Skip to contentSkip to conversation

No idle HZ aka dynticks i386 for 2.6.16-rc4

PATCH In review Con Kolivas 2 participants 2 messages Feb 25, 2006

Subsystems (5)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 No idle HZ aka dynticks i386 for 2.6.16-rc4

I've only committed minor changes to dynticks since the last public release.

The smp idle handler was modified to wake up the APIC timer of the last cpu going to sleep when the PIC timer is skipping.

One arch specific function was moved out of common code (noted by Thomas Renninger thanks).

The fix for sys_nanosleep being converted to hrtimers breaking next_timer_interrupt was added by Tony Lindgren.

Latest rolled up patch, split patch series and tarball available here:
http://ck.kolivas.org/patches/dyn-ticks/

Uniprocessor dynticks has been working flawlessly for some time.

Known issues:
SMP still isn't playing well. Too many ticks still occur on the irq0 cpu and eventually other cpus seem to go into a sustained long skip cycle (after a day or so.) I haven't been able to track the cause of either of these and would appreciate anyone's outside help if they're knowledgeable in this area.

Rolled up patch v060225 included here.

Cheers,
Con
---
Rolled up dynticks patch including arch independent code and i386 implementation.

 arch/i386/Kconfig                       |   21 ++
 arch/i386/defconfig                     |    1
 arch/i386/kernel/Makefile               |    1
 arch/i386/kernel/apic.c                 |   34 +++
 arch/i386/kernel/dyntick.c              |  279 +++++++++++++++++++++++++++++
 arch/i386/kernel/irq.c                  |    4
 arch/i386/kernel/process.c              |    4
 arch/i386/kernel/smp.c                  |    9
 arch/i386/kernel/time.c                 |   19 +-
 arch/i386/kernel/timers/timer_cyclone.c |    4
 arch/i386/kernel/timers/timer_hpet.c    |    4
 arch/i386/kernel/timers/timer_none.c    |    3
 arch/i386/kernel/timers/timer_pit.c     |   22 ++
 arch/i386/kernel/timers/timer_pm.c      |   32 +--
 arch/i386/kernel/timers/timer_tsc.c     |   64 ++----
 drivers/acpi/Kconfig                    |    2
 drivers/acpi/processor_idle.c           |  108 +++++++++--
 drivers/input/serio/i8042.h             |    2
 include/acpi/processor.h                |    3
 include/asm-i386/apic.h                 |    2
 include/asm-i386/dyntick.h              |   70 +++++++
 include/asm-i386/timer.h                |   37 +++
 include/linux/dyntick.h                 |  170 +++++++++++++++++
 include/linux/hrtimer.h                 |    1
 include/linux/timer.h                   |    1
 kernel/Makefile                         |    2
 kernel/dyntick.c                        |  304 ++++++++++++++++++++++++++++++++
 kernel/hrtimer.c                        |   73 +++++++
 kernel/timer.c                          |   21 ++
 kernel/timer_top.c                      |  234 ++++++++++++++++++++++++
 lib/Kconfig.debug                       |   13 +
 31 files changed, 1459 insertions(+), 85 deletions(-)
include/linux/timer.h +1 -0
Index: linux-2.6.16-rc4-dt/include/linux/timer.h
===================================================================
@@ -97,5 +97,6 @@ static inline void add_timer(struct time
 extern void init_timers(void);
 extern void run_local_timers(void);
 extern int it_real_fn(void *);
+extern void conditional_run_local_timers(void);
 
 #endif
kernel/Makefile +2 -0
Index: linux-2.6.16-rc4-dt/kernel/Makefile
===================================================================
@@ -34,6 +34,8 @@ obj-$(CONFIG_DETECT_SOFTLOCKUP) += softl
 obj-$(CONFIG_GENERIC_HARDIRQS) += irq/
 obj-$(CONFIG_SECCOMP) += seccomp.o
 obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o
+obj-$(CONFIG_NO_IDLE_HZ) += dyntick.o
+obj-$(CONFIG_TIMER_INFO) += timer_top.o
 
 ifneq ($(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER),y)
 # According to Alan Modra <[email protected]>, the -fno-omit-frame-pointer is
kernel/timer.c +21 -0
Index: linux-2.6.16-rc4-dt/kernel/timer.c
===================================================================
@@ -34,6 +34,7 @@
 #include <linux/cpu.h>
 #include <linux/syscalls.h>
 #include <linux/delay.h>
+#include <linux/dyntick.h>
 
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
@@ -478,6 +479,7 @@ static inline void __run_timers(tvec_bas
 }
 
