator(&itc_interpolator);
}
/* Setup the CPU local timer tick */
ia64_cpu_local_tick();
}
static struct irqaction timer_irqaction = {
.handler = timer_interrupt,
.flags = IRQF_DISABLED,
.name = "timer"
};
void __devinit ia64_disable_timer(void)
{
ia64_set_itv(1 << 16);
}
void __init
time_init (void)
{
register_percpu_irq(IA64_TIMER_VECTOR, &timer_irqaction);
efi_gettimeofday(&xtime);
ia64_init_itm();
/*
* Initialize wall_to_monotonic such that adding it to xtime will yield zero, the
* tv_nsec field must be normalized (i.e., 0 <= nsec < NSEC_PER_SEC).
*/
set_normalized_timespec(&wall_to_monotonic, -xtime.tv_sec, -xtime.tv_nsec);
}
/*
* Generic udelay assumes that if preemption is allowed and the thread
* migrates to another CPU, that the ITC values are synchronized across
* all CPUs.
*/
static void
ia64_itc_udelay (unsigned long usecs)
{
unsigned long start = ia64_get_itc();
unsigned long end = start + usecs*local_cpu_data->cyc_per_usec;
while (time_before(ia64_get_itc(), end))
cpu_relax();
}
void (*ia64_udelay)(unsigned long usecs) = &ia64_itc_udelay;
void
udelay (unsigned long usecs)
{
(*ia64_udelay)(usecs);
}
EXPORT_SYMBOL(udelay);
static unsigned long long ia64_itc_printk_clock(void)
{
if (ia64_get_kr(IA64_KR_PER_CPU_DATA))
return sched_clock();
return 0;
}
static unsigned long long ia64_default_printk_clock(void)
{
return (unsigned long long)(jiffies_64 - INITIAL_JIFFIES) *
(1000000000/HZ);
}
unsigned long long (*ia64_printk_clock)(void) = &ia64_default_printk_clock;
unsigned long long printk_clock(void)
{
return ia64_printk_clock();
}
void __init
ia64_setup_printk_clock(void)
{
if (!(sal_platform_features & IA64_SAL_PLATFORM_FEATURE_ITC_DRIFT))
ia64_printk_clock = ia64_itc_printk_clock;
}