 #ifdef CONFIG_NO_IDLE_HZ
+
 /*
  * Find out when the next timer event is due to happen. This
  * is used on S/390 to stop all activity when a cpus is idle.
@@ -489,9 +491,15 @@ unsigned long next_timer_interrupt(void)
 	struct list_head *list;
 	struct timer_list *nte;
 	unsigned long expires;
+	unsigned long hr_expires = jiffies + 10 * HZ;	/* Anything far ahead */
 	tvec_t *varray[4];
 	int i, j;
 
+	/* Look for timer events in hrtimer. */
+	if ((hrtimer_next_jiffie(&hr_expires) == 0)
+		&& (time_before(hr_expires, jiffies + 2)))
+			return hr_expires;
+
 	base = &__get_cpu_var(tvec_bases);
 	spin_lock(&base->t_base.lock);
 	expires = base->timer_jiffies + (LONG_MAX >> 1);
@@ -541,7 +549,13 @@ found:
 				expires = nte->expires;
 		}
 	}
+	account_timer((unsigned long)nte->function, nte->data);
+
 	spin_unlock(&base->t_base.lock);
+
+	if (time_before(hr_expires, expires))
+		expires = hr_expires;
+
 	return expires;
 }
 #endif
@@ -900,6 +914,13 @@ void run_local_timers(void)
 	raise_softirq(TIMER_SOFTIRQ);
 }
 
+void conditional_run_local_timers(void)
+{
+	tvec_base_t *base  = &__get_cpu_var(tvec_bases);
+
+	if (base->timer_jiffies != jiffies)
+		run_local_timers();
+}
 /*
  * Called by the timer interrupt. xtime_lock must already be taken
  * by the timer IRQ!
include/linux/dyntick.h new file +170 -0

170 lines not on this page · raw message

kernel/dyntick.c new file +304 -0

304 lines not on this page · raw message

arch/i386/kernel/time.c +13 -6
Index: linux-2.6.16-rc4-dt/arch/i386/kernel/time.c
===================================================================
@@ -46,6 +46,7 @@
 #include <linux/bcd.h>
 #include <linux/efi.h>
 #include <linux/mca.h>
+#include <linux/dyntick.h>
 
 #include <asm/io.h>
 #include <asm/smp.h>
@@ -56,6 +57,7 @@
 #include <asm/uaccess.h>
 #include <asm/processor.h>
 #include <asm/timer.h>
+#include <asm/dyntick.h>
 
 #include "mach_time.h"
 
@@ -245,7 +247,7 @@ EXPORT_SYMBOL(profile_pc);
  * timer_interrupt() needs to keep up the real-time clock,
  * as well as call the "do_timer()" routine every clocktick
  */
-static inline void do_timer_interrupt(int irq, struct pt_regs *regs)
+static inline void do_timer_interrupt(int irq, struct pt_regs *regs, int lost)
 {
 #ifdef CONFIG_X86_IO_APIC
 	if (timer_ack) {
@@ -263,7 +265,8 @@ static inline void do_timer_interrupt(in
 	}
 #endif
 
-	do_timer_interrupt_hook(regs);
+	if (!dyntick_enabled() || lost)
+		do_timer_interrupt_hook(regs);
 
 
 	if (MCA_bus) {
@@ -288,6 +291,8 @@ static inline void do_timer_interrupt(in
  */
 irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
+	int lost;
+
 	/*
 	 * Here we are in the timer irq handler. We just have irqs locally
 	 * disabled but we don't know if the timer_bh is running on the other
@@ -297,9 +302,9 @@ irqreturn_t timer_interrupt(int irq, voi
 	 */
 	write_seqlock(&xtime_lock);
 
-	cur_timer->mark_offset();
- 
-	do_timer_interrupt(irq, regs);
+	lost = cur_timer->mark_offset();
+
+	do_timer_interrupt(irq, regs, lost);
 
 	write_sequnlock(&xtime_lock);
 
@@ -431,7 +436,7 @@ static struct sysdev_class timer_sysclas
 
 
 /* XXX this driverfs stuff should probably go elsewhere later -john */
-static struct sys_device device_timer = {
+struct sys_device device_timer = {
 	.id	= 0,
 	.cls	= &timer_sysclass,
 };
@@ -487,5 +492,7 @@ void __init time_init(void)
 	cur_timer = select_timer();
 	printk(KERN_INFO "Using %s for high-res timesource\n",cur_timer->name);
 
+	dyntick_time_init(cur_timer);
+
 	time_init_hook();
 }
arch/i386/kernel/timers/timer_cyclone.c +3 -1
Index: linux-2.6.16-rc4-dt/arch/i386/kernel/timers/timer_cyclone.c
===================================================================
@@ -45,7 +45,7 @@ static seqlock_t monotonic_lock = SEQLOC
 	} while (high != cyclone_timer[1]);
 
 
-static void mark_offset_cyclone(void)
+static int mark_offset_cyclone(void)
 {
 	unsigned long lost, delay;
 	unsigned long delta = last_cyclone_low;
@@ -101,6 +101,8 @@ static void mark_offset_cyclone(void)
 	 */
 	if (lost && abs(delay - delay_at_last_interrupt) > (900000/HZ))
 		jiffies_64++;
+
+	return lost;
 }
 
 static unsigned long get_offset_cyclone(void)
arch/i386/kernel/timers/timer_hpet.c +3 -1
Index: linux-2.6.16-rc4-dt/arch/i386/kernel/timers/timer_hpet.c
===================================================================
@@ -101,7 +101,7 @@ static unsigned long get_offset_hpet(voi
 	return edx;
 }
 
-static void mark_offset_hpet(void)
+static int mark_offset_hpet(void)
 {
 	unsigned long long this_offset, last_offset;
 	unsigned long offset;
@@ -124,6 +124,8 @@ static void mark_offset_hpet(void)
 	this_offset = ((unsigned long long)last_tsc_high<<32)|last_tsc_low;
 	monotonic_base += cycles_2_ns(this_offset - last_offset);
 	write_sequnlock(&monotonic_lock);
+
+	return 1;
 }
 
 static void delay_hpet(unsigned long loops)
arch/i386/kernel/timers/timer_none.c +2 -1
Index: linux-2.6.16-rc4-dt/arch/i386/kernel/timers/timer_none.c
===================================================================
@@ -1,9 +1,10 @@
 #include <linux/init.h>
 #include <asm/timer.h>
 
-static void mark_offset_none(void)
+static int mark_offset_none(void)
 {
 	/* nothing needed */
+	return 1;
 }
 
 static unsigned long get_offset_none(void)
arch/i386/kernel/timers/timer_pit.c +21 -1
Index: linux-2.6.16-rc4-dt/arch/i386/kernel/timers/timer_pit.c
===================================================================
@@ -32,9 +32,10 @@ static int __init init_pit(char* overrid
 	return 0;
 }
 
-static void mark_offset_pit(void)
+static int mark_offset_pit(void)
 {
 	/* nothing needed */
+	return 1;
 }
 
 static unsigned long long monotonic_clock_pit(void)
@@ -148,6 +149,25 @@ static unsigned long get_offset_pit(void
 	return count;
 }
 
+/*
+ * Reprograms the next timer interrupt
+ * PIT timer reprogramming code taken from APM code.
+ * Note that PIT timer is a 16-bit timer.
+ * Called with irqs already disabled.
+ */
+void reprogram_pit_timer(unsigned long jiffies_to_skip)
+{
+	int skip = jiffies_to_skip * LATCH;
+
+	if (skip > 0xffff)
+		skip = 0xffff;
+
+	spin_lock(&i8253_lock);
+	outb_p(0x34, PIT_MODE);		/* binary, mode 2, LSB/MSB, ch 0 */
+	outb_p(skip & 0xff, PIT_CH0);	/* LSB */
+	outb(skip >> 8, PIT_CH0);	/* MSB */
+	spin_unlock(&i8253_lock);
+}
 
 /* tsc timer_opts struct */
 struct timer_opts timer_pit = {
arch/i386/kernel/timers/timer_pm.c +15 -17
Index: linux-2.6.16-rc4-dt/arch/i386/kernel/timers/timer_pm.c
===================================================================
@@ -15,6 +15,7 @@
 #include <linux/module.h>
 #include <linux/device.h>
 #include <linux/init.h>
+#include <linux/dyntick.h>
 #include <asm/types.h>
 #include <asm/timer.h>
 #include <asm/smp.h>
@@ -28,7 +29,7 @@
 #define PMTMR_TICKS_PER_SEC 3579545
 #define PMTMR_EXPECTED_RATE \
   ((CALIBRATE_LATCH * (PMTMR_TICKS_PER_SEC >> 10)) / (CLOCK_TICK_RATE>>10))
-
+#define PMTMR_TICKS_PER_JIFFY (PMTMR_EXPECTED_RATE / (CALIBRATE_LATCH/LATCH))
 
 /* The I/O port the PMTMR resides at.
  * The location is detected during setup_arch(),
@@ -127,7 +128,13 @@ pm_good:
 	if (verify_pmtmr_rate() != 0)
 		return -ENODEV;
 
+	printk("Using %u PM timer ticks per jiffy \n", PMTMR_TICKS_PER_JIFFY);
+
+	offset_tick = read_pmtmr();
+	setup_pit_timer();
+
 	init_cpu_khz();
+	set_dyntick_limits(((0xFFFFFF / 1000000) * 286 * HZ) >> 10, 0);
 	return 0;
 }
 
@@ -148,10 +155,9 @@ static inline u32 cyc2us(u32 cycles)
  * this gets called during each timer interrupt
  *   - Called while holding the writer xtime_lock
  */
-static void mark_offset_pmtmr(void)
+static int mark_offset_pmtmr(void)
 {
 	u32 lost, delta, last_offset;
-	static int first_run = 1;
 	last_offset = offset_tick;
 
 	write_seqlock(&monotonic_lock);
@@ -161,29 +167,20 @@ static void mark_offset_pmtmr(void)
 	/* calculate tick interval */
 	delta = (offset_tick - last_offset) & ACPI_PM_MASK;
 
-	/* convert to usecs */
-	delta = cyc2us(delta);
-
 	/* update the monotonic base value */
-	monotonic_base += delta * NSEC_PER_USEC;
+	monotonic_base += cyc2us(delta) * NSEC_PER_USEC;
 	write_sequnlock(&monotonic_lock);
 
 	/* convert to ticks */
 	delta += offset_delay;
-	lost = delta / (USEC_PER_SEC / HZ);
-	offset_delay = delta % (USEC_PER_SEC / HZ);
-
+	lost = delta / PMTMR_TICKS_PER_JIFFY;
+	offset_delay = delta % PMTMR_TICKS_PER_JIFFY;
 
 	/* compensate for lost ticks */
 	if (lost >= 2)
 		jiffies_64 += lost - 1;
 
-	/* don't calculate delay for first run,
-	   or if we've got less then a tick */
-	if (first_run || (lost < 1)) {
-		first_run = 0;
-		offset_delay = 0;
-	}
+	return lost;
 }
 
 static int pmtmr_resume(void)
@@ -192,6 +189,7 @@ static int pmtmr_resume(void)
 	/* Assume this is the last mark offset time */
 	offset_tick = read_pmtmr();
 	write_sequnlock(&monotonic_lock);
+	offset_delay = 0;
 	return 0;
 }
 
@@ -243,7 +241,7 @@ static unsigned long get_offset_pmtmr(vo
 	now = read_pmtmr();
 	delta = (now - offset)&ACPI_PM_MASK;
 
-	return (unsigned long) offset_delay + cyc2us(delta);
+	return (unsigned long) cyc2us(delta + offset_delay);
 }
 
 
arch/i386/kernel/timers/timer_tsc.c +28 -36

143 lines not on this page · raw message

include/asm-i386/timer.h +36 -1
Index: linux-2.6.16-rc4-dt/include/asm-i386/timer.h
===================================================================
@@ -1,5 +1,6 @@
 #ifndef _ASMi386_TIMER_H
 #define _ASMi386_TIMER_H
+#include <linux/jiffies.h>
 #include <linux/init.h>
 #include <linux/pm.h>
 
@@ -19,7 +20,7 @@
  */
 struct timer_opts {
 	char* name;
-	void (*mark_offset)(void);
+	int (*mark_offset)(void);
 	unsigned long (*get_offset)(void);
 	unsigned long long (*monotonic_clock)(void);
 	void (*delay)(unsigned long);
@@ -38,6 +39,7 @@ struct init_timer_opts {
 extern struct timer_opts* __init select_timer(void);
 extern void clock_fallback(void);
 void setup_pit_timer(void);
+extern void reprogram_pit_timer(unsigned long jiffies_to_skip);
 
 /* Modifiers for buggy PIT handling */
 
@@ -67,4 +69,37 @@ extern unsigned long calibrate_tsc_hpet(
 #ifdef CONFIG_X86_PM_TIMER
 extern struct init_timer_opts timer_pmtmr_init;
 #endif
+#ifdef CONFIG_NO_IDLE_HZ
+static inline void tsc_sanity_check(int lost)
+{
+}
+#else /* CONFIG_NO_IDLE_HZ */
+extern void cpufreq_delayed_get(void);
+
+static inline void tsc_sanity_check(int lost)
+{
+	static int lost_count = 0;
+
+	if (lost >= 2) {
+		jiffies_64 += lost-1;
+
+		/* sanity check to ensure we're not always losing ticks */
+		if (lost_count++ > 100) {
+			printk(KERN_WARNING "Losing too many ticks!\n");
+			printk(KERN_WARNING "TSC cannot be used as a timesource.  \n");
+			printk(KERN_WARNING "Possible reasons for this are:\n");
+			printk(KERN_WARNING "  You're running with Speedstep,\n");
+			printk(KERN_WARNING "  You don't have DMA enabled for your hard disk (see hdparm),\n");
+			printk(KERN_WARNING "  Incorrect TSC synchronization on an SMP system (see dmesg).\n");
+			printk(KERN_WARNING "Falling back to a sane timesource now.\n");
+
+			clock_fallback();
+		}
+		/* ... but give the TSC a fair chance */
+		if (lost_count > 25)
+			cpufreq_delayed_get();
+	} else
+		lost_count = 0;
+}
+#endif /* CONFIG_NO_IDLE_HZ */
 #endif
drivers/acpi/Kconfig +2 -0
Index: linux-2.6.16-rc4-dt/drivers/acpi/Kconfig
===================================================================
@@ -298,6 +298,8 @@ config X86_PM_TIMER
 	  voltage scaling, unlike the commonly used Time Stamp Counter
 	  (TSC) timing source.
 
+	  This timer is selected by dyntick (NO_IDLE_HZ).
+
 	  So, if you see messages like 'Losing too many ticks!' in the
 	  kernel logs, and/or you are using this on a notebook which
 	  does not yet have an HPET, you should say "Y" here.
arch/i386/defconfig +1 -0
Index: linux-2.6.16-rc4-dt/arch/i386/defconfig
===================================================================
@@ -91,6 +91,7 @@ CONFIG_X86_INTEL_USERCOPY=y
 CONFIG_X86_USE_PPRO_CHECKSUM=y
 # CONFIG_HPET_TIMER is not set
 # CONFIG_HPET_EMULATE_RTC is not set
+# CONFIG_NO_IDLE_HZ is not set
 CONFIG_SMP=y
 CONFIG_NR_CPUS=8
 CONFIG_SCHED_SMT=y
arch/i386/Kconfig +21 -0
Index: linux-2.6.16-rc4-dt/arch/i386/Kconfig
===================================================================
@@ -173,6 +173,27 @@ config HPET_EMULATE_RTC
 	depends on HPET_TIMER && RTC=y
 	default y
 
+config NO_IDLE_HZ
+	bool "Dynamic Tick Timer - Skip timer ticks during idle"
+	depends on EXPERIMENTAL && X86_32
+	select X86_PM_TIMER
+	select ACPI
+	help
+	  This option enables support for skipping timer ticks when the
+	  processor is idle. During system load, timer is continuous.
+	  This option saves power, as it allows the system to stay in
+	  idle mode longer. Currently the only supported timer is ACPI PM
+	  timer.
+
+	  Note that you can disable dynamic tick timer either by
+	  passing dyntick=disable command line option, or via sysfs:
+
+	  # echo 0 > /sys/devices/system/timer/timer0/dyntick
+
+	  Most users wishing to lower their power usage while retaining
+	  low latencies will most likely want to say Y here in combination
+	  with a high HZ value (eg 1000).
+
 config SMP
 	bool "Symmetric multi-processing support"
 	---help---
arch/i386/kernel/apic.c +32 -2
Index: linux-2.6.16-rc4-dt/arch/i386/kernel/apic.c
===================================================================
@@ -27,6 +27,7 @@
 #include <linux/sysdev.h>
 #include <linux/cpu.h>
 #include <linux/module.h>
+#include <linux/dyntick.h>
 
 #include <asm/atomic.h>
 #include <asm/smp.h>
@@ -36,11 +37,13 @@
 #include <asm/arch_hooks.h>
 #include <asm/hpet.h>
 #include <asm/i8253.h>
+#include <asm/dyntick.h>
 
 #include <mach_apic.h>
 #include <mach_ipi.h>
 
 #include "io_ports.h"
+#include "do_timer.h"
 
 /*
  * cpu_mask that denotes the CPUs that needs timer interrupt coming in as
@@ -938,6 +941,8 @@ void (*wait_timer_tick)(void) __devinitd
 
 #define APIC_DIVISOR 16
 
+static u32 apic_timer_val __read_mostly;
+
 static void __setup_APIC_LVTT(unsigned int clocks)
 {
 	unsigned int lvtt_value, tmp_value, ver;
@@ -961,7 +966,9 @@ static void __setup_APIC_LVTT(unsigned i
 				& ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE))
 				| APIC_TDR_DIV_16);
 
-	apic_write_around(APIC_TMICT, clocks/APIC_DIVISOR);
+	apic_timer_val = clocks / APIC_DIVISOR;
+
+	apic_write_around(APIC_TMICT, apic_timer_val);
 }
 
 static void __devinit setup_APIC_timer(unsigned int clocks)
@@ -981,6 +988,17 @@ static void __devinit setup_APIC_timer(u
 }
 
 /*
+ * Used by NO_IDLE_HZ to skip ticks on idle CPUs. Called with IRQs already
+ * disabled
+ */
+void reprogram_apic_timer(unsigned long count)
+{
+	count = count * apic_timer_val;
+	apic_read(APIC_TMICT);
+	apic_write_around(APIC_TMICT, count);
+}
+
+/*
  * In this function we calibrate APIC bus clocks to the external
  * timer. Unfortunately we cannot use jiffies and the timer irq
  * to calibrate, since some later bootup code depends on getting
@@ -1075,6 +1093,10 @@ void __init setup_boot_APIC_clock(void)
 	 */
 	setup_APIC_timer(calibration_result);
 
+	setup_dyntick_use_apic();
+	set_dyntick_limits((0xFFFFFFFF / calibration_result) * APIC_DIVISOR,
+		2);
+
 	local_irq_restore(flags);
 }
 
@@ -1144,8 +1166,10 @@ EXPORT_SYMBOL(switch_ipi_to_APIC_timer);
  * value into /proc/profile.
  */
 
-inline void smp_local_timer_interrupt(struct pt_regs * regs)
+void smp_local_timer_interrupt(struct pt_regs * regs)
 {
+	dyntick_interrupt(regs);
+
 	profile_tick(CPU_PROFILING, regs);
 #ifdef CONFIG_SMP
 	update_process_times(user_mode_vm(regs));
@@ -1241,6 +1265,9 @@ fastcall void smp_spurious_interrupt(str
 	unsigned long v;
 
 	irq_enter();
+
+	dyntick_interrupt(regs);
+
 	/*
 	 * Check if this really is a spurious interrupt and ACK it
 	 * if it is a vectored one.  Just in case...
@@ -1265,6 +1292,9 @@ fastcall void smp_error_interrupt(struct
 	unsigned long v, v1;
 
 	irq_enter();
+
+	dyntick_interrupt(regs);
+
 	/* First tickle the hardware, only then report what went on. -- REW */
 	v = apic_read(APIC_ESR);
 	apic_write(APIC_ESR, 0);
arch/i386/kernel/dyntick.c new file +279 -0

279 lines not on this page · raw message

arch/i386/kernel/irq.c +4 -0
Index: linux-2.6.16-rc4-dt/arch/i386/kernel/irq.c
===================================================================
@@ -18,6 +18,8 @@
 #include <linux/notifier.h>
 #include <linux/cpu.h>
 #include <linux/delay.h>
+#include <linux/dyntick.h>
+#include <asm/dyntick.h>
 
 DEFINE_PER_CPU(irq_cpustat_t, irq_stat) ____cacheline_internodealigned_in_smp;
 EXPORT_PER_CPU_SYMBOL(irq_stat);
@@ -76,6 +78,8 @@ fastcall unsigned int do_IRQ(struct pt_r
 	}
 #endif
 
+	dyntick_interrupt(regs);
+
 #ifdef CONFIG_4KSTACKS
 
 	curctx = (union irq_ctx *) current_thread_info();
arch/i386/kernel/Makefile +1 -0
Index: linux-2.6.16-rc4-dt/arch/i386/kernel/Makefile
===================================================================
@@ -33,6 +33,7 @@ obj-$(CONFIG_MODULES)		+= module.o
 obj-y				+= sysenter.o vsyscall.o
 obj-$(CONFIG_ACPI_SRAT) 	+= srat.o
 obj-$(CONFIG_HPET_TIMER) 	+= time_hpet.o
+obj-$(CONFIG_NO_IDLE_HZ) 	+= dyntick.o
 obj-$(CONFIG_EFI) 		+= efi.o efi_stub.o
 obj-$(CONFIG_DOUBLEFAULT) 	+= doublefault.o
 obj-$(CONFIG_VM86)		+= vm86.o
arch/i386/kernel/process.c +4 -0
Index: linux-2.6.16-rc4-dt/arch/i386/kernel/process.c
===================================================================
@@ -39,6 +39,7 @@
 #include <linux/ptrace.h>
 #include <linux/random.h>
 #include <linux/kprobes.h>
+#include <linux/dyntick.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -57,6 +58,7 @@
 
 #include <asm/tlbflush.h>
 #include <asm/cpu.h>
+#include <asm/dyntick.h>
 
 asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
 
@@ -195,6 +197,8 @@ void cpu_idle(void)
 				play_dead();
 
 			__get_cpu_var(irq_stat).idle_timestamp = jiffies;
+			idle_reprogram_timer();
+
 			idle();
 		}
 		preempt_enable_no_resched();
arch/i386/kernel/smp.c +9 -0
Index: linux-2.6.16-rc4-dt/arch/i386/kernel/smp.c
===================================================================
@@ -20,9 +20,11 @@
 #include <linux/interrupt.h>
 #include <linux/cpu.h>
 #include <linux/module.h>
+#include <linux/dyntick.h>
 
 #include <asm/mtrr.h>
 #include <asm/tlbflush.h>
+#include <asm/dyntick.h>
 #include <mach_apic.h>
 
 /*
@@ -313,6 +315,8 @@ fastcall void smp_invalidate_interrupt(s
 {
 	unsigned long cpu;
 
+	dyntick_interrupt(regs);
+
 	cpu = get_cpu();
 
 	if (!cpu_isset(cpu, flush_cpumask))
@@ -600,6 +604,8 @@ void smp_send_stop(void)
 fastcall void smp_reschedule_interrupt(struct pt_regs *regs)
 {
 	ack_APIC_irq();
+
+	dyntick_interrupt(regs);
 }
 
 fastcall void smp_call_function_interrupt(struct pt_regs *regs)
@@ -609,6 +615,9 @@ fastcall void smp_call_function_interrup
 	int wait = call_data->wait;
 
 	ack_APIC_irq();
+
+	dyntick_interrupt(regs);
+
 	/*
 	 * Notify initiating CPU that I've grabbed the data and am
 	 * about to execute the function
include/asm-i386/apic.h +2 -0
Index: linux-2.6.16-rc4-dt/include/asm-i386/apic.h
===================================================================
@@ -121,6 +121,7 @@ extern void nmi_watchdog_tick (struct pt
 extern int APIC_init_uniprocessor (void);
 extern void disable_APIC_timer(void);
 extern void enable_APIC_timer(void);
+extern void reprogram_apic_timer(unsigned long count);
 
 extern void enable_NMI_through_LVT0 (void * dummy);
 
@@ -139,6 +140,7 @@ void switch_ipi_to_APIC_timer(void *cpum
 
 #else /* !CONFIG_X86_LOCAL_APIC */
 static inline void lapic_shutdown(void) { }
+static inline void reprogram_apic_timer(unsigned long count) { }
 
 #endif /* !CONFIG_X86_LOCAL_APIC */
 
include/asm-i386/dyntick.h new file +70 -0

70 lines not on this page · raw message

kernel/timer_top.c new file +234 -0

234 lines not on this page · raw message

lib/Kconfig.debug +13 -0
Index: linux-2.6.16-rc4-dt/lib/Kconfig.debug
===================================================================
@@ -77,6 +77,19 @@ config SCHEDSTATS
 	  application, you can say N to avoid the very slight overhead
 	  this adds.
 
+config TIMER_INFO
+	bool "Collect kernel timers statistics"
+	depends on DEBUG_KERNEL && PROC_FS && NO_IDLE_HZ
+	help
+	  If you say Y here, additional code will be inserted into the
+	  timer routines to collect statistics about kernel timers being
+	  reprogrammed through dynamic ticks feature. The statistics
+	  will be provided in /proc/timer_info and the behavior of this
+	  feature can be controlled through /proc/timer_input.
+	  The goal is to offer some output to let user applications show
+	  timer pattern usage and allow some tuning in them to
+	  maximise idle time.
+
 config DEBUG_SLAB
 	bool "Debug memory allocations"
 	depends on DEBUG_KERNEL && SLAB
drivers/input/serio/i8042.h +1 -1
Index: linux-2.6.16-rc4-dt/drivers/input/serio/i8042.h
===================================================================
@@ -44,7 +44,7 @@
  * polling.
  */
 
-#define I8042_POLL_PERIOD	HZ/20
+#define I8042_POLL_PERIOD	HZ/5
 
 /*
  * Status register bits.
drivers/acpi/processor_idle.c +90 -18

192 lines not on this page · raw message

include/acpi/processor.h +3 -0
Index: linux-2.6.16-rc4-dt/include/acpi/processor.h
===================================================================
@@ -51,6 +51,7 @@ struct acpi_processor_cx {
 	u32 latency_ticks;
 	u32 power;
 	u32 usage;
+	u64 time;
 	struct acpi_processor_cx_policy promotion;
 	struct acpi_processor_cx_policy demotion;
 };
@@ -60,8 +61,10 @@ struct acpi_processor_power {
 	unsigned long bm_check_timestamp;
 	u32 default_state;
 	u32 bm_activity;
+	u16 last_sleep;
 	int count;
 	struct acpi_processor_cx states[ACPI_PROCESSOR_MAX_POWER];
+	struct acpi_processor_cx *pre_bm_state;
 };
 
 /* Performance Management */
kernel/hrtimer.c +73 -0

79 lines not on this page · raw message

include/linux/hrtimer.h +1 -0
Index: linux-2.6.16-rc4-dt/include/linux/hrtimer.h
===================================================================
@@ -115,6 +115,7 @@ extern int hrtimer_try_to_cancel(struct 
 /* Query timers: */
 extern ktime_t hrtimer_get_remaining(const struct hrtimer *timer);
 extern int hrtimer_get_res(const clockid_t which_clock, struct timespec *tp);
+extern int hrtimer_next_jiffie(unsigned long *next_jiffie);
 
 static inline int hrtimer_active(const struct hrtimer *timer)
 {
#1 lore raw
Quoted 9 lines · Con Kolivas
include/linux/dyntick.h new file +9 -0
+struct dyntick_timer {
+	spinlock_t lock;
+
+	/* dyntick init */
+	int (*arch_init)(void);
+	/* Enables dynamic tick */
+	int (*arch_enable)(void);
+	/* Disables dynamic tick */
+	int (*arch_disable)(void);

include/linux/dyntick.h line 44

Quoted 10 lines
include/linux/dyntick.h new file +10 -0
+	/* Reprograms the timer */
+	void (*arch_reprogram)(unsigned long);
+	/* Function called when all cpus are idle, passing the idle duration */
+	void (*arch_all_cpus_idle)(unsigned int);
+
+	unsigned short state;		/* Current state */
+	unsigned int min_skip;		/* Min number of ticks to skip */
+	unsigned int max_skip;		/* Max number of ticks to skip */
+	unsigned long tick;		/* The next earliest tick */
+};

If you make min_skip a short here, the structure will pack nicely and save 6 bytes of padding on 64 bit machines. Alternatively, keep it an int and put it in the padding following the spinlock to save the whole 8 bytes.

kernel/dyntick.c line 49

Quoted 7 lines
kernel/dyntick.c new file +6 -0
+int dyntick_skipping(void)
+{
+	int ret = (get_cpu_var(dyn_cpu).next_tick > jiffies);
+
+	put_cpu_var(dyn_cpu);
+	return ret;

This looks wrong... Shouldn't it be time_after()? Otherwise it seems to not work when jiffies wraps.

kernel/dyntick.c line 72

Quoted 9 lines
kernel/dyntick.c new file +9 -0
+int dyntick_current_skip(void)
+{
+	int ret = 0;
+
+	if (get_cpu_var(dyn_cpu).next_tick > jiffies)
+		ret = __get_cpu_var(dyn_cpu).skip;
+	put_cpu_var(dyn_cpu);
+	return ret;
+}

Ditto.

kernel/dyntick.c line 87

Quoted 12 lines
kernel/dyntick.c new file +12 -0
+/*
+ * Returns the next scheduled dyntick if we are skipping ticks.
+ */
+unsigned long dyntick_next_tick(void)
+{
+	unsigned long next = 0;
+
+	if (get_cpu_var(dyn_cpu).next_tick > jiffies)
+		next = __get_cpu_var(dyn_cpu).next_tick;
+	put_cpu_var(dyn_cpu);
+	return next;
+}

Ditto.

Quoted 4 lines
kernel/dyntick.c new file +4 -0
+/*
+ * dyn_early_reprogram is used to reprogram an earlier tick than is currently
+ * set by timer_dyn_reprogram.
+ * dyn_early_reprogram allows other code such as the acpi idle code to

kernel/dyntick.c line 160

Quoted 10 lines
kernel/dyntick.c new file +10 -0
+ * program an earlier tick than the one already chosen by timer_dyn_reprogram.
+ * It only reprograms it if the tick is earlier than the next one planned.
+ */
+void dyn_early_reprogram(unsigned int delta)
+{
+	unsigned long flags, tick = jiffies + delta;
+
+	if (tick >= get_cpu_var(dyn_cpu).next_tick &&
+		__get_cpu_var(dyn_cpu).next_tick > jiffies)
+			goto put_out;

Ditto.

The SMP case requires a bit more thorough reading... It seems there are a few places that call test_nohz_cpu() without taking the spinlock. That could be the race causing missed ticks on smp. Cheers,

-ben

Signature · 2 lines
"Ladies and gentlemen, I'm sorry to interrupt, but the police are here 
and they've asked us to stop the party."  Don't Email: <[email protected]>